Delocation in the Manufacturing Sectors in the EU: A Regional Overview

29
Electronic copy available at: http://ssrn.com/abstract=1714301 DELOCATION IN THE MANUFACTURING SECTORS IN THE EU. A REGIONAL OVERVIEW M.ª Elisa Álvarez López† Rafael Myro Sánchez†† Josefa Vega Crespo† Abstract The question of delocation in main industrial regions across the EU is approached here, and its effects on jobs losses estimated. Seventy five regions are selected following the double criteria of size of the manufacturing industries and high per capita income levels. Taken together, the selected regions account for 65% of total manufacturing GVA in the EU-15. Delocation affected half of them in the period 2000-2005 when competition in the international markets rose and shows higher intensity in most of the British, French and Italian regions although its aggregate effect on employment seems have being offset for growth in services sectors. On the other hand, regions located in the continental area with more market potential have the highest concentration of location effects. Therefore delocation has changed the location of European manufacturing industry, benefitting those latter regions and perhaps bringing higher spatial concentration. JEL Classification: L6, R3, O4. Key words: manufacturing industry, delocation, regions, European Union † University of Valladolid, Department of Applied Economy, Avda. Valle Esgueva 6, 47011 Valladolid (e- mail: [email protected] ; [email protected] ). †† Complutense University of Madrid, Department of Applied Economy II, Campus de Somosaguas s/n, Pozuelo de Alarcón, 28223 Madrid (e-mail: [email protected] ).

Transcript of Delocation in the Manufacturing Sectors in the EU: A Regional Overview

Electronic copy available at httpssrncomabstract=1714301

DELOCATION IN THE MANUFACTURING SECTORS IN THE EU A

REGIONAL OVERVIEW

Mordf Elisa Aacutelvarez Loacutepezdagger

Rafael Myro Saacutenchezdaggerdagger

Josefa Vega Crespodagger

Abstract

The question of delocation in main industrial regions across the EU is approached

here and its effects on jobs losses estimated Seventy five regions are selected following

the double criteria of size of the manufacturing industries and high per capita income

levels Taken together the selected regions account for 65 of total manufacturing GVA

in the EU-15 Delocation affected half of them in the period 2000-2005 when competition

in the international markets rose and shows higher intensity in most of the British French

and Italian regions although its aggregate effect on employment seems have being offset

for growth in services sectors On the other hand regions located in the continental area

with more market potential have the highest concentration of location effects Therefore

delocation has changed the location of European manufacturing industry benefitting those

latter regions and perhaps bringing higher spatial concentration

JEL Classification L6 R3 O4

Key words manufacturing industry delocation regions European Union

dagger University of Valladolid Department of Applied Economy Avda Valle Esgueva 6 47011 Valladolid (e-mail elsaecouvaes josefaecouvaes) daggerdagger Complutense University of Madrid Department of Applied Economy II Campus de Somosaguas sn Pozuelo de Alarcoacuten 28223 Madrid (e-mail Rmyrocceeucmes)

Electronic copy available at httpssrncomabstract=1714301

Delocation in the manufacturing sectors in the EU A regional overview

1- Introduction

Throughout the last decade delocation of manufacturing activities has increased very

rapidly mainly in the developed countries due both to a more globalized and competitive

international environment and the emergence of China India and other big countries as

new industrial powers with the help of a large list of multinational corporations which have

been locating new plants in their territories since around 1990

We give the name `delocationacute to a process going beyond the process of moving

companies offshore as it also includes the closure of plants as a result of fierce

competition following the idea suggested by Baldwing and Robert-Nicoud who defined

delocation as `a loss of manufacturing jobs to trading partnersacute in the presence of a process

of opening up to foreign competition (Baldwin and Robert-Nicoud 2000) In our view

delocation refers to manufacturing activities as a whole not only to companies exporting

jobs as off-shoring does

While off-shoring of companies has received great attention from researchers in the

last years1 delocation of manufacturing activities has not perhaps because the latter is not

shown as different from the changes in production and trade patterns deriving from

extended international trade (Bhagwathi et al 2004 Grossman and Rossi-Hansberg

2006) or can be just considered as a special case of the location theory now arising with

the development of economic geography following the seminal work by Krugman and

Venables (1990) Krugman (1999) Fujita Krugman and Venables (1999) and Puga

(1999)2

The aim of this paper is to measure the effect on labour employment of the

delocation process in the manufacturing sectors throughout the main industrial regions in

the EU from 2000 to 2005 As such a process affects each manufacturing section in a very

different way the analysis by branches is indicated but data availability is an obstacle to

taking this path suggesting instead an initial view at the aggregate level Even with that

restriction it will be an important task to obtain the required data

In order to register the delocation patterns in European regions this paper takes as

reference those territories with a strong industrial sector high level of economic

1 Cfr pe Antragraves and Helpman (2005) Bjerring (2006) Helpman et al (2008) 2 Relationship between new geography and location are summarized in Puga (2002)

2

Delocation in the manufacturing sectors in the EU A regional overview

development and homogeneous space dimension So the sample contains only

geographical areas at NUTS 2 level3 whose industrial production represents at least 04

of total EU manufacturing GVA and at the same time has a per capita income of over

90 of the EU-15 average Therefore they are developed regions that have a powerful

industry at the Community level

Consistently the resulting list from applying to the set of EU-27 territories the three

already mentioned selection criteria includes a total of seventy-five regions of thirteen

Member States distributed as follows 23 from Germany 14 the United Kingdom 10 from

France 7 from Italy 4 Holland 4 Sweden 3 Spain 2 from Austria Belgium Denmark

and Finland 1 Greek and 1 Irish On the whole they accounted for 64 of real GVA in

2005 and almost three-fifths of manufacturing employment in the EU-15 (60 and 45

respectively taking as reference the EU-27) The complete catalogue of regions grouped

by country and their characteristics are set out in Appendixes 1 and 2 As is shown there

regions included in the sample are principally in sizes over 05 of EU-15 industrial

production and above the average EU-15 per capita income although a significant number

of them lay below those levels particularly in the per capita income Looking more closely

at the industrial size although most of the NUTS 2 examined move in around 05 there

are a group of fifteen regions located in Italy Germany Ireland France and Spain with a

strong industrial sector (more than 1) among them Lombardy in Italy with about 4 of

total EU-15 manufacturing GVA

After this first introductory section the paper is organized as follows In a second

section the model to capture the delocation patterns is introduced Then in a third section

the data sources are commented on In section four we try to asses the impact of delocation

in each region in terms of jobs affected and examine whether their location patterns have

altered its position in the European industrial scene contrasting country trends with

regional performances Concluding remarks round up the paper

3 The term NUTS corresponds to the French acronym for Nomenclature of Statistical Territorial Units used by the European Union This classification has a hierarchic structure at three levels which among other factors comes from demographic thresholds In particular the NUTS2 level covers regions of an average size between 08 to 3 million inhabitants In Spain the nineteen units included in this level coincide with the different AACC more than Ceuta and Melilla

3

Delocation in the manufacturing sectors in the EU A regional overview

2- Measuring delocation

As was posed in the introduction following Baldwin and Robert-Nicoud (2000)

delocation of activities has to be assessed in terms of losses of productive activity within

each region The variables which approximate these losses are gross added value and

number of jobs or total employment The first one is the most suitable since the latter is

dependent on advances registered in labour productivity mainly in the face of strong

competitive pressures forcing companies to achieve greater efficiency gains

In spite of that jobs continue to be a useful measure since its calculation is simpler

and is often more reliable In addition this indicator has received greater attention and is

more easily interpreted by the analysts and especially the social partners

Nevertheless if evolution of jobs is chosen as a measure of delocation it is necessary

to discount the effect on it of an increase in labour productivity (which reduces the need for

labour) and of the economic cycle (which may reduce or expand the existing employment)

The remaining reduction in the number of jobs measures the delocation effect on

employment

Furthermore when average values for quite a long time interval are taken it is

possible to ignore almost completely the impact of the cycle that otherwise would be

estimated using econometric techniques to isolate the trend In this way the delocation

effect can be just approximated after deducting the change in jobs caused by the increase in

labour productivity

Therefore the change in total employment may be split up into two effects one of

them due to the increased of labour productivity and the other to a location effect -

delocation if it is negative- Box 1 shows that decomposition

In fact the location effect reflects the impact on jobs of an increase or decrease in

value added estimated through the primitive rate of labour productivity and as has been

mentioned above may be positive (location) or negative (delocation) In the first case an

expansion in activity has taken place while in the second a reduction of the productive

scope has occurred

4

Delocation in the manufacturing sectors in the EU A regional overview

∆ N = GVA N(0) ⎯ π N(0)

Delocalisation ProductivityEffect Effect

Box 1 GVA= Labor Productivity Employment GVA = π N where π represents productivity and N employment

N = GVA - π

∆ N = N N(0)

N = GVAπ

where a hat over the variable denotes its rate of change

and and and

and

and and

Delocation of activities that can result from this calculation is compatible with the

absence of off-shoring companies and it may be due as noted above to the closure of

establishments as a result of their being uncompetitive In the same way the location of

activities in a territory is compatible with off-shoring of companies

Thus analysis of delocation does not necessarily tell us much about the importance

of firmsrsquo off-shoring a matter that has to be studied on a different basis However when

off-shoring reaches a large-scale dimension it affects the extension of manufacturing

industries

3- Database

As was said before to achieve aggregate data on manufacturing activities by regions

even just those of real GVA and total employment has demanded a laborious task of

collecting information provided by Eurostat (REGIO database) and the National Statistical

Offices of member countries to which the different regions belong as well as the

estimation in most cases of the product in real terms

Looking first at the GVA since the data offered by REGIO include those related to

energy and mining only access to the National Statistics of every member country has

allowed us to isolate the manufacturing sectors Nevertheless additional work has been

applied to transform the data of GVA into real values as only Germany Spain Finland

5

Delocation in the manufacturing sectors in the EU A regional overview

Holland and Italy provide such information or at least indexes of volume In the rest of the

countries real values of GVA have been estimated by applying the national price deflators

to the regional series at current values

As regards total employment significant discrepancies between the data published

by Eurostat and that country itself (particularly in the United Kingdom) are found as well

as the gaps found in several NUTS 2 of some Member States such as in Germany forced

to complete regional series and correct such deficiencies with the help of the information

coming from the National Statistics Offices

To sum up most of the information used (especially production data) comes from

National Statistical Offices but this is commonly close to the Eurostat database as this is

built mainly with data coming from the regional accounts of each country

By the way one of the problems arising from taking the National Statistics as the

main source of data is that the time intervals for which they provide information do not

always coincide That explains that despite having more recent data from some areas the

analysis must conclude in 2005

4- Delocation of manufacturing industries in European regions

As noted above in the introduction the group of regions examined reached just over

64 of EU-15 industrial GVA and about 60 of jobs in 2005 figures slightly lower than

in 2000 Thus despite nearly half of regionsrsquo industrial output at constant prices having

achieved positive growth rates the group accumulated throughout the period a decrease of

about 1 similar to EU-15 average4 Regarding employment the evolution has been even

more negative Throughout the first five years of the current century destruction of jobs in

the aggregate industry has been a common feature in the vast majority (90) of the

seventy-five NUTS 2 examined as well as for each of the Member States where they are

located except Spain accumulating the total sample a decline of close to 8

4 It must be noted that the EU average has been calculated from the aggregate industrial GVA at constant prices of thirteen countries included in the sample Thus Portugal and Luxembourg have been excluded and the values from other countries have been estimated by adding figures of all their regions So the above mentioned growth rate differs from that provided by Eurostat exhibiting a positive increase of 46

6

Delocation in the manufacturing sectors in the EU A regional overview

Focusing on employment the most dynamic areas are located mostly in France

Spain Austria and Italy while the Netherlands and especially the British regions exhibit

the largest job losses (Table 1)

Registered unemployment seems to be related to good performance in labour

productivity pointing to the fierce competition faced by the manufacturing sector in the

analyzed period Thus higher job losses are found in regions with stronger labour

productivity increases although some French and most of the British regions escaped from

this rule as large losses in employment are not accompanied by significant gains in labour

productivity suggesting delocation effects of a particular intensity

That reveals a very different growth path in employment and labour productivity by

regions differences which can be explored in a more detail to isolate the attractive regions

to locate manufacturing activities from the others characterized by a delocation process

To go deeply into this latter question Table 1 contains the results of estimating the

impact on total employment of location and delocation forces in the manufacturing sectors

across the seventy-five regions included in the sample over the period 2001-2005 by using

the procedure proposed in the second section that is discounting from the total change in

jobs the reduction corresponding to the increase in labour productivity Therefore the

location effect shows the effect on employment of GVA change calculated by applying the

initial rate of labour productivity

In about half of the regions with job losses all of those considered in the

Netherlands Finland Greece and Ireland three-quarters of the Germans two British two

Swedish and Madrid in Spain the decline in employment has been due entirely to

increased labour productivity so a positive effect of location appears This behaviour

extends to Finnish German and Swedish economies as a whole as well as Greece Ireland

and Austria Therefore all those regions have proved attractive for the location of

manufacturing firms

7

Delocation in the manufacturing sectors in the EU A regional overview

8

Productivity Location Productivity Locationeffect effect effect effect

at22 Steiermark 372 -108 480 36 -11 47at31 Oberoumlsterreich 097 -837 934 06 -56 62

Austria -1104 -2649 1545 -18 -42 25be21 Prov Antwerpen -1010 079 -1089 -70 05 -75be23 Prov Oost-Vlaanderen -750 -627 -123 -74 -61 -12

Belgium -5740 -2219 -3521 -84 -32 -51de11 Stuttgart -3157 -8244 5087 -48 -125 77de12 Karlsruhe -2036 -3585 1550 -57 -101 44de13 Freiburg -1291 -1301 011 -45 -45 00de14 Tuumlbingen -585 -1805 1219 -23 -72 49de21 Oberbayern -1630 -7976 6346 -37 -179 142de22 Niederbayern -490 -3270 2781 -33 -218 186de23 Oberpfalz -538 -1716 1178 -38 -120 83de24 Oberfranken -1957 -3131 1175 -116 -186 70de25 Mittelfranken -1184 -1770 586 -53 -80 26de26 Unterfranken -819 -3157 2338 -49 -189 140de27 Schwaben -932 -3335 2403 -41 -148 106de60 Hamburg -760 -1222 463 -60 -97 37de71 Darmstadt -5343 -4720 -623 -149 -132 -17de73 Kassel -1059 -1396 336 -78 -103 25de91 Braunschweig -947 -500 -447 -47 -25 -22de92 Hannover -1900 -1544 -356 -106 -86 -20dea1 Duumlsseldorf -6272 -8787 2515 -126 -177 51dea2 Koumlln -5428 -4971 -457 -150 -137 -13dea4 Detmold -3026 -3112 086 -107 -110 03dea5 Arnsberg -5824 -5502 -322 -129 -122 -07deb3 Rheinhessen-Pfalz -2099 -3016 917 -106 -152 46dec0 Saarland -360 -1810 1450 -32 -160 128def0 Schleswig-Holstein -1867 -2485 617 -101 -135 33

Germany -60300 -112275 51975 -74 -138 64dk01 Hovedstaden -1300 -1216 -084 -134 -125 -09dk04 Midtjylland -2000 -1622 -378 -146 -118 -28

Denmark -6100 -3428 -2672 -130 -73 -57es21 Pais Vasco 1840 -331 2171 74 -13 87es30 Comunidad de Madrid -1060 -2180 1120 -31 -63 32es51 Cataluntildea 660 -1914 2574 09 -25 34

Spain 13970 -3927 17897 47 -13 60fi18 Etelauml-Suomi -1807 -7159 5352 -86 -339 253fi19 Laumlnsi-Suomi -701 -3886 3185 -54 -297 244

Finland -1798 -12006 10208 -43 -286 243fr10 Icircle de France 12652 17868 -5216 218 307 -90fr23 Haute-Normandie -2418 -1093 -1325 -171 -77 -94fr24 Centre -2016 -120 -1896 -106 -06 -100fr42 Alsace -2489 -988 -1501 -154 -61 -93fr51 Pays de la Loire -217 1794 -2011 -08 67 -75fr52 Bretagne -415 -215 -200 -23 -12 -11fr61 Aquitaine -199 742 -941 -14 50 -64fr62 Midi-Pyreacuteneacutees 209 3318 -3109 15 245 -229fr71 Rhocircne-Alpes -2919 2759 -5678 -61 57 -118fr82 Provence-Alpes-Cocircte dAzur -097 262 -359 -06 18 -24

France -15306 31566 -46872 -40 83 -123gr30 Attiki -360 -5072 4712 -17 -234 217

Greece -730 -11989 11259 -14 -222 208ie02 Southern and Eastern -1840 -9985 8145 -78 -426 347

Ireland -2420 -12274 9854 -76 -387 311itc1 Piemonte -6869 2539 -9407 -129 48 -177itc4 Lombardia -7940 -7682 -258 -63 -61 -02itd3 Veneto -3967 -211 -3756 -62 -03 -58itd5 Emilia-Romagna -860 -840 -020 -16 -16 00ite1 Toscana -3620 078 -3697 -98 02 -101ite3 Marche 164 -555 719 08 -28 36ite4 Lazio -733 956 -1689 -36 47 -82

Italy -21761 6310 -28071 -45 13 -58nl22 Gelderland -2164 -2506 341 -178 -206 28nl32 Noord-Holland -1555 -2034 479 -138 -180 42nl33 Zuid-Holland -1734 -3980 2246 -129 -296 167nl41 Noord-Brabant -3539 -5515 1976 -173 -269 96

Netherlands -14767 -24690 9923 -161 -270 108se11 Stockholm -1570 -4183 2613 -136 -363 227se12 Oumlstra Mellansverige -2170 -2387 217 -151 -166 15se22 Sydsverige -1010 -690 -320 -91 -62 -29se23 Vaumlstsverige -460 134 -594 -26 07 -33

Sweden -7500 -9400 1900 -95 -119 24ukd3 Greater Manchester -4900 -1907 -2993 -232 -90 -142ukd4 Lancashire -4600 -5082 482 -317 -350 33uke4 West Yorkshire -3700 -1192 -2508 -186 -60 -126ukf1 Derbyshire and Nottinghamshire -3700 -4965 1265 -179 -240 61ukf2 Leicestershire Rutland and Northan -4400 -3607 -793 -233 -191 -42ukg3 West Midlands -8700 -4999 -3701 -313 -180 -133ukh1 East Anglia -2500 -1292 -1208 -137 -71 -66uki1 Inner London -1919 -1381 -538 -142 -102 -40uki2 Outer London -4481 1106 -5587 -271 67 -338ukj1 Berkshire Bucks and Oxfordshire -4600 -3898 -702 -246 -208 -38ukj3 Hampshire and Isle of Wight -1100 -689 -411 -81 -51 -30ukk1 Gloucestershire Wiltshire and Brist -3400 -1253 -2147 -169 -62 -107ukm2 Eastern Scotland -1100 1402 -2502 -91 116 -207ukm3 South Western Scotland -3400 -2107 -1293 -207 -128 -79

United Kingdom -89800 -48574 -41226 -196 -106 -90

Table 1 Importance of manufacturing delocation in European regions 2001-2005 (Decomposition of change in employment)

Total effect

Source Elaborated from Eurostat and National Statistical offices

Number of jobs (thousands)

Total effect

Percent share of 2000 employmentRegions

Delocation in the manufacturing sectors in the EU A regional overview

By contrast thirty-eight remaining regions that also show reductions in

manufacturing employment have been affected by delocation processes to different

extents although only in a small number of them have they been accompanied by

decreases in labour productivity which could make the prospects for economic growth

tougher (four from France Piamonte Toscana and Lazio in Italy Outer London and

Eastern Scotland in the United Kingdom the Swedish Vaumlstsverige and Antwerp in

Belgium)

Regions with positive location effects are shown in Map 1 Most of them are large

industrial regions and are located in the traditional industrial growth centres5 those

characterized by high market potential (Combes and Overman 2003) although there are

also some intermediate and peripheral territories mainly located in the north of Italy

United Kingdom and Spain and in the south of Sweden and Norway This means that the

delocation process has contributed to concentrating the European manufacturing industries

in that privileged area of faster growth apart from some other changes inside every

country

To compare now the intensity of such effects of location and delocation across

regions avoiding the differences they have in size we built a relative measure of them the

ratio of job losses in the period to the volume of employment in the first year 2000 (Table

1)

Following this ratio delocation effects are revealed to be particularly strong in

France United Kingdom Italy and in one region of Belgium with declines in

employment exceeding 20 in some of them

Conversely the regions of Finland and several of the German and the Dutch ones

exhibit strong location effects

In any case as was pointed above the importance of location effects seems to lie in

labour productivity gains In fact higher increases in productivity (over 15) are found

precisely in those regions of Ireland Finland Sweden Netherlands and Germany besides

the Greek Attiki where the location effect has in general reached its greatest extent

(Figure 1)

5 Note that the map of winning regions is clearly influenced by the large number (30) of German NUTS 2 in the sample

9

Delocation in the manufacturing sectors in the EU A regional overview

fi19

fi18

ie02

dec0

gr30

se11

nl33

de21

de22

de26

de25

de12

de14

de11

dea4

dea1de73

def0

de60

deb3

de27

de23

itd5

ite3

se12

nl22

es 30

ukd4

es51

nl32

at22

ukf1

at31

es21

Map 1 Regions with positive location effect ( of 2000 employment)

Above 10 0 to 10

nl41

de13de27

Nevertheless productivity also grows faster in most of the regions belonging to the

United Kingdom Belgium and Denmark in spite of the strong delocation effects registered

10

Delocation in the manufacturing sectors in the EU A regional overview

which disturbs the above-mentioned positive relationship suggesting the harsh process of

industrial restructuring involved

Figure 1 Location effect and productivity growth in select regions manufacturing sector 2001-2005 (percentages)

UKi2

EU 15

FR6 2

UKm2

ITc1

UKg3UKd 3

UKe4

UKk1FR71F R

FR10UKm3

UKj1UKf2

UKd 4

UKi1

D K

DK01

UKh1U K

FR2 3FR4 2

FR24ITe1

BE2 1ITe4FR51

FR61

ITITd 3

SE2 2FR8 2

SE23

NL2 2

UKf1DE2 4

NL3 2DEa1

DEf0

UKj3DE9 1 DE92

DEa2DE71DK0 4

S EDE25

DE6 0DE12

A T

DE14

ITe3

AT2 2

AT3 1ESD E

DE 11DE2 3

NL41N LDE2 7

DEc0

NL3 3

DE2 6DE2 1

ES3 0ES51

SE11

DE2 2

ES21

GR

FI19F I

FI18

IE

GR3 0

B E

DE73

DEb 3

IE0 2

ITd 5 DE13SE12

FR52ITc4BE2 3

DEa4DEa5

y = 07x - 59R2 = 06

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

40

-35 -30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30 35 40 45

Productivity grothw Source Elaborated from Eurostat and National Statistical offices

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

11

Delocation in the manufacturing sectors in the EU A regional overview

In contrast the productivity decline in half of the French and Italian regions albeit

with a positive impact on employment is largely offset by destruction of industrial

activity except in Icircle de France and Midi -Pyreacuteneacutees the two territories in which

productivity has a more negative trajectory

These results closely match those reported by Cuadrado Roura et al (2000) based on

the growth rates for regional productivity and employment during the period 1980-1993

grouping European regions in four typologies comparing the growth rates of these two

variables with their respective European averages

According to this approach and including the location effect in the analysis the

group of regions that are examined in this work would be categorized as follows (Table 2)

The first group includes areas with productivity and employment growth rates above

sample average and show positive location effect (Quadrant 1) So as has been underlined

in that work possibly this reflects the fact that the technological and organizational

changes introduced during the period would have been successful promoting the

diversification and specialization in activities with high growth levels and attracting new

investments which are less labour-intensive

In a second stage are those regions that as the above mentioned authors suggest

have implemented restructuring processes to eliminate the most inefficient outputs

achieving higher levels of productivity and in about half of them increased industrial

production at the expense of a notable decline in employment (Quadrant 4) They named

this typology `restructuring via productivityacute

With a few exceptions (those where the location effect is negative) the regions in

both groups are the winners of industrial dynamic during the period Most of them are

large industrial regions and are located in the traditional industrial growth centres As

demonstrated above they are mainly Finish German and Dutch as well as Greek and Irish

regions

In contrast another large group of regions due to different factors6 have opted for

more labour-intensive industrial models which in a few cases (especially in Spanish and

Austrian regions) have been accompanied by output increases so they can also be included

6 See Cuadrado et al (2000)

12

Delocation in the manufacturing sectors in the EU A regional overview

in the winners (Map 1) while in other the result has been a loss of productive activity

(Quadrant 2)

(1) (2)Employment Productivity Employment Productivity

de11 Stuttgart -48 125 at22 Steiermark 36 11de12 Karlsruhe -57 101 at31 Oberoumlsterreich 06 56de21 Oberbayern -37 179 be21 Prov Antwerpen -70 -05de22 Niederbayern -33 218 be23 Prov Oost-Vlaanderen -74 61de23 Oberpfalz -38 120 de13 Freiburg -45 45de25 Mittelfranken -53 80 de14 Tuumlbingen -23 72de26 Unterfranken -49 189 de91 Braunschweig -47 25de27 Schwaben -41 148 es21 Pais Vasco 74 13de60 Hamburg -60 97 es30 Comunidad de Madrid -31 63dec0 Saarland -32 160 es51 Cataluntildea 09 25fi19 Laumlnsi-Suomi -54 297 fr10 Icircle de France 218 -307gr30 Attiki -17 234 fr51 Pays de la Loire -08 -67

fr52 Bretagne -23 12fr61 Aquitaine -14 -50fr62 Midi-Pyreacuteneacutees 15 -245fr71 Rhocircne-Alpes -61 -57fr82 Provence-Alpes-Cocircte dAzur -06 -18itc4 Lombardia -63 61itd3 Veneto -62 03itd5 Emilia-Romagna -16 16ite3 Marche 08 28ite4 Lazio -36 -47se23 Vaumlstsverige -26 -07

(4) (3)Employment Productivity Employment Productivity

de24 Oberfranken -116 186 fr24 Centre -106 06de71 Darmstadt -149 132 fr42 Alsace -154 61de73 Kassel -78 103 itc1 Piemonte -129 -48de92 Hannover -106 86 ite1 Toscana -98 -02dea1 Duumlsseldorf -126 177 se22 Sydsverige -91 62dea2 Koumlln -150 137 uke4 West Yorkshire -186 60dea4 Detmold -107 110 ukh1 East Anglia -137 71dea5 Arnsberg -129 122 uki2 Outer London -271 -67deb3 Rheinhessen-Pfalz -106 152 ukj3 Hampshire and Isle of Wight -81 51def0 Schleswig-Holstein -101 135 ukk1 Gloucestershire Wiltshire -169 62dk01 Hovedstaden -134 125 ukm2 Eastern Scotland -91 -116dk04 Midtjylland -146 118fi18 Etelauml-Suomi -86 339fr23 Haute-Normandie -171 77ie02 Southern and Eastern -78 426nl22 Gelderland -178 206nl32 Noord-Holland -138 180nl33 Zuid-Holland -129 296nl41 Noord-Brabant -173 269se11 Stockholm -136 363se12 Oumlstra Mellansverige -151 166ukd3 Greater Manchester -232 90ukd4 Lancashire -317 350ukf1 Derbyshire and Nottinghamshire -179 240ukf2 Leicestershire Rutland and Northants -233 191ukg3 West Midlands -313 180uki1 Inner London -142 102ukj1 Berkshire Bucks and Oxfordshire -246 208ukm3 South Western Scotland -207 128

Source Elaborated from Eurostat and national Statistical officesNote Shaded areas indicate regions with positive location effect

Em

ploy

men

t gro

wth

lt A

vera

ge v

alue

Productivity growth gt Average value Productivity growth lt Average value

Em

ploy

men

t gro

wth

gt A

vera

ge v

alue

Table 2 Employment and productivity growth rates in select regions manufacturing sector 2001-2005 (Averages of all regions +74 productivity -77 employment)

13

Delocation in the manufacturing sectors in the EU A regional overview

Finally the worst results can be seen in a fourth type of region with growth rates of

GVA and productivity below average and serious job destruction (Quadrant 3) This would

occur in regions with problems in restructuring their manufacturing sector or with low

levels of activity So together with those that are either restructuring via productivity or

via employment but without achieving a positive location effect they are clearly the

losers These include mainly British French Italian Belgian and Danish regions

To sum up these results confirm the direct relationship between growth in labour

efficiency and progress in the creation of productive industrial activity as pointed out

above and which can be more clearly by looking at Figure 1 In other words regions with

higher levels of productivity growth in general show a greater attractiveness for new

manufacturing firms

These differences in the behaviour of productivity and employment across regions

have led to remarkable disparities in rates of change of their industrial output during the

period Thus while a few increased their real manufacturing GVA by more than 20 in

the less dynamic ones the loss of industrial output is between a sixth and a third of the

initial value So significant changes had taken place in the relative position of some of

them (Figure 2)

Growth is led by a group of twenty regions located in Ireland Finland Greece

Germany Sweden Netherlands Spain and Austria with real GVA growth rates that are

more than five points above average As a result they increased their share in the industrial

output of the sample Looking at the complete list we see that the growth deviation from

the mean exhibits positive values in just over half of the regions examined among them

seven of the ten with the largest industrial scale in 2005 Thus the top ten regions have

increased their share from 32 to 33 percent in the last five years

Conversely among the less dynamic regions are found a great number of British and

some French and Italian ones which consequently have fallen back in the ranking of

NUTS 2 arranged by their contribution to aggregate manufacturing output

Between both groups there are approximately twenty regions mainly in Germany

with growth rates not far from average so their positions have hardly changed during the

last five years

14

Delocation in the manufacturing sectors in the EU A regional overview

Figure 2 Regional shares in the total sample manufacturing GVA(percentage at 2000 prices)

ITc4

UKf2

UKm3

Uke4

UKf1

UKk1

UKh1

AT22

ITe1

ES21

ES51

FR71

FR62

IT e4

DE24

DE26

SE23

FR61

FR51

FR82

SE22

ITe3

AT31

DK04

DE11

DE23

GR30

DEc0

ES30DEb3DEf0

DEa4

DE92

BE21

DE73

IE02

ITd3

FR52

DE91

DE13DE25

DE 12DE60

DE27

DE21

DE22

BE23DEa2

DE71

DE 14

SE12

ITd5

SE11

FI19

FR10

FR 24

ITe1

FI18

DK01 DEa5

FR4 2

DEa1NL32

NL3 3

FR2 3

NL22

NL 41

UKd3

UKj3

UKi1

UKm2

UKj1

UKd4

UKi2

Ukg3

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

Share in 2005

Source Elaborated from Eurostat and National Statistical offices

Rat

e of

cha

nge

200

1-20

05

70403530252015100500

Finally when the question is analyzed at Member State level (Table 1) it is possible

to check out the disparities among the above mentioned regions Thus while in some

15

Delocation in the manufacturing sectors in the EU A regional overview

countries such as Spain manufacturing growth has been clearly labour-intensive

something which can be extended to its regions as also happens in Greece and Austria

which both boast positive location effect and a job destruction rate far below the average

in others such as Holland Sweden and to a lesser extent Belgium Ireland and Finland

increasing of industrial activity is only due by advances in productivity so has been

accompanied by heavy job losses and these are patterns repeated in most of their regions

with noticeable relevance in the Netherlands As for the United Kingdom Belgium and

Denmark all share declining industrial production together with strong job cuts a pattern

that is replicated in many of their regions In this situation are the bulk of French and

Italian NUTS 2 just like the countries to which they belong although in this case job

destruction has been much less pronounced and below the group average Close to this

mean is Germany as well as most of their regions

Thus the analysis at Member States level also confirms the positive relationship

between the advance of industrial productivity and the attractiveness for new

manufacturing firms As demonstrated for their regions Ireland Finland Netherlands and

Greece following at quite distance from Germany are Member States showing a better

performance in productivity and with high location effects (see Figure 1)

In fact similarities between performances in the regions and in the Member States

they belong to suggest that the trend in regional industrial employment is to some extent

influenced by belonging to the same country in other words the locational attractions of

each area and its effort to improve productivity is partly due to national specificities or

idiosyncratic factors In fact as shown in Appendix 3 most European industrialized

regions follow the location patterns of the country they belong to reproducing its

productivity behaviour too In summary the evolution of these variables in each region

seems to have a clear `country effectacute that is different behaviours motivated by specific

factors andor differentiated industrial policies

For example in order to correctly interpret the results for Spanish regions it can not

be overlooked that during the period analyzed as in the whole country regions have

absorbed large numbers of immigrants which has boosted the proliferation of lower

productivity and lower wages in industrial activities However it may be thought that this

phenomenon has tended to conceal productivity increases in larger and better equipped

16

Delocation in the manufacturing sectors in the EU A regional overview

firms which could have been significant in the light of their results7 On the other hand

countries like Finland or Ireland with the lowest population increase have opted for high-

tech manufacturing sectors as a way to improve national competitiveness8 Nonetheless

this positive location effect has not reached in these countriesrsquo NUTS 2 enough magnitude

to offset completely the job losses arising from productivity improvements

On the other hand this `national effectacute has also played an important role in

explaining the economic outcomes of European regions in terms of per capita GDP in

accordance with the results obtained by Cuadrado-Roura (2001) using the data available

for 109 EU regions over the 1977-1994 period

Table 3 Export growth in key sectors by countries 2000-2005 (percentages)Impact ITon Total

Spain 65 91 72 158 38 111 04Germany 116 138 61 111 111 130 14United Kingdom 09 85 29 32 23 61 02France -09 102 25 94 118 78 -01Ireland -09 -102 01 25 -34 77 -03Finland 47 152 45 49 56 73 13Italy 40 88 41 90 56 89 02EU-15 70 141 59 91 67 111 11Source Elaborated from Comtrade

Shoes TotalIT Auto Textile Clothes

This importance of the `national effectacute should allow us to make a better approach to

the determinants of above-mentioned delocation patterns in future research One of such

determinants might be found in the different importance and performance by countries of

some manufacturing branches particularly affected by the increase in global competition as

their development has being receiving strong support in emerging countries from both

domestic policies and the location of big multinational companies This is the case of IT

manufacturing sectors or that of automobiles clothes textiles and shoes (Sachwald 2004)

Table 3 shows how the exports of these key sectors have evolved in some of the European

countries considered here It is amazing to see the poor results registered by IT sectors in

United Kingdom France and Ireland compared to the other countries mentioned Germany

in particular This evolution could be behind the stronger effects of delocation in France

and the United Kingdom although it must have produced similar results in the case of

Ireland Furthermore on such a basis what has happened in Italy would remain

7 For details see Myro et al (2008) 8 See Aacutelvarez Loacutepez et al (2007)

17

Delocation in the manufacturing sectors in the EU A regional overview

unexplained as the selected key sectors do not show a bad performance in terms of exports

in the latter country

One reason for such contradictory results lies in the possibility that production

instead of export reflects accurately what has been happening in every sector in terms of

activity However Aacutelvarez et al (2007) show very similar patterns in production and

export specialization from 1995 to 2005 except in the case of Ireland which thus appears

as a very particular one This is because IT specialization in production increases at the

same time as it decreases in exports Perhaps the strong growth in domestic demand

prevents IT exports from growing The opposite case could be that of Italy where slow

growth in domestic demand might have been pushing up IT export9

Anyway further developments on the determinants of delocation patterns require the

preparation of disaggregated data on production a task beyond the objectives of this paper

That could allow us to update the available manufacturing specialization analysis As the

two countries exhibiting higher delocation effects France and the United Kingdom

showed the lowest rates of specialization compared with the EU (Milderfart and Overman

2002 Combes and Overman 2003) something could have changed in the last few years

which would merit a fresh look based on more recent data

On the other hand it is now clear that insofar as delocation in manufacturing

activities has affected each country in a different way a relocation process inside the EU

and between countries has taken place This process deserved to be deeply analyzed

including a perspective by sectors updating what we know about changes in the

distribution of manufacturing industry in the EU and offering the possibility of exploring

whether the European Common Policy has influenced them in some way as was already

suggested by Milderfart-Knarvik and Overman (2002)

9 In Finland IT activities grows much more in production than in exports but in both aspects a positive growth is registered

18

Delocation in the manufacturing sectors in the EU A regional overview

19

Figure 3 Employment growth rates in selected regions 2001-2005 manufacturing and total economy (percentages)

Ukg3

UKi2

UKd4

UKj1

UKm2

UKi1 UKj3

UKd3

NL 41

NL22

FR23

NL33

NL32

DEa1

FR42

DEa5

DK01

FI18

ITe1

FR 24FR10

FI19

SE11

ITd5FR52

DE 14

DE71

DEa2

BE23

DE22

DE21

DE27

IC c4

DE60DE 12

DE25

DE13

DE9 1SE12

ITd3

IE02

DE73

BE21

DE92DEa4

DEf0

DEb3

ES30

DEc0

GR30

DE23

DE11

DK04

AT31

ITe3

SE22

FR82

FR51FR61

SE2 3

DE26

DE24

IT e4

FR62

FR71

ES51

ES2 1

ITe1

AT22

UKh1

UKk1

UKf1

Uke4

UKm3

UKf2

-20

-15

-10

-5

0

5

10

15

20

25

-40 -30 -20 -10 0 10 20 30

Manufacturing

Source Elaborated from Eurostat and National Statistical offices

Tota

l em

ploy

men

t

Delocation in the manufacturing sectors in the EU A regional overview

One last question that deserves consideration is the effect of industrial delocation

process on the regional economies Surprisingly manufacturing job losses have not been

translated into total employment which exhibit positive growth for most of the chosen

regions with the exceptions of some Germans and all the Netherlands (Figure 3) That

means the regions have been able to compensate the negative effect on the employment

from shrinking manufacturing allocating more resources inside the services sectors In fact

a substitution of manufacturing by services seems to have happened as are the region

showing more delocation in manufacturing those that increase in a higher measure the

employment in the service sectors that is the French British and Italian However the

relation is not accurate as there are not a strict quantitative correlation region by region

between rates of delocation and increase in the employment in the service activities

These good results in terms of general employment shown by the regions affected in

a higher measure by manufacturing delocation suggest the convenience of a more general

analysis of regional economic growth beyond the manufacturing sector as some regions

strongly shrinking manufacturing activities in favour of those of services might be leading

an interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

5- Final remarks

Comparative analysis of location patterns for a large sample of European regions -

characterized by having an upper-middle level income and a strong industrial sector-

during the last few years has revealed that only half of them have shown locational

attractions for manufacturing activities The regional patterns found clearly show the

influence of `national effectacute the regions showing location effects corresponding to most

of those located in Austria Germany the Netherlands Ireland Finland Sweden Spain and

Greece

On the contrary the delocation process has affected another half of the regions

included most of them French British and Italian Therefore delocation has contributed to

concentrating the European manufacturing industry in the areas with higher market

potential in the centre of continental Europe It has contributed furthermore to relocating

manufacturing activities inside every country something that requires further analysis

from future research

20

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Electronic copy available at httpssrncomabstract=1714301

Delocation in the manufacturing sectors in the EU A regional overview

1- Introduction

Throughout the last decade delocation of manufacturing activities has increased very

rapidly mainly in the developed countries due both to a more globalized and competitive

international environment and the emergence of China India and other big countries as

new industrial powers with the help of a large list of multinational corporations which have

been locating new plants in their territories since around 1990

We give the name `delocationacute to a process going beyond the process of moving

companies offshore as it also includes the closure of plants as a result of fierce

competition following the idea suggested by Baldwing and Robert-Nicoud who defined

delocation as `a loss of manufacturing jobs to trading partnersacute in the presence of a process

of opening up to foreign competition (Baldwin and Robert-Nicoud 2000) In our view

delocation refers to manufacturing activities as a whole not only to companies exporting

jobs as off-shoring does

While off-shoring of companies has received great attention from researchers in the

last years1 delocation of manufacturing activities has not perhaps because the latter is not

shown as different from the changes in production and trade patterns deriving from

extended international trade (Bhagwathi et al 2004 Grossman and Rossi-Hansberg

2006) or can be just considered as a special case of the location theory now arising with

the development of economic geography following the seminal work by Krugman and

Venables (1990) Krugman (1999) Fujita Krugman and Venables (1999) and Puga

(1999)2

The aim of this paper is to measure the effect on labour employment of the

delocation process in the manufacturing sectors throughout the main industrial regions in

the EU from 2000 to 2005 As such a process affects each manufacturing section in a very

different way the analysis by branches is indicated but data availability is an obstacle to

taking this path suggesting instead an initial view at the aggregate level Even with that

restriction it will be an important task to obtain the required data

In order to register the delocation patterns in European regions this paper takes as

reference those territories with a strong industrial sector high level of economic

1 Cfr pe Antragraves and Helpman (2005) Bjerring (2006) Helpman et al (2008) 2 Relationship between new geography and location are summarized in Puga (2002)

2

Delocation in the manufacturing sectors in the EU A regional overview

development and homogeneous space dimension So the sample contains only

geographical areas at NUTS 2 level3 whose industrial production represents at least 04

of total EU manufacturing GVA and at the same time has a per capita income of over

90 of the EU-15 average Therefore they are developed regions that have a powerful

industry at the Community level

Consistently the resulting list from applying to the set of EU-27 territories the three

already mentioned selection criteria includes a total of seventy-five regions of thirteen

Member States distributed as follows 23 from Germany 14 the United Kingdom 10 from

France 7 from Italy 4 Holland 4 Sweden 3 Spain 2 from Austria Belgium Denmark

and Finland 1 Greek and 1 Irish On the whole they accounted for 64 of real GVA in

2005 and almost three-fifths of manufacturing employment in the EU-15 (60 and 45

respectively taking as reference the EU-27) The complete catalogue of regions grouped

by country and their characteristics are set out in Appendixes 1 and 2 As is shown there

regions included in the sample are principally in sizes over 05 of EU-15 industrial

production and above the average EU-15 per capita income although a significant number

of them lay below those levels particularly in the per capita income Looking more closely

at the industrial size although most of the NUTS 2 examined move in around 05 there

are a group of fifteen regions located in Italy Germany Ireland France and Spain with a

strong industrial sector (more than 1) among them Lombardy in Italy with about 4 of

total EU-15 manufacturing GVA

After this first introductory section the paper is organized as follows In a second

section the model to capture the delocation patterns is introduced Then in a third section

the data sources are commented on In section four we try to asses the impact of delocation

in each region in terms of jobs affected and examine whether their location patterns have

altered its position in the European industrial scene contrasting country trends with

regional performances Concluding remarks round up the paper

3 The term NUTS corresponds to the French acronym for Nomenclature of Statistical Territorial Units used by the European Union This classification has a hierarchic structure at three levels which among other factors comes from demographic thresholds In particular the NUTS2 level covers regions of an average size between 08 to 3 million inhabitants In Spain the nineteen units included in this level coincide with the different AACC more than Ceuta and Melilla

3

Delocation in the manufacturing sectors in the EU A regional overview

2- Measuring delocation

As was posed in the introduction following Baldwin and Robert-Nicoud (2000)

delocation of activities has to be assessed in terms of losses of productive activity within

each region The variables which approximate these losses are gross added value and

number of jobs or total employment The first one is the most suitable since the latter is

dependent on advances registered in labour productivity mainly in the face of strong

competitive pressures forcing companies to achieve greater efficiency gains

In spite of that jobs continue to be a useful measure since its calculation is simpler

and is often more reliable In addition this indicator has received greater attention and is

more easily interpreted by the analysts and especially the social partners

Nevertheless if evolution of jobs is chosen as a measure of delocation it is necessary

to discount the effect on it of an increase in labour productivity (which reduces the need for

labour) and of the economic cycle (which may reduce or expand the existing employment)

The remaining reduction in the number of jobs measures the delocation effect on

employment

Furthermore when average values for quite a long time interval are taken it is

possible to ignore almost completely the impact of the cycle that otherwise would be

estimated using econometric techniques to isolate the trend In this way the delocation

effect can be just approximated after deducting the change in jobs caused by the increase in

labour productivity

Therefore the change in total employment may be split up into two effects one of

them due to the increased of labour productivity and the other to a location effect -

delocation if it is negative- Box 1 shows that decomposition

In fact the location effect reflects the impact on jobs of an increase or decrease in

value added estimated through the primitive rate of labour productivity and as has been

mentioned above may be positive (location) or negative (delocation) In the first case an

expansion in activity has taken place while in the second a reduction of the productive

scope has occurred

4

Delocation in the manufacturing sectors in the EU A regional overview

∆ N = GVA N(0) ⎯ π N(0)

Delocalisation ProductivityEffect Effect

Box 1 GVA= Labor Productivity Employment GVA = π N where π represents productivity and N employment

N = GVA - π

∆ N = N N(0)

N = GVAπ

where a hat over the variable denotes its rate of change

and and and

and

and and

Delocation of activities that can result from this calculation is compatible with the

absence of off-shoring companies and it may be due as noted above to the closure of

establishments as a result of their being uncompetitive In the same way the location of

activities in a territory is compatible with off-shoring of companies

Thus analysis of delocation does not necessarily tell us much about the importance

of firmsrsquo off-shoring a matter that has to be studied on a different basis However when

off-shoring reaches a large-scale dimension it affects the extension of manufacturing

industries

3- Database

As was said before to achieve aggregate data on manufacturing activities by regions

even just those of real GVA and total employment has demanded a laborious task of

collecting information provided by Eurostat (REGIO database) and the National Statistical

Offices of member countries to which the different regions belong as well as the

estimation in most cases of the product in real terms

Looking first at the GVA since the data offered by REGIO include those related to

energy and mining only access to the National Statistics of every member country has

allowed us to isolate the manufacturing sectors Nevertheless additional work has been

applied to transform the data of GVA into real values as only Germany Spain Finland

5

Delocation in the manufacturing sectors in the EU A regional overview

Holland and Italy provide such information or at least indexes of volume In the rest of the

countries real values of GVA have been estimated by applying the national price deflators

to the regional series at current values

As regards total employment significant discrepancies between the data published

by Eurostat and that country itself (particularly in the United Kingdom) are found as well

as the gaps found in several NUTS 2 of some Member States such as in Germany forced

to complete regional series and correct such deficiencies with the help of the information

coming from the National Statistics Offices

To sum up most of the information used (especially production data) comes from

National Statistical Offices but this is commonly close to the Eurostat database as this is

built mainly with data coming from the regional accounts of each country

By the way one of the problems arising from taking the National Statistics as the

main source of data is that the time intervals for which they provide information do not

always coincide That explains that despite having more recent data from some areas the

analysis must conclude in 2005

4- Delocation of manufacturing industries in European regions

As noted above in the introduction the group of regions examined reached just over

64 of EU-15 industrial GVA and about 60 of jobs in 2005 figures slightly lower than

in 2000 Thus despite nearly half of regionsrsquo industrial output at constant prices having

achieved positive growth rates the group accumulated throughout the period a decrease of

about 1 similar to EU-15 average4 Regarding employment the evolution has been even

more negative Throughout the first five years of the current century destruction of jobs in

the aggregate industry has been a common feature in the vast majority (90) of the

seventy-five NUTS 2 examined as well as for each of the Member States where they are

located except Spain accumulating the total sample a decline of close to 8

4 It must be noted that the EU average has been calculated from the aggregate industrial GVA at constant prices of thirteen countries included in the sample Thus Portugal and Luxembourg have been excluded and the values from other countries have been estimated by adding figures of all their regions So the above mentioned growth rate differs from that provided by Eurostat exhibiting a positive increase of 46

6

Delocation in the manufacturing sectors in the EU A regional overview

Focusing on employment the most dynamic areas are located mostly in France

Spain Austria and Italy while the Netherlands and especially the British regions exhibit

the largest job losses (Table 1)

Registered unemployment seems to be related to good performance in labour

productivity pointing to the fierce competition faced by the manufacturing sector in the

analyzed period Thus higher job losses are found in regions with stronger labour

productivity increases although some French and most of the British regions escaped from

this rule as large losses in employment are not accompanied by significant gains in labour

productivity suggesting delocation effects of a particular intensity

That reveals a very different growth path in employment and labour productivity by

regions differences which can be explored in a more detail to isolate the attractive regions

to locate manufacturing activities from the others characterized by a delocation process

To go deeply into this latter question Table 1 contains the results of estimating the

impact on total employment of location and delocation forces in the manufacturing sectors

across the seventy-five regions included in the sample over the period 2001-2005 by using

the procedure proposed in the second section that is discounting from the total change in

jobs the reduction corresponding to the increase in labour productivity Therefore the

location effect shows the effect on employment of GVA change calculated by applying the

initial rate of labour productivity

In about half of the regions with job losses all of those considered in the

Netherlands Finland Greece and Ireland three-quarters of the Germans two British two

Swedish and Madrid in Spain the decline in employment has been due entirely to

increased labour productivity so a positive effect of location appears This behaviour

extends to Finnish German and Swedish economies as a whole as well as Greece Ireland

and Austria Therefore all those regions have proved attractive for the location of

manufacturing firms

7

Delocation in the manufacturing sectors in the EU A regional overview

8

Productivity Location Productivity Locationeffect effect effect effect

at22 Steiermark 372 -108 480 36 -11 47at31 Oberoumlsterreich 097 -837 934 06 -56 62

Austria -1104 -2649 1545 -18 -42 25be21 Prov Antwerpen -1010 079 -1089 -70 05 -75be23 Prov Oost-Vlaanderen -750 -627 -123 -74 -61 -12

Belgium -5740 -2219 -3521 -84 -32 -51de11 Stuttgart -3157 -8244 5087 -48 -125 77de12 Karlsruhe -2036 -3585 1550 -57 -101 44de13 Freiburg -1291 -1301 011 -45 -45 00de14 Tuumlbingen -585 -1805 1219 -23 -72 49de21 Oberbayern -1630 -7976 6346 -37 -179 142de22 Niederbayern -490 -3270 2781 -33 -218 186de23 Oberpfalz -538 -1716 1178 -38 -120 83de24 Oberfranken -1957 -3131 1175 -116 -186 70de25 Mittelfranken -1184 -1770 586 -53 -80 26de26 Unterfranken -819 -3157 2338 -49 -189 140de27 Schwaben -932 -3335 2403 -41 -148 106de60 Hamburg -760 -1222 463 -60 -97 37de71 Darmstadt -5343 -4720 -623 -149 -132 -17de73 Kassel -1059 -1396 336 -78 -103 25de91 Braunschweig -947 -500 -447 -47 -25 -22de92 Hannover -1900 -1544 -356 -106 -86 -20dea1 Duumlsseldorf -6272 -8787 2515 -126 -177 51dea2 Koumlln -5428 -4971 -457 -150 -137 -13dea4 Detmold -3026 -3112 086 -107 -110 03dea5 Arnsberg -5824 -5502 -322 -129 -122 -07deb3 Rheinhessen-Pfalz -2099 -3016 917 -106 -152 46dec0 Saarland -360 -1810 1450 -32 -160 128def0 Schleswig-Holstein -1867 -2485 617 -101 -135 33

Germany -60300 -112275 51975 -74 -138 64dk01 Hovedstaden -1300 -1216 -084 -134 -125 -09dk04 Midtjylland -2000 -1622 -378 -146 -118 -28

Denmark -6100 -3428 -2672 -130 -73 -57es21 Pais Vasco 1840 -331 2171 74 -13 87es30 Comunidad de Madrid -1060 -2180 1120 -31 -63 32es51 Cataluntildea 660 -1914 2574 09 -25 34

Spain 13970 -3927 17897 47 -13 60fi18 Etelauml-Suomi -1807 -7159 5352 -86 -339 253fi19 Laumlnsi-Suomi -701 -3886 3185 -54 -297 244

Finland -1798 -12006 10208 -43 -286 243fr10 Icircle de France 12652 17868 -5216 218 307 -90fr23 Haute-Normandie -2418 -1093 -1325 -171 -77 -94fr24 Centre -2016 -120 -1896 -106 -06 -100fr42 Alsace -2489 -988 -1501 -154 -61 -93fr51 Pays de la Loire -217 1794 -2011 -08 67 -75fr52 Bretagne -415 -215 -200 -23 -12 -11fr61 Aquitaine -199 742 -941 -14 50 -64fr62 Midi-Pyreacuteneacutees 209 3318 -3109 15 245 -229fr71 Rhocircne-Alpes -2919 2759 -5678 -61 57 -118fr82 Provence-Alpes-Cocircte dAzur -097 262 -359 -06 18 -24

France -15306 31566 -46872 -40 83 -123gr30 Attiki -360 -5072 4712 -17 -234 217

Greece -730 -11989 11259 -14 -222 208ie02 Southern and Eastern -1840 -9985 8145 -78 -426 347

Ireland -2420 -12274 9854 -76 -387 311itc1 Piemonte -6869 2539 -9407 -129 48 -177itc4 Lombardia -7940 -7682 -258 -63 -61 -02itd3 Veneto -3967 -211 -3756 -62 -03 -58itd5 Emilia-Romagna -860 -840 -020 -16 -16 00ite1 Toscana -3620 078 -3697 -98 02 -101ite3 Marche 164 -555 719 08 -28 36ite4 Lazio -733 956 -1689 -36 47 -82

Italy -21761 6310 -28071 -45 13 -58nl22 Gelderland -2164 -2506 341 -178 -206 28nl32 Noord-Holland -1555 -2034 479 -138 -180 42nl33 Zuid-Holland -1734 -3980 2246 -129 -296 167nl41 Noord-Brabant -3539 -5515 1976 -173 -269 96

Netherlands -14767 -24690 9923 -161 -270 108se11 Stockholm -1570 -4183 2613 -136 -363 227se12 Oumlstra Mellansverige -2170 -2387 217 -151 -166 15se22 Sydsverige -1010 -690 -320 -91 -62 -29se23 Vaumlstsverige -460 134 -594 -26 07 -33

Sweden -7500 -9400 1900 -95 -119 24ukd3 Greater Manchester -4900 -1907 -2993 -232 -90 -142ukd4 Lancashire -4600 -5082 482 -317 -350 33uke4 West Yorkshire -3700 -1192 -2508 -186 -60 -126ukf1 Derbyshire and Nottinghamshire -3700 -4965 1265 -179 -240 61ukf2 Leicestershire Rutland and Northan -4400 -3607 -793 -233 -191 -42ukg3 West Midlands -8700 -4999 -3701 -313 -180 -133ukh1 East Anglia -2500 -1292 -1208 -137 -71 -66uki1 Inner London -1919 -1381 -538 -142 -102 -40uki2 Outer London -4481 1106 -5587 -271 67 -338ukj1 Berkshire Bucks and Oxfordshire -4600 -3898 -702 -246 -208 -38ukj3 Hampshire and Isle of Wight -1100 -689 -411 -81 -51 -30ukk1 Gloucestershire Wiltshire and Brist -3400 -1253 -2147 -169 -62 -107ukm2 Eastern Scotland -1100 1402 -2502 -91 116 -207ukm3 South Western Scotland -3400 -2107 -1293 -207 -128 -79

United Kingdom -89800 -48574 -41226 -196 -106 -90

Table 1 Importance of manufacturing delocation in European regions 2001-2005 (Decomposition of change in employment)

Total effect

Source Elaborated from Eurostat and National Statistical offices

Number of jobs (thousands)

Total effect

Percent share of 2000 employmentRegions

Delocation in the manufacturing sectors in the EU A regional overview

By contrast thirty-eight remaining regions that also show reductions in

manufacturing employment have been affected by delocation processes to different

extents although only in a small number of them have they been accompanied by

decreases in labour productivity which could make the prospects for economic growth

tougher (four from France Piamonte Toscana and Lazio in Italy Outer London and

Eastern Scotland in the United Kingdom the Swedish Vaumlstsverige and Antwerp in

Belgium)

Regions with positive location effects are shown in Map 1 Most of them are large

industrial regions and are located in the traditional industrial growth centres5 those

characterized by high market potential (Combes and Overman 2003) although there are

also some intermediate and peripheral territories mainly located in the north of Italy

United Kingdom and Spain and in the south of Sweden and Norway This means that the

delocation process has contributed to concentrating the European manufacturing industries

in that privileged area of faster growth apart from some other changes inside every

country

To compare now the intensity of such effects of location and delocation across

regions avoiding the differences they have in size we built a relative measure of them the

ratio of job losses in the period to the volume of employment in the first year 2000 (Table

1)

Following this ratio delocation effects are revealed to be particularly strong in

France United Kingdom Italy and in one region of Belgium with declines in

employment exceeding 20 in some of them

Conversely the regions of Finland and several of the German and the Dutch ones

exhibit strong location effects

In any case as was pointed above the importance of location effects seems to lie in

labour productivity gains In fact higher increases in productivity (over 15) are found

precisely in those regions of Ireland Finland Sweden Netherlands and Germany besides

the Greek Attiki where the location effect has in general reached its greatest extent

(Figure 1)

5 Note that the map of winning regions is clearly influenced by the large number (30) of German NUTS 2 in the sample

9

Delocation in the manufacturing sectors in the EU A regional overview

fi19

fi18

ie02

dec0

gr30

se11

nl33

de21

de22

de26

de25

de12

de14

de11

dea4

dea1de73

def0

de60

deb3

de27

de23

itd5

ite3

se12

nl22

es 30

ukd4

es51

nl32

at22

ukf1

at31

es21

Map 1 Regions with positive location effect ( of 2000 employment)

Above 10 0 to 10

nl41

de13de27

Nevertheless productivity also grows faster in most of the regions belonging to the

United Kingdom Belgium and Denmark in spite of the strong delocation effects registered

10

Delocation in the manufacturing sectors in the EU A regional overview

which disturbs the above-mentioned positive relationship suggesting the harsh process of

industrial restructuring involved

Figure 1 Location effect and productivity growth in select regions manufacturing sector 2001-2005 (percentages)

UKi2

EU 15

FR6 2

UKm2

ITc1

UKg3UKd 3

UKe4

UKk1FR71F R

FR10UKm3

UKj1UKf2

UKd 4

UKi1

D K

DK01

UKh1U K

FR2 3FR4 2

FR24ITe1

BE2 1ITe4FR51

FR61

ITITd 3

SE2 2FR8 2

SE23

NL2 2

UKf1DE2 4

NL3 2DEa1

DEf0

UKj3DE9 1 DE92

DEa2DE71DK0 4

S EDE25

DE6 0DE12

A T

DE14

ITe3

AT2 2

AT3 1ESD E

DE 11DE2 3

NL41N LDE2 7

DEc0

NL3 3

DE2 6DE2 1

ES3 0ES51

SE11

DE2 2

ES21

GR

FI19F I

FI18

IE

GR3 0

B E

DE73

DEb 3

IE0 2

ITd 5 DE13SE12

FR52ITc4BE2 3

DEa4DEa5

y = 07x - 59R2 = 06

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

40

-35 -30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30 35 40 45

Productivity grothw Source Elaborated from Eurostat and National Statistical offices

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

11

Delocation in the manufacturing sectors in the EU A regional overview

In contrast the productivity decline in half of the French and Italian regions albeit

with a positive impact on employment is largely offset by destruction of industrial

activity except in Icircle de France and Midi -Pyreacuteneacutees the two territories in which

productivity has a more negative trajectory

These results closely match those reported by Cuadrado Roura et al (2000) based on

the growth rates for regional productivity and employment during the period 1980-1993

grouping European regions in four typologies comparing the growth rates of these two

variables with their respective European averages

According to this approach and including the location effect in the analysis the

group of regions that are examined in this work would be categorized as follows (Table 2)

The first group includes areas with productivity and employment growth rates above

sample average and show positive location effect (Quadrant 1) So as has been underlined

in that work possibly this reflects the fact that the technological and organizational

changes introduced during the period would have been successful promoting the

diversification and specialization in activities with high growth levels and attracting new

investments which are less labour-intensive

In a second stage are those regions that as the above mentioned authors suggest

have implemented restructuring processes to eliminate the most inefficient outputs

achieving higher levels of productivity and in about half of them increased industrial

production at the expense of a notable decline in employment (Quadrant 4) They named

this typology `restructuring via productivityacute

With a few exceptions (those where the location effect is negative) the regions in

both groups are the winners of industrial dynamic during the period Most of them are

large industrial regions and are located in the traditional industrial growth centres As

demonstrated above they are mainly Finish German and Dutch as well as Greek and Irish

regions

In contrast another large group of regions due to different factors6 have opted for

more labour-intensive industrial models which in a few cases (especially in Spanish and

Austrian regions) have been accompanied by output increases so they can also be included

6 See Cuadrado et al (2000)

12

Delocation in the manufacturing sectors in the EU A regional overview

in the winners (Map 1) while in other the result has been a loss of productive activity

(Quadrant 2)

(1) (2)Employment Productivity Employment Productivity

de11 Stuttgart -48 125 at22 Steiermark 36 11de12 Karlsruhe -57 101 at31 Oberoumlsterreich 06 56de21 Oberbayern -37 179 be21 Prov Antwerpen -70 -05de22 Niederbayern -33 218 be23 Prov Oost-Vlaanderen -74 61de23 Oberpfalz -38 120 de13 Freiburg -45 45de25 Mittelfranken -53 80 de14 Tuumlbingen -23 72de26 Unterfranken -49 189 de91 Braunschweig -47 25de27 Schwaben -41 148 es21 Pais Vasco 74 13de60 Hamburg -60 97 es30 Comunidad de Madrid -31 63dec0 Saarland -32 160 es51 Cataluntildea 09 25fi19 Laumlnsi-Suomi -54 297 fr10 Icircle de France 218 -307gr30 Attiki -17 234 fr51 Pays de la Loire -08 -67

fr52 Bretagne -23 12fr61 Aquitaine -14 -50fr62 Midi-Pyreacuteneacutees 15 -245fr71 Rhocircne-Alpes -61 -57fr82 Provence-Alpes-Cocircte dAzur -06 -18itc4 Lombardia -63 61itd3 Veneto -62 03itd5 Emilia-Romagna -16 16ite3 Marche 08 28ite4 Lazio -36 -47se23 Vaumlstsverige -26 -07

(4) (3)Employment Productivity Employment Productivity

de24 Oberfranken -116 186 fr24 Centre -106 06de71 Darmstadt -149 132 fr42 Alsace -154 61de73 Kassel -78 103 itc1 Piemonte -129 -48de92 Hannover -106 86 ite1 Toscana -98 -02dea1 Duumlsseldorf -126 177 se22 Sydsverige -91 62dea2 Koumlln -150 137 uke4 West Yorkshire -186 60dea4 Detmold -107 110 ukh1 East Anglia -137 71dea5 Arnsberg -129 122 uki2 Outer London -271 -67deb3 Rheinhessen-Pfalz -106 152 ukj3 Hampshire and Isle of Wight -81 51def0 Schleswig-Holstein -101 135 ukk1 Gloucestershire Wiltshire -169 62dk01 Hovedstaden -134 125 ukm2 Eastern Scotland -91 -116dk04 Midtjylland -146 118fi18 Etelauml-Suomi -86 339fr23 Haute-Normandie -171 77ie02 Southern and Eastern -78 426nl22 Gelderland -178 206nl32 Noord-Holland -138 180nl33 Zuid-Holland -129 296nl41 Noord-Brabant -173 269se11 Stockholm -136 363se12 Oumlstra Mellansverige -151 166ukd3 Greater Manchester -232 90ukd4 Lancashire -317 350ukf1 Derbyshire and Nottinghamshire -179 240ukf2 Leicestershire Rutland and Northants -233 191ukg3 West Midlands -313 180uki1 Inner London -142 102ukj1 Berkshire Bucks and Oxfordshire -246 208ukm3 South Western Scotland -207 128

Source Elaborated from Eurostat and national Statistical officesNote Shaded areas indicate regions with positive location effect

Em

ploy

men

t gro

wth

lt A

vera

ge v

alue

Productivity growth gt Average value Productivity growth lt Average value

Em

ploy

men

t gro

wth

gt A

vera

ge v

alue

Table 2 Employment and productivity growth rates in select regions manufacturing sector 2001-2005 (Averages of all regions +74 productivity -77 employment)

13

Delocation in the manufacturing sectors in the EU A regional overview

Finally the worst results can be seen in a fourth type of region with growth rates of

GVA and productivity below average and serious job destruction (Quadrant 3) This would

occur in regions with problems in restructuring their manufacturing sector or with low

levels of activity So together with those that are either restructuring via productivity or

via employment but without achieving a positive location effect they are clearly the

losers These include mainly British French Italian Belgian and Danish regions

To sum up these results confirm the direct relationship between growth in labour

efficiency and progress in the creation of productive industrial activity as pointed out

above and which can be more clearly by looking at Figure 1 In other words regions with

higher levels of productivity growth in general show a greater attractiveness for new

manufacturing firms

These differences in the behaviour of productivity and employment across regions

have led to remarkable disparities in rates of change of their industrial output during the

period Thus while a few increased their real manufacturing GVA by more than 20 in

the less dynamic ones the loss of industrial output is between a sixth and a third of the

initial value So significant changes had taken place in the relative position of some of

them (Figure 2)

Growth is led by a group of twenty regions located in Ireland Finland Greece

Germany Sweden Netherlands Spain and Austria with real GVA growth rates that are

more than five points above average As a result they increased their share in the industrial

output of the sample Looking at the complete list we see that the growth deviation from

the mean exhibits positive values in just over half of the regions examined among them

seven of the ten with the largest industrial scale in 2005 Thus the top ten regions have

increased their share from 32 to 33 percent in the last five years

Conversely among the less dynamic regions are found a great number of British and

some French and Italian ones which consequently have fallen back in the ranking of

NUTS 2 arranged by their contribution to aggregate manufacturing output

Between both groups there are approximately twenty regions mainly in Germany

with growth rates not far from average so their positions have hardly changed during the

last five years

14

Delocation in the manufacturing sectors in the EU A regional overview

Figure 2 Regional shares in the total sample manufacturing GVA(percentage at 2000 prices)

ITc4

UKf2

UKm3

Uke4

UKf1

UKk1

UKh1

AT22

ITe1

ES21

ES51

FR71

FR62

IT e4

DE24

DE26

SE23

FR61

FR51

FR82

SE22

ITe3

AT31

DK04

DE11

DE23

GR30

DEc0

ES30DEb3DEf0

DEa4

DE92

BE21

DE73

IE02

ITd3

FR52

DE91

DE13DE25

DE 12DE60

DE27

DE21

DE22

BE23DEa2

DE71

DE 14

SE12

ITd5

SE11

FI19

FR10

FR 24

ITe1

FI18

DK01 DEa5

FR4 2

DEa1NL32

NL3 3

FR2 3

NL22

NL 41

UKd3

UKj3

UKi1

UKm2

UKj1

UKd4

UKi2

Ukg3

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

Share in 2005

Source Elaborated from Eurostat and National Statistical offices

Rat

e of

cha

nge

200

1-20

05

70403530252015100500

Finally when the question is analyzed at Member State level (Table 1) it is possible

to check out the disparities among the above mentioned regions Thus while in some

15

Delocation in the manufacturing sectors in the EU A regional overview

countries such as Spain manufacturing growth has been clearly labour-intensive

something which can be extended to its regions as also happens in Greece and Austria

which both boast positive location effect and a job destruction rate far below the average

in others such as Holland Sweden and to a lesser extent Belgium Ireland and Finland

increasing of industrial activity is only due by advances in productivity so has been

accompanied by heavy job losses and these are patterns repeated in most of their regions

with noticeable relevance in the Netherlands As for the United Kingdom Belgium and

Denmark all share declining industrial production together with strong job cuts a pattern

that is replicated in many of their regions In this situation are the bulk of French and

Italian NUTS 2 just like the countries to which they belong although in this case job

destruction has been much less pronounced and below the group average Close to this

mean is Germany as well as most of their regions

Thus the analysis at Member States level also confirms the positive relationship

between the advance of industrial productivity and the attractiveness for new

manufacturing firms As demonstrated for their regions Ireland Finland Netherlands and

Greece following at quite distance from Germany are Member States showing a better

performance in productivity and with high location effects (see Figure 1)

In fact similarities between performances in the regions and in the Member States

they belong to suggest that the trend in regional industrial employment is to some extent

influenced by belonging to the same country in other words the locational attractions of

each area and its effort to improve productivity is partly due to national specificities or

idiosyncratic factors In fact as shown in Appendix 3 most European industrialized

regions follow the location patterns of the country they belong to reproducing its

productivity behaviour too In summary the evolution of these variables in each region

seems to have a clear `country effectacute that is different behaviours motivated by specific

factors andor differentiated industrial policies

For example in order to correctly interpret the results for Spanish regions it can not

be overlooked that during the period analyzed as in the whole country regions have

absorbed large numbers of immigrants which has boosted the proliferation of lower

productivity and lower wages in industrial activities However it may be thought that this

phenomenon has tended to conceal productivity increases in larger and better equipped

16

Delocation in the manufacturing sectors in the EU A regional overview

firms which could have been significant in the light of their results7 On the other hand

countries like Finland or Ireland with the lowest population increase have opted for high-

tech manufacturing sectors as a way to improve national competitiveness8 Nonetheless

this positive location effect has not reached in these countriesrsquo NUTS 2 enough magnitude

to offset completely the job losses arising from productivity improvements

On the other hand this `national effectacute has also played an important role in

explaining the economic outcomes of European regions in terms of per capita GDP in

accordance with the results obtained by Cuadrado-Roura (2001) using the data available

for 109 EU regions over the 1977-1994 period

Table 3 Export growth in key sectors by countries 2000-2005 (percentages)Impact ITon Total

Spain 65 91 72 158 38 111 04Germany 116 138 61 111 111 130 14United Kingdom 09 85 29 32 23 61 02France -09 102 25 94 118 78 -01Ireland -09 -102 01 25 -34 77 -03Finland 47 152 45 49 56 73 13Italy 40 88 41 90 56 89 02EU-15 70 141 59 91 67 111 11Source Elaborated from Comtrade

Shoes TotalIT Auto Textile Clothes

This importance of the `national effectacute should allow us to make a better approach to

the determinants of above-mentioned delocation patterns in future research One of such

determinants might be found in the different importance and performance by countries of

some manufacturing branches particularly affected by the increase in global competition as

their development has being receiving strong support in emerging countries from both

domestic policies and the location of big multinational companies This is the case of IT

manufacturing sectors or that of automobiles clothes textiles and shoes (Sachwald 2004)

Table 3 shows how the exports of these key sectors have evolved in some of the European

countries considered here It is amazing to see the poor results registered by IT sectors in

United Kingdom France and Ireland compared to the other countries mentioned Germany

in particular This evolution could be behind the stronger effects of delocation in France

and the United Kingdom although it must have produced similar results in the case of

Ireland Furthermore on such a basis what has happened in Italy would remain

7 For details see Myro et al (2008) 8 See Aacutelvarez Loacutepez et al (2007)

17

Delocation in the manufacturing sectors in the EU A regional overview

unexplained as the selected key sectors do not show a bad performance in terms of exports

in the latter country

One reason for such contradictory results lies in the possibility that production

instead of export reflects accurately what has been happening in every sector in terms of

activity However Aacutelvarez et al (2007) show very similar patterns in production and

export specialization from 1995 to 2005 except in the case of Ireland which thus appears

as a very particular one This is because IT specialization in production increases at the

same time as it decreases in exports Perhaps the strong growth in domestic demand

prevents IT exports from growing The opposite case could be that of Italy where slow

growth in domestic demand might have been pushing up IT export9

Anyway further developments on the determinants of delocation patterns require the

preparation of disaggregated data on production a task beyond the objectives of this paper

That could allow us to update the available manufacturing specialization analysis As the

two countries exhibiting higher delocation effects France and the United Kingdom

showed the lowest rates of specialization compared with the EU (Milderfart and Overman

2002 Combes and Overman 2003) something could have changed in the last few years

which would merit a fresh look based on more recent data

On the other hand it is now clear that insofar as delocation in manufacturing

activities has affected each country in a different way a relocation process inside the EU

and between countries has taken place This process deserved to be deeply analyzed

including a perspective by sectors updating what we know about changes in the

distribution of manufacturing industry in the EU and offering the possibility of exploring

whether the European Common Policy has influenced them in some way as was already

suggested by Milderfart-Knarvik and Overman (2002)

9 In Finland IT activities grows much more in production than in exports but in both aspects a positive growth is registered

18

Delocation in the manufacturing sectors in the EU A regional overview

19

Figure 3 Employment growth rates in selected regions 2001-2005 manufacturing and total economy (percentages)

Ukg3

UKi2

UKd4

UKj1

UKm2

UKi1 UKj3

UKd3

NL 41

NL22

FR23

NL33

NL32

DEa1

FR42

DEa5

DK01

FI18

ITe1

FR 24FR10

FI19

SE11

ITd5FR52

DE 14

DE71

DEa2

BE23

DE22

DE21

DE27

IC c4

DE60DE 12

DE25

DE13

DE9 1SE12

ITd3

IE02

DE73

BE21

DE92DEa4

DEf0

DEb3

ES30

DEc0

GR30

DE23

DE11

DK04

AT31

ITe3

SE22

FR82

FR51FR61

SE2 3

DE26

DE24

IT e4

FR62

FR71

ES51

ES2 1

ITe1

AT22

UKh1

UKk1

UKf1

Uke4

UKm3

UKf2

-20

-15

-10

-5

0

5

10

15

20

25

-40 -30 -20 -10 0 10 20 30

Manufacturing

Source Elaborated from Eurostat and National Statistical offices

Tota

l em

ploy

men

t

Delocation in the manufacturing sectors in the EU A regional overview

One last question that deserves consideration is the effect of industrial delocation

process on the regional economies Surprisingly manufacturing job losses have not been

translated into total employment which exhibit positive growth for most of the chosen

regions with the exceptions of some Germans and all the Netherlands (Figure 3) That

means the regions have been able to compensate the negative effect on the employment

from shrinking manufacturing allocating more resources inside the services sectors In fact

a substitution of manufacturing by services seems to have happened as are the region

showing more delocation in manufacturing those that increase in a higher measure the

employment in the service sectors that is the French British and Italian However the

relation is not accurate as there are not a strict quantitative correlation region by region

between rates of delocation and increase in the employment in the service activities

These good results in terms of general employment shown by the regions affected in

a higher measure by manufacturing delocation suggest the convenience of a more general

analysis of regional economic growth beyond the manufacturing sector as some regions

strongly shrinking manufacturing activities in favour of those of services might be leading

an interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

5- Final remarks

Comparative analysis of location patterns for a large sample of European regions -

characterized by having an upper-middle level income and a strong industrial sector-

during the last few years has revealed that only half of them have shown locational

attractions for manufacturing activities The regional patterns found clearly show the

influence of `national effectacute the regions showing location effects corresponding to most

of those located in Austria Germany the Netherlands Ireland Finland Sweden Spain and

Greece

On the contrary the delocation process has affected another half of the regions

included most of them French British and Italian Therefore delocation has contributed to

concentrating the European manufacturing industry in the areas with higher market

potential in the centre of continental Europe It has contributed furthermore to relocating

manufacturing activities inside every country something that requires further analysis

from future research

20

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

development and homogeneous space dimension So the sample contains only

geographical areas at NUTS 2 level3 whose industrial production represents at least 04

of total EU manufacturing GVA and at the same time has a per capita income of over

90 of the EU-15 average Therefore they are developed regions that have a powerful

industry at the Community level

Consistently the resulting list from applying to the set of EU-27 territories the three

already mentioned selection criteria includes a total of seventy-five regions of thirteen

Member States distributed as follows 23 from Germany 14 the United Kingdom 10 from

France 7 from Italy 4 Holland 4 Sweden 3 Spain 2 from Austria Belgium Denmark

and Finland 1 Greek and 1 Irish On the whole they accounted for 64 of real GVA in

2005 and almost three-fifths of manufacturing employment in the EU-15 (60 and 45

respectively taking as reference the EU-27) The complete catalogue of regions grouped

by country and their characteristics are set out in Appendixes 1 and 2 As is shown there

regions included in the sample are principally in sizes over 05 of EU-15 industrial

production and above the average EU-15 per capita income although a significant number

of them lay below those levels particularly in the per capita income Looking more closely

at the industrial size although most of the NUTS 2 examined move in around 05 there

are a group of fifteen regions located in Italy Germany Ireland France and Spain with a

strong industrial sector (more than 1) among them Lombardy in Italy with about 4 of

total EU-15 manufacturing GVA

After this first introductory section the paper is organized as follows In a second

section the model to capture the delocation patterns is introduced Then in a third section

the data sources are commented on In section four we try to asses the impact of delocation

in each region in terms of jobs affected and examine whether their location patterns have

altered its position in the European industrial scene contrasting country trends with

regional performances Concluding remarks round up the paper

3 The term NUTS corresponds to the French acronym for Nomenclature of Statistical Territorial Units used by the European Union This classification has a hierarchic structure at three levels which among other factors comes from demographic thresholds In particular the NUTS2 level covers regions of an average size between 08 to 3 million inhabitants In Spain the nineteen units included in this level coincide with the different AACC more than Ceuta and Melilla

3

Delocation in the manufacturing sectors in the EU A regional overview

2- Measuring delocation

As was posed in the introduction following Baldwin and Robert-Nicoud (2000)

delocation of activities has to be assessed in terms of losses of productive activity within

each region The variables which approximate these losses are gross added value and

number of jobs or total employment The first one is the most suitable since the latter is

dependent on advances registered in labour productivity mainly in the face of strong

competitive pressures forcing companies to achieve greater efficiency gains

In spite of that jobs continue to be a useful measure since its calculation is simpler

and is often more reliable In addition this indicator has received greater attention and is

more easily interpreted by the analysts and especially the social partners

Nevertheless if evolution of jobs is chosen as a measure of delocation it is necessary

to discount the effect on it of an increase in labour productivity (which reduces the need for

labour) and of the economic cycle (which may reduce or expand the existing employment)

The remaining reduction in the number of jobs measures the delocation effect on

employment

Furthermore when average values for quite a long time interval are taken it is

possible to ignore almost completely the impact of the cycle that otherwise would be

estimated using econometric techniques to isolate the trend In this way the delocation

effect can be just approximated after deducting the change in jobs caused by the increase in

labour productivity

Therefore the change in total employment may be split up into two effects one of

them due to the increased of labour productivity and the other to a location effect -

delocation if it is negative- Box 1 shows that decomposition

In fact the location effect reflects the impact on jobs of an increase or decrease in

value added estimated through the primitive rate of labour productivity and as has been

mentioned above may be positive (location) or negative (delocation) In the first case an

expansion in activity has taken place while in the second a reduction of the productive

scope has occurred

4

Delocation in the manufacturing sectors in the EU A regional overview

∆ N = GVA N(0) ⎯ π N(0)

Delocalisation ProductivityEffect Effect

Box 1 GVA= Labor Productivity Employment GVA = π N where π represents productivity and N employment

N = GVA - π

∆ N = N N(0)

N = GVAπ

where a hat over the variable denotes its rate of change

and and and

and

and and

Delocation of activities that can result from this calculation is compatible with the

absence of off-shoring companies and it may be due as noted above to the closure of

establishments as a result of their being uncompetitive In the same way the location of

activities in a territory is compatible with off-shoring of companies

Thus analysis of delocation does not necessarily tell us much about the importance

of firmsrsquo off-shoring a matter that has to be studied on a different basis However when

off-shoring reaches a large-scale dimension it affects the extension of manufacturing

industries

3- Database

As was said before to achieve aggregate data on manufacturing activities by regions

even just those of real GVA and total employment has demanded a laborious task of

collecting information provided by Eurostat (REGIO database) and the National Statistical

Offices of member countries to which the different regions belong as well as the

estimation in most cases of the product in real terms

Looking first at the GVA since the data offered by REGIO include those related to

energy and mining only access to the National Statistics of every member country has

allowed us to isolate the manufacturing sectors Nevertheless additional work has been

applied to transform the data of GVA into real values as only Germany Spain Finland

5

Delocation in the manufacturing sectors in the EU A regional overview

Holland and Italy provide such information or at least indexes of volume In the rest of the

countries real values of GVA have been estimated by applying the national price deflators

to the regional series at current values

As regards total employment significant discrepancies between the data published

by Eurostat and that country itself (particularly in the United Kingdom) are found as well

as the gaps found in several NUTS 2 of some Member States such as in Germany forced

to complete regional series and correct such deficiencies with the help of the information

coming from the National Statistics Offices

To sum up most of the information used (especially production data) comes from

National Statistical Offices but this is commonly close to the Eurostat database as this is

built mainly with data coming from the regional accounts of each country

By the way one of the problems arising from taking the National Statistics as the

main source of data is that the time intervals for which they provide information do not

always coincide That explains that despite having more recent data from some areas the

analysis must conclude in 2005

4- Delocation of manufacturing industries in European regions

As noted above in the introduction the group of regions examined reached just over

64 of EU-15 industrial GVA and about 60 of jobs in 2005 figures slightly lower than

in 2000 Thus despite nearly half of regionsrsquo industrial output at constant prices having

achieved positive growth rates the group accumulated throughout the period a decrease of

about 1 similar to EU-15 average4 Regarding employment the evolution has been even

more negative Throughout the first five years of the current century destruction of jobs in

the aggregate industry has been a common feature in the vast majority (90) of the

seventy-five NUTS 2 examined as well as for each of the Member States where they are

located except Spain accumulating the total sample a decline of close to 8

4 It must be noted that the EU average has been calculated from the aggregate industrial GVA at constant prices of thirteen countries included in the sample Thus Portugal and Luxembourg have been excluded and the values from other countries have been estimated by adding figures of all their regions So the above mentioned growth rate differs from that provided by Eurostat exhibiting a positive increase of 46

6

Delocation in the manufacturing sectors in the EU A regional overview

Focusing on employment the most dynamic areas are located mostly in France

Spain Austria and Italy while the Netherlands and especially the British regions exhibit

the largest job losses (Table 1)

Registered unemployment seems to be related to good performance in labour

productivity pointing to the fierce competition faced by the manufacturing sector in the

analyzed period Thus higher job losses are found in regions with stronger labour

productivity increases although some French and most of the British regions escaped from

this rule as large losses in employment are not accompanied by significant gains in labour

productivity suggesting delocation effects of a particular intensity

That reveals a very different growth path in employment and labour productivity by

regions differences which can be explored in a more detail to isolate the attractive regions

to locate manufacturing activities from the others characterized by a delocation process

To go deeply into this latter question Table 1 contains the results of estimating the

impact on total employment of location and delocation forces in the manufacturing sectors

across the seventy-five regions included in the sample over the period 2001-2005 by using

the procedure proposed in the second section that is discounting from the total change in

jobs the reduction corresponding to the increase in labour productivity Therefore the

location effect shows the effect on employment of GVA change calculated by applying the

initial rate of labour productivity

In about half of the regions with job losses all of those considered in the

Netherlands Finland Greece and Ireland three-quarters of the Germans two British two

Swedish and Madrid in Spain the decline in employment has been due entirely to

increased labour productivity so a positive effect of location appears This behaviour

extends to Finnish German and Swedish economies as a whole as well as Greece Ireland

and Austria Therefore all those regions have proved attractive for the location of

manufacturing firms

7

Delocation in the manufacturing sectors in the EU A regional overview

8

Productivity Location Productivity Locationeffect effect effect effect

at22 Steiermark 372 -108 480 36 -11 47at31 Oberoumlsterreich 097 -837 934 06 -56 62

Austria -1104 -2649 1545 -18 -42 25be21 Prov Antwerpen -1010 079 -1089 -70 05 -75be23 Prov Oost-Vlaanderen -750 -627 -123 -74 -61 -12

Belgium -5740 -2219 -3521 -84 -32 -51de11 Stuttgart -3157 -8244 5087 -48 -125 77de12 Karlsruhe -2036 -3585 1550 -57 -101 44de13 Freiburg -1291 -1301 011 -45 -45 00de14 Tuumlbingen -585 -1805 1219 -23 -72 49de21 Oberbayern -1630 -7976 6346 -37 -179 142de22 Niederbayern -490 -3270 2781 -33 -218 186de23 Oberpfalz -538 -1716 1178 -38 -120 83de24 Oberfranken -1957 -3131 1175 -116 -186 70de25 Mittelfranken -1184 -1770 586 -53 -80 26de26 Unterfranken -819 -3157 2338 -49 -189 140de27 Schwaben -932 -3335 2403 -41 -148 106de60 Hamburg -760 -1222 463 -60 -97 37de71 Darmstadt -5343 -4720 -623 -149 -132 -17de73 Kassel -1059 -1396 336 -78 -103 25de91 Braunschweig -947 -500 -447 -47 -25 -22de92 Hannover -1900 -1544 -356 -106 -86 -20dea1 Duumlsseldorf -6272 -8787 2515 -126 -177 51dea2 Koumlln -5428 -4971 -457 -150 -137 -13dea4 Detmold -3026 -3112 086 -107 -110 03dea5 Arnsberg -5824 -5502 -322 -129 -122 -07deb3 Rheinhessen-Pfalz -2099 -3016 917 -106 -152 46dec0 Saarland -360 -1810 1450 -32 -160 128def0 Schleswig-Holstein -1867 -2485 617 -101 -135 33

Germany -60300 -112275 51975 -74 -138 64dk01 Hovedstaden -1300 -1216 -084 -134 -125 -09dk04 Midtjylland -2000 -1622 -378 -146 -118 -28

Denmark -6100 -3428 -2672 -130 -73 -57es21 Pais Vasco 1840 -331 2171 74 -13 87es30 Comunidad de Madrid -1060 -2180 1120 -31 -63 32es51 Cataluntildea 660 -1914 2574 09 -25 34

Spain 13970 -3927 17897 47 -13 60fi18 Etelauml-Suomi -1807 -7159 5352 -86 -339 253fi19 Laumlnsi-Suomi -701 -3886 3185 -54 -297 244

Finland -1798 -12006 10208 -43 -286 243fr10 Icircle de France 12652 17868 -5216 218 307 -90fr23 Haute-Normandie -2418 -1093 -1325 -171 -77 -94fr24 Centre -2016 -120 -1896 -106 -06 -100fr42 Alsace -2489 -988 -1501 -154 -61 -93fr51 Pays de la Loire -217 1794 -2011 -08 67 -75fr52 Bretagne -415 -215 -200 -23 -12 -11fr61 Aquitaine -199 742 -941 -14 50 -64fr62 Midi-Pyreacuteneacutees 209 3318 -3109 15 245 -229fr71 Rhocircne-Alpes -2919 2759 -5678 -61 57 -118fr82 Provence-Alpes-Cocircte dAzur -097 262 -359 -06 18 -24

France -15306 31566 -46872 -40 83 -123gr30 Attiki -360 -5072 4712 -17 -234 217

Greece -730 -11989 11259 -14 -222 208ie02 Southern and Eastern -1840 -9985 8145 -78 -426 347

Ireland -2420 -12274 9854 -76 -387 311itc1 Piemonte -6869 2539 -9407 -129 48 -177itc4 Lombardia -7940 -7682 -258 -63 -61 -02itd3 Veneto -3967 -211 -3756 -62 -03 -58itd5 Emilia-Romagna -860 -840 -020 -16 -16 00ite1 Toscana -3620 078 -3697 -98 02 -101ite3 Marche 164 -555 719 08 -28 36ite4 Lazio -733 956 -1689 -36 47 -82

Italy -21761 6310 -28071 -45 13 -58nl22 Gelderland -2164 -2506 341 -178 -206 28nl32 Noord-Holland -1555 -2034 479 -138 -180 42nl33 Zuid-Holland -1734 -3980 2246 -129 -296 167nl41 Noord-Brabant -3539 -5515 1976 -173 -269 96

Netherlands -14767 -24690 9923 -161 -270 108se11 Stockholm -1570 -4183 2613 -136 -363 227se12 Oumlstra Mellansverige -2170 -2387 217 -151 -166 15se22 Sydsverige -1010 -690 -320 -91 -62 -29se23 Vaumlstsverige -460 134 -594 -26 07 -33

Sweden -7500 -9400 1900 -95 -119 24ukd3 Greater Manchester -4900 -1907 -2993 -232 -90 -142ukd4 Lancashire -4600 -5082 482 -317 -350 33uke4 West Yorkshire -3700 -1192 -2508 -186 -60 -126ukf1 Derbyshire and Nottinghamshire -3700 -4965 1265 -179 -240 61ukf2 Leicestershire Rutland and Northan -4400 -3607 -793 -233 -191 -42ukg3 West Midlands -8700 -4999 -3701 -313 -180 -133ukh1 East Anglia -2500 -1292 -1208 -137 -71 -66uki1 Inner London -1919 -1381 -538 -142 -102 -40uki2 Outer London -4481 1106 -5587 -271 67 -338ukj1 Berkshire Bucks and Oxfordshire -4600 -3898 -702 -246 -208 -38ukj3 Hampshire and Isle of Wight -1100 -689 -411 -81 -51 -30ukk1 Gloucestershire Wiltshire and Brist -3400 -1253 -2147 -169 -62 -107ukm2 Eastern Scotland -1100 1402 -2502 -91 116 -207ukm3 South Western Scotland -3400 -2107 -1293 -207 -128 -79

United Kingdom -89800 -48574 -41226 -196 -106 -90

Table 1 Importance of manufacturing delocation in European regions 2001-2005 (Decomposition of change in employment)

Total effect

Source Elaborated from Eurostat and National Statistical offices

Number of jobs (thousands)

Total effect

Percent share of 2000 employmentRegions

Delocation in the manufacturing sectors in the EU A regional overview

By contrast thirty-eight remaining regions that also show reductions in

manufacturing employment have been affected by delocation processes to different

extents although only in a small number of them have they been accompanied by

decreases in labour productivity which could make the prospects for economic growth

tougher (four from France Piamonte Toscana and Lazio in Italy Outer London and

Eastern Scotland in the United Kingdom the Swedish Vaumlstsverige and Antwerp in

Belgium)

Regions with positive location effects are shown in Map 1 Most of them are large

industrial regions and are located in the traditional industrial growth centres5 those

characterized by high market potential (Combes and Overman 2003) although there are

also some intermediate and peripheral territories mainly located in the north of Italy

United Kingdom and Spain and in the south of Sweden and Norway This means that the

delocation process has contributed to concentrating the European manufacturing industries

in that privileged area of faster growth apart from some other changes inside every

country

To compare now the intensity of such effects of location and delocation across

regions avoiding the differences they have in size we built a relative measure of them the

ratio of job losses in the period to the volume of employment in the first year 2000 (Table

1)

Following this ratio delocation effects are revealed to be particularly strong in

France United Kingdom Italy and in one region of Belgium with declines in

employment exceeding 20 in some of them

Conversely the regions of Finland and several of the German and the Dutch ones

exhibit strong location effects

In any case as was pointed above the importance of location effects seems to lie in

labour productivity gains In fact higher increases in productivity (over 15) are found

precisely in those regions of Ireland Finland Sweden Netherlands and Germany besides

the Greek Attiki where the location effect has in general reached its greatest extent

(Figure 1)

5 Note that the map of winning regions is clearly influenced by the large number (30) of German NUTS 2 in the sample

9

Delocation in the manufacturing sectors in the EU A regional overview

fi19

fi18

ie02

dec0

gr30

se11

nl33

de21

de22

de26

de25

de12

de14

de11

dea4

dea1de73

def0

de60

deb3

de27

de23

itd5

ite3

se12

nl22

es 30

ukd4

es51

nl32

at22

ukf1

at31

es21

Map 1 Regions with positive location effect ( of 2000 employment)

Above 10 0 to 10

nl41

de13de27

Nevertheless productivity also grows faster in most of the regions belonging to the

United Kingdom Belgium and Denmark in spite of the strong delocation effects registered

10

Delocation in the manufacturing sectors in the EU A regional overview

which disturbs the above-mentioned positive relationship suggesting the harsh process of

industrial restructuring involved

Figure 1 Location effect and productivity growth in select regions manufacturing sector 2001-2005 (percentages)

UKi2

EU 15

FR6 2

UKm2

ITc1

UKg3UKd 3

UKe4

UKk1FR71F R

FR10UKm3

UKj1UKf2

UKd 4

UKi1

D K

DK01

UKh1U K

FR2 3FR4 2

FR24ITe1

BE2 1ITe4FR51

FR61

ITITd 3

SE2 2FR8 2

SE23

NL2 2

UKf1DE2 4

NL3 2DEa1

DEf0

UKj3DE9 1 DE92

DEa2DE71DK0 4

S EDE25

DE6 0DE12

A T

DE14

ITe3

AT2 2

AT3 1ESD E

DE 11DE2 3

NL41N LDE2 7

DEc0

NL3 3

DE2 6DE2 1

ES3 0ES51

SE11

DE2 2

ES21

GR

FI19F I

FI18

IE

GR3 0

B E

DE73

DEb 3

IE0 2

ITd 5 DE13SE12

FR52ITc4BE2 3

DEa4DEa5

y = 07x - 59R2 = 06

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

40

-35 -30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30 35 40 45

Productivity grothw Source Elaborated from Eurostat and National Statistical offices

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

11

Delocation in the manufacturing sectors in the EU A regional overview

In contrast the productivity decline in half of the French and Italian regions albeit

with a positive impact on employment is largely offset by destruction of industrial

activity except in Icircle de France and Midi -Pyreacuteneacutees the two territories in which

productivity has a more negative trajectory

These results closely match those reported by Cuadrado Roura et al (2000) based on

the growth rates for regional productivity and employment during the period 1980-1993

grouping European regions in four typologies comparing the growth rates of these two

variables with their respective European averages

According to this approach and including the location effect in the analysis the

group of regions that are examined in this work would be categorized as follows (Table 2)

The first group includes areas with productivity and employment growth rates above

sample average and show positive location effect (Quadrant 1) So as has been underlined

in that work possibly this reflects the fact that the technological and organizational

changes introduced during the period would have been successful promoting the

diversification and specialization in activities with high growth levels and attracting new

investments which are less labour-intensive

In a second stage are those regions that as the above mentioned authors suggest

have implemented restructuring processes to eliminate the most inefficient outputs

achieving higher levels of productivity and in about half of them increased industrial

production at the expense of a notable decline in employment (Quadrant 4) They named

this typology `restructuring via productivityacute

With a few exceptions (those where the location effect is negative) the regions in

both groups are the winners of industrial dynamic during the period Most of them are

large industrial regions and are located in the traditional industrial growth centres As

demonstrated above they are mainly Finish German and Dutch as well as Greek and Irish

regions

In contrast another large group of regions due to different factors6 have opted for

more labour-intensive industrial models which in a few cases (especially in Spanish and

Austrian regions) have been accompanied by output increases so they can also be included

6 See Cuadrado et al (2000)

12

Delocation in the manufacturing sectors in the EU A regional overview

in the winners (Map 1) while in other the result has been a loss of productive activity

(Quadrant 2)

(1) (2)Employment Productivity Employment Productivity

de11 Stuttgart -48 125 at22 Steiermark 36 11de12 Karlsruhe -57 101 at31 Oberoumlsterreich 06 56de21 Oberbayern -37 179 be21 Prov Antwerpen -70 -05de22 Niederbayern -33 218 be23 Prov Oost-Vlaanderen -74 61de23 Oberpfalz -38 120 de13 Freiburg -45 45de25 Mittelfranken -53 80 de14 Tuumlbingen -23 72de26 Unterfranken -49 189 de91 Braunschweig -47 25de27 Schwaben -41 148 es21 Pais Vasco 74 13de60 Hamburg -60 97 es30 Comunidad de Madrid -31 63dec0 Saarland -32 160 es51 Cataluntildea 09 25fi19 Laumlnsi-Suomi -54 297 fr10 Icircle de France 218 -307gr30 Attiki -17 234 fr51 Pays de la Loire -08 -67

fr52 Bretagne -23 12fr61 Aquitaine -14 -50fr62 Midi-Pyreacuteneacutees 15 -245fr71 Rhocircne-Alpes -61 -57fr82 Provence-Alpes-Cocircte dAzur -06 -18itc4 Lombardia -63 61itd3 Veneto -62 03itd5 Emilia-Romagna -16 16ite3 Marche 08 28ite4 Lazio -36 -47se23 Vaumlstsverige -26 -07

(4) (3)Employment Productivity Employment Productivity

de24 Oberfranken -116 186 fr24 Centre -106 06de71 Darmstadt -149 132 fr42 Alsace -154 61de73 Kassel -78 103 itc1 Piemonte -129 -48de92 Hannover -106 86 ite1 Toscana -98 -02dea1 Duumlsseldorf -126 177 se22 Sydsverige -91 62dea2 Koumlln -150 137 uke4 West Yorkshire -186 60dea4 Detmold -107 110 ukh1 East Anglia -137 71dea5 Arnsberg -129 122 uki2 Outer London -271 -67deb3 Rheinhessen-Pfalz -106 152 ukj3 Hampshire and Isle of Wight -81 51def0 Schleswig-Holstein -101 135 ukk1 Gloucestershire Wiltshire -169 62dk01 Hovedstaden -134 125 ukm2 Eastern Scotland -91 -116dk04 Midtjylland -146 118fi18 Etelauml-Suomi -86 339fr23 Haute-Normandie -171 77ie02 Southern and Eastern -78 426nl22 Gelderland -178 206nl32 Noord-Holland -138 180nl33 Zuid-Holland -129 296nl41 Noord-Brabant -173 269se11 Stockholm -136 363se12 Oumlstra Mellansverige -151 166ukd3 Greater Manchester -232 90ukd4 Lancashire -317 350ukf1 Derbyshire and Nottinghamshire -179 240ukf2 Leicestershire Rutland and Northants -233 191ukg3 West Midlands -313 180uki1 Inner London -142 102ukj1 Berkshire Bucks and Oxfordshire -246 208ukm3 South Western Scotland -207 128

Source Elaborated from Eurostat and national Statistical officesNote Shaded areas indicate regions with positive location effect

Em

ploy

men

t gro

wth

lt A

vera

ge v

alue

Productivity growth gt Average value Productivity growth lt Average value

Em

ploy

men

t gro

wth

gt A

vera

ge v

alue

Table 2 Employment and productivity growth rates in select regions manufacturing sector 2001-2005 (Averages of all regions +74 productivity -77 employment)

13

Delocation in the manufacturing sectors in the EU A regional overview

Finally the worst results can be seen in a fourth type of region with growth rates of

GVA and productivity below average and serious job destruction (Quadrant 3) This would

occur in regions with problems in restructuring their manufacturing sector or with low

levels of activity So together with those that are either restructuring via productivity or

via employment but without achieving a positive location effect they are clearly the

losers These include mainly British French Italian Belgian and Danish regions

To sum up these results confirm the direct relationship between growth in labour

efficiency and progress in the creation of productive industrial activity as pointed out

above and which can be more clearly by looking at Figure 1 In other words regions with

higher levels of productivity growth in general show a greater attractiveness for new

manufacturing firms

These differences in the behaviour of productivity and employment across regions

have led to remarkable disparities in rates of change of their industrial output during the

period Thus while a few increased their real manufacturing GVA by more than 20 in

the less dynamic ones the loss of industrial output is between a sixth and a third of the

initial value So significant changes had taken place in the relative position of some of

them (Figure 2)

Growth is led by a group of twenty regions located in Ireland Finland Greece

Germany Sweden Netherlands Spain and Austria with real GVA growth rates that are

more than five points above average As a result they increased their share in the industrial

output of the sample Looking at the complete list we see that the growth deviation from

the mean exhibits positive values in just over half of the regions examined among them

seven of the ten with the largest industrial scale in 2005 Thus the top ten regions have

increased their share from 32 to 33 percent in the last five years

Conversely among the less dynamic regions are found a great number of British and

some French and Italian ones which consequently have fallen back in the ranking of

NUTS 2 arranged by their contribution to aggregate manufacturing output

Between both groups there are approximately twenty regions mainly in Germany

with growth rates not far from average so their positions have hardly changed during the

last five years

14

Delocation in the manufacturing sectors in the EU A regional overview

Figure 2 Regional shares in the total sample manufacturing GVA(percentage at 2000 prices)

ITc4

UKf2

UKm3

Uke4

UKf1

UKk1

UKh1

AT22

ITe1

ES21

ES51

FR71

FR62

IT e4

DE24

DE26

SE23

FR61

FR51

FR82

SE22

ITe3

AT31

DK04

DE11

DE23

GR30

DEc0

ES30DEb3DEf0

DEa4

DE92

BE21

DE73

IE02

ITd3

FR52

DE91

DE13DE25

DE 12DE60

DE27

DE21

DE22

BE23DEa2

DE71

DE 14

SE12

ITd5

SE11

FI19

FR10

FR 24

ITe1

FI18

DK01 DEa5

FR4 2

DEa1NL32

NL3 3

FR2 3

NL22

NL 41

UKd3

UKj3

UKi1

UKm2

UKj1

UKd4

UKi2

Ukg3

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

Share in 2005

Source Elaborated from Eurostat and National Statistical offices

Rat

e of

cha

nge

200

1-20

05

70403530252015100500

Finally when the question is analyzed at Member State level (Table 1) it is possible

to check out the disparities among the above mentioned regions Thus while in some

15

Delocation in the manufacturing sectors in the EU A regional overview

countries such as Spain manufacturing growth has been clearly labour-intensive

something which can be extended to its regions as also happens in Greece and Austria

which both boast positive location effect and a job destruction rate far below the average

in others such as Holland Sweden and to a lesser extent Belgium Ireland and Finland

increasing of industrial activity is only due by advances in productivity so has been

accompanied by heavy job losses and these are patterns repeated in most of their regions

with noticeable relevance in the Netherlands As for the United Kingdom Belgium and

Denmark all share declining industrial production together with strong job cuts a pattern

that is replicated in many of their regions In this situation are the bulk of French and

Italian NUTS 2 just like the countries to which they belong although in this case job

destruction has been much less pronounced and below the group average Close to this

mean is Germany as well as most of their regions

Thus the analysis at Member States level also confirms the positive relationship

between the advance of industrial productivity and the attractiveness for new

manufacturing firms As demonstrated for their regions Ireland Finland Netherlands and

Greece following at quite distance from Germany are Member States showing a better

performance in productivity and with high location effects (see Figure 1)

In fact similarities between performances in the regions and in the Member States

they belong to suggest that the trend in regional industrial employment is to some extent

influenced by belonging to the same country in other words the locational attractions of

each area and its effort to improve productivity is partly due to national specificities or

idiosyncratic factors In fact as shown in Appendix 3 most European industrialized

regions follow the location patterns of the country they belong to reproducing its

productivity behaviour too In summary the evolution of these variables in each region

seems to have a clear `country effectacute that is different behaviours motivated by specific

factors andor differentiated industrial policies

For example in order to correctly interpret the results for Spanish regions it can not

be overlooked that during the period analyzed as in the whole country regions have

absorbed large numbers of immigrants which has boosted the proliferation of lower

productivity and lower wages in industrial activities However it may be thought that this

phenomenon has tended to conceal productivity increases in larger and better equipped

16

Delocation in the manufacturing sectors in the EU A regional overview

firms which could have been significant in the light of their results7 On the other hand

countries like Finland or Ireland with the lowest population increase have opted for high-

tech manufacturing sectors as a way to improve national competitiveness8 Nonetheless

this positive location effect has not reached in these countriesrsquo NUTS 2 enough magnitude

to offset completely the job losses arising from productivity improvements

On the other hand this `national effectacute has also played an important role in

explaining the economic outcomes of European regions in terms of per capita GDP in

accordance with the results obtained by Cuadrado-Roura (2001) using the data available

for 109 EU regions over the 1977-1994 period

Table 3 Export growth in key sectors by countries 2000-2005 (percentages)Impact ITon Total

Spain 65 91 72 158 38 111 04Germany 116 138 61 111 111 130 14United Kingdom 09 85 29 32 23 61 02France -09 102 25 94 118 78 -01Ireland -09 -102 01 25 -34 77 -03Finland 47 152 45 49 56 73 13Italy 40 88 41 90 56 89 02EU-15 70 141 59 91 67 111 11Source Elaborated from Comtrade

Shoes TotalIT Auto Textile Clothes

This importance of the `national effectacute should allow us to make a better approach to

the determinants of above-mentioned delocation patterns in future research One of such

determinants might be found in the different importance and performance by countries of

some manufacturing branches particularly affected by the increase in global competition as

their development has being receiving strong support in emerging countries from both

domestic policies and the location of big multinational companies This is the case of IT

manufacturing sectors or that of automobiles clothes textiles and shoes (Sachwald 2004)

Table 3 shows how the exports of these key sectors have evolved in some of the European

countries considered here It is amazing to see the poor results registered by IT sectors in

United Kingdom France and Ireland compared to the other countries mentioned Germany

in particular This evolution could be behind the stronger effects of delocation in France

and the United Kingdom although it must have produced similar results in the case of

Ireland Furthermore on such a basis what has happened in Italy would remain

7 For details see Myro et al (2008) 8 See Aacutelvarez Loacutepez et al (2007)

17

Delocation in the manufacturing sectors in the EU A regional overview

unexplained as the selected key sectors do not show a bad performance in terms of exports

in the latter country

One reason for such contradictory results lies in the possibility that production

instead of export reflects accurately what has been happening in every sector in terms of

activity However Aacutelvarez et al (2007) show very similar patterns in production and

export specialization from 1995 to 2005 except in the case of Ireland which thus appears

as a very particular one This is because IT specialization in production increases at the

same time as it decreases in exports Perhaps the strong growth in domestic demand

prevents IT exports from growing The opposite case could be that of Italy where slow

growth in domestic demand might have been pushing up IT export9

Anyway further developments on the determinants of delocation patterns require the

preparation of disaggregated data on production a task beyond the objectives of this paper

That could allow us to update the available manufacturing specialization analysis As the

two countries exhibiting higher delocation effects France and the United Kingdom

showed the lowest rates of specialization compared with the EU (Milderfart and Overman

2002 Combes and Overman 2003) something could have changed in the last few years

which would merit a fresh look based on more recent data

On the other hand it is now clear that insofar as delocation in manufacturing

activities has affected each country in a different way a relocation process inside the EU

and between countries has taken place This process deserved to be deeply analyzed

including a perspective by sectors updating what we know about changes in the

distribution of manufacturing industry in the EU and offering the possibility of exploring

whether the European Common Policy has influenced them in some way as was already

suggested by Milderfart-Knarvik and Overman (2002)

9 In Finland IT activities grows much more in production than in exports but in both aspects a positive growth is registered

18

Delocation in the manufacturing sectors in the EU A regional overview

19

Figure 3 Employment growth rates in selected regions 2001-2005 manufacturing and total economy (percentages)

Ukg3

UKi2

UKd4

UKj1

UKm2

UKi1 UKj3

UKd3

NL 41

NL22

FR23

NL33

NL32

DEa1

FR42

DEa5

DK01

FI18

ITe1

FR 24FR10

FI19

SE11

ITd5FR52

DE 14

DE71

DEa2

BE23

DE22

DE21

DE27

IC c4

DE60DE 12

DE25

DE13

DE9 1SE12

ITd3

IE02

DE73

BE21

DE92DEa4

DEf0

DEb3

ES30

DEc0

GR30

DE23

DE11

DK04

AT31

ITe3

SE22

FR82

FR51FR61

SE2 3

DE26

DE24

IT e4

FR62

FR71

ES51

ES2 1

ITe1

AT22

UKh1

UKk1

UKf1

Uke4

UKm3

UKf2

-20

-15

-10

-5

0

5

10

15

20

25

-40 -30 -20 -10 0 10 20 30

Manufacturing

Source Elaborated from Eurostat and National Statistical offices

Tota

l em

ploy

men

t

Delocation in the manufacturing sectors in the EU A regional overview

One last question that deserves consideration is the effect of industrial delocation

process on the regional economies Surprisingly manufacturing job losses have not been

translated into total employment which exhibit positive growth for most of the chosen

regions with the exceptions of some Germans and all the Netherlands (Figure 3) That

means the regions have been able to compensate the negative effect on the employment

from shrinking manufacturing allocating more resources inside the services sectors In fact

a substitution of manufacturing by services seems to have happened as are the region

showing more delocation in manufacturing those that increase in a higher measure the

employment in the service sectors that is the French British and Italian However the

relation is not accurate as there are not a strict quantitative correlation region by region

between rates of delocation and increase in the employment in the service activities

These good results in terms of general employment shown by the regions affected in

a higher measure by manufacturing delocation suggest the convenience of a more general

analysis of regional economic growth beyond the manufacturing sector as some regions

strongly shrinking manufacturing activities in favour of those of services might be leading

an interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

5- Final remarks

Comparative analysis of location patterns for a large sample of European regions -

characterized by having an upper-middle level income and a strong industrial sector-

during the last few years has revealed that only half of them have shown locational

attractions for manufacturing activities The regional patterns found clearly show the

influence of `national effectacute the regions showing location effects corresponding to most

of those located in Austria Germany the Netherlands Ireland Finland Sweden Spain and

Greece

On the contrary the delocation process has affected another half of the regions

included most of them French British and Italian Therefore delocation has contributed to

concentrating the European manufacturing industry in the areas with higher market

potential in the centre of continental Europe It has contributed furthermore to relocating

manufacturing activities inside every country something that requires further analysis

from future research

20

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

2- Measuring delocation

As was posed in the introduction following Baldwin and Robert-Nicoud (2000)

delocation of activities has to be assessed in terms of losses of productive activity within

each region The variables which approximate these losses are gross added value and

number of jobs or total employment The first one is the most suitable since the latter is

dependent on advances registered in labour productivity mainly in the face of strong

competitive pressures forcing companies to achieve greater efficiency gains

In spite of that jobs continue to be a useful measure since its calculation is simpler

and is often more reliable In addition this indicator has received greater attention and is

more easily interpreted by the analysts and especially the social partners

Nevertheless if evolution of jobs is chosen as a measure of delocation it is necessary

to discount the effect on it of an increase in labour productivity (which reduces the need for

labour) and of the economic cycle (which may reduce or expand the existing employment)

The remaining reduction in the number of jobs measures the delocation effect on

employment

Furthermore when average values for quite a long time interval are taken it is

possible to ignore almost completely the impact of the cycle that otherwise would be

estimated using econometric techniques to isolate the trend In this way the delocation

effect can be just approximated after deducting the change in jobs caused by the increase in

labour productivity

Therefore the change in total employment may be split up into two effects one of

them due to the increased of labour productivity and the other to a location effect -

delocation if it is negative- Box 1 shows that decomposition

In fact the location effect reflects the impact on jobs of an increase or decrease in

value added estimated through the primitive rate of labour productivity and as has been

mentioned above may be positive (location) or negative (delocation) In the first case an

expansion in activity has taken place while in the second a reduction of the productive

scope has occurred

4

Delocation in the manufacturing sectors in the EU A regional overview

∆ N = GVA N(0) ⎯ π N(0)

Delocalisation ProductivityEffect Effect

Box 1 GVA= Labor Productivity Employment GVA = π N where π represents productivity and N employment

N = GVA - π

∆ N = N N(0)

N = GVAπ

where a hat over the variable denotes its rate of change

and and and

and

and and

Delocation of activities that can result from this calculation is compatible with the

absence of off-shoring companies and it may be due as noted above to the closure of

establishments as a result of their being uncompetitive In the same way the location of

activities in a territory is compatible with off-shoring of companies

Thus analysis of delocation does not necessarily tell us much about the importance

of firmsrsquo off-shoring a matter that has to be studied on a different basis However when

off-shoring reaches a large-scale dimension it affects the extension of manufacturing

industries

3- Database

As was said before to achieve aggregate data on manufacturing activities by regions

even just those of real GVA and total employment has demanded a laborious task of

collecting information provided by Eurostat (REGIO database) and the National Statistical

Offices of member countries to which the different regions belong as well as the

estimation in most cases of the product in real terms

Looking first at the GVA since the data offered by REGIO include those related to

energy and mining only access to the National Statistics of every member country has

allowed us to isolate the manufacturing sectors Nevertheless additional work has been

applied to transform the data of GVA into real values as only Germany Spain Finland

5

Delocation in the manufacturing sectors in the EU A regional overview

Holland and Italy provide such information or at least indexes of volume In the rest of the

countries real values of GVA have been estimated by applying the national price deflators

to the regional series at current values

As regards total employment significant discrepancies between the data published

by Eurostat and that country itself (particularly in the United Kingdom) are found as well

as the gaps found in several NUTS 2 of some Member States such as in Germany forced

to complete regional series and correct such deficiencies with the help of the information

coming from the National Statistics Offices

To sum up most of the information used (especially production data) comes from

National Statistical Offices but this is commonly close to the Eurostat database as this is

built mainly with data coming from the regional accounts of each country

By the way one of the problems arising from taking the National Statistics as the

main source of data is that the time intervals for which they provide information do not

always coincide That explains that despite having more recent data from some areas the

analysis must conclude in 2005

4- Delocation of manufacturing industries in European regions

As noted above in the introduction the group of regions examined reached just over

64 of EU-15 industrial GVA and about 60 of jobs in 2005 figures slightly lower than

in 2000 Thus despite nearly half of regionsrsquo industrial output at constant prices having

achieved positive growth rates the group accumulated throughout the period a decrease of

about 1 similar to EU-15 average4 Regarding employment the evolution has been even

more negative Throughout the first five years of the current century destruction of jobs in

the aggregate industry has been a common feature in the vast majority (90) of the

seventy-five NUTS 2 examined as well as for each of the Member States where they are

located except Spain accumulating the total sample a decline of close to 8

4 It must be noted that the EU average has been calculated from the aggregate industrial GVA at constant prices of thirteen countries included in the sample Thus Portugal and Luxembourg have been excluded and the values from other countries have been estimated by adding figures of all their regions So the above mentioned growth rate differs from that provided by Eurostat exhibiting a positive increase of 46

6

Delocation in the manufacturing sectors in the EU A regional overview

Focusing on employment the most dynamic areas are located mostly in France

Spain Austria and Italy while the Netherlands and especially the British regions exhibit

the largest job losses (Table 1)

Registered unemployment seems to be related to good performance in labour

productivity pointing to the fierce competition faced by the manufacturing sector in the

analyzed period Thus higher job losses are found in regions with stronger labour

productivity increases although some French and most of the British regions escaped from

this rule as large losses in employment are not accompanied by significant gains in labour

productivity suggesting delocation effects of a particular intensity

That reveals a very different growth path in employment and labour productivity by

regions differences which can be explored in a more detail to isolate the attractive regions

to locate manufacturing activities from the others characterized by a delocation process

To go deeply into this latter question Table 1 contains the results of estimating the

impact on total employment of location and delocation forces in the manufacturing sectors

across the seventy-five regions included in the sample over the period 2001-2005 by using

the procedure proposed in the second section that is discounting from the total change in

jobs the reduction corresponding to the increase in labour productivity Therefore the

location effect shows the effect on employment of GVA change calculated by applying the

initial rate of labour productivity

In about half of the regions with job losses all of those considered in the

Netherlands Finland Greece and Ireland three-quarters of the Germans two British two

Swedish and Madrid in Spain the decline in employment has been due entirely to

increased labour productivity so a positive effect of location appears This behaviour

extends to Finnish German and Swedish economies as a whole as well as Greece Ireland

and Austria Therefore all those regions have proved attractive for the location of

manufacturing firms

7

Delocation in the manufacturing sectors in the EU A regional overview

8

Productivity Location Productivity Locationeffect effect effect effect

at22 Steiermark 372 -108 480 36 -11 47at31 Oberoumlsterreich 097 -837 934 06 -56 62

Austria -1104 -2649 1545 -18 -42 25be21 Prov Antwerpen -1010 079 -1089 -70 05 -75be23 Prov Oost-Vlaanderen -750 -627 -123 -74 -61 -12

Belgium -5740 -2219 -3521 -84 -32 -51de11 Stuttgart -3157 -8244 5087 -48 -125 77de12 Karlsruhe -2036 -3585 1550 -57 -101 44de13 Freiburg -1291 -1301 011 -45 -45 00de14 Tuumlbingen -585 -1805 1219 -23 -72 49de21 Oberbayern -1630 -7976 6346 -37 -179 142de22 Niederbayern -490 -3270 2781 -33 -218 186de23 Oberpfalz -538 -1716 1178 -38 -120 83de24 Oberfranken -1957 -3131 1175 -116 -186 70de25 Mittelfranken -1184 -1770 586 -53 -80 26de26 Unterfranken -819 -3157 2338 -49 -189 140de27 Schwaben -932 -3335 2403 -41 -148 106de60 Hamburg -760 -1222 463 -60 -97 37de71 Darmstadt -5343 -4720 -623 -149 -132 -17de73 Kassel -1059 -1396 336 -78 -103 25de91 Braunschweig -947 -500 -447 -47 -25 -22de92 Hannover -1900 -1544 -356 -106 -86 -20dea1 Duumlsseldorf -6272 -8787 2515 -126 -177 51dea2 Koumlln -5428 -4971 -457 -150 -137 -13dea4 Detmold -3026 -3112 086 -107 -110 03dea5 Arnsberg -5824 -5502 -322 -129 -122 -07deb3 Rheinhessen-Pfalz -2099 -3016 917 -106 -152 46dec0 Saarland -360 -1810 1450 -32 -160 128def0 Schleswig-Holstein -1867 -2485 617 -101 -135 33

Germany -60300 -112275 51975 -74 -138 64dk01 Hovedstaden -1300 -1216 -084 -134 -125 -09dk04 Midtjylland -2000 -1622 -378 -146 -118 -28

Denmark -6100 -3428 -2672 -130 -73 -57es21 Pais Vasco 1840 -331 2171 74 -13 87es30 Comunidad de Madrid -1060 -2180 1120 -31 -63 32es51 Cataluntildea 660 -1914 2574 09 -25 34

Spain 13970 -3927 17897 47 -13 60fi18 Etelauml-Suomi -1807 -7159 5352 -86 -339 253fi19 Laumlnsi-Suomi -701 -3886 3185 -54 -297 244

Finland -1798 -12006 10208 -43 -286 243fr10 Icircle de France 12652 17868 -5216 218 307 -90fr23 Haute-Normandie -2418 -1093 -1325 -171 -77 -94fr24 Centre -2016 -120 -1896 -106 -06 -100fr42 Alsace -2489 -988 -1501 -154 -61 -93fr51 Pays de la Loire -217 1794 -2011 -08 67 -75fr52 Bretagne -415 -215 -200 -23 -12 -11fr61 Aquitaine -199 742 -941 -14 50 -64fr62 Midi-Pyreacuteneacutees 209 3318 -3109 15 245 -229fr71 Rhocircne-Alpes -2919 2759 -5678 -61 57 -118fr82 Provence-Alpes-Cocircte dAzur -097 262 -359 -06 18 -24

France -15306 31566 -46872 -40 83 -123gr30 Attiki -360 -5072 4712 -17 -234 217

Greece -730 -11989 11259 -14 -222 208ie02 Southern and Eastern -1840 -9985 8145 -78 -426 347

Ireland -2420 -12274 9854 -76 -387 311itc1 Piemonte -6869 2539 -9407 -129 48 -177itc4 Lombardia -7940 -7682 -258 -63 -61 -02itd3 Veneto -3967 -211 -3756 -62 -03 -58itd5 Emilia-Romagna -860 -840 -020 -16 -16 00ite1 Toscana -3620 078 -3697 -98 02 -101ite3 Marche 164 -555 719 08 -28 36ite4 Lazio -733 956 -1689 -36 47 -82

Italy -21761 6310 -28071 -45 13 -58nl22 Gelderland -2164 -2506 341 -178 -206 28nl32 Noord-Holland -1555 -2034 479 -138 -180 42nl33 Zuid-Holland -1734 -3980 2246 -129 -296 167nl41 Noord-Brabant -3539 -5515 1976 -173 -269 96

Netherlands -14767 -24690 9923 -161 -270 108se11 Stockholm -1570 -4183 2613 -136 -363 227se12 Oumlstra Mellansverige -2170 -2387 217 -151 -166 15se22 Sydsverige -1010 -690 -320 -91 -62 -29se23 Vaumlstsverige -460 134 -594 -26 07 -33

Sweden -7500 -9400 1900 -95 -119 24ukd3 Greater Manchester -4900 -1907 -2993 -232 -90 -142ukd4 Lancashire -4600 -5082 482 -317 -350 33uke4 West Yorkshire -3700 -1192 -2508 -186 -60 -126ukf1 Derbyshire and Nottinghamshire -3700 -4965 1265 -179 -240 61ukf2 Leicestershire Rutland and Northan -4400 -3607 -793 -233 -191 -42ukg3 West Midlands -8700 -4999 -3701 -313 -180 -133ukh1 East Anglia -2500 -1292 -1208 -137 -71 -66uki1 Inner London -1919 -1381 -538 -142 -102 -40uki2 Outer London -4481 1106 -5587 -271 67 -338ukj1 Berkshire Bucks and Oxfordshire -4600 -3898 -702 -246 -208 -38ukj3 Hampshire and Isle of Wight -1100 -689 -411 -81 -51 -30ukk1 Gloucestershire Wiltshire and Brist -3400 -1253 -2147 -169 -62 -107ukm2 Eastern Scotland -1100 1402 -2502 -91 116 -207ukm3 South Western Scotland -3400 -2107 -1293 -207 -128 -79

United Kingdom -89800 -48574 -41226 -196 -106 -90

Table 1 Importance of manufacturing delocation in European regions 2001-2005 (Decomposition of change in employment)

Total effect

Source Elaborated from Eurostat and National Statistical offices

Number of jobs (thousands)

Total effect

Percent share of 2000 employmentRegions

Delocation in the manufacturing sectors in the EU A regional overview

By contrast thirty-eight remaining regions that also show reductions in

manufacturing employment have been affected by delocation processes to different

extents although only in a small number of them have they been accompanied by

decreases in labour productivity which could make the prospects for economic growth

tougher (four from France Piamonte Toscana and Lazio in Italy Outer London and

Eastern Scotland in the United Kingdom the Swedish Vaumlstsverige and Antwerp in

Belgium)

Regions with positive location effects are shown in Map 1 Most of them are large

industrial regions and are located in the traditional industrial growth centres5 those

characterized by high market potential (Combes and Overman 2003) although there are

also some intermediate and peripheral territories mainly located in the north of Italy

United Kingdom and Spain and in the south of Sweden and Norway This means that the

delocation process has contributed to concentrating the European manufacturing industries

in that privileged area of faster growth apart from some other changes inside every

country

To compare now the intensity of such effects of location and delocation across

regions avoiding the differences they have in size we built a relative measure of them the

ratio of job losses in the period to the volume of employment in the first year 2000 (Table

1)

Following this ratio delocation effects are revealed to be particularly strong in

France United Kingdom Italy and in one region of Belgium with declines in

employment exceeding 20 in some of them

Conversely the regions of Finland and several of the German and the Dutch ones

exhibit strong location effects

In any case as was pointed above the importance of location effects seems to lie in

labour productivity gains In fact higher increases in productivity (over 15) are found

precisely in those regions of Ireland Finland Sweden Netherlands and Germany besides

the Greek Attiki where the location effect has in general reached its greatest extent

(Figure 1)

5 Note that the map of winning regions is clearly influenced by the large number (30) of German NUTS 2 in the sample

9

Delocation in the manufacturing sectors in the EU A regional overview

fi19

fi18

ie02

dec0

gr30

se11

nl33

de21

de22

de26

de25

de12

de14

de11

dea4

dea1de73

def0

de60

deb3

de27

de23

itd5

ite3

se12

nl22

es 30

ukd4

es51

nl32

at22

ukf1

at31

es21

Map 1 Regions with positive location effect ( of 2000 employment)

Above 10 0 to 10

nl41

de13de27

Nevertheless productivity also grows faster in most of the regions belonging to the

United Kingdom Belgium and Denmark in spite of the strong delocation effects registered

10

Delocation in the manufacturing sectors in the EU A regional overview

which disturbs the above-mentioned positive relationship suggesting the harsh process of

industrial restructuring involved

Figure 1 Location effect and productivity growth in select regions manufacturing sector 2001-2005 (percentages)

UKi2

EU 15

FR6 2

UKm2

ITc1

UKg3UKd 3

UKe4

UKk1FR71F R

FR10UKm3

UKj1UKf2

UKd 4

UKi1

D K

DK01

UKh1U K

FR2 3FR4 2

FR24ITe1

BE2 1ITe4FR51

FR61

ITITd 3

SE2 2FR8 2

SE23

NL2 2

UKf1DE2 4

NL3 2DEa1

DEf0

UKj3DE9 1 DE92

DEa2DE71DK0 4

S EDE25

DE6 0DE12

A T

DE14

ITe3

AT2 2

AT3 1ESD E

DE 11DE2 3

NL41N LDE2 7

DEc0

NL3 3

DE2 6DE2 1

ES3 0ES51

SE11

DE2 2

ES21

GR

FI19F I

FI18

IE

GR3 0

B E

DE73

DEb 3

IE0 2

ITd 5 DE13SE12

FR52ITc4BE2 3

DEa4DEa5

y = 07x - 59R2 = 06

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

40

-35 -30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30 35 40 45

Productivity grothw Source Elaborated from Eurostat and National Statistical offices

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

11

Delocation in the manufacturing sectors in the EU A regional overview

In contrast the productivity decline in half of the French and Italian regions albeit

with a positive impact on employment is largely offset by destruction of industrial

activity except in Icircle de France and Midi -Pyreacuteneacutees the two territories in which

productivity has a more negative trajectory

These results closely match those reported by Cuadrado Roura et al (2000) based on

the growth rates for regional productivity and employment during the period 1980-1993

grouping European regions in four typologies comparing the growth rates of these two

variables with their respective European averages

According to this approach and including the location effect in the analysis the

group of regions that are examined in this work would be categorized as follows (Table 2)

The first group includes areas with productivity and employment growth rates above

sample average and show positive location effect (Quadrant 1) So as has been underlined

in that work possibly this reflects the fact that the technological and organizational

changes introduced during the period would have been successful promoting the

diversification and specialization in activities with high growth levels and attracting new

investments which are less labour-intensive

In a second stage are those regions that as the above mentioned authors suggest

have implemented restructuring processes to eliminate the most inefficient outputs

achieving higher levels of productivity and in about half of them increased industrial

production at the expense of a notable decline in employment (Quadrant 4) They named

this typology `restructuring via productivityacute

With a few exceptions (those where the location effect is negative) the regions in

both groups are the winners of industrial dynamic during the period Most of them are

large industrial regions and are located in the traditional industrial growth centres As

demonstrated above they are mainly Finish German and Dutch as well as Greek and Irish

regions

In contrast another large group of regions due to different factors6 have opted for

more labour-intensive industrial models which in a few cases (especially in Spanish and

Austrian regions) have been accompanied by output increases so they can also be included

6 See Cuadrado et al (2000)

12

Delocation in the manufacturing sectors in the EU A regional overview

in the winners (Map 1) while in other the result has been a loss of productive activity

(Quadrant 2)

(1) (2)Employment Productivity Employment Productivity

de11 Stuttgart -48 125 at22 Steiermark 36 11de12 Karlsruhe -57 101 at31 Oberoumlsterreich 06 56de21 Oberbayern -37 179 be21 Prov Antwerpen -70 -05de22 Niederbayern -33 218 be23 Prov Oost-Vlaanderen -74 61de23 Oberpfalz -38 120 de13 Freiburg -45 45de25 Mittelfranken -53 80 de14 Tuumlbingen -23 72de26 Unterfranken -49 189 de91 Braunschweig -47 25de27 Schwaben -41 148 es21 Pais Vasco 74 13de60 Hamburg -60 97 es30 Comunidad de Madrid -31 63dec0 Saarland -32 160 es51 Cataluntildea 09 25fi19 Laumlnsi-Suomi -54 297 fr10 Icircle de France 218 -307gr30 Attiki -17 234 fr51 Pays de la Loire -08 -67

fr52 Bretagne -23 12fr61 Aquitaine -14 -50fr62 Midi-Pyreacuteneacutees 15 -245fr71 Rhocircne-Alpes -61 -57fr82 Provence-Alpes-Cocircte dAzur -06 -18itc4 Lombardia -63 61itd3 Veneto -62 03itd5 Emilia-Romagna -16 16ite3 Marche 08 28ite4 Lazio -36 -47se23 Vaumlstsverige -26 -07

(4) (3)Employment Productivity Employment Productivity

de24 Oberfranken -116 186 fr24 Centre -106 06de71 Darmstadt -149 132 fr42 Alsace -154 61de73 Kassel -78 103 itc1 Piemonte -129 -48de92 Hannover -106 86 ite1 Toscana -98 -02dea1 Duumlsseldorf -126 177 se22 Sydsverige -91 62dea2 Koumlln -150 137 uke4 West Yorkshire -186 60dea4 Detmold -107 110 ukh1 East Anglia -137 71dea5 Arnsberg -129 122 uki2 Outer London -271 -67deb3 Rheinhessen-Pfalz -106 152 ukj3 Hampshire and Isle of Wight -81 51def0 Schleswig-Holstein -101 135 ukk1 Gloucestershire Wiltshire -169 62dk01 Hovedstaden -134 125 ukm2 Eastern Scotland -91 -116dk04 Midtjylland -146 118fi18 Etelauml-Suomi -86 339fr23 Haute-Normandie -171 77ie02 Southern and Eastern -78 426nl22 Gelderland -178 206nl32 Noord-Holland -138 180nl33 Zuid-Holland -129 296nl41 Noord-Brabant -173 269se11 Stockholm -136 363se12 Oumlstra Mellansverige -151 166ukd3 Greater Manchester -232 90ukd4 Lancashire -317 350ukf1 Derbyshire and Nottinghamshire -179 240ukf2 Leicestershire Rutland and Northants -233 191ukg3 West Midlands -313 180uki1 Inner London -142 102ukj1 Berkshire Bucks and Oxfordshire -246 208ukm3 South Western Scotland -207 128

Source Elaborated from Eurostat and national Statistical officesNote Shaded areas indicate regions with positive location effect

Em

ploy

men

t gro

wth

lt A

vera

ge v

alue

Productivity growth gt Average value Productivity growth lt Average value

Em

ploy

men

t gro

wth

gt A

vera

ge v

alue

Table 2 Employment and productivity growth rates in select regions manufacturing sector 2001-2005 (Averages of all regions +74 productivity -77 employment)

13

Delocation in the manufacturing sectors in the EU A regional overview

Finally the worst results can be seen in a fourth type of region with growth rates of

GVA and productivity below average and serious job destruction (Quadrant 3) This would

occur in regions with problems in restructuring their manufacturing sector or with low

levels of activity So together with those that are either restructuring via productivity or

via employment but without achieving a positive location effect they are clearly the

losers These include mainly British French Italian Belgian and Danish regions

To sum up these results confirm the direct relationship between growth in labour

efficiency and progress in the creation of productive industrial activity as pointed out

above and which can be more clearly by looking at Figure 1 In other words regions with

higher levels of productivity growth in general show a greater attractiveness for new

manufacturing firms

These differences in the behaviour of productivity and employment across regions

have led to remarkable disparities in rates of change of their industrial output during the

period Thus while a few increased their real manufacturing GVA by more than 20 in

the less dynamic ones the loss of industrial output is between a sixth and a third of the

initial value So significant changes had taken place in the relative position of some of

them (Figure 2)

Growth is led by a group of twenty regions located in Ireland Finland Greece

Germany Sweden Netherlands Spain and Austria with real GVA growth rates that are

more than five points above average As a result they increased their share in the industrial

output of the sample Looking at the complete list we see that the growth deviation from

the mean exhibits positive values in just over half of the regions examined among them

seven of the ten with the largest industrial scale in 2005 Thus the top ten regions have

increased their share from 32 to 33 percent in the last five years

Conversely among the less dynamic regions are found a great number of British and

some French and Italian ones which consequently have fallen back in the ranking of

NUTS 2 arranged by their contribution to aggregate manufacturing output

Between both groups there are approximately twenty regions mainly in Germany

with growth rates not far from average so their positions have hardly changed during the

last five years

14

Delocation in the manufacturing sectors in the EU A regional overview

Figure 2 Regional shares in the total sample manufacturing GVA(percentage at 2000 prices)

ITc4

UKf2

UKm3

Uke4

UKf1

UKk1

UKh1

AT22

ITe1

ES21

ES51

FR71

FR62

IT e4

DE24

DE26

SE23

FR61

FR51

FR82

SE22

ITe3

AT31

DK04

DE11

DE23

GR30

DEc0

ES30DEb3DEf0

DEa4

DE92

BE21

DE73

IE02

ITd3

FR52

DE91

DE13DE25

DE 12DE60

DE27

DE21

DE22

BE23DEa2

DE71

DE 14

SE12

ITd5

SE11

FI19

FR10

FR 24

ITe1

FI18

DK01 DEa5

FR4 2

DEa1NL32

NL3 3

FR2 3

NL22

NL 41

UKd3

UKj3

UKi1

UKm2

UKj1

UKd4

UKi2

Ukg3

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

Share in 2005

Source Elaborated from Eurostat and National Statistical offices

Rat

e of

cha

nge

200

1-20

05

70403530252015100500

Finally when the question is analyzed at Member State level (Table 1) it is possible

to check out the disparities among the above mentioned regions Thus while in some

15

Delocation in the manufacturing sectors in the EU A regional overview

countries such as Spain manufacturing growth has been clearly labour-intensive

something which can be extended to its regions as also happens in Greece and Austria

which both boast positive location effect and a job destruction rate far below the average

in others such as Holland Sweden and to a lesser extent Belgium Ireland and Finland

increasing of industrial activity is only due by advances in productivity so has been

accompanied by heavy job losses and these are patterns repeated in most of their regions

with noticeable relevance in the Netherlands As for the United Kingdom Belgium and

Denmark all share declining industrial production together with strong job cuts a pattern

that is replicated in many of their regions In this situation are the bulk of French and

Italian NUTS 2 just like the countries to which they belong although in this case job

destruction has been much less pronounced and below the group average Close to this

mean is Germany as well as most of their regions

Thus the analysis at Member States level also confirms the positive relationship

between the advance of industrial productivity and the attractiveness for new

manufacturing firms As demonstrated for their regions Ireland Finland Netherlands and

Greece following at quite distance from Germany are Member States showing a better

performance in productivity and with high location effects (see Figure 1)

In fact similarities between performances in the regions and in the Member States

they belong to suggest that the trend in regional industrial employment is to some extent

influenced by belonging to the same country in other words the locational attractions of

each area and its effort to improve productivity is partly due to national specificities or

idiosyncratic factors In fact as shown in Appendix 3 most European industrialized

regions follow the location patterns of the country they belong to reproducing its

productivity behaviour too In summary the evolution of these variables in each region

seems to have a clear `country effectacute that is different behaviours motivated by specific

factors andor differentiated industrial policies

For example in order to correctly interpret the results for Spanish regions it can not

be overlooked that during the period analyzed as in the whole country regions have

absorbed large numbers of immigrants which has boosted the proliferation of lower

productivity and lower wages in industrial activities However it may be thought that this

phenomenon has tended to conceal productivity increases in larger and better equipped

16

Delocation in the manufacturing sectors in the EU A regional overview

firms which could have been significant in the light of their results7 On the other hand

countries like Finland or Ireland with the lowest population increase have opted for high-

tech manufacturing sectors as a way to improve national competitiveness8 Nonetheless

this positive location effect has not reached in these countriesrsquo NUTS 2 enough magnitude

to offset completely the job losses arising from productivity improvements

On the other hand this `national effectacute has also played an important role in

explaining the economic outcomes of European regions in terms of per capita GDP in

accordance with the results obtained by Cuadrado-Roura (2001) using the data available

for 109 EU regions over the 1977-1994 period

Table 3 Export growth in key sectors by countries 2000-2005 (percentages)Impact ITon Total

Spain 65 91 72 158 38 111 04Germany 116 138 61 111 111 130 14United Kingdom 09 85 29 32 23 61 02France -09 102 25 94 118 78 -01Ireland -09 -102 01 25 -34 77 -03Finland 47 152 45 49 56 73 13Italy 40 88 41 90 56 89 02EU-15 70 141 59 91 67 111 11Source Elaborated from Comtrade

Shoes TotalIT Auto Textile Clothes

This importance of the `national effectacute should allow us to make a better approach to

the determinants of above-mentioned delocation patterns in future research One of such

determinants might be found in the different importance and performance by countries of

some manufacturing branches particularly affected by the increase in global competition as

their development has being receiving strong support in emerging countries from both

domestic policies and the location of big multinational companies This is the case of IT

manufacturing sectors or that of automobiles clothes textiles and shoes (Sachwald 2004)

Table 3 shows how the exports of these key sectors have evolved in some of the European

countries considered here It is amazing to see the poor results registered by IT sectors in

United Kingdom France and Ireland compared to the other countries mentioned Germany

in particular This evolution could be behind the stronger effects of delocation in France

and the United Kingdom although it must have produced similar results in the case of

Ireland Furthermore on such a basis what has happened in Italy would remain

7 For details see Myro et al (2008) 8 See Aacutelvarez Loacutepez et al (2007)

17

Delocation in the manufacturing sectors in the EU A regional overview

unexplained as the selected key sectors do not show a bad performance in terms of exports

in the latter country

One reason for such contradictory results lies in the possibility that production

instead of export reflects accurately what has been happening in every sector in terms of

activity However Aacutelvarez et al (2007) show very similar patterns in production and

export specialization from 1995 to 2005 except in the case of Ireland which thus appears

as a very particular one This is because IT specialization in production increases at the

same time as it decreases in exports Perhaps the strong growth in domestic demand

prevents IT exports from growing The opposite case could be that of Italy where slow

growth in domestic demand might have been pushing up IT export9

Anyway further developments on the determinants of delocation patterns require the

preparation of disaggregated data on production a task beyond the objectives of this paper

That could allow us to update the available manufacturing specialization analysis As the

two countries exhibiting higher delocation effects France and the United Kingdom

showed the lowest rates of specialization compared with the EU (Milderfart and Overman

2002 Combes and Overman 2003) something could have changed in the last few years

which would merit a fresh look based on more recent data

On the other hand it is now clear that insofar as delocation in manufacturing

activities has affected each country in a different way a relocation process inside the EU

and between countries has taken place This process deserved to be deeply analyzed

including a perspective by sectors updating what we know about changes in the

distribution of manufacturing industry in the EU and offering the possibility of exploring

whether the European Common Policy has influenced them in some way as was already

suggested by Milderfart-Knarvik and Overman (2002)

9 In Finland IT activities grows much more in production than in exports but in both aspects a positive growth is registered

18

Delocation in the manufacturing sectors in the EU A regional overview

19

Figure 3 Employment growth rates in selected regions 2001-2005 manufacturing and total economy (percentages)

Ukg3

UKi2

UKd4

UKj1

UKm2

UKi1 UKj3

UKd3

NL 41

NL22

FR23

NL33

NL32

DEa1

FR42

DEa5

DK01

FI18

ITe1

FR 24FR10

FI19

SE11

ITd5FR52

DE 14

DE71

DEa2

BE23

DE22

DE21

DE27

IC c4

DE60DE 12

DE25

DE13

DE9 1SE12

ITd3

IE02

DE73

BE21

DE92DEa4

DEf0

DEb3

ES30

DEc0

GR30

DE23

DE11

DK04

AT31

ITe3

SE22

FR82

FR51FR61

SE2 3

DE26

DE24

IT e4

FR62

FR71

ES51

ES2 1

ITe1

AT22

UKh1

UKk1

UKf1

Uke4

UKm3

UKf2

-20

-15

-10

-5

0

5

10

15

20

25

-40 -30 -20 -10 0 10 20 30

Manufacturing

Source Elaborated from Eurostat and National Statistical offices

Tota

l em

ploy

men

t

Delocation in the manufacturing sectors in the EU A regional overview

One last question that deserves consideration is the effect of industrial delocation

process on the regional economies Surprisingly manufacturing job losses have not been

translated into total employment which exhibit positive growth for most of the chosen

regions with the exceptions of some Germans and all the Netherlands (Figure 3) That

means the regions have been able to compensate the negative effect on the employment

from shrinking manufacturing allocating more resources inside the services sectors In fact

a substitution of manufacturing by services seems to have happened as are the region

showing more delocation in manufacturing those that increase in a higher measure the

employment in the service sectors that is the French British and Italian However the

relation is not accurate as there are not a strict quantitative correlation region by region

between rates of delocation and increase in the employment in the service activities

These good results in terms of general employment shown by the regions affected in

a higher measure by manufacturing delocation suggest the convenience of a more general

analysis of regional economic growth beyond the manufacturing sector as some regions

strongly shrinking manufacturing activities in favour of those of services might be leading

an interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

5- Final remarks

Comparative analysis of location patterns for a large sample of European regions -

characterized by having an upper-middle level income and a strong industrial sector-

during the last few years has revealed that only half of them have shown locational

attractions for manufacturing activities The regional patterns found clearly show the

influence of `national effectacute the regions showing location effects corresponding to most

of those located in Austria Germany the Netherlands Ireland Finland Sweden Spain and

Greece

On the contrary the delocation process has affected another half of the regions

included most of them French British and Italian Therefore delocation has contributed to

concentrating the European manufacturing industry in the areas with higher market

potential in the centre of continental Europe It has contributed furthermore to relocating

manufacturing activities inside every country something that requires further analysis

from future research

20

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

∆ N = GVA N(0) ⎯ π N(0)

Delocalisation ProductivityEffect Effect

Box 1 GVA= Labor Productivity Employment GVA = π N where π represents productivity and N employment

N = GVA - π

∆ N = N N(0)

N = GVAπ

where a hat over the variable denotes its rate of change

and and and

and

and and

Delocation of activities that can result from this calculation is compatible with the

absence of off-shoring companies and it may be due as noted above to the closure of

establishments as a result of their being uncompetitive In the same way the location of

activities in a territory is compatible with off-shoring of companies

Thus analysis of delocation does not necessarily tell us much about the importance

of firmsrsquo off-shoring a matter that has to be studied on a different basis However when

off-shoring reaches a large-scale dimension it affects the extension of manufacturing

industries

3- Database

As was said before to achieve aggregate data on manufacturing activities by regions

even just those of real GVA and total employment has demanded a laborious task of

collecting information provided by Eurostat (REGIO database) and the National Statistical

Offices of member countries to which the different regions belong as well as the

estimation in most cases of the product in real terms

Looking first at the GVA since the data offered by REGIO include those related to

energy and mining only access to the National Statistics of every member country has

allowed us to isolate the manufacturing sectors Nevertheless additional work has been

applied to transform the data of GVA into real values as only Germany Spain Finland

5

Delocation in the manufacturing sectors in the EU A regional overview

Holland and Italy provide such information or at least indexes of volume In the rest of the

countries real values of GVA have been estimated by applying the national price deflators

to the regional series at current values

As regards total employment significant discrepancies between the data published

by Eurostat and that country itself (particularly in the United Kingdom) are found as well

as the gaps found in several NUTS 2 of some Member States such as in Germany forced

to complete regional series and correct such deficiencies with the help of the information

coming from the National Statistics Offices

To sum up most of the information used (especially production data) comes from

National Statistical Offices but this is commonly close to the Eurostat database as this is

built mainly with data coming from the regional accounts of each country

By the way one of the problems arising from taking the National Statistics as the

main source of data is that the time intervals for which they provide information do not

always coincide That explains that despite having more recent data from some areas the

analysis must conclude in 2005

4- Delocation of manufacturing industries in European regions

As noted above in the introduction the group of regions examined reached just over

64 of EU-15 industrial GVA and about 60 of jobs in 2005 figures slightly lower than

in 2000 Thus despite nearly half of regionsrsquo industrial output at constant prices having

achieved positive growth rates the group accumulated throughout the period a decrease of

about 1 similar to EU-15 average4 Regarding employment the evolution has been even

more negative Throughout the first five years of the current century destruction of jobs in

the aggregate industry has been a common feature in the vast majority (90) of the

seventy-five NUTS 2 examined as well as for each of the Member States where they are

located except Spain accumulating the total sample a decline of close to 8

4 It must be noted that the EU average has been calculated from the aggregate industrial GVA at constant prices of thirteen countries included in the sample Thus Portugal and Luxembourg have been excluded and the values from other countries have been estimated by adding figures of all their regions So the above mentioned growth rate differs from that provided by Eurostat exhibiting a positive increase of 46

6

Delocation in the manufacturing sectors in the EU A regional overview

Focusing on employment the most dynamic areas are located mostly in France

Spain Austria and Italy while the Netherlands and especially the British regions exhibit

the largest job losses (Table 1)

Registered unemployment seems to be related to good performance in labour

productivity pointing to the fierce competition faced by the manufacturing sector in the

analyzed period Thus higher job losses are found in regions with stronger labour

productivity increases although some French and most of the British regions escaped from

this rule as large losses in employment are not accompanied by significant gains in labour

productivity suggesting delocation effects of a particular intensity

That reveals a very different growth path in employment and labour productivity by

regions differences which can be explored in a more detail to isolate the attractive regions

to locate manufacturing activities from the others characterized by a delocation process

To go deeply into this latter question Table 1 contains the results of estimating the

impact on total employment of location and delocation forces in the manufacturing sectors

across the seventy-five regions included in the sample over the period 2001-2005 by using

the procedure proposed in the second section that is discounting from the total change in

jobs the reduction corresponding to the increase in labour productivity Therefore the

location effect shows the effect on employment of GVA change calculated by applying the

initial rate of labour productivity

In about half of the regions with job losses all of those considered in the

Netherlands Finland Greece and Ireland three-quarters of the Germans two British two

Swedish and Madrid in Spain the decline in employment has been due entirely to

increased labour productivity so a positive effect of location appears This behaviour

extends to Finnish German and Swedish economies as a whole as well as Greece Ireland

and Austria Therefore all those regions have proved attractive for the location of

manufacturing firms

7

Delocation in the manufacturing sectors in the EU A regional overview

8

Productivity Location Productivity Locationeffect effect effect effect

at22 Steiermark 372 -108 480 36 -11 47at31 Oberoumlsterreich 097 -837 934 06 -56 62

Austria -1104 -2649 1545 -18 -42 25be21 Prov Antwerpen -1010 079 -1089 -70 05 -75be23 Prov Oost-Vlaanderen -750 -627 -123 -74 -61 -12

Belgium -5740 -2219 -3521 -84 -32 -51de11 Stuttgart -3157 -8244 5087 -48 -125 77de12 Karlsruhe -2036 -3585 1550 -57 -101 44de13 Freiburg -1291 -1301 011 -45 -45 00de14 Tuumlbingen -585 -1805 1219 -23 -72 49de21 Oberbayern -1630 -7976 6346 -37 -179 142de22 Niederbayern -490 -3270 2781 -33 -218 186de23 Oberpfalz -538 -1716 1178 -38 -120 83de24 Oberfranken -1957 -3131 1175 -116 -186 70de25 Mittelfranken -1184 -1770 586 -53 -80 26de26 Unterfranken -819 -3157 2338 -49 -189 140de27 Schwaben -932 -3335 2403 -41 -148 106de60 Hamburg -760 -1222 463 -60 -97 37de71 Darmstadt -5343 -4720 -623 -149 -132 -17de73 Kassel -1059 -1396 336 -78 -103 25de91 Braunschweig -947 -500 -447 -47 -25 -22de92 Hannover -1900 -1544 -356 -106 -86 -20dea1 Duumlsseldorf -6272 -8787 2515 -126 -177 51dea2 Koumlln -5428 -4971 -457 -150 -137 -13dea4 Detmold -3026 -3112 086 -107 -110 03dea5 Arnsberg -5824 -5502 -322 -129 -122 -07deb3 Rheinhessen-Pfalz -2099 -3016 917 -106 -152 46dec0 Saarland -360 -1810 1450 -32 -160 128def0 Schleswig-Holstein -1867 -2485 617 -101 -135 33

Germany -60300 -112275 51975 -74 -138 64dk01 Hovedstaden -1300 -1216 -084 -134 -125 -09dk04 Midtjylland -2000 -1622 -378 -146 -118 -28

Denmark -6100 -3428 -2672 -130 -73 -57es21 Pais Vasco 1840 -331 2171 74 -13 87es30 Comunidad de Madrid -1060 -2180 1120 -31 -63 32es51 Cataluntildea 660 -1914 2574 09 -25 34

Spain 13970 -3927 17897 47 -13 60fi18 Etelauml-Suomi -1807 -7159 5352 -86 -339 253fi19 Laumlnsi-Suomi -701 -3886 3185 -54 -297 244

Finland -1798 -12006 10208 -43 -286 243fr10 Icircle de France 12652 17868 -5216 218 307 -90fr23 Haute-Normandie -2418 -1093 -1325 -171 -77 -94fr24 Centre -2016 -120 -1896 -106 -06 -100fr42 Alsace -2489 -988 -1501 -154 -61 -93fr51 Pays de la Loire -217 1794 -2011 -08 67 -75fr52 Bretagne -415 -215 -200 -23 -12 -11fr61 Aquitaine -199 742 -941 -14 50 -64fr62 Midi-Pyreacuteneacutees 209 3318 -3109 15 245 -229fr71 Rhocircne-Alpes -2919 2759 -5678 -61 57 -118fr82 Provence-Alpes-Cocircte dAzur -097 262 -359 -06 18 -24

France -15306 31566 -46872 -40 83 -123gr30 Attiki -360 -5072 4712 -17 -234 217

Greece -730 -11989 11259 -14 -222 208ie02 Southern and Eastern -1840 -9985 8145 -78 -426 347

Ireland -2420 -12274 9854 -76 -387 311itc1 Piemonte -6869 2539 -9407 -129 48 -177itc4 Lombardia -7940 -7682 -258 -63 -61 -02itd3 Veneto -3967 -211 -3756 -62 -03 -58itd5 Emilia-Romagna -860 -840 -020 -16 -16 00ite1 Toscana -3620 078 -3697 -98 02 -101ite3 Marche 164 -555 719 08 -28 36ite4 Lazio -733 956 -1689 -36 47 -82

Italy -21761 6310 -28071 -45 13 -58nl22 Gelderland -2164 -2506 341 -178 -206 28nl32 Noord-Holland -1555 -2034 479 -138 -180 42nl33 Zuid-Holland -1734 -3980 2246 -129 -296 167nl41 Noord-Brabant -3539 -5515 1976 -173 -269 96

Netherlands -14767 -24690 9923 -161 -270 108se11 Stockholm -1570 -4183 2613 -136 -363 227se12 Oumlstra Mellansverige -2170 -2387 217 -151 -166 15se22 Sydsverige -1010 -690 -320 -91 -62 -29se23 Vaumlstsverige -460 134 -594 -26 07 -33

Sweden -7500 -9400 1900 -95 -119 24ukd3 Greater Manchester -4900 -1907 -2993 -232 -90 -142ukd4 Lancashire -4600 -5082 482 -317 -350 33uke4 West Yorkshire -3700 -1192 -2508 -186 -60 -126ukf1 Derbyshire and Nottinghamshire -3700 -4965 1265 -179 -240 61ukf2 Leicestershire Rutland and Northan -4400 -3607 -793 -233 -191 -42ukg3 West Midlands -8700 -4999 -3701 -313 -180 -133ukh1 East Anglia -2500 -1292 -1208 -137 -71 -66uki1 Inner London -1919 -1381 -538 -142 -102 -40uki2 Outer London -4481 1106 -5587 -271 67 -338ukj1 Berkshire Bucks and Oxfordshire -4600 -3898 -702 -246 -208 -38ukj3 Hampshire and Isle of Wight -1100 -689 -411 -81 -51 -30ukk1 Gloucestershire Wiltshire and Brist -3400 -1253 -2147 -169 -62 -107ukm2 Eastern Scotland -1100 1402 -2502 -91 116 -207ukm3 South Western Scotland -3400 -2107 -1293 -207 -128 -79

United Kingdom -89800 -48574 -41226 -196 -106 -90

Table 1 Importance of manufacturing delocation in European regions 2001-2005 (Decomposition of change in employment)

Total effect

Source Elaborated from Eurostat and National Statistical offices

Number of jobs (thousands)

Total effect

Percent share of 2000 employmentRegions

Delocation in the manufacturing sectors in the EU A regional overview

By contrast thirty-eight remaining regions that also show reductions in

manufacturing employment have been affected by delocation processes to different

extents although only in a small number of them have they been accompanied by

decreases in labour productivity which could make the prospects for economic growth

tougher (four from France Piamonte Toscana and Lazio in Italy Outer London and

Eastern Scotland in the United Kingdom the Swedish Vaumlstsverige and Antwerp in

Belgium)

Regions with positive location effects are shown in Map 1 Most of them are large

industrial regions and are located in the traditional industrial growth centres5 those

characterized by high market potential (Combes and Overman 2003) although there are

also some intermediate and peripheral territories mainly located in the north of Italy

United Kingdom and Spain and in the south of Sweden and Norway This means that the

delocation process has contributed to concentrating the European manufacturing industries

in that privileged area of faster growth apart from some other changes inside every

country

To compare now the intensity of such effects of location and delocation across

regions avoiding the differences they have in size we built a relative measure of them the

ratio of job losses in the period to the volume of employment in the first year 2000 (Table

1)

Following this ratio delocation effects are revealed to be particularly strong in

France United Kingdom Italy and in one region of Belgium with declines in

employment exceeding 20 in some of them

Conversely the regions of Finland and several of the German and the Dutch ones

exhibit strong location effects

In any case as was pointed above the importance of location effects seems to lie in

labour productivity gains In fact higher increases in productivity (over 15) are found

precisely in those regions of Ireland Finland Sweden Netherlands and Germany besides

the Greek Attiki where the location effect has in general reached its greatest extent

(Figure 1)

5 Note that the map of winning regions is clearly influenced by the large number (30) of German NUTS 2 in the sample

9

Delocation in the manufacturing sectors in the EU A regional overview

fi19

fi18

ie02

dec0

gr30

se11

nl33

de21

de22

de26

de25

de12

de14

de11

dea4

dea1de73

def0

de60

deb3

de27

de23

itd5

ite3

se12

nl22

es 30

ukd4

es51

nl32

at22

ukf1

at31

es21

Map 1 Regions with positive location effect ( of 2000 employment)

Above 10 0 to 10

nl41

de13de27

Nevertheless productivity also grows faster in most of the regions belonging to the

United Kingdom Belgium and Denmark in spite of the strong delocation effects registered

10

Delocation in the manufacturing sectors in the EU A regional overview

which disturbs the above-mentioned positive relationship suggesting the harsh process of

industrial restructuring involved

Figure 1 Location effect and productivity growth in select regions manufacturing sector 2001-2005 (percentages)

UKi2

EU 15

FR6 2

UKm2

ITc1

UKg3UKd 3

UKe4

UKk1FR71F R

FR10UKm3

UKj1UKf2

UKd 4

UKi1

D K

DK01

UKh1U K

FR2 3FR4 2

FR24ITe1

BE2 1ITe4FR51

FR61

ITITd 3

SE2 2FR8 2

SE23

NL2 2

UKf1DE2 4

NL3 2DEa1

DEf0

UKj3DE9 1 DE92

DEa2DE71DK0 4

S EDE25

DE6 0DE12

A T

DE14

ITe3

AT2 2

AT3 1ESD E

DE 11DE2 3

NL41N LDE2 7

DEc0

NL3 3

DE2 6DE2 1

ES3 0ES51

SE11

DE2 2

ES21

GR

FI19F I

FI18

IE

GR3 0

B E

DE73

DEb 3

IE0 2

ITd 5 DE13SE12

FR52ITc4BE2 3

DEa4DEa5

y = 07x - 59R2 = 06

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

40

-35 -30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30 35 40 45

Productivity grothw Source Elaborated from Eurostat and National Statistical offices

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

11

Delocation in the manufacturing sectors in the EU A regional overview

In contrast the productivity decline in half of the French and Italian regions albeit

with a positive impact on employment is largely offset by destruction of industrial

activity except in Icircle de France and Midi -Pyreacuteneacutees the two territories in which

productivity has a more negative trajectory

These results closely match those reported by Cuadrado Roura et al (2000) based on

the growth rates for regional productivity and employment during the period 1980-1993

grouping European regions in four typologies comparing the growth rates of these two

variables with their respective European averages

According to this approach and including the location effect in the analysis the

group of regions that are examined in this work would be categorized as follows (Table 2)

The first group includes areas with productivity and employment growth rates above

sample average and show positive location effect (Quadrant 1) So as has been underlined

in that work possibly this reflects the fact that the technological and organizational

changes introduced during the period would have been successful promoting the

diversification and specialization in activities with high growth levels and attracting new

investments which are less labour-intensive

In a second stage are those regions that as the above mentioned authors suggest

have implemented restructuring processes to eliminate the most inefficient outputs

achieving higher levels of productivity and in about half of them increased industrial

production at the expense of a notable decline in employment (Quadrant 4) They named

this typology `restructuring via productivityacute

With a few exceptions (those where the location effect is negative) the regions in

both groups are the winners of industrial dynamic during the period Most of them are

large industrial regions and are located in the traditional industrial growth centres As

demonstrated above they are mainly Finish German and Dutch as well as Greek and Irish

regions

In contrast another large group of regions due to different factors6 have opted for

more labour-intensive industrial models which in a few cases (especially in Spanish and

Austrian regions) have been accompanied by output increases so they can also be included

6 See Cuadrado et al (2000)

12

Delocation in the manufacturing sectors in the EU A regional overview

in the winners (Map 1) while in other the result has been a loss of productive activity

(Quadrant 2)

(1) (2)Employment Productivity Employment Productivity

de11 Stuttgart -48 125 at22 Steiermark 36 11de12 Karlsruhe -57 101 at31 Oberoumlsterreich 06 56de21 Oberbayern -37 179 be21 Prov Antwerpen -70 -05de22 Niederbayern -33 218 be23 Prov Oost-Vlaanderen -74 61de23 Oberpfalz -38 120 de13 Freiburg -45 45de25 Mittelfranken -53 80 de14 Tuumlbingen -23 72de26 Unterfranken -49 189 de91 Braunschweig -47 25de27 Schwaben -41 148 es21 Pais Vasco 74 13de60 Hamburg -60 97 es30 Comunidad de Madrid -31 63dec0 Saarland -32 160 es51 Cataluntildea 09 25fi19 Laumlnsi-Suomi -54 297 fr10 Icircle de France 218 -307gr30 Attiki -17 234 fr51 Pays de la Loire -08 -67

fr52 Bretagne -23 12fr61 Aquitaine -14 -50fr62 Midi-Pyreacuteneacutees 15 -245fr71 Rhocircne-Alpes -61 -57fr82 Provence-Alpes-Cocircte dAzur -06 -18itc4 Lombardia -63 61itd3 Veneto -62 03itd5 Emilia-Romagna -16 16ite3 Marche 08 28ite4 Lazio -36 -47se23 Vaumlstsverige -26 -07

(4) (3)Employment Productivity Employment Productivity

de24 Oberfranken -116 186 fr24 Centre -106 06de71 Darmstadt -149 132 fr42 Alsace -154 61de73 Kassel -78 103 itc1 Piemonte -129 -48de92 Hannover -106 86 ite1 Toscana -98 -02dea1 Duumlsseldorf -126 177 se22 Sydsverige -91 62dea2 Koumlln -150 137 uke4 West Yorkshire -186 60dea4 Detmold -107 110 ukh1 East Anglia -137 71dea5 Arnsberg -129 122 uki2 Outer London -271 -67deb3 Rheinhessen-Pfalz -106 152 ukj3 Hampshire and Isle of Wight -81 51def0 Schleswig-Holstein -101 135 ukk1 Gloucestershire Wiltshire -169 62dk01 Hovedstaden -134 125 ukm2 Eastern Scotland -91 -116dk04 Midtjylland -146 118fi18 Etelauml-Suomi -86 339fr23 Haute-Normandie -171 77ie02 Southern and Eastern -78 426nl22 Gelderland -178 206nl32 Noord-Holland -138 180nl33 Zuid-Holland -129 296nl41 Noord-Brabant -173 269se11 Stockholm -136 363se12 Oumlstra Mellansverige -151 166ukd3 Greater Manchester -232 90ukd4 Lancashire -317 350ukf1 Derbyshire and Nottinghamshire -179 240ukf2 Leicestershire Rutland and Northants -233 191ukg3 West Midlands -313 180uki1 Inner London -142 102ukj1 Berkshire Bucks and Oxfordshire -246 208ukm3 South Western Scotland -207 128

Source Elaborated from Eurostat and national Statistical officesNote Shaded areas indicate regions with positive location effect

Em

ploy

men

t gro

wth

lt A

vera

ge v

alue

Productivity growth gt Average value Productivity growth lt Average value

Em

ploy

men

t gro

wth

gt A

vera

ge v

alue

Table 2 Employment and productivity growth rates in select regions manufacturing sector 2001-2005 (Averages of all regions +74 productivity -77 employment)

13

Delocation in the manufacturing sectors in the EU A regional overview

Finally the worst results can be seen in a fourth type of region with growth rates of

GVA and productivity below average and serious job destruction (Quadrant 3) This would

occur in regions with problems in restructuring their manufacturing sector or with low

levels of activity So together with those that are either restructuring via productivity or

via employment but without achieving a positive location effect they are clearly the

losers These include mainly British French Italian Belgian and Danish regions

To sum up these results confirm the direct relationship between growth in labour

efficiency and progress in the creation of productive industrial activity as pointed out

above and which can be more clearly by looking at Figure 1 In other words regions with

higher levels of productivity growth in general show a greater attractiveness for new

manufacturing firms

These differences in the behaviour of productivity and employment across regions

have led to remarkable disparities in rates of change of their industrial output during the

period Thus while a few increased their real manufacturing GVA by more than 20 in

the less dynamic ones the loss of industrial output is between a sixth and a third of the

initial value So significant changes had taken place in the relative position of some of

them (Figure 2)

Growth is led by a group of twenty regions located in Ireland Finland Greece

Germany Sweden Netherlands Spain and Austria with real GVA growth rates that are

more than five points above average As a result they increased their share in the industrial

output of the sample Looking at the complete list we see that the growth deviation from

the mean exhibits positive values in just over half of the regions examined among them

seven of the ten with the largest industrial scale in 2005 Thus the top ten regions have

increased their share from 32 to 33 percent in the last five years

Conversely among the less dynamic regions are found a great number of British and

some French and Italian ones which consequently have fallen back in the ranking of

NUTS 2 arranged by their contribution to aggregate manufacturing output

Between both groups there are approximately twenty regions mainly in Germany

with growth rates not far from average so their positions have hardly changed during the

last five years

14

Delocation in the manufacturing sectors in the EU A regional overview

Figure 2 Regional shares in the total sample manufacturing GVA(percentage at 2000 prices)

ITc4

UKf2

UKm3

Uke4

UKf1

UKk1

UKh1

AT22

ITe1

ES21

ES51

FR71

FR62

IT e4

DE24

DE26

SE23

FR61

FR51

FR82

SE22

ITe3

AT31

DK04

DE11

DE23

GR30

DEc0

ES30DEb3DEf0

DEa4

DE92

BE21

DE73

IE02

ITd3

FR52

DE91

DE13DE25

DE 12DE60

DE27

DE21

DE22

BE23DEa2

DE71

DE 14

SE12

ITd5

SE11

FI19

FR10

FR 24

ITe1

FI18

DK01 DEa5

FR4 2

DEa1NL32

NL3 3

FR2 3

NL22

NL 41

UKd3

UKj3

UKi1

UKm2

UKj1

UKd4

UKi2

Ukg3

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

Share in 2005

Source Elaborated from Eurostat and National Statistical offices

Rat

e of

cha

nge

200

1-20

05

70403530252015100500

Finally when the question is analyzed at Member State level (Table 1) it is possible

to check out the disparities among the above mentioned regions Thus while in some

15

Delocation in the manufacturing sectors in the EU A regional overview

countries such as Spain manufacturing growth has been clearly labour-intensive

something which can be extended to its regions as also happens in Greece and Austria

which both boast positive location effect and a job destruction rate far below the average

in others such as Holland Sweden and to a lesser extent Belgium Ireland and Finland

increasing of industrial activity is only due by advances in productivity so has been

accompanied by heavy job losses and these are patterns repeated in most of their regions

with noticeable relevance in the Netherlands As for the United Kingdom Belgium and

Denmark all share declining industrial production together with strong job cuts a pattern

that is replicated in many of their regions In this situation are the bulk of French and

Italian NUTS 2 just like the countries to which they belong although in this case job

destruction has been much less pronounced and below the group average Close to this

mean is Germany as well as most of their regions

Thus the analysis at Member States level also confirms the positive relationship

between the advance of industrial productivity and the attractiveness for new

manufacturing firms As demonstrated for their regions Ireland Finland Netherlands and

Greece following at quite distance from Germany are Member States showing a better

performance in productivity and with high location effects (see Figure 1)

In fact similarities between performances in the regions and in the Member States

they belong to suggest that the trend in regional industrial employment is to some extent

influenced by belonging to the same country in other words the locational attractions of

each area and its effort to improve productivity is partly due to national specificities or

idiosyncratic factors In fact as shown in Appendix 3 most European industrialized

regions follow the location patterns of the country they belong to reproducing its

productivity behaviour too In summary the evolution of these variables in each region

seems to have a clear `country effectacute that is different behaviours motivated by specific

factors andor differentiated industrial policies

For example in order to correctly interpret the results for Spanish regions it can not

be overlooked that during the period analyzed as in the whole country regions have

absorbed large numbers of immigrants which has boosted the proliferation of lower

productivity and lower wages in industrial activities However it may be thought that this

phenomenon has tended to conceal productivity increases in larger and better equipped

16

Delocation in the manufacturing sectors in the EU A regional overview

firms which could have been significant in the light of their results7 On the other hand

countries like Finland or Ireland with the lowest population increase have opted for high-

tech manufacturing sectors as a way to improve national competitiveness8 Nonetheless

this positive location effect has not reached in these countriesrsquo NUTS 2 enough magnitude

to offset completely the job losses arising from productivity improvements

On the other hand this `national effectacute has also played an important role in

explaining the economic outcomes of European regions in terms of per capita GDP in

accordance with the results obtained by Cuadrado-Roura (2001) using the data available

for 109 EU regions over the 1977-1994 period

Table 3 Export growth in key sectors by countries 2000-2005 (percentages)Impact ITon Total

Spain 65 91 72 158 38 111 04Germany 116 138 61 111 111 130 14United Kingdom 09 85 29 32 23 61 02France -09 102 25 94 118 78 -01Ireland -09 -102 01 25 -34 77 -03Finland 47 152 45 49 56 73 13Italy 40 88 41 90 56 89 02EU-15 70 141 59 91 67 111 11Source Elaborated from Comtrade

Shoes TotalIT Auto Textile Clothes

This importance of the `national effectacute should allow us to make a better approach to

the determinants of above-mentioned delocation patterns in future research One of such

determinants might be found in the different importance and performance by countries of

some manufacturing branches particularly affected by the increase in global competition as

their development has being receiving strong support in emerging countries from both

domestic policies and the location of big multinational companies This is the case of IT

manufacturing sectors or that of automobiles clothes textiles and shoes (Sachwald 2004)

Table 3 shows how the exports of these key sectors have evolved in some of the European

countries considered here It is amazing to see the poor results registered by IT sectors in

United Kingdom France and Ireland compared to the other countries mentioned Germany

in particular This evolution could be behind the stronger effects of delocation in France

and the United Kingdom although it must have produced similar results in the case of

Ireland Furthermore on such a basis what has happened in Italy would remain

7 For details see Myro et al (2008) 8 See Aacutelvarez Loacutepez et al (2007)

17

Delocation in the manufacturing sectors in the EU A regional overview

unexplained as the selected key sectors do not show a bad performance in terms of exports

in the latter country

One reason for such contradictory results lies in the possibility that production

instead of export reflects accurately what has been happening in every sector in terms of

activity However Aacutelvarez et al (2007) show very similar patterns in production and

export specialization from 1995 to 2005 except in the case of Ireland which thus appears

as a very particular one This is because IT specialization in production increases at the

same time as it decreases in exports Perhaps the strong growth in domestic demand

prevents IT exports from growing The opposite case could be that of Italy where slow

growth in domestic demand might have been pushing up IT export9

Anyway further developments on the determinants of delocation patterns require the

preparation of disaggregated data on production a task beyond the objectives of this paper

That could allow us to update the available manufacturing specialization analysis As the

two countries exhibiting higher delocation effects France and the United Kingdom

showed the lowest rates of specialization compared with the EU (Milderfart and Overman

2002 Combes and Overman 2003) something could have changed in the last few years

which would merit a fresh look based on more recent data

On the other hand it is now clear that insofar as delocation in manufacturing

activities has affected each country in a different way a relocation process inside the EU

and between countries has taken place This process deserved to be deeply analyzed

including a perspective by sectors updating what we know about changes in the

distribution of manufacturing industry in the EU and offering the possibility of exploring

whether the European Common Policy has influenced them in some way as was already

suggested by Milderfart-Knarvik and Overman (2002)

9 In Finland IT activities grows much more in production than in exports but in both aspects a positive growth is registered

18

Delocation in the manufacturing sectors in the EU A regional overview

19

Figure 3 Employment growth rates in selected regions 2001-2005 manufacturing and total economy (percentages)

Ukg3

UKi2

UKd4

UKj1

UKm2

UKi1 UKj3

UKd3

NL 41

NL22

FR23

NL33

NL32

DEa1

FR42

DEa5

DK01

FI18

ITe1

FR 24FR10

FI19

SE11

ITd5FR52

DE 14

DE71

DEa2

BE23

DE22

DE21

DE27

IC c4

DE60DE 12

DE25

DE13

DE9 1SE12

ITd3

IE02

DE73

BE21

DE92DEa4

DEf0

DEb3

ES30

DEc0

GR30

DE23

DE11

DK04

AT31

ITe3

SE22

FR82

FR51FR61

SE2 3

DE26

DE24

IT e4

FR62

FR71

ES51

ES2 1

ITe1

AT22

UKh1

UKk1

UKf1

Uke4

UKm3

UKf2

-20

-15

-10

-5

0

5

10

15

20

25

-40 -30 -20 -10 0 10 20 30

Manufacturing

Source Elaborated from Eurostat and National Statistical offices

Tota

l em

ploy

men

t

Delocation in the manufacturing sectors in the EU A regional overview

One last question that deserves consideration is the effect of industrial delocation

process on the regional economies Surprisingly manufacturing job losses have not been

translated into total employment which exhibit positive growth for most of the chosen

regions with the exceptions of some Germans and all the Netherlands (Figure 3) That

means the regions have been able to compensate the negative effect on the employment

from shrinking manufacturing allocating more resources inside the services sectors In fact

a substitution of manufacturing by services seems to have happened as are the region

showing more delocation in manufacturing those that increase in a higher measure the

employment in the service sectors that is the French British and Italian However the

relation is not accurate as there are not a strict quantitative correlation region by region

between rates of delocation and increase in the employment in the service activities

These good results in terms of general employment shown by the regions affected in

a higher measure by manufacturing delocation suggest the convenience of a more general

analysis of regional economic growth beyond the manufacturing sector as some regions

strongly shrinking manufacturing activities in favour of those of services might be leading

an interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

5- Final remarks

Comparative analysis of location patterns for a large sample of European regions -

characterized by having an upper-middle level income and a strong industrial sector-

during the last few years has revealed that only half of them have shown locational

attractions for manufacturing activities The regional patterns found clearly show the

influence of `national effectacute the regions showing location effects corresponding to most

of those located in Austria Germany the Netherlands Ireland Finland Sweden Spain and

Greece

On the contrary the delocation process has affected another half of the regions

included most of them French British and Italian Therefore delocation has contributed to

concentrating the European manufacturing industry in the areas with higher market

potential in the centre of continental Europe It has contributed furthermore to relocating

manufacturing activities inside every country something that requires further analysis

from future research

20

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

Holland and Italy provide such information or at least indexes of volume In the rest of the

countries real values of GVA have been estimated by applying the national price deflators

to the regional series at current values

As regards total employment significant discrepancies between the data published

by Eurostat and that country itself (particularly in the United Kingdom) are found as well

as the gaps found in several NUTS 2 of some Member States such as in Germany forced

to complete regional series and correct such deficiencies with the help of the information

coming from the National Statistics Offices

To sum up most of the information used (especially production data) comes from

National Statistical Offices but this is commonly close to the Eurostat database as this is

built mainly with data coming from the regional accounts of each country

By the way one of the problems arising from taking the National Statistics as the

main source of data is that the time intervals for which they provide information do not

always coincide That explains that despite having more recent data from some areas the

analysis must conclude in 2005

4- Delocation of manufacturing industries in European regions

As noted above in the introduction the group of regions examined reached just over

64 of EU-15 industrial GVA and about 60 of jobs in 2005 figures slightly lower than

in 2000 Thus despite nearly half of regionsrsquo industrial output at constant prices having

achieved positive growth rates the group accumulated throughout the period a decrease of

about 1 similar to EU-15 average4 Regarding employment the evolution has been even

more negative Throughout the first five years of the current century destruction of jobs in

the aggregate industry has been a common feature in the vast majority (90) of the

seventy-five NUTS 2 examined as well as for each of the Member States where they are

located except Spain accumulating the total sample a decline of close to 8

4 It must be noted that the EU average has been calculated from the aggregate industrial GVA at constant prices of thirteen countries included in the sample Thus Portugal and Luxembourg have been excluded and the values from other countries have been estimated by adding figures of all their regions So the above mentioned growth rate differs from that provided by Eurostat exhibiting a positive increase of 46

6

Delocation in the manufacturing sectors in the EU A regional overview

Focusing on employment the most dynamic areas are located mostly in France

Spain Austria and Italy while the Netherlands and especially the British regions exhibit

the largest job losses (Table 1)

Registered unemployment seems to be related to good performance in labour

productivity pointing to the fierce competition faced by the manufacturing sector in the

analyzed period Thus higher job losses are found in regions with stronger labour

productivity increases although some French and most of the British regions escaped from

this rule as large losses in employment are not accompanied by significant gains in labour

productivity suggesting delocation effects of a particular intensity

That reveals a very different growth path in employment and labour productivity by

regions differences which can be explored in a more detail to isolate the attractive regions

to locate manufacturing activities from the others characterized by a delocation process

To go deeply into this latter question Table 1 contains the results of estimating the

impact on total employment of location and delocation forces in the manufacturing sectors

across the seventy-five regions included in the sample over the period 2001-2005 by using

the procedure proposed in the second section that is discounting from the total change in

jobs the reduction corresponding to the increase in labour productivity Therefore the

location effect shows the effect on employment of GVA change calculated by applying the

initial rate of labour productivity

In about half of the regions with job losses all of those considered in the

Netherlands Finland Greece and Ireland three-quarters of the Germans two British two

Swedish and Madrid in Spain the decline in employment has been due entirely to

increased labour productivity so a positive effect of location appears This behaviour

extends to Finnish German and Swedish economies as a whole as well as Greece Ireland

and Austria Therefore all those regions have proved attractive for the location of

manufacturing firms

7

Delocation in the manufacturing sectors in the EU A regional overview

8

Productivity Location Productivity Locationeffect effect effect effect

at22 Steiermark 372 -108 480 36 -11 47at31 Oberoumlsterreich 097 -837 934 06 -56 62

Austria -1104 -2649 1545 -18 -42 25be21 Prov Antwerpen -1010 079 -1089 -70 05 -75be23 Prov Oost-Vlaanderen -750 -627 -123 -74 -61 -12

Belgium -5740 -2219 -3521 -84 -32 -51de11 Stuttgart -3157 -8244 5087 -48 -125 77de12 Karlsruhe -2036 -3585 1550 -57 -101 44de13 Freiburg -1291 -1301 011 -45 -45 00de14 Tuumlbingen -585 -1805 1219 -23 -72 49de21 Oberbayern -1630 -7976 6346 -37 -179 142de22 Niederbayern -490 -3270 2781 -33 -218 186de23 Oberpfalz -538 -1716 1178 -38 -120 83de24 Oberfranken -1957 -3131 1175 -116 -186 70de25 Mittelfranken -1184 -1770 586 -53 -80 26de26 Unterfranken -819 -3157 2338 -49 -189 140de27 Schwaben -932 -3335 2403 -41 -148 106de60 Hamburg -760 -1222 463 -60 -97 37de71 Darmstadt -5343 -4720 -623 -149 -132 -17de73 Kassel -1059 -1396 336 -78 -103 25de91 Braunschweig -947 -500 -447 -47 -25 -22de92 Hannover -1900 -1544 -356 -106 -86 -20dea1 Duumlsseldorf -6272 -8787 2515 -126 -177 51dea2 Koumlln -5428 -4971 -457 -150 -137 -13dea4 Detmold -3026 -3112 086 -107 -110 03dea5 Arnsberg -5824 -5502 -322 -129 -122 -07deb3 Rheinhessen-Pfalz -2099 -3016 917 -106 -152 46dec0 Saarland -360 -1810 1450 -32 -160 128def0 Schleswig-Holstein -1867 -2485 617 -101 -135 33

Germany -60300 -112275 51975 -74 -138 64dk01 Hovedstaden -1300 -1216 -084 -134 -125 -09dk04 Midtjylland -2000 -1622 -378 -146 -118 -28

Denmark -6100 -3428 -2672 -130 -73 -57es21 Pais Vasco 1840 -331 2171 74 -13 87es30 Comunidad de Madrid -1060 -2180 1120 -31 -63 32es51 Cataluntildea 660 -1914 2574 09 -25 34

Spain 13970 -3927 17897 47 -13 60fi18 Etelauml-Suomi -1807 -7159 5352 -86 -339 253fi19 Laumlnsi-Suomi -701 -3886 3185 -54 -297 244

Finland -1798 -12006 10208 -43 -286 243fr10 Icircle de France 12652 17868 -5216 218 307 -90fr23 Haute-Normandie -2418 -1093 -1325 -171 -77 -94fr24 Centre -2016 -120 -1896 -106 -06 -100fr42 Alsace -2489 -988 -1501 -154 -61 -93fr51 Pays de la Loire -217 1794 -2011 -08 67 -75fr52 Bretagne -415 -215 -200 -23 -12 -11fr61 Aquitaine -199 742 -941 -14 50 -64fr62 Midi-Pyreacuteneacutees 209 3318 -3109 15 245 -229fr71 Rhocircne-Alpes -2919 2759 -5678 -61 57 -118fr82 Provence-Alpes-Cocircte dAzur -097 262 -359 -06 18 -24

France -15306 31566 -46872 -40 83 -123gr30 Attiki -360 -5072 4712 -17 -234 217

Greece -730 -11989 11259 -14 -222 208ie02 Southern and Eastern -1840 -9985 8145 -78 -426 347

Ireland -2420 -12274 9854 -76 -387 311itc1 Piemonte -6869 2539 -9407 -129 48 -177itc4 Lombardia -7940 -7682 -258 -63 -61 -02itd3 Veneto -3967 -211 -3756 -62 -03 -58itd5 Emilia-Romagna -860 -840 -020 -16 -16 00ite1 Toscana -3620 078 -3697 -98 02 -101ite3 Marche 164 -555 719 08 -28 36ite4 Lazio -733 956 -1689 -36 47 -82

Italy -21761 6310 -28071 -45 13 -58nl22 Gelderland -2164 -2506 341 -178 -206 28nl32 Noord-Holland -1555 -2034 479 -138 -180 42nl33 Zuid-Holland -1734 -3980 2246 -129 -296 167nl41 Noord-Brabant -3539 -5515 1976 -173 -269 96

Netherlands -14767 -24690 9923 -161 -270 108se11 Stockholm -1570 -4183 2613 -136 -363 227se12 Oumlstra Mellansverige -2170 -2387 217 -151 -166 15se22 Sydsverige -1010 -690 -320 -91 -62 -29se23 Vaumlstsverige -460 134 -594 -26 07 -33

Sweden -7500 -9400 1900 -95 -119 24ukd3 Greater Manchester -4900 -1907 -2993 -232 -90 -142ukd4 Lancashire -4600 -5082 482 -317 -350 33uke4 West Yorkshire -3700 -1192 -2508 -186 -60 -126ukf1 Derbyshire and Nottinghamshire -3700 -4965 1265 -179 -240 61ukf2 Leicestershire Rutland and Northan -4400 -3607 -793 -233 -191 -42ukg3 West Midlands -8700 -4999 -3701 -313 -180 -133ukh1 East Anglia -2500 -1292 -1208 -137 -71 -66uki1 Inner London -1919 -1381 -538 -142 -102 -40uki2 Outer London -4481 1106 -5587 -271 67 -338ukj1 Berkshire Bucks and Oxfordshire -4600 -3898 -702 -246 -208 -38ukj3 Hampshire and Isle of Wight -1100 -689 -411 -81 -51 -30ukk1 Gloucestershire Wiltshire and Brist -3400 -1253 -2147 -169 -62 -107ukm2 Eastern Scotland -1100 1402 -2502 -91 116 -207ukm3 South Western Scotland -3400 -2107 -1293 -207 -128 -79

United Kingdom -89800 -48574 -41226 -196 -106 -90

Table 1 Importance of manufacturing delocation in European regions 2001-2005 (Decomposition of change in employment)

Total effect

Source Elaborated from Eurostat and National Statistical offices

Number of jobs (thousands)

Total effect

Percent share of 2000 employmentRegions

Delocation in the manufacturing sectors in the EU A regional overview

By contrast thirty-eight remaining regions that also show reductions in

manufacturing employment have been affected by delocation processes to different

extents although only in a small number of them have they been accompanied by

decreases in labour productivity which could make the prospects for economic growth

tougher (four from France Piamonte Toscana and Lazio in Italy Outer London and

Eastern Scotland in the United Kingdom the Swedish Vaumlstsverige and Antwerp in

Belgium)

Regions with positive location effects are shown in Map 1 Most of them are large

industrial regions and are located in the traditional industrial growth centres5 those

characterized by high market potential (Combes and Overman 2003) although there are

also some intermediate and peripheral territories mainly located in the north of Italy

United Kingdom and Spain and in the south of Sweden and Norway This means that the

delocation process has contributed to concentrating the European manufacturing industries

in that privileged area of faster growth apart from some other changes inside every

country

To compare now the intensity of such effects of location and delocation across

regions avoiding the differences they have in size we built a relative measure of them the

ratio of job losses in the period to the volume of employment in the first year 2000 (Table

1)

Following this ratio delocation effects are revealed to be particularly strong in

France United Kingdom Italy and in one region of Belgium with declines in

employment exceeding 20 in some of them

Conversely the regions of Finland and several of the German and the Dutch ones

exhibit strong location effects

In any case as was pointed above the importance of location effects seems to lie in

labour productivity gains In fact higher increases in productivity (over 15) are found

precisely in those regions of Ireland Finland Sweden Netherlands and Germany besides

the Greek Attiki where the location effect has in general reached its greatest extent

(Figure 1)

5 Note that the map of winning regions is clearly influenced by the large number (30) of German NUTS 2 in the sample

9

Delocation in the manufacturing sectors in the EU A regional overview

fi19

fi18

ie02

dec0

gr30

se11

nl33

de21

de22

de26

de25

de12

de14

de11

dea4

dea1de73

def0

de60

deb3

de27

de23

itd5

ite3

se12

nl22

es 30

ukd4

es51

nl32

at22

ukf1

at31

es21

Map 1 Regions with positive location effect ( of 2000 employment)

Above 10 0 to 10

nl41

de13de27

Nevertheless productivity also grows faster in most of the regions belonging to the

United Kingdom Belgium and Denmark in spite of the strong delocation effects registered

10

Delocation in the manufacturing sectors in the EU A regional overview

which disturbs the above-mentioned positive relationship suggesting the harsh process of

industrial restructuring involved

Figure 1 Location effect and productivity growth in select regions manufacturing sector 2001-2005 (percentages)

UKi2

EU 15

FR6 2

UKm2

ITc1

UKg3UKd 3

UKe4

UKk1FR71F R

FR10UKm3

UKj1UKf2

UKd 4

UKi1

D K

DK01

UKh1U K

FR2 3FR4 2

FR24ITe1

BE2 1ITe4FR51

FR61

ITITd 3

SE2 2FR8 2

SE23

NL2 2

UKf1DE2 4

NL3 2DEa1

DEf0

UKj3DE9 1 DE92

DEa2DE71DK0 4

S EDE25

DE6 0DE12

A T

DE14

ITe3

AT2 2

AT3 1ESD E

DE 11DE2 3

NL41N LDE2 7

DEc0

NL3 3

DE2 6DE2 1

ES3 0ES51

SE11

DE2 2

ES21

GR

FI19F I

FI18

IE

GR3 0

B E

DE73

DEb 3

IE0 2

ITd 5 DE13SE12

FR52ITc4BE2 3

DEa4DEa5

y = 07x - 59R2 = 06

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

40

-35 -30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30 35 40 45

Productivity grothw Source Elaborated from Eurostat and National Statistical offices

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

11

Delocation in the manufacturing sectors in the EU A regional overview

In contrast the productivity decline in half of the French and Italian regions albeit

with a positive impact on employment is largely offset by destruction of industrial

activity except in Icircle de France and Midi -Pyreacuteneacutees the two territories in which

productivity has a more negative trajectory

These results closely match those reported by Cuadrado Roura et al (2000) based on

the growth rates for regional productivity and employment during the period 1980-1993

grouping European regions in four typologies comparing the growth rates of these two

variables with their respective European averages

According to this approach and including the location effect in the analysis the

group of regions that are examined in this work would be categorized as follows (Table 2)

The first group includes areas with productivity and employment growth rates above

sample average and show positive location effect (Quadrant 1) So as has been underlined

in that work possibly this reflects the fact that the technological and organizational

changes introduced during the period would have been successful promoting the

diversification and specialization in activities with high growth levels and attracting new

investments which are less labour-intensive

In a second stage are those regions that as the above mentioned authors suggest

have implemented restructuring processes to eliminate the most inefficient outputs

achieving higher levels of productivity and in about half of them increased industrial

production at the expense of a notable decline in employment (Quadrant 4) They named

this typology `restructuring via productivityacute

With a few exceptions (those where the location effect is negative) the regions in

both groups are the winners of industrial dynamic during the period Most of them are

large industrial regions and are located in the traditional industrial growth centres As

demonstrated above they are mainly Finish German and Dutch as well as Greek and Irish

regions

In contrast another large group of regions due to different factors6 have opted for

more labour-intensive industrial models which in a few cases (especially in Spanish and

Austrian regions) have been accompanied by output increases so they can also be included

6 See Cuadrado et al (2000)

12

Delocation in the manufacturing sectors in the EU A regional overview

in the winners (Map 1) while in other the result has been a loss of productive activity

(Quadrant 2)

(1) (2)Employment Productivity Employment Productivity

de11 Stuttgart -48 125 at22 Steiermark 36 11de12 Karlsruhe -57 101 at31 Oberoumlsterreich 06 56de21 Oberbayern -37 179 be21 Prov Antwerpen -70 -05de22 Niederbayern -33 218 be23 Prov Oost-Vlaanderen -74 61de23 Oberpfalz -38 120 de13 Freiburg -45 45de25 Mittelfranken -53 80 de14 Tuumlbingen -23 72de26 Unterfranken -49 189 de91 Braunschweig -47 25de27 Schwaben -41 148 es21 Pais Vasco 74 13de60 Hamburg -60 97 es30 Comunidad de Madrid -31 63dec0 Saarland -32 160 es51 Cataluntildea 09 25fi19 Laumlnsi-Suomi -54 297 fr10 Icircle de France 218 -307gr30 Attiki -17 234 fr51 Pays de la Loire -08 -67

fr52 Bretagne -23 12fr61 Aquitaine -14 -50fr62 Midi-Pyreacuteneacutees 15 -245fr71 Rhocircne-Alpes -61 -57fr82 Provence-Alpes-Cocircte dAzur -06 -18itc4 Lombardia -63 61itd3 Veneto -62 03itd5 Emilia-Romagna -16 16ite3 Marche 08 28ite4 Lazio -36 -47se23 Vaumlstsverige -26 -07

(4) (3)Employment Productivity Employment Productivity

de24 Oberfranken -116 186 fr24 Centre -106 06de71 Darmstadt -149 132 fr42 Alsace -154 61de73 Kassel -78 103 itc1 Piemonte -129 -48de92 Hannover -106 86 ite1 Toscana -98 -02dea1 Duumlsseldorf -126 177 se22 Sydsverige -91 62dea2 Koumlln -150 137 uke4 West Yorkshire -186 60dea4 Detmold -107 110 ukh1 East Anglia -137 71dea5 Arnsberg -129 122 uki2 Outer London -271 -67deb3 Rheinhessen-Pfalz -106 152 ukj3 Hampshire and Isle of Wight -81 51def0 Schleswig-Holstein -101 135 ukk1 Gloucestershire Wiltshire -169 62dk01 Hovedstaden -134 125 ukm2 Eastern Scotland -91 -116dk04 Midtjylland -146 118fi18 Etelauml-Suomi -86 339fr23 Haute-Normandie -171 77ie02 Southern and Eastern -78 426nl22 Gelderland -178 206nl32 Noord-Holland -138 180nl33 Zuid-Holland -129 296nl41 Noord-Brabant -173 269se11 Stockholm -136 363se12 Oumlstra Mellansverige -151 166ukd3 Greater Manchester -232 90ukd4 Lancashire -317 350ukf1 Derbyshire and Nottinghamshire -179 240ukf2 Leicestershire Rutland and Northants -233 191ukg3 West Midlands -313 180uki1 Inner London -142 102ukj1 Berkshire Bucks and Oxfordshire -246 208ukm3 South Western Scotland -207 128

Source Elaborated from Eurostat and national Statistical officesNote Shaded areas indicate regions with positive location effect

Em

ploy

men

t gro

wth

lt A

vera

ge v

alue

Productivity growth gt Average value Productivity growth lt Average value

Em

ploy

men

t gro

wth

gt A

vera

ge v

alue

Table 2 Employment and productivity growth rates in select regions manufacturing sector 2001-2005 (Averages of all regions +74 productivity -77 employment)

13

Delocation in the manufacturing sectors in the EU A regional overview

Finally the worst results can be seen in a fourth type of region with growth rates of

GVA and productivity below average and serious job destruction (Quadrant 3) This would

occur in regions with problems in restructuring their manufacturing sector or with low

levels of activity So together with those that are either restructuring via productivity or

via employment but without achieving a positive location effect they are clearly the

losers These include mainly British French Italian Belgian and Danish regions

To sum up these results confirm the direct relationship between growth in labour

efficiency and progress in the creation of productive industrial activity as pointed out

above and which can be more clearly by looking at Figure 1 In other words regions with

higher levels of productivity growth in general show a greater attractiveness for new

manufacturing firms

These differences in the behaviour of productivity and employment across regions

have led to remarkable disparities in rates of change of their industrial output during the

period Thus while a few increased their real manufacturing GVA by more than 20 in

the less dynamic ones the loss of industrial output is between a sixth and a third of the

initial value So significant changes had taken place in the relative position of some of

them (Figure 2)

Growth is led by a group of twenty regions located in Ireland Finland Greece

Germany Sweden Netherlands Spain and Austria with real GVA growth rates that are

more than five points above average As a result they increased their share in the industrial

output of the sample Looking at the complete list we see that the growth deviation from

the mean exhibits positive values in just over half of the regions examined among them

seven of the ten with the largest industrial scale in 2005 Thus the top ten regions have

increased their share from 32 to 33 percent in the last five years

Conversely among the less dynamic regions are found a great number of British and

some French and Italian ones which consequently have fallen back in the ranking of

NUTS 2 arranged by their contribution to aggregate manufacturing output

Between both groups there are approximately twenty regions mainly in Germany

with growth rates not far from average so their positions have hardly changed during the

last five years

14

Delocation in the manufacturing sectors in the EU A regional overview

Figure 2 Regional shares in the total sample manufacturing GVA(percentage at 2000 prices)

ITc4

UKf2

UKm3

Uke4

UKf1

UKk1

UKh1

AT22

ITe1

ES21

ES51

FR71

FR62

IT e4

DE24

DE26

SE23

FR61

FR51

FR82

SE22

ITe3

AT31

DK04

DE11

DE23

GR30

DEc0

ES30DEb3DEf0

DEa4

DE92

BE21

DE73

IE02

ITd3

FR52

DE91

DE13DE25

DE 12DE60

DE27

DE21

DE22

BE23DEa2

DE71

DE 14

SE12

ITd5

SE11

FI19

FR10

FR 24

ITe1

FI18

DK01 DEa5

FR4 2

DEa1NL32

NL3 3

FR2 3

NL22

NL 41

UKd3

UKj3

UKi1

UKm2

UKj1

UKd4

UKi2

Ukg3

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

Share in 2005

Source Elaborated from Eurostat and National Statistical offices

Rat

e of

cha

nge

200

1-20

05

70403530252015100500

Finally when the question is analyzed at Member State level (Table 1) it is possible

to check out the disparities among the above mentioned regions Thus while in some

15

Delocation in the manufacturing sectors in the EU A regional overview

countries such as Spain manufacturing growth has been clearly labour-intensive

something which can be extended to its regions as also happens in Greece and Austria

which both boast positive location effect and a job destruction rate far below the average

in others such as Holland Sweden and to a lesser extent Belgium Ireland and Finland

increasing of industrial activity is only due by advances in productivity so has been

accompanied by heavy job losses and these are patterns repeated in most of their regions

with noticeable relevance in the Netherlands As for the United Kingdom Belgium and

Denmark all share declining industrial production together with strong job cuts a pattern

that is replicated in many of their regions In this situation are the bulk of French and

Italian NUTS 2 just like the countries to which they belong although in this case job

destruction has been much less pronounced and below the group average Close to this

mean is Germany as well as most of their regions

Thus the analysis at Member States level also confirms the positive relationship

between the advance of industrial productivity and the attractiveness for new

manufacturing firms As demonstrated for their regions Ireland Finland Netherlands and

Greece following at quite distance from Germany are Member States showing a better

performance in productivity and with high location effects (see Figure 1)

In fact similarities between performances in the regions and in the Member States

they belong to suggest that the trend in regional industrial employment is to some extent

influenced by belonging to the same country in other words the locational attractions of

each area and its effort to improve productivity is partly due to national specificities or

idiosyncratic factors In fact as shown in Appendix 3 most European industrialized

regions follow the location patterns of the country they belong to reproducing its

productivity behaviour too In summary the evolution of these variables in each region

seems to have a clear `country effectacute that is different behaviours motivated by specific

factors andor differentiated industrial policies

For example in order to correctly interpret the results for Spanish regions it can not

be overlooked that during the period analyzed as in the whole country regions have

absorbed large numbers of immigrants which has boosted the proliferation of lower

productivity and lower wages in industrial activities However it may be thought that this

phenomenon has tended to conceal productivity increases in larger and better equipped

16

Delocation in the manufacturing sectors in the EU A regional overview

firms which could have been significant in the light of their results7 On the other hand

countries like Finland or Ireland with the lowest population increase have opted for high-

tech manufacturing sectors as a way to improve national competitiveness8 Nonetheless

this positive location effect has not reached in these countriesrsquo NUTS 2 enough magnitude

to offset completely the job losses arising from productivity improvements

On the other hand this `national effectacute has also played an important role in

explaining the economic outcomes of European regions in terms of per capita GDP in

accordance with the results obtained by Cuadrado-Roura (2001) using the data available

for 109 EU regions over the 1977-1994 period

Table 3 Export growth in key sectors by countries 2000-2005 (percentages)Impact ITon Total

Spain 65 91 72 158 38 111 04Germany 116 138 61 111 111 130 14United Kingdom 09 85 29 32 23 61 02France -09 102 25 94 118 78 -01Ireland -09 -102 01 25 -34 77 -03Finland 47 152 45 49 56 73 13Italy 40 88 41 90 56 89 02EU-15 70 141 59 91 67 111 11Source Elaborated from Comtrade

Shoes TotalIT Auto Textile Clothes

This importance of the `national effectacute should allow us to make a better approach to

the determinants of above-mentioned delocation patterns in future research One of such

determinants might be found in the different importance and performance by countries of

some manufacturing branches particularly affected by the increase in global competition as

their development has being receiving strong support in emerging countries from both

domestic policies and the location of big multinational companies This is the case of IT

manufacturing sectors or that of automobiles clothes textiles and shoes (Sachwald 2004)

Table 3 shows how the exports of these key sectors have evolved in some of the European

countries considered here It is amazing to see the poor results registered by IT sectors in

United Kingdom France and Ireland compared to the other countries mentioned Germany

in particular This evolution could be behind the stronger effects of delocation in France

and the United Kingdom although it must have produced similar results in the case of

Ireland Furthermore on such a basis what has happened in Italy would remain

7 For details see Myro et al (2008) 8 See Aacutelvarez Loacutepez et al (2007)

17

Delocation in the manufacturing sectors in the EU A regional overview

unexplained as the selected key sectors do not show a bad performance in terms of exports

in the latter country

One reason for such contradictory results lies in the possibility that production

instead of export reflects accurately what has been happening in every sector in terms of

activity However Aacutelvarez et al (2007) show very similar patterns in production and

export specialization from 1995 to 2005 except in the case of Ireland which thus appears

as a very particular one This is because IT specialization in production increases at the

same time as it decreases in exports Perhaps the strong growth in domestic demand

prevents IT exports from growing The opposite case could be that of Italy where slow

growth in domestic demand might have been pushing up IT export9

Anyway further developments on the determinants of delocation patterns require the

preparation of disaggregated data on production a task beyond the objectives of this paper

That could allow us to update the available manufacturing specialization analysis As the

two countries exhibiting higher delocation effects France and the United Kingdom

showed the lowest rates of specialization compared with the EU (Milderfart and Overman

2002 Combes and Overman 2003) something could have changed in the last few years

which would merit a fresh look based on more recent data

On the other hand it is now clear that insofar as delocation in manufacturing

activities has affected each country in a different way a relocation process inside the EU

and between countries has taken place This process deserved to be deeply analyzed

including a perspective by sectors updating what we know about changes in the

distribution of manufacturing industry in the EU and offering the possibility of exploring

whether the European Common Policy has influenced them in some way as was already

suggested by Milderfart-Knarvik and Overman (2002)

9 In Finland IT activities grows much more in production than in exports but in both aspects a positive growth is registered

18

Delocation in the manufacturing sectors in the EU A regional overview

19

Figure 3 Employment growth rates in selected regions 2001-2005 manufacturing and total economy (percentages)

Ukg3

UKi2

UKd4

UKj1

UKm2

UKi1 UKj3

UKd3

NL 41

NL22

FR23

NL33

NL32

DEa1

FR42

DEa5

DK01

FI18

ITe1

FR 24FR10

FI19

SE11

ITd5FR52

DE 14

DE71

DEa2

BE23

DE22

DE21

DE27

IC c4

DE60DE 12

DE25

DE13

DE9 1SE12

ITd3

IE02

DE73

BE21

DE92DEa4

DEf0

DEb3

ES30

DEc0

GR30

DE23

DE11

DK04

AT31

ITe3

SE22

FR82

FR51FR61

SE2 3

DE26

DE24

IT e4

FR62

FR71

ES51

ES2 1

ITe1

AT22

UKh1

UKk1

UKf1

Uke4

UKm3

UKf2

-20

-15

-10

-5

0

5

10

15

20

25

-40 -30 -20 -10 0 10 20 30

Manufacturing

Source Elaborated from Eurostat and National Statistical offices

Tota

l em

ploy

men

t

Delocation in the manufacturing sectors in the EU A regional overview

One last question that deserves consideration is the effect of industrial delocation

process on the regional economies Surprisingly manufacturing job losses have not been

translated into total employment which exhibit positive growth for most of the chosen

regions with the exceptions of some Germans and all the Netherlands (Figure 3) That

means the regions have been able to compensate the negative effect on the employment

from shrinking manufacturing allocating more resources inside the services sectors In fact

a substitution of manufacturing by services seems to have happened as are the region

showing more delocation in manufacturing those that increase in a higher measure the

employment in the service sectors that is the French British and Italian However the

relation is not accurate as there are not a strict quantitative correlation region by region

between rates of delocation and increase in the employment in the service activities

These good results in terms of general employment shown by the regions affected in

a higher measure by manufacturing delocation suggest the convenience of a more general

analysis of regional economic growth beyond the manufacturing sector as some regions

strongly shrinking manufacturing activities in favour of those of services might be leading

an interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

5- Final remarks

Comparative analysis of location patterns for a large sample of European regions -

characterized by having an upper-middle level income and a strong industrial sector-

during the last few years has revealed that only half of them have shown locational

attractions for manufacturing activities The regional patterns found clearly show the

influence of `national effectacute the regions showing location effects corresponding to most

of those located in Austria Germany the Netherlands Ireland Finland Sweden Spain and

Greece

On the contrary the delocation process has affected another half of the regions

included most of them French British and Italian Therefore delocation has contributed to

concentrating the European manufacturing industry in the areas with higher market

potential in the centre of continental Europe It has contributed furthermore to relocating

manufacturing activities inside every country something that requires further analysis

from future research

20

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

Focusing on employment the most dynamic areas are located mostly in France

Spain Austria and Italy while the Netherlands and especially the British regions exhibit

the largest job losses (Table 1)

Registered unemployment seems to be related to good performance in labour

productivity pointing to the fierce competition faced by the manufacturing sector in the

analyzed period Thus higher job losses are found in regions with stronger labour

productivity increases although some French and most of the British regions escaped from

this rule as large losses in employment are not accompanied by significant gains in labour

productivity suggesting delocation effects of a particular intensity

That reveals a very different growth path in employment and labour productivity by

regions differences which can be explored in a more detail to isolate the attractive regions

to locate manufacturing activities from the others characterized by a delocation process

To go deeply into this latter question Table 1 contains the results of estimating the

impact on total employment of location and delocation forces in the manufacturing sectors

across the seventy-five regions included in the sample over the period 2001-2005 by using

the procedure proposed in the second section that is discounting from the total change in

jobs the reduction corresponding to the increase in labour productivity Therefore the

location effect shows the effect on employment of GVA change calculated by applying the

initial rate of labour productivity

In about half of the regions with job losses all of those considered in the

Netherlands Finland Greece and Ireland three-quarters of the Germans two British two

Swedish and Madrid in Spain the decline in employment has been due entirely to

increased labour productivity so a positive effect of location appears This behaviour

extends to Finnish German and Swedish economies as a whole as well as Greece Ireland

and Austria Therefore all those regions have proved attractive for the location of

manufacturing firms

7

Delocation in the manufacturing sectors in the EU A regional overview

8

Productivity Location Productivity Locationeffect effect effect effect

at22 Steiermark 372 -108 480 36 -11 47at31 Oberoumlsterreich 097 -837 934 06 -56 62

Austria -1104 -2649 1545 -18 -42 25be21 Prov Antwerpen -1010 079 -1089 -70 05 -75be23 Prov Oost-Vlaanderen -750 -627 -123 -74 -61 -12

Belgium -5740 -2219 -3521 -84 -32 -51de11 Stuttgart -3157 -8244 5087 -48 -125 77de12 Karlsruhe -2036 -3585 1550 -57 -101 44de13 Freiburg -1291 -1301 011 -45 -45 00de14 Tuumlbingen -585 -1805 1219 -23 -72 49de21 Oberbayern -1630 -7976 6346 -37 -179 142de22 Niederbayern -490 -3270 2781 -33 -218 186de23 Oberpfalz -538 -1716 1178 -38 -120 83de24 Oberfranken -1957 -3131 1175 -116 -186 70de25 Mittelfranken -1184 -1770 586 -53 -80 26de26 Unterfranken -819 -3157 2338 -49 -189 140de27 Schwaben -932 -3335 2403 -41 -148 106de60 Hamburg -760 -1222 463 -60 -97 37de71 Darmstadt -5343 -4720 -623 -149 -132 -17de73 Kassel -1059 -1396 336 -78 -103 25de91 Braunschweig -947 -500 -447 -47 -25 -22de92 Hannover -1900 -1544 -356 -106 -86 -20dea1 Duumlsseldorf -6272 -8787 2515 -126 -177 51dea2 Koumlln -5428 -4971 -457 -150 -137 -13dea4 Detmold -3026 -3112 086 -107 -110 03dea5 Arnsberg -5824 -5502 -322 -129 -122 -07deb3 Rheinhessen-Pfalz -2099 -3016 917 -106 -152 46dec0 Saarland -360 -1810 1450 -32 -160 128def0 Schleswig-Holstein -1867 -2485 617 -101 -135 33

Germany -60300 -112275 51975 -74 -138 64dk01 Hovedstaden -1300 -1216 -084 -134 -125 -09dk04 Midtjylland -2000 -1622 -378 -146 -118 -28

Denmark -6100 -3428 -2672 -130 -73 -57es21 Pais Vasco 1840 -331 2171 74 -13 87es30 Comunidad de Madrid -1060 -2180 1120 -31 -63 32es51 Cataluntildea 660 -1914 2574 09 -25 34

Spain 13970 -3927 17897 47 -13 60fi18 Etelauml-Suomi -1807 -7159 5352 -86 -339 253fi19 Laumlnsi-Suomi -701 -3886 3185 -54 -297 244

Finland -1798 -12006 10208 -43 -286 243fr10 Icircle de France 12652 17868 -5216 218 307 -90fr23 Haute-Normandie -2418 -1093 -1325 -171 -77 -94fr24 Centre -2016 -120 -1896 -106 -06 -100fr42 Alsace -2489 -988 -1501 -154 -61 -93fr51 Pays de la Loire -217 1794 -2011 -08 67 -75fr52 Bretagne -415 -215 -200 -23 -12 -11fr61 Aquitaine -199 742 -941 -14 50 -64fr62 Midi-Pyreacuteneacutees 209 3318 -3109 15 245 -229fr71 Rhocircne-Alpes -2919 2759 -5678 -61 57 -118fr82 Provence-Alpes-Cocircte dAzur -097 262 -359 -06 18 -24

France -15306 31566 -46872 -40 83 -123gr30 Attiki -360 -5072 4712 -17 -234 217

Greece -730 -11989 11259 -14 -222 208ie02 Southern and Eastern -1840 -9985 8145 -78 -426 347

Ireland -2420 -12274 9854 -76 -387 311itc1 Piemonte -6869 2539 -9407 -129 48 -177itc4 Lombardia -7940 -7682 -258 -63 -61 -02itd3 Veneto -3967 -211 -3756 -62 -03 -58itd5 Emilia-Romagna -860 -840 -020 -16 -16 00ite1 Toscana -3620 078 -3697 -98 02 -101ite3 Marche 164 -555 719 08 -28 36ite4 Lazio -733 956 -1689 -36 47 -82

Italy -21761 6310 -28071 -45 13 -58nl22 Gelderland -2164 -2506 341 -178 -206 28nl32 Noord-Holland -1555 -2034 479 -138 -180 42nl33 Zuid-Holland -1734 -3980 2246 -129 -296 167nl41 Noord-Brabant -3539 -5515 1976 -173 -269 96

Netherlands -14767 -24690 9923 -161 -270 108se11 Stockholm -1570 -4183 2613 -136 -363 227se12 Oumlstra Mellansverige -2170 -2387 217 -151 -166 15se22 Sydsverige -1010 -690 -320 -91 -62 -29se23 Vaumlstsverige -460 134 -594 -26 07 -33

Sweden -7500 -9400 1900 -95 -119 24ukd3 Greater Manchester -4900 -1907 -2993 -232 -90 -142ukd4 Lancashire -4600 -5082 482 -317 -350 33uke4 West Yorkshire -3700 -1192 -2508 -186 -60 -126ukf1 Derbyshire and Nottinghamshire -3700 -4965 1265 -179 -240 61ukf2 Leicestershire Rutland and Northan -4400 -3607 -793 -233 -191 -42ukg3 West Midlands -8700 -4999 -3701 -313 -180 -133ukh1 East Anglia -2500 -1292 -1208 -137 -71 -66uki1 Inner London -1919 -1381 -538 -142 -102 -40uki2 Outer London -4481 1106 -5587 -271 67 -338ukj1 Berkshire Bucks and Oxfordshire -4600 -3898 -702 -246 -208 -38ukj3 Hampshire and Isle of Wight -1100 -689 -411 -81 -51 -30ukk1 Gloucestershire Wiltshire and Brist -3400 -1253 -2147 -169 -62 -107ukm2 Eastern Scotland -1100 1402 -2502 -91 116 -207ukm3 South Western Scotland -3400 -2107 -1293 -207 -128 -79

United Kingdom -89800 -48574 -41226 -196 -106 -90

Table 1 Importance of manufacturing delocation in European regions 2001-2005 (Decomposition of change in employment)

Total effect

Source Elaborated from Eurostat and National Statistical offices

Number of jobs (thousands)

Total effect

Percent share of 2000 employmentRegions

Delocation in the manufacturing sectors in the EU A regional overview

By contrast thirty-eight remaining regions that also show reductions in

manufacturing employment have been affected by delocation processes to different

extents although only in a small number of them have they been accompanied by

decreases in labour productivity which could make the prospects for economic growth

tougher (four from France Piamonte Toscana and Lazio in Italy Outer London and

Eastern Scotland in the United Kingdom the Swedish Vaumlstsverige and Antwerp in

Belgium)

Regions with positive location effects are shown in Map 1 Most of them are large

industrial regions and are located in the traditional industrial growth centres5 those

characterized by high market potential (Combes and Overman 2003) although there are

also some intermediate and peripheral territories mainly located in the north of Italy

United Kingdom and Spain and in the south of Sweden and Norway This means that the

delocation process has contributed to concentrating the European manufacturing industries

in that privileged area of faster growth apart from some other changes inside every

country

To compare now the intensity of such effects of location and delocation across

regions avoiding the differences they have in size we built a relative measure of them the

ratio of job losses in the period to the volume of employment in the first year 2000 (Table

1)

Following this ratio delocation effects are revealed to be particularly strong in

France United Kingdom Italy and in one region of Belgium with declines in

employment exceeding 20 in some of them

Conversely the regions of Finland and several of the German and the Dutch ones

exhibit strong location effects

In any case as was pointed above the importance of location effects seems to lie in

labour productivity gains In fact higher increases in productivity (over 15) are found

precisely in those regions of Ireland Finland Sweden Netherlands and Germany besides

the Greek Attiki where the location effect has in general reached its greatest extent

(Figure 1)

5 Note that the map of winning regions is clearly influenced by the large number (30) of German NUTS 2 in the sample

9

Delocation in the manufacturing sectors in the EU A regional overview

fi19

fi18

ie02

dec0

gr30

se11

nl33

de21

de22

de26

de25

de12

de14

de11

dea4

dea1de73

def0

de60

deb3

de27

de23

itd5

ite3

se12

nl22

es 30

ukd4

es51

nl32

at22

ukf1

at31

es21

Map 1 Regions with positive location effect ( of 2000 employment)

Above 10 0 to 10

nl41

de13de27

Nevertheless productivity also grows faster in most of the regions belonging to the

United Kingdom Belgium and Denmark in spite of the strong delocation effects registered

10

Delocation in the manufacturing sectors in the EU A regional overview

which disturbs the above-mentioned positive relationship suggesting the harsh process of

industrial restructuring involved

Figure 1 Location effect and productivity growth in select regions manufacturing sector 2001-2005 (percentages)

UKi2

EU 15

FR6 2

UKm2

ITc1

UKg3UKd 3

UKe4

UKk1FR71F R

FR10UKm3

UKj1UKf2

UKd 4

UKi1

D K

DK01

UKh1U K

FR2 3FR4 2

FR24ITe1

BE2 1ITe4FR51

FR61

ITITd 3

SE2 2FR8 2

SE23

NL2 2

UKf1DE2 4

NL3 2DEa1

DEf0

UKj3DE9 1 DE92

DEa2DE71DK0 4

S EDE25

DE6 0DE12

A T

DE14

ITe3

AT2 2

AT3 1ESD E

DE 11DE2 3

NL41N LDE2 7

DEc0

NL3 3

DE2 6DE2 1

ES3 0ES51

SE11

DE2 2

ES21

GR

FI19F I

FI18

IE

GR3 0

B E

DE73

DEb 3

IE0 2

ITd 5 DE13SE12

FR52ITc4BE2 3

DEa4DEa5

y = 07x - 59R2 = 06

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

40

-35 -30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30 35 40 45

Productivity grothw Source Elaborated from Eurostat and National Statistical offices

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

11

Delocation in the manufacturing sectors in the EU A regional overview

In contrast the productivity decline in half of the French and Italian regions albeit

with a positive impact on employment is largely offset by destruction of industrial

activity except in Icircle de France and Midi -Pyreacuteneacutees the two territories in which

productivity has a more negative trajectory

These results closely match those reported by Cuadrado Roura et al (2000) based on

the growth rates for regional productivity and employment during the period 1980-1993

grouping European regions in four typologies comparing the growth rates of these two

variables with their respective European averages

According to this approach and including the location effect in the analysis the

group of regions that are examined in this work would be categorized as follows (Table 2)

The first group includes areas with productivity and employment growth rates above

sample average and show positive location effect (Quadrant 1) So as has been underlined

in that work possibly this reflects the fact that the technological and organizational

changes introduced during the period would have been successful promoting the

diversification and specialization in activities with high growth levels and attracting new

investments which are less labour-intensive

In a second stage are those regions that as the above mentioned authors suggest

have implemented restructuring processes to eliminate the most inefficient outputs

achieving higher levels of productivity and in about half of them increased industrial

production at the expense of a notable decline in employment (Quadrant 4) They named

this typology `restructuring via productivityacute

With a few exceptions (those where the location effect is negative) the regions in

both groups are the winners of industrial dynamic during the period Most of them are

large industrial regions and are located in the traditional industrial growth centres As

demonstrated above they are mainly Finish German and Dutch as well as Greek and Irish

regions

In contrast another large group of regions due to different factors6 have opted for

more labour-intensive industrial models which in a few cases (especially in Spanish and

Austrian regions) have been accompanied by output increases so they can also be included

6 See Cuadrado et al (2000)

12

Delocation in the manufacturing sectors in the EU A regional overview

in the winners (Map 1) while in other the result has been a loss of productive activity

(Quadrant 2)

(1) (2)Employment Productivity Employment Productivity

de11 Stuttgart -48 125 at22 Steiermark 36 11de12 Karlsruhe -57 101 at31 Oberoumlsterreich 06 56de21 Oberbayern -37 179 be21 Prov Antwerpen -70 -05de22 Niederbayern -33 218 be23 Prov Oost-Vlaanderen -74 61de23 Oberpfalz -38 120 de13 Freiburg -45 45de25 Mittelfranken -53 80 de14 Tuumlbingen -23 72de26 Unterfranken -49 189 de91 Braunschweig -47 25de27 Schwaben -41 148 es21 Pais Vasco 74 13de60 Hamburg -60 97 es30 Comunidad de Madrid -31 63dec0 Saarland -32 160 es51 Cataluntildea 09 25fi19 Laumlnsi-Suomi -54 297 fr10 Icircle de France 218 -307gr30 Attiki -17 234 fr51 Pays de la Loire -08 -67

fr52 Bretagne -23 12fr61 Aquitaine -14 -50fr62 Midi-Pyreacuteneacutees 15 -245fr71 Rhocircne-Alpes -61 -57fr82 Provence-Alpes-Cocircte dAzur -06 -18itc4 Lombardia -63 61itd3 Veneto -62 03itd5 Emilia-Romagna -16 16ite3 Marche 08 28ite4 Lazio -36 -47se23 Vaumlstsverige -26 -07

(4) (3)Employment Productivity Employment Productivity

de24 Oberfranken -116 186 fr24 Centre -106 06de71 Darmstadt -149 132 fr42 Alsace -154 61de73 Kassel -78 103 itc1 Piemonte -129 -48de92 Hannover -106 86 ite1 Toscana -98 -02dea1 Duumlsseldorf -126 177 se22 Sydsverige -91 62dea2 Koumlln -150 137 uke4 West Yorkshire -186 60dea4 Detmold -107 110 ukh1 East Anglia -137 71dea5 Arnsberg -129 122 uki2 Outer London -271 -67deb3 Rheinhessen-Pfalz -106 152 ukj3 Hampshire and Isle of Wight -81 51def0 Schleswig-Holstein -101 135 ukk1 Gloucestershire Wiltshire -169 62dk01 Hovedstaden -134 125 ukm2 Eastern Scotland -91 -116dk04 Midtjylland -146 118fi18 Etelauml-Suomi -86 339fr23 Haute-Normandie -171 77ie02 Southern and Eastern -78 426nl22 Gelderland -178 206nl32 Noord-Holland -138 180nl33 Zuid-Holland -129 296nl41 Noord-Brabant -173 269se11 Stockholm -136 363se12 Oumlstra Mellansverige -151 166ukd3 Greater Manchester -232 90ukd4 Lancashire -317 350ukf1 Derbyshire and Nottinghamshire -179 240ukf2 Leicestershire Rutland and Northants -233 191ukg3 West Midlands -313 180uki1 Inner London -142 102ukj1 Berkshire Bucks and Oxfordshire -246 208ukm3 South Western Scotland -207 128

Source Elaborated from Eurostat and national Statistical officesNote Shaded areas indicate regions with positive location effect

Em

ploy

men

t gro

wth

lt A

vera

ge v

alue

Productivity growth gt Average value Productivity growth lt Average value

Em

ploy

men

t gro

wth

gt A

vera

ge v

alue

Table 2 Employment and productivity growth rates in select regions manufacturing sector 2001-2005 (Averages of all regions +74 productivity -77 employment)

13

Delocation in the manufacturing sectors in the EU A regional overview

Finally the worst results can be seen in a fourth type of region with growth rates of

GVA and productivity below average and serious job destruction (Quadrant 3) This would

occur in regions with problems in restructuring their manufacturing sector or with low

levels of activity So together with those that are either restructuring via productivity or

via employment but without achieving a positive location effect they are clearly the

losers These include mainly British French Italian Belgian and Danish regions

To sum up these results confirm the direct relationship between growth in labour

efficiency and progress in the creation of productive industrial activity as pointed out

above and which can be more clearly by looking at Figure 1 In other words regions with

higher levels of productivity growth in general show a greater attractiveness for new

manufacturing firms

These differences in the behaviour of productivity and employment across regions

have led to remarkable disparities in rates of change of their industrial output during the

period Thus while a few increased their real manufacturing GVA by more than 20 in

the less dynamic ones the loss of industrial output is between a sixth and a third of the

initial value So significant changes had taken place in the relative position of some of

them (Figure 2)

Growth is led by a group of twenty regions located in Ireland Finland Greece

Germany Sweden Netherlands Spain and Austria with real GVA growth rates that are

more than five points above average As a result they increased their share in the industrial

output of the sample Looking at the complete list we see that the growth deviation from

the mean exhibits positive values in just over half of the regions examined among them

seven of the ten with the largest industrial scale in 2005 Thus the top ten regions have

increased their share from 32 to 33 percent in the last five years

Conversely among the less dynamic regions are found a great number of British and

some French and Italian ones which consequently have fallen back in the ranking of

NUTS 2 arranged by their contribution to aggregate manufacturing output

Between both groups there are approximately twenty regions mainly in Germany

with growth rates not far from average so their positions have hardly changed during the

last five years

14

Delocation in the manufacturing sectors in the EU A regional overview

Figure 2 Regional shares in the total sample manufacturing GVA(percentage at 2000 prices)

ITc4

UKf2

UKm3

Uke4

UKf1

UKk1

UKh1

AT22

ITe1

ES21

ES51

FR71

FR62

IT e4

DE24

DE26

SE23

FR61

FR51

FR82

SE22

ITe3

AT31

DK04

DE11

DE23

GR30

DEc0

ES30DEb3DEf0

DEa4

DE92

BE21

DE73

IE02

ITd3

FR52

DE91

DE13DE25

DE 12DE60

DE27

DE21

DE22

BE23DEa2

DE71

DE 14

SE12

ITd5

SE11

FI19

FR10

FR 24

ITe1

FI18

DK01 DEa5

FR4 2

DEa1NL32

NL3 3

FR2 3

NL22

NL 41

UKd3

UKj3

UKi1

UKm2

UKj1

UKd4

UKi2

Ukg3

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

Share in 2005

Source Elaborated from Eurostat and National Statistical offices

Rat

e of

cha

nge

200

1-20

05

70403530252015100500

Finally when the question is analyzed at Member State level (Table 1) it is possible

to check out the disparities among the above mentioned regions Thus while in some

15

Delocation in the manufacturing sectors in the EU A regional overview

countries such as Spain manufacturing growth has been clearly labour-intensive

something which can be extended to its regions as also happens in Greece and Austria

which both boast positive location effect and a job destruction rate far below the average

in others such as Holland Sweden and to a lesser extent Belgium Ireland and Finland

increasing of industrial activity is only due by advances in productivity so has been

accompanied by heavy job losses and these are patterns repeated in most of their regions

with noticeable relevance in the Netherlands As for the United Kingdom Belgium and

Denmark all share declining industrial production together with strong job cuts a pattern

that is replicated in many of their regions In this situation are the bulk of French and

Italian NUTS 2 just like the countries to which they belong although in this case job

destruction has been much less pronounced and below the group average Close to this

mean is Germany as well as most of their regions

Thus the analysis at Member States level also confirms the positive relationship

between the advance of industrial productivity and the attractiveness for new

manufacturing firms As demonstrated for their regions Ireland Finland Netherlands and

Greece following at quite distance from Germany are Member States showing a better

performance in productivity and with high location effects (see Figure 1)

In fact similarities between performances in the regions and in the Member States

they belong to suggest that the trend in regional industrial employment is to some extent

influenced by belonging to the same country in other words the locational attractions of

each area and its effort to improve productivity is partly due to national specificities or

idiosyncratic factors In fact as shown in Appendix 3 most European industrialized

regions follow the location patterns of the country they belong to reproducing its

productivity behaviour too In summary the evolution of these variables in each region

seems to have a clear `country effectacute that is different behaviours motivated by specific

factors andor differentiated industrial policies

For example in order to correctly interpret the results for Spanish regions it can not

be overlooked that during the period analyzed as in the whole country regions have

absorbed large numbers of immigrants which has boosted the proliferation of lower

productivity and lower wages in industrial activities However it may be thought that this

phenomenon has tended to conceal productivity increases in larger and better equipped

16

Delocation in the manufacturing sectors in the EU A regional overview

firms which could have been significant in the light of their results7 On the other hand

countries like Finland or Ireland with the lowest population increase have opted for high-

tech manufacturing sectors as a way to improve national competitiveness8 Nonetheless

this positive location effect has not reached in these countriesrsquo NUTS 2 enough magnitude

to offset completely the job losses arising from productivity improvements

On the other hand this `national effectacute has also played an important role in

explaining the economic outcomes of European regions in terms of per capita GDP in

accordance with the results obtained by Cuadrado-Roura (2001) using the data available

for 109 EU regions over the 1977-1994 period

Table 3 Export growth in key sectors by countries 2000-2005 (percentages)Impact ITon Total

Spain 65 91 72 158 38 111 04Germany 116 138 61 111 111 130 14United Kingdom 09 85 29 32 23 61 02France -09 102 25 94 118 78 -01Ireland -09 -102 01 25 -34 77 -03Finland 47 152 45 49 56 73 13Italy 40 88 41 90 56 89 02EU-15 70 141 59 91 67 111 11Source Elaborated from Comtrade

Shoes TotalIT Auto Textile Clothes

This importance of the `national effectacute should allow us to make a better approach to

the determinants of above-mentioned delocation patterns in future research One of such

determinants might be found in the different importance and performance by countries of

some manufacturing branches particularly affected by the increase in global competition as

their development has being receiving strong support in emerging countries from both

domestic policies and the location of big multinational companies This is the case of IT

manufacturing sectors or that of automobiles clothes textiles and shoes (Sachwald 2004)

Table 3 shows how the exports of these key sectors have evolved in some of the European

countries considered here It is amazing to see the poor results registered by IT sectors in

United Kingdom France and Ireland compared to the other countries mentioned Germany

in particular This evolution could be behind the stronger effects of delocation in France

and the United Kingdom although it must have produced similar results in the case of

Ireland Furthermore on such a basis what has happened in Italy would remain

7 For details see Myro et al (2008) 8 See Aacutelvarez Loacutepez et al (2007)

17

Delocation in the manufacturing sectors in the EU A regional overview

unexplained as the selected key sectors do not show a bad performance in terms of exports

in the latter country

One reason for such contradictory results lies in the possibility that production

instead of export reflects accurately what has been happening in every sector in terms of

activity However Aacutelvarez et al (2007) show very similar patterns in production and

export specialization from 1995 to 2005 except in the case of Ireland which thus appears

as a very particular one This is because IT specialization in production increases at the

same time as it decreases in exports Perhaps the strong growth in domestic demand

prevents IT exports from growing The opposite case could be that of Italy where slow

growth in domestic demand might have been pushing up IT export9

Anyway further developments on the determinants of delocation patterns require the

preparation of disaggregated data on production a task beyond the objectives of this paper

That could allow us to update the available manufacturing specialization analysis As the

two countries exhibiting higher delocation effects France and the United Kingdom

showed the lowest rates of specialization compared with the EU (Milderfart and Overman

2002 Combes and Overman 2003) something could have changed in the last few years

which would merit a fresh look based on more recent data

On the other hand it is now clear that insofar as delocation in manufacturing

activities has affected each country in a different way a relocation process inside the EU

and between countries has taken place This process deserved to be deeply analyzed

including a perspective by sectors updating what we know about changes in the

distribution of manufacturing industry in the EU and offering the possibility of exploring

whether the European Common Policy has influenced them in some way as was already

suggested by Milderfart-Knarvik and Overman (2002)

9 In Finland IT activities grows much more in production than in exports but in both aspects a positive growth is registered

18

Delocation in the manufacturing sectors in the EU A regional overview

19

Figure 3 Employment growth rates in selected regions 2001-2005 manufacturing and total economy (percentages)

Ukg3

UKi2

UKd4

UKj1

UKm2

UKi1 UKj3

UKd3

NL 41

NL22

FR23

NL33

NL32

DEa1

FR42

DEa5

DK01

FI18

ITe1

FR 24FR10

FI19

SE11

ITd5FR52

DE 14

DE71

DEa2

BE23

DE22

DE21

DE27

IC c4

DE60DE 12

DE25

DE13

DE9 1SE12

ITd3

IE02

DE73

BE21

DE92DEa4

DEf0

DEb3

ES30

DEc0

GR30

DE23

DE11

DK04

AT31

ITe3

SE22

FR82

FR51FR61

SE2 3

DE26

DE24

IT e4

FR62

FR71

ES51

ES2 1

ITe1

AT22

UKh1

UKk1

UKf1

Uke4

UKm3

UKf2

-20

-15

-10

-5

0

5

10

15

20

25

-40 -30 -20 -10 0 10 20 30

Manufacturing

Source Elaborated from Eurostat and National Statistical offices

Tota

l em

ploy

men

t

Delocation in the manufacturing sectors in the EU A regional overview

One last question that deserves consideration is the effect of industrial delocation

process on the regional economies Surprisingly manufacturing job losses have not been

translated into total employment which exhibit positive growth for most of the chosen

regions with the exceptions of some Germans and all the Netherlands (Figure 3) That

means the regions have been able to compensate the negative effect on the employment

from shrinking manufacturing allocating more resources inside the services sectors In fact

a substitution of manufacturing by services seems to have happened as are the region

showing more delocation in manufacturing those that increase in a higher measure the

employment in the service sectors that is the French British and Italian However the

relation is not accurate as there are not a strict quantitative correlation region by region

between rates of delocation and increase in the employment in the service activities

These good results in terms of general employment shown by the regions affected in

a higher measure by manufacturing delocation suggest the convenience of a more general

analysis of regional economic growth beyond the manufacturing sector as some regions

strongly shrinking manufacturing activities in favour of those of services might be leading

an interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

5- Final remarks

Comparative analysis of location patterns for a large sample of European regions -

characterized by having an upper-middle level income and a strong industrial sector-

during the last few years has revealed that only half of them have shown locational

attractions for manufacturing activities The regional patterns found clearly show the

influence of `national effectacute the regions showing location effects corresponding to most

of those located in Austria Germany the Netherlands Ireland Finland Sweden Spain and

Greece

On the contrary the delocation process has affected another half of the regions

included most of them French British and Italian Therefore delocation has contributed to

concentrating the European manufacturing industry in the areas with higher market

potential in the centre of continental Europe It has contributed furthermore to relocating

manufacturing activities inside every country something that requires further analysis

from future research

20

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

8

Productivity Location Productivity Locationeffect effect effect effect

at22 Steiermark 372 -108 480 36 -11 47at31 Oberoumlsterreich 097 -837 934 06 -56 62

Austria -1104 -2649 1545 -18 -42 25be21 Prov Antwerpen -1010 079 -1089 -70 05 -75be23 Prov Oost-Vlaanderen -750 -627 -123 -74 -61 -12

Belgium -5740 -2219 -3521 -84 -32 -51de11 Stuttgart -3157 -8244 5087 -48 -125 77de12 Karlsruhe -2036 -3585 1550 -57 -101 44de13 Freiburg -1291 -1301 011 -45 -45 00de14 Tuumlbingen -585 -1805 1219 -23 -72 49de21 Oberbayern -1630 -7976 6346 -37 -179 142de22 Niederbayern -490 -3270 2781 -33 -218 186de23 Oberpfalz -538 -1716 1178 -38 -120 83de24 Oberfranken -1957 -3131 1175 -116 -186 70de25 Mittelfranken -1184 -1770 586 -53 -80 26de26 Unterfranken -819 -3157 2338 -49 -189 140de27 Schwaben -932 -3335 2403 -41 -148 106de60 Hamburg -760 -1222 463 -60 -97 37de71 Darmstadt -5343 -4720 -623 -149 -132 -17de73 Kassel -1059 -1396 336 -78 -103 25de91 Braunschweig -947 -500 -447 -47 -25 -22de92 Hannover -1900 -1544 -356 -106 -86 -20dea1 Duumlsseldorf -6272 -8787 2515 -126 -177 51dea2 Koumlln -5428 -4971 -457 -150 -137 -13dea4 Detmold -3026 -3112 086 -107 -110 03dea5 Arnsberg -5824 -5502 -322 -129 -122 -07deb3 Rheinhessen-Pfalz -2099 -3016 917 -106 -152 46dec0 Saarland -360 -1810 1450 -32 -160 128def0 Schleswig-Holstein -1867 -2485 617 -101 -135 33

Germany -60300 -112275 51975 -74 -138 64dk01 Hovedstaden -1300 -1216 -084 -134 -125 -09dk04 Midtjylland -2000 -1622 -378 -146 -118 -28

Denmark -6100 -3428 -2672 -130 -73 -57es21 Pais Vasco 1840 -331 2171 74 -13 87es30 Comunidad de Madrid -1060 -2180 1120 -31 -63 32es51 Cataluntildea 660 -1914 2574 09 -25 34

Spain 13970 -3927 17897 47 -13 60fi18 Etelauml-Suomi -1807 -7159 5352 -86 -339 253fi19 Laumlnsi-Suomi -701 -3886 3185 -54 -297 244

Finland -1798 -12006 10208 -43 -286 243fr10 Icircle de France 12652 17868 -5216 218 307 -90fr23 Haute-Normandie -2418 -1093 -1325 -171 -77 -94fr24 Centre -2016 -120 -1896 -106 -06 -100fr42 Alsace -2489 -988 -1501 -154 -61 -93fr51 Pays de la Loire -217 1794 -2011 -08 67 -75fr52 Bretagne -415 -215 -200 -23 -12 -11fr61 Aquitaine -199 742 -941 -14 50 -64fr62 Midi-Pyreacuteneacutees 209 3318 -3109 15 245 -229fr71 Rhocircne-Alpes -2919 2759 -5678 -61 57 -118fr82 Provence-Alpes-Cocircte dAzur -097 262 -359 -06 18 -24

France -15306 31566 -46872 -40 83 -123gr30 Attiki -360 -5072 4712 -17 -234 217

Greece -730 -11989 11259 -14 -222 208ie02 Southern and Eastern -1840 -9985 8145 -78 -426 347

Ireland -2420 -12274 9854 -76 -387 311itc1 Piemonte -6869 2539 -9407 -129 48 -177itc4 Lombardia -7940 -7682 -258 -63 -61 -02itd3 Veneto -3967 -211 -3756 -62 -03 -58itd5 Emilia-Romagna -860 -840 -020 -16 -16 00ite1 Toscana -3620 078 -3697 -98 02 -101ite3 Marche 164 -555 719 08 -28 36ite4 Lazio -733 956 -1689 -36 47 -82

Italy -21761 6310 -28071 -45 13 -58nl22 Gelderland -2164 -2506 341 -178 -206 28nl32 Noord-Holland -1555 -2034 479 -138 -180 42nl33 Zuid-Holland -1734 -3980 2246 -129 -296 167nl41 Noord-Brabant -3539 -5515 1976 -173 -269 96

Netherlands -14767 -24690 9923 -161 -270 108se11 Stockholm -1570 -4183 2613 -136 -363 227se12 Oumlstra Mellansverige -2170 -2387 217 -151 -166 15se22 Sydsverige -1010 -690 -320 -91 -62 -29se23 Vaumlstsverige -460 134 -594 -26 07 -33

Sweden -7500 -9400 1900 -95 -119 24ukd3 Greater Manchester -4900 -1907 -2993 -232 -90 -142ukd4 Lancashire -4600 -5082 482 -317 -350 33uke4 West Yorkshire -3700 -1192 -2508 -186 -60 -126ukf1 Derbyshire and Nottinghamshire -3700 -4965 1265 -179 -240 61ukf2 Leicestershire Rutland and Northan -4400 -3607 -793 -233 -191 -42ukg3 West Midlands -8700 -4999 -3701 -313 -180 -133ukh1 East Anglia -2500 -1292 -1208 -137 -71 -66uki1 Inner London -1919 -1381 -538 -142 -102 -40uki2 Outer London -4481 1106 -5587 -271 67 -338ukj1 Berkshire Bucks and Oxfordshire -4600 -3898 -702 -246 -208 -38ukj3 Hampshire and Isle of Wight -1100 -689 -411 -81 -51 -30ukk1 Gloucestershire Wiltshire and Brist -3400 -1253 -2147 -169 -62 -107ukm2 Eastern Scotland -1100 1402 -2502 -91 116 -207ukm3 South Western Scotland -3400 -2107 -1293 -207 -128 -79

United Kingdom -89800 -48574 -41226 -196 -106 -90

Table 1 Importance of manufacturing delocation in European regions 2001-2005 (Decomposition of change in employment)

Total effect

Source Elaborated from Eurostat and National Statistical offices

Number of jobs (thousands)

Total effect

Percent share of 2000 employmentRegions

Delocation in the manufacturing sectors in the EU A regional overview

By contrast thirty-eight remaining regions that also show reductions in

manufacturing employment have been affected by delocation processes to different

extents although only in a small number of them have they been accompanied by

decreases in labour productivity which could make the prospects for economic growth

tougher (four from France Piamonte Toscana and Lazio in Italy Outer London and

Eastern Scotland in the United Kingdom the Swedish Vaumlstsverige and Antwerp in

Belgium)

Regions with positive location effects are shown in Map 1 Most of them are large

industrial regions and are located in the traditional industrial growth centres5 those

characterized by high market potential (Combes and Overman 2003) although there are

also some intermediate and peripheral territories mainly located in the north of Italy

United Kingdom and Spain and in the south of Sweden and Norway This means that the

delocation process has contributed to concentrating the European manufacturing industries

in that privileged area of faster growth apart from some other changes inside every

country

To compare now the intensity of such effects of location and delocation across

regions avoiding the differences they have in size we built a relative measure of them the

ratio of job losses in the period to the volume of employment in the first year 2000 (Table

1)

Following this ratio delocation effects are revealed to be particularly strong in

France United Kingdom Italy and in one region of Belgium with declines in

employment exceeding 20 in some of them

Conversely the regions of Finland and several of the German and the Dutch ones

exhibit strong location effects

In any case as was pointed above the importance of location effects seems to lie in

labour productivity gains In fact higher increases in productivity (over 15) are found

precisely in those regions of Ireland Finland Sweden Netherlands and Germany besides

the Greek Attiki where the location effect has in general reached its greatest extent

(Figure 1)

5 Note that the map of winning regions is clearly influenced by the large number (30) of German NUTS 2 in the sample

9

Delocation in the manufacturing sectors in the EU A regional overview

fi19

fi18

ie02

dec0

gr30

se11

nl33

de21

de22

de26

de25

de12

de14

de11

dea4

dea1de73

def0

de60

deb3

de27

de23

itd5

ite3

se12

nl22

es 30

ukd4

es51

nl32

at22

ukf1

at31

es21

Map 1 Regions with positive location effect ( of 2000 employment)

Above 10 0 to 10

nl41

de13de27

Nevertheless productivity also grows faster in most of the regions belonging to the

United Kingdom Belgium and Denmark in spite of the strong delocation effects registered

10

Delocation in the manufacturing sectors in the EU A regional overview

which disturbs the above-mentioned positive relationship suggesting the harsh process of

industrial restructuring involved

Figure 1 Location effect and productivity growth in select regions manufacturing sector 2001-2005 (percentages)

UKi2

EU 15

FR6 2

UKm2

ITc1

UKg3UKd 3

UKe4

UKk1FR71F R

FR10UKm3

UKj1UKf2

UKd 4

UKi1

D K

DK01

UKh1U K

FR2 3FR4 2

FR24ITe1

BE2 1ITe4FR51

FR61

ITITd 3

SE2 2FR8 2

SE23

NL2 2

UKf1DE2 4

NL3 2DEa1

DEf0

UKj3DE9 1 DE92

DEa2DE71DK0 4

S EDE25

DE6 0DE12

A T

DE14

ITe3

AT2 2

AT3 1ESD E

DE 11DE2 3

NL41N LDE2 7

DEc0

NL3 3

DE2 6DE2 1

ES3 0ES51

SE11

DE2 2

ES21

GR

FI19F I

FI18

IE

GR3 0

B E

DE73

DEb 3

IE0 2

ITd 5 DE13SE12

FR52ITc4BE2 3

DEa4DEa5

y = 07x - 59R2 = 06

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

40

-35 -30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30 35 40 45

Productivity grothw Source Elaborated from Eurostat and National Statistical offices

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

11

Delocation in the manufacturing sectors in the EU A regional overview

In contrast the productivity decline in half of the French and Italian regions albeit

with a positive impact on employment is largely offset by destruction of industrial

activity except in Icircle de France and Midi -Pyreacuteneacutees the two territories in which

productivity has a more negative trajectory

These results closely match those reported by Cuadrado Roura et al (2000) based on

the growth rates for regional productivity and employment during the period 1980-1993

grouping European regions in four typologies comparing the growth rates of these two

variables with their respective European averages

According to this approach and including the location effect in the analysis the

group of regions that are examined in this work would be categorized as follows (Table 2)

The first group includes areas with productivity and employment growth rates above

sample average and show positive location effect (Quadrant 1) So as has been underlined

in that work possibly this reflects the fact that the technological and organizational

changes introduced during the period would have been successful promoting the

diversification and specialization in activities with high growth levels and attracting new

investments which are less labour-intensive

In a second stage are those regions that as the above mentioned authors suggest

have implemented restructuring processes to eliminate the most inefficient outputs

achieving higher levels of productivity and in about half of them increased industrial

production at the expense of a notable decline in employment (Quadrant 4) They named

this typology `restructuring via productivityacute

With a few exceptions (those where the location effect is negative) the regions in

both groups are the winners of industrial dynamic during the period Most of them are

large industrial regions and are located in the traditional industrial growth centres As

demonstrated above they are mainly Finish German and Dutch as well as Greek and Irish

regions

In contrast another large group of regions due to different factors6 have opted for

more labour-intensive industrial models which in a few cases (especially in Spanish and

Austrian regions) have been accompanied by output increases so they can also be included

6 See Cuadrado et al (2000)

12

Delocation in the manufacturing sectors in the EU A regional overview

in the winners (Map 1) while in other the result has been a loss of productive activity

(Quadrant 2)

(1) (2)Employment Productivity Employment Productivity

de11 Stuttgart -48 125 at22 Steiermark 36 11de12 Karlsruhe -57 101 at31 Oberoumlsterreich 06 56de21 Oberbayern -37 179 be21 Prov Antwerpen -70 -05de22 Niederbayern -33 218 be23 Prov Oost-Vlaanderen -74 61de23 Oberpfalz -38 120 de13 Freiburg -45 45de25 Mittelfranken -53 80 de14 Tuumlbingen -23 72de26 Unterfranken -49 189 de91 Braunschweig -47 25de27 Schwaben -41 148 es21 Pais Vasco 74 13de60 Hamburg -60 97 es30 Comunidad de Madrid -31 63dec0 Saarland -32 160 es51 Cataluntildea 09 25fi19 Laumlnsi-Suomi -54 297 fr10 Icircle de France 218 -307gr30 Attiki -17 234 fr51 Pays de la Loire -08 -67

fr52 Bretagne -23 12fr61 Aquitaine -14 -50fr62 Midi-Pyreacuteneacutees 15 -245fr71 Rhocircne-Alpes -61 -57fr82 Provence-Alpes-Cocircte dAzur -06 -18itc4 Lombardia -63 61itd3 Veneto -62 03itd5 Emilia-Romagna -16 16ite3 Marche 08 28ite4 Lazio -36 -47se23 Vaumlstsverige -26 -07

(4) (3)Employment Productivity Employment Productivity

de24 Oberfranken -116 186 fr24 Centre -106 06de71 Darmstadt -149 132 fr42 Alsace -154 61de73 Kassel -78 103 itc1 Piemonte -129 -48de92 Hannover -106 86 ite1 Toscana -98 -02dea1 Duumlsseldorf -126 177 se22 Sydsverige -91 62dea2 Koumlln -150 137 uke4 West Yorkshire -186 60dea4 Detmold -107 110 ukh1 East Anglia -137 71dea5 Arnsberg -129 122 uki2 Outer London -271 -67deb3 Rheinhessen-Pfalz -106 152 ukj3 Hampshire and Isle of Wight -81 51def0 Schleswig-Holstein -101 135 ukk1 Gloucestershire Wiltshire -169 62dk01 Hovedstaden -134 125 ukm2 Eastern Scotland -91 -116dk04 Midtjylland -146 118fi18 Etelauml-Suomi -86 339fr23 Haute-Normandie -171 77ie02 Southern and Eastern -78 426nl22 Gelderland -178 206nl32 Noord-Holland -138 180nl33 Zuid-Holland -129 296nl41 Noord-Brabant -173 269se11 Stockholm -136 363se12 Oumlstra Mellansverige -151 166ukd3 Greater Manchester -232 90ukd4 Lancashire -317 350ukf1 Derbyshire and Nottinghamshire -179 240ukf2 Leicestershire Rutland and Northants -233 191ukg3 West Midlands -313 180uki1 Inner London -142 102ukj1 Berkshire Bucks and Oxfordshire -246 208ukm3 South Western Scotland -207 128

Source Elaborated from Eurostat and national Statistical officesNote Shaded areas indicate regions with positive location effect

Em

ploy

men

t gro

wth

lt A

vera

ge v

alue

Productivity growth gt Average value Productivity growth lt Average value

Em

ploy

men

t gro

wth

gt A

vera

ge v

alue

Table 2 Employment and productivity growth rates in select regions manufacturing sector 2001-2005 (Averages of all regions +74 productivity -77 employment)

13

Delocation in the manufacturing sectors in the EU A regional overview

Finally the worst results can be seen in a fourth type of region with growth rates of

GVA and productivity below average and serious job destruction (Quadrant 3) This would

occur in regions with problems in restructuring their manufacturing sector or with low

levels of activity So together with those that are either restructuring via productivity or

via employment but without achieving a positive location effect they are clearly the

losers These include mainly British French Italian Belgian and Danish regions

To sum up these results confirm the direct relationship between growth in labour

efficiency and progress in the creation of productive industrial activity as pointed out

above and which can be more clearly by looking at Figure 1 In other words regions with

higher levels of productivity growth in general show a greater attractiveness for new

manufacturing firms

These differences in the behaviour of productivity and employment across regions

have led to remarkable disparities in rates of change of their industrial output during the

period Thus while a few increased their real manufacturing GVA by more than 20 in

the less dynamic ones the loss of industrial output is between a sixth and a third of the

initial value So significant changes had taken place in the relative position of some of

them (Figure 2)

Growth is led by a group of twenty regions located in Ireland Finland Greece

Germany Sweden Netherlands Spain and Austria with real GVA growth rates that are

more than five points above average As a result they increased their share in the industrial

output of the sample Looking at the complete list we see that the growth deviation from

the mean exhibits positive values in just over half of the regions examined among them

seven of the ten with the largest industrial scale in 2005 Thus the top ten regions have

increased their share from 32 to 33 percent in the last five years

Conversely among the less dynamic regions are found a great number of British and

some French and Italian ones which consequently have fallen back in the ranking of

NUTS 2 arranged by their contribution to aggregate manufacturing output

Between both groups there are approximately twenty regions mainly in Germany

with growth rates not far from average so their positions have hardly changed during the

last five years

14

Delocation in the manufacturing sectors in the EU A regional overview

Figure 2 Regional shares in the total sample manufacturing GVA(percentage at 2000 prices)

ITc4

UKf2

UKm3

Uke4

UKf1

UKk1

UKh1

AT22

ITe1

ES21

ES51

FR71

FR62

IT e4

DE24

DE26

SE23

FR61

FR51

FR82

SE22

ITe3

AT31

DK04

DE11

DE23

GR30

DEc0

ES30DEb3DEf0

DEa4

DE92

BE21

DE73

IE02

ITd3

FR52

DE91

DE13DE25

DE 12DE60

DE27

DE21

DE22

BE23DEa2

DE71

DE 14

SE12

ITd5

SE11

FI19

FR10

FR 24

ITe1

FI18

DK01 DEa5

FR4 2

DEa1NL32

NL3 3

FR2 3

NL22

NL 41

UKd3

UKj3

UKi1

UKm2

UKj1

UKd4

UKi2

Ukg3

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

Share in 2005

Source Elaborated from Eurostat and National Statistical offices

Rat

e of

cha

nge

200

1-20

05

70403530252015100500

Finally when the question is analyzed at Member State level (Table 1) it is possible

to check out the disparities among the above mentioned regions Thus while in some

15

Delocation in the manufacturing sectors in the EU A regional overview

countries such as Spain manufacturing growth has been clearly labour-intensive

something which can be extended to its regions as also happens in Greece and Austria

which both boast positive location effect and a job destruction rate far below the average

in others such as Holland Sweden and to a lesser extent Belgium Ireland and Finland

increasing of industrial activity is only due by advances in productivity so has been

accompanied by heavy job losses and these are patterns repeated in most of their regions

with noticeable relevance in the Netherlands As for the United Kingdom Belgium and

Denmark all share declining industrial production together with strong job cuts a pattern

that is replicated in many of their regions In this situation are the bulk of French and

Italian NUTS 2 just like the countries to which they belong although in this case job

destruction has been much less pronounced and below the group average Close to this

mean is Germany as well as most of their regions

Thus the analysis at Member States level also confirms the positive relationship

between the advance of industrial productivity and the attractiveness for new

manufacturing firms As demonstrated for their regions Ireland Finland Netherlands and

Greece following at quite distance from Germany are Member States showing a better

performance in productivity and with high location effects (see Figure 1)

In fact similarities between performances in the regions and in the Member States

they belong to suggest that the trend in regional industrial employment is to some extent

influenced by belonging to the same country in other words the locational attractions of

each area and its effort to improve productivity is partly due to national specificities or

idiosyncratic factors In fact as shown in Appendix 3 most European industrialized

regions follow the location patterns of the country they belong to reproducing its

productivity behaviour too In summary the evolution of these variables in each region

seems to have a clear `country effectacute that is different behaviours motivated by specific

factors andor differentiated industrial policies

For example in order to correctly interpret the results for Spanish regions it can not

be overlooked that during the period analyzed as in the whole country regions have

absorbed large numbers of immigrants which has boosted the proliferation of lower

productivity and lower wages in industrial activities However it may be thought that this

phenomenon has tended to conceal productivity increases in larger and better equipped

16

Delocation in the manufacturing sectors in the EU A regional overview

firms which could have been significant in the light of their results7 On the other hand

countries like Finland or Ireland with the lowest population increase have opted for high-

tech manufacturing sectors as a way to improve national competitiveness8 Nonetheless

this positive location effect has not reached in these countriesrsquo NUTS 2 enough magnitude

to offset completely the job losses arising from productivity improvements

On the other hand this `national effectacute has also played an important role in

explaining the economic outcomes of European regions in terms of per capita GDP in

accordance with the results obtained by Cuadrado-Roura (2001) using the data available

for 109 EU regions over the 1977-1994 period

Table 3 Export growth in key sectors by countries 2000-2005 (percentages)Impact ITon Total

Spain 65 91 72 158 38 111 04Germany 116 138 61 111 111 130 14United Kingdom 09 85 29 32 23 61 02France -09 102 25 94 118 78 -01Ireland -09 -102 01 25 -34 77 -03Finland 47 152 45 49 56 73 13Italy 40 88 41 90 56 89 02EU-15 70 141 59 91 67 111 11Source Elaborated from Comtrade

Shoes TotalIT Auto Textile Clothes

This importance of the `national effectacute should allow us to make a better approach to

the determinants of above-mentioned delocation patterns in future research One of such

determinants might be found in the different importance and performance by countries of

some manufacturing branches particularly affected by the increase in global competition as

their development has being receiving strong support in emerging countries from both

domestic policies and the location of big multinational companies This is the case of IT

manufacturing sectors or that of automobiles clothes textiles and shoes (Sachwald 2004)

Table 3 shows how the exports of these key sectors have evolved in some of the European

countries considered here It is amazing to see the poor results registered by IT sectors in

United Kingdom France and Ireland compared to the other countries mentioned Germany

in particular This evolution could be behind the stronger effects of delocation in France

and the United Kingdom although it must have produced similar results in the case of

Ireland Furthermore on such a basis what has happened in Italy would remain

7 For details see Myro et al (2008) 8 See Aacutelvarez Loacutepez et al (2007)

17

Delocation in the manufacturing sectors in the EU A regional overview

unexplained as the selected key sectors do not show a bad performance in terms of exports

in the latter country

One reason for such contradictory results lies in the possibility that production

instead of export reflects accurately what has been happening in every sector in terms of

activity However Aacutelvarez et al (2007) show very similar patterns in production and

export specialization from 1995 to 2005 except in the case of Ireland which thus appears

as a very particular one This is because IT specialization in production increases at the

same time as it decreases in exports Perhaps the strong growth in domestic demand

prevents IT exports from growing The opposite case could be that of Italy where slow

growth in domestic demand might have been pushing up IT export9

Anyway further developments on the determinants of delocation patterns require the

preparation of disaggregated data on production a task beyond the objectives of this paper

That could allow us to update the available manufacturing specialization analysis As the

two countries exhibiting higher delocation effects France and the United Kingdom

showed the lowest rates of specialization compared with the EU (Milderfart and Overman

2002 Combes and Overman 2003) something could have changed in the last few years

which would merit a fresh look based on more recent data

On the other hand it is now clear that insofar as delocation in manufacturing

activities has affected each country in a different way a relocation process inside the EU

and between countries has taken place This process deserved to be deeply analyzed

including a perspective by sectors updating what we know about changes in the

distribution of manufacturing industry in the EU and offering the possibility of exploring

whether the European Common Policy has influenced them in some way as was already

suggested by Milderfart-Knarvik and Overman (2002)

9 In Finland IT activities grows much more in production than in exports but in both aspects a positive growth is registered

18

Delocation in the manufacturing sectors in the EU A regional overview

19

Figure 3 Employment growth rates in selected regions 2001-2005 manufacturing and total economy (percentages)

Ukg3

UKi2

UKd4

UKj1

UKm2

UKi1 UKj3

UKd3

NL 41

NL22

FR23

NL33

NL32

DEa1

FR42

DEa5

DK01

FI18

ITe1

FR 24FR10

FI19

SE11

ITd5FR52

DE 14

DE71

DEa2

BE23

DE22

DE21

DE27

IC c4

DE60DE 12

DE25

DE13

DE9 1SE12

ITd3

IE02

DE73

BE21

DE92DEa4

DEf0

DEb3

ES30

DEc0

GR30

DE23

DE11

DK04

AT31

ITe3

SE22

FR82

FR51FR61

SE2 3

DE26

DE24

IT e4

FR62

FR71

ES51

ES2 1

ITe1

AT22

UKh1

UKk1

UKf1

Uke4

UKm3

UKf2

-20

-15

-10

-5

0

5

10

15

20

25

-40 -30 -20 -10 0 10 20 30

Manufacturing

Source Elaborated from Eurostat and National Statistical offices

Tota

l em

ploy

men

t

Delocation in the manufacturing sectors in the EU A regional overview

One last question that deserves consideration is the effect of industrial delocation

process on the regional economies Surprisingly manufacturing job losses have not been

translated into total employment which exhibit positive growth for most of the chosen

regions with the exceptions of some Germans and all the Netherlands (Figure 3) That

means the regions have been able to compensate the negative effect on the employment

from shrinking manufacturing allocating more resources inside the services sectors In fact

a substitution of manufacturing by services seems to have happened as are the region

showing more delocation in manufacturing those that increase in a higher measure the

employment in the service sectors that is the French British and Italian However the

relation is not accurate as there are not a strict quantitative correlation region by region

between rates of delocation and increase in the employment in the service activities

These good results in terms of general employment shown by the regions affected in

a higher measure by manufacturing delocation suggest the convenience of a more general

analysis of regional economic growth beyond the manufacturing sector as some regions

strongly shrinking manufacturing activities in favour of those of services might be leading

an interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

5- Final remarks

Comparative analysis of location patterns for a large sample of European regions -

characterized by having an upper-middle level income and a strong industrial sector-

during the last few years has revealed that only half of them have shown locational

attractions for manufacturing activities The regional patterns found clearly show the

influence of `national effectacute the regions showing location effects corresponding to most

of those located in Austria Germany the Netherlands Ireland Finland Sweden Spain and

Greece

On the contrary the delocation process has affected another half of the regions

included most of them French British and Italian Therefore delocation has contributed to

concentrating the European manufacturing industry in the areas with higher market

potential in the centre of continental Europe It has contributed furthermore to relocating

manufacturing activities inside every country something that requires further analysis

from future research

20

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

By contrast thirty-eight remaining regions that also show reductions in

manufacturing employment have been affected by delocation processes to different

extents although only in a small number of them have they been accompanied by

decreases in labour productivity which could make the prospects for economic growth

tougher (four from France Piamonte Toscana and Lazio in Italy Outer London and

Eastern Scotland in the United Kingdom the Swedish Vaumlstsverige and Antwerp in

Belgium)

Regions with positive location effects are shown in Map 1 Most of them are large

industrial regions and are located in the traditional industrial growth centres5 those

characterized by high market potential (Combes and Overman 2003) although there are

also some intermediate and peripheral territories mainly located in the north of Italy

United Kingdom and Spain and in the south of Sweden and Norway This means that the

delocation process has contributed to concentrating the European manufacturing industries

in that privileged area of faster growth apart from some other changes inside every

country

To compare now the intensity of such effects of location and delocation across

regions avoiding the differences they have in size we built a relative measure of them the

ratio of job losses in the period to the volume of employment in the first year 2000 (Table

1)

Following this ratio delocation effects are revealed to be particularly strong in

France United Kingdom Italy and in one region of Belgium with declines in

employment exceeding 20 in some of them

Conversely the regions of Finland and several of the German and the Dutch ones

exhibit strong location effects

In any case as was pointed above the importance of location effects seems to lie in

labour productivity gains In fact higher increases in productivity (over 15) are found

precisely in those regions of Ireland Finland Sweden Netherlands and Germany besides

the Greek Attiki where the location effect has in general reached its greatest extent

(Figure 1)

5 Note that the map of winning regions is clearly influenced by the large number (30) of German NUTS 2 in the sample

9

Delocation in the manufacturing sectors in the EU A regional overview

fi19

fi18

ie02

dec0

gr30

se11

nl33

de21

de22

de26

de25

de12

de14

de11

dea4

dea1de73

def0

de60

deb3

de27

de23

itd5

ite3

se12

nl22

es 30

ukd4

es51

nl32

at22

ukf1

at31

es21

Map 1 Regions with positive location effect ( of 2000 employment)

Above 10 0 to 10

nl41

de13de27

Nevertheless productivity also grows faster in most of the regions belonging to the

United Kingdom Belgium and Denmark in spite of the strong delocation effects registered

10

Delocation in the manufacturing sectors in the EU A regional overview

which disturbs the above-mentioned positive relationship suggesting the harsh process of

industrial restructuring involved

Figure 1 Location effect and productivity growth in select regions manufacturing sector 2001-2005 (percentages)

UKi2

EU 15

FR6 2

UKm2

ITc1

UKg3UKd 3

UKe4

UKk1FR71F R

FR10UKm3

UKj1UKf2

UKd 4

UKi1

D K

DK01

UKh1U K

FR2 3FR4 2

FR24ITe1

BE2 1ITe4FR51

FR61

ITITd 3

SE2 2FR8 2

SE23

NL2 2

UKf1DE2 4

NL3 2DEa1

DEf0

UKj3DE9 1 DE92

DEa2DE71DK0 4

S EDE25

DE6 0DE12

A T

DE14

ITe3

AT2 2

AT3 1ESD E

DE 11DE2 3

NL41N LDE2 7

DEc0

NL3 3

DE2 6DE2 1

ES3 0ES51

SE11

DE2 2

ES21

GR

FI19F I

FI18

IE

GR3 0

B E

DE73

DEb 3

IE0 2

ITd 5 DE13SE12

FR52ITc4BE2 3

DEa4DEa5

y = 07x - 59R2 = 06

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

40

-35 -30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30 35 40 45

Productivity grothw Source Elaborated from Eurostat and National Statistical offices

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

11

Delocation in the manufacturing sectors in the EU A regional overview

In contrast the productivity decline in half of the French and Italian regions albeit

with a positive impact on employment is largely offset by destruction of industrial

activity except in Icircle de France and Midi -Pyreacuteneacutees the two territories in which

productivity has a more negative trajectory

These results closely match those reported by Cuadrado Roura et al (2000) based on

the growth rates for regional productivity and employment during the period 1980-1993

grouping European regions in four typologies comparing the growth rates of these two

variables with their respective European averages

According to this approach and including the location effect in the analysis the

group of regions that are examined in this work would be categorized as follows (Table 2)

The first group includes areas with productivity and employment growth rates above

sample average and show positive location effect (Quadrant 1) So as has been underlined

in that work possibly this reflects the fact that the technological and organizational

changes introduced during the period would have been successful promoting the

diversification and specialization in activities with high growth levels and attracting new

investments which are less labour-intensive

In a second stage are those regions that as the above mentioned authors suggest

have implemented restructuring processes to eliminate the most inefficient outputs

achieving higher levels of productivity and in about half of them increased industrial

production at the expense of a notable decline in employment (Quadrant 4) They named

this typology `restructuring via productivityacute

With a few exceptions (those where the location effect is negative) the regions in

both groups are the winners of industrial dynamic during the period Most of them are

large industrial regions and are located in the traditional industrial growth centres As

demonstrated above they are mainly Finish German and Dutch as well as Greek and Irish

regions

In contrast another large group of regions due to different factors6 have opted for

more labour-intensive industrial models which in a few cases (especially in Spanish and

Austrian regions) have been accompanied by output increases so they can also be included

6 See Cuadrado et al (2000)

12

Delocation in the manufacturing sectors in the EU A regional overview

in the winners (Map 1) while in other the result has been a loss of productive activity

(Quadrant 2)

(1) (2)Employment Productivity Employment Productivity

de11 Stuttgart -48 125 at22 Steiermark 36 11de12 Karlsruhe -57 101 at31 Oberoumlsterreich 06 56de21 Oberbayern -37 179 be21 Prov Antwerpen -70 -05de22 Niederbayern -33 218 be23 Prov Oost-Vlaanderen -74 61de23 Oberpfalz -38 120 de13 Freiburg -45 45de25 Mittelfranken -53 80 de14 Tuumlbingen -23 72de26 Unterfranken -49 189 de91 Braunschweig -47 25de27 Schwaben -41 148 es21 Pais Vasco 74 13de60 Hamburg -60 97 es30 Comunidad de Madrid -31 63dec0 Saarland -32 160 es51 Cataluntildea 09 25fi19 Laumlnsi-Suomi -54 297 fr10 Icircle de France 218 -307gr30 Attiki -17 234 fr51 Pays de la Loire -08 -67

fr52 Bretagne -23 12fr61 Aquitaine -14 -50fr62 Midi-Pyreacuteneacutees 15 -245fr71 Rhocircne-Alpes -61 -57fr82 Provence-Alpes-Cocircte dAzur -06 -18itc4 Lombardia -63 61itd3 Veneto -62 03itd5 Emilia-Romagna -16 16ite3 Marche 08 28ite4 Lazio -36 -47se23 Vaumlstsverige -26 -07

(4) (3)Employment Productivity Employment Productivity

de24 Oberfranken -116 186 fr24 Centre -106 06de71 Darmstadt -149 132 fr42 Alsace -154 61de73 Kassel -78 103 itc1 Piemonte -129 -48de92 Hannover -106 86 ite1 Toscana -98 -02dea1 Duumlsseldorf -126 177 se22 Sydsverige -91 62dea2 Koumlln -150 137 uke4 West Yorkshire -186 60dea4 Detmold -107 110 ukh1 East Anglia -137 71dea5 Arnsberg -129 122 uki2 Outer London -271 -67deb3 Rheinhessen-Pfalz -106 152 ukj3 Hampshire and Isle of Wight -81 51def0 Schleswig-Holstein -101 135 ukk1 Gloucestershire Wiltshire -169 62dk01 Hovedstaden -134 125 ukm2 Eastern Scotland -91 -116dk04 Midtjylland -146 118fi18 Etelauml-Suomi -86 339fr23 Haute-Normandie -171 77ie02 Southern and Eastern -78 426nl22 Gelderland -178 206nl32 Noord-Holland -138 180nl33 Zuid-Holland -129 296nl41 Noord-Brabant -173 269se11 Stockholm -136 363se12 Oumlstra Mellansverige -151 166ukd3 Greater Manchester -232 90ukd4 Lancashire -317 350ukf1 Derbyshire and Nottinghamshire -179 240ukf2 Leicestershire Rutland and Northants -233 191ukg3 West Midlands -313 180uki1 Inner London -142 102ukj1 Berkshire Bucks and Oxfordshire -246 208ukm3 South Western Scotland -207 128

Source Elaborated from Eurostat and national Statistical officesNote Shaded areas indicate regions with positive location effect

Em

ploy

men

t gro

wth

lt A

vera

ge v

alue

Productivity growth gt Average value Productivity growth lt Average value

Em

ploy

men

t gro

wth

gt A

vera

ge v

alue

Table 2 Employment and productivity growth rates in select regions manufacturing sector 2001-2005 (Averages of all regions +74 productivity -77 employment)

13

Delocation in the manufacturing sectors in the EU A regional overview

Finally the worst results can be seen in a fourth type of region with growth rates of

GVA and productivity below average and serious job destruction (Quadrant 3) This would

occur in regions with problems in restructuring their manufacturing sector or with low

levels of activity So together with those that are either restructuring via productivity or

via employment but without achieving a positive location effect they are clearly the

losers These include mainly British French Italian Belgian and Danish regions

To sum up these results confirm the direct relationship between growth in labour

efficiency and progress in the creation of productive industrial activity as pointed out

above and which can be more clearly by looking at Figure 1 In other words regions with

higher levels of productivity growth in general show a greater attractiveness for new

manufacturing firms

These differences in the behaviour of productivity and employment across regions

have led to remarkable disparities in rates of change of their industrial output during the

period Thus while a few increased their real manufacturing GVA by more than 20 in

the less dynamic ones the loss of industrial output is between a sixth and a third of the

initial value So significant changes had taken place in the relative position of some of

them (Figure 2)

Growth is led by a group of twenty regions located in Ireland Finland Greece

Germany Sweden Netherlands Spain and Austria with real GVA growth rates that are

more than five points above average As a result they increased their share in the industrial

output of the sample Looking at the complete list we see that the growth deviation from

the mean exhibits positive values in just over half of the regions examined among them

seven of the ten with the largest industrial scale in 2005 Thus the top ten regions have

increased their share from 32 to 33 percent in the last five years

Conversely among the less dynamic regions are found a great number of British and

some French and Italian ones which consequently have fallen back in the ranking of

NUTS 2 arranged by their contribution to aggregate manufacturing output

Between both groups there are approximately twenty regions mainly in Germany

with growth rates not far from average so their positions have hardly changed during the

last five years

14

Delocation in the manufacturing sectors in the EU A regional overview

Figure 2 Regional shares in the total sample manufacturing GVA(percentage at 2000 prices)

ITc4

UKf2

UKm3

Uke4

UKf1

UKk1

UKh1

AT22

ITe1

ES21

ES51

FR71

FR62

IT e4

DE24

DE26

SE23

FR61

FR51

FR82

SE22

ITe3

AT31

DK04

DE11

DE23

GR30

DEc0

ES30DEb3DEf0

DEa4

DE92

BE21

DE73

IE02

ITd3

FR52

DE91

DE13DE25

DE 12DE60

DE27

DE21

DE22

BE23DEa2

DE71

DE 14

SE12

ITd5

SE11

FI19

FR10

FR 24

ITe1

FI18

DK01 DEa5

FR4 2

DEa1NL32

NL3 3

FR2 3

NL22

NL 41

UKd3

UKj3

UKi1

UKm2

UKj1

UKd4

UKi2

Ukg3

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

Share in 2005

Source Elaborated from Eurostat and National Statistical offices

Rat

e of

cha

nge

200

1-20

05

70403530252015100500

Finally when the question is analyzed at Member State level (Table 1) it is possible

to check out the disparities among the above mentioned regions Thus while in some

15

Delocation in the manufacturing sectors in the EU A regional overview

countries such as Spain manufacturing growth has been clearly labour-intensive

something which can be extended to its regions as also happens in Greece and Austria

which both boast positive location effect and a job destruction rate far below the average

in others such as Holland Sweden and to a lesser extent Belgium Ireland and Finland

increasing of industrial activity is only due by advances in productivity so has been

accompanied by heavy job losses and these are patterns repeated in most of their regions

with noticeable relevance in the Netherlands As for the United Kingdom Belgium and

Denmark all share declining industrial production together with strong job cuts a pattern

that is replicated in many of their regions In this situation are the bulk of French and

Italian NUTS 2 just like the countries to which they belong although in this case job

destruction has been much less pronounced and below the group average Close to this

mean is Germany as well as most of their regions

Thus the analysis at Member States level also confirms the positive relationship

between the advance of industrial productivity and the attractiveness for new

manufacturing firms As demonstrated for their regions Ireland Finland Netherlands and

Greece following at quite distance from Germany are Member States showing a better

performance in productivity and with high location effects (see Figure 1)

In fact similarities between performances in the regions and in the Member States

they belong to suggest that the trend in regional industrial employment is to some extent

influenced by belonging to the same country in other words the locational attractions of

each area and its effort to improve productivity is partly due to national specificities or

idiosyncratic factors In fact as shown in Appendix 3 most European industrialized

regions follow the location patterns of the country they belong to reproducing its

productivity behaviour too In summary the evolution of these variables in each region

seems to have a clear `country effectacute that is different behaviours motivated by specific

factors andor differentiated industrial policies

For example in order to correctly interpret the results for Spanish regions it can not

be overlooked that during the period analyzed as in the whole country regions have

absorbed large numbers of immigrants which has boosted the proliferation of lower

productivity and lower wages in industrial activities However it may be thought that this

phenomenon has tended to conceal productivity increases in larger and better equipped

16

Delocation in the manufacturing sectors in the EU A regional overview

firms which could have been significant in the light of their results7 On the other hand

countries like Finland or Ireland with the lowest population increase have opted for high-

tech manufacturing sectors as a way to improve national competitiveness8 Nonetheless

this positive location effect has not reached in these countriesrsquo NUTS 2 enough magnitude

to offset completely the job losses arising from productivity improvements

On the other hand this `national effectacute has also played an important role in

explaining the economic outcomes of European regions in terms of per capita GDP in

accordance with the results obtained by Cuadrado-Roura (2001) using the data available

for 109 EU regions over the 1977-1994 period

Table 3 Export growth in key sectors by countries 2000-2005 (percentages)Impact ITon Total

Spain 65 91 72 158 38 111 04Germany 116 138 61 111 111 130 14United Kingdom 09 85 29 32 23 61 02France -09 102 25 94 118 78 -01Ireland -09 -102 01 25 -34 77 -03Finland 47 152 45 49 56 73 13Italy 40 88 41 90 56 89 02EU-15 70 141 59 91 67 111 11Source Elaborated from Comtrade

Shoes TotalIT Auto Textile Clothes

This importance of the `national effectacute should allow us to make a better approach to

the determinants of above-mentioned delocation patterns in future research One of such

determinants might be found in the different importance and performance by countries of

some manufacturing branches particularly affected by the increase in global competition as

their development has being receiving strong support in emerging countries from both

domestic policies and the location of big multinational companies This is the case of IT

manufacturing sectors or that of automobiles clothes textiles and shoes (Sachwald 2004)

Table 3 shows how the exports of these key sectors have evolved in some of the European

countries considered here It is amazing to see the poor results registered by IT sectors in

United Kingdom France and Ireland compared to the other countries mentioned Germany

in particular This evolution could be behind the stronger effects of delocation in France

and the United Kingdom although it must have produced similar results in the case of

Ireland Furthermore on such a basis what has happened in Italy would remain

7 For details see Myro et al (2008) 8 See Aacutelvarez Loacutepez et al (2007)

17

Delocation in the manufacturing sectors in the EU A regional overview

unexplained as the selected key sectors do not show a bad performance in terms of exports

in the latter country

One reason for such contradictory results lies in the possibility that production

instead of export reflects accurately what has been happening in every sector in terms of

activity However Aacutelvarez et al (2007) show very similar patterns in production and

export specialization from 1995 to 2005 except in the case of Ireland which thus appears

as a very particular one This is because IT specialization in production increases at the

same time as it decreases in exports Perhaps the strong growth in domestic demand

prevents IT exports from growing The opposite case could be that of Italy where slow

growth in domestic demand might have been pushing up IT export9

Anyway further developments on the determinants of delocation patterns require the

preparation of disaggregated data on production a task beyond the objectives of this paper

That could allow us to update the available manufacturing specialization analysis As the

two countries exhibiting higher delocation effects France and the United Kingdom

showed the lowest rates of specialization compared with the EU (Milderfart and Overman

2002 Combes and Overman 2003) something could have changed in the last few years

which would merit a fresh look based on more recent data

On the other hand it is now clear that insofar as delocation in manufacturing

activities has affected each country in a different way a relocation process inside the EU

and between countries has taken place This process deserved to be deeply analyzed

including a perspective by sectors updating what we know about changes in the

distribution of manufacturing industry in the EU and offering the possibility of exploring

whether the European Common Policy has influenced them in some way as was already

suggested by Milderfart-Knarvik and Overman (2002)

9 In Finland IT activities grows much more in production than in exports but in both aspects a positive growth is registered

18

Delocation in the manufacturing sectors in the EU A regional overview

19

Figure 3 Employment growth rates in selected regions 2001-2005 manufacturing and total economy (percentages)

Ukg3

UKi2

UKd4

UKj1

UKm2

UKi1 UKj3

UKd3

NL 41

NL22

FR23

NL33

NL32

DEa1

FR42

DEa5

DK01

FI18

ITe1

FR 24FR10

FI19

SE11

ITd5FR52

DE 14

DE71

DEa2

BE23

DE22

DE21

DE27

IC c4

DE60DE 12

DE25

DE13

DE9 1SE12

ITd3

IE02

DE73

BE21

DE92DEa4

DEf0

DEb3

ES30

DEc0

GR30

DE23

DE11

DK04

AT31

ITe3

SE22

FR82

FR51FR61

SE2 3

DE26

DE24

IT e4

FR62

FR71

ES51

ES2 1

ITe1

AT22

UKh1

UKk1

UKf1

Uke4

UKm3

UKf2

-20

-15

-10

-5

0

5

10

15

20

25

-40 -30 -20 -10 0 10 20 30

Manufacturing

Source Elaborated from Eurostat and National Statistical offices

Tota

l em

ploy

men

t

Delocation in the manufacturing sectors in the EU A regional overview

One last question that deserves consideration is the effect of industrial delocation

process on the regional economies Surprisingly manufacturing job losses have not been

translated into total employment which exhibit positive growth for most of the chosen

regions with the exceptions of some Germans and all the Netherlands (Figure 3) That

means the regions have been able to compensate the negative effect on the employment

from shrinking manufacturing allocating more resources inside the services sectors In fact

a substitution of manufacturing by services seems to have happened as are the region

showing more delocation in manufacturing those that increase in a higher measure the

employment in the service sectors that is the French British and Italian However the

relation is not accurate as there are not a strict quantitative correlation region by region

between rates of delocation and increase in the employment in the service activities

These good results in terms of general employment shown by the regions affected in

a higher measure by manufacturing delocation suggest the convenience of a more general

analysis of regional economic growth beyond the manufacturing sector as some regions

strongly shrinking manufacturing activities in favour of those of services might be leading

an interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

5- Final remarks

Comparative analysis of location patterns for a large sample of European regions -

characterized by having an upper-middle level income and a strong industrial sector-

during the last few years has revealed that only half of them have shown locational

attractions for manufacturing activities The regional patterns found clearly show the

influence of `national effectacute the regions showing location effects corresponding to most

of those located in Austria Germany the Netherlands Ireland Finland Sweden Spain and

Greece

On the contrary the delocation process has affected another half of the regions

included most of them French British and Italian Therefore delocation has contributed to

concentrating the European manufacturing industry in the areas with higher market

potential in the centre of continental Europe It has contributed furthermore to relocating

manufacturing activities inside every country something that requires further analysis

from future research

20

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

fi19

fi18

ie02

dec0

gr30

se11

nl33

de21

de22

de26

de25

de12

de14

de11

dea4

dea1de73

def0

de60

deb3

de27

de23

itd5

ite3

se12

nl22

es 30

ukd4

es51

nl32

at22

ukf1

at31

es21

Map 1 Regions with positive location effect ( of 2000 employment)

Above 10 0 to 10

nl41

de13de27

Nevertheless productivity also grows faster in most of the regions belonging to the

United Kingdom Belgium and Denmark in spite of the strong delocation effects registered

10

Delocation in the manufacturing sectors in the EU A regional overview

which disturbs the above-mentioned positive relationship suggesting the harsh process of

industrial restructuring involved

Figure 1 Location effect and productivity growth in select regions manufacturing sector 2001-2005 (percentages)

UKi2

EU 15

FR6 2

UKm2

ITc1

UKg3UKd 3

UKe4

UKk1FR71F R

FR10UKm3

UKj1UKf2

UKd 4

UKi1

D K

DK01

UKh1U K

FR2 3FR4 2

FR24ITe1

BE2 1ITe4FR51

FR61

ITITd 3

SE2 2FR8 2

SE23

NL2 2

UKf1DE2 4

NL3 2DEa1

DEf0

UKj3DE9 1 DE92

DEa2DE71DK0 4

S EDE25

DE6 0DE12

A T

DE14

ITe3

AT2 2

AT3 1ESD E

DE 11DE2 3

NL41N LDE2 7

DEc0

NL3 3

DE2 6DE2 1

ES3 0ES51

SE11

DE2 2

ES21

GR

FI19F I

FI18

IE

GR3 0

B E

DE73

DEb 3

IE0 2

ITd 5 DE13SE12

FR52ITc4BE2 3

DEa4DEa5

y = 07x - 59R2 = 06

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

40

-35 -30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30 35 40 45

Productivity grothw Source Elaborated from Eurostat and National Statistical offices

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

11

Delocation in the manufacturing sectors in the EU A regional overview

In contrast the productivity decline in half of the French and Italian regions albeit

with a positive impact on employment is largely offset by destruction of industrial

activity except in Icircle de France and Midi -Pyreacuteneacutees the two territories in which

productivity has a more negative trajectory

These results closely match those reported by Cuadrado Roura et al (2000) based on

the growth rates for regional productivity and employment during the period 1980-1993

grouping European regions in four typologies comparing the growth rates of these two

variables with their respective European averages

According to this approach and including the location effect in the analysis the

group of regions that are examined in this work would be categorized as follows (Table 2)

The first group includes areas with productivity and employment growth rates above

sample average and show positive location effect (Quadrant 1) So as has been underlined

in that work possibly this reflects the fact that the technological and organizational

changes introduced during the period would have been successful promoting the

diversification and specialization in activities with high growth levels and attracting new

investments which are less labour-intensive

In a second stage are those regions that as the above mentioned authors suggest

have implemented restructuring processes to eliminate the most inefficient outputs

achieving higher levels of productivity and in about half of them increased industrial

production at the expense of a notable decline in employment (Quadrant 4) They named

this typology `restructuring via productivityacute

With a few exceptions (those where the location effect is negative) the regions in

both groups are the winners of industrial dynamic during the period Most of them are

large industrial regions and are located in the traditional industrial growth centres As

demonstrated above they are mainly Finish German and Dutch as well as Greek and Irish

regions

In contrast another large group of regions due to different factors6 have opted for

more labour-intensive industrial models which in a few cases (especially in Spanish and

Austrian regions) have been accompanied by output increases so they can also be included

6 See Cuadrado et al (2000)

12

Delocation in the manufacturing sectors in the EU A regional overview

in the winners (Map 1) while in other the result has been a loss of productive activity

(Quadrant 2)

(1) (2)Employment Productivity Employment Productivity

de11 Stuttgart -48 125 at22 Steiermark 36 11de12 Karlsruhe -57 101 at31 Oberoumlsterreich 06 56de21 Oberbayern -37 179 be21 Prov Antwerpen -70 -05de22 Niederbayern -33 218 be23 Prov Oost-Vlaanderen -74 61de23 Oberpfalz -38 120 de13 Freiburg -45 45de25 Mittelfranken -53 80 de14 Tuumlbingen -23 72de26 Unterfranken -49 189 de91 Braunschweig -47 25de27 Schwaben -41 148 es21 Pais Vasco 74 13de60 Hamburg -60 97 es30 Comunidad de Madrid -31 63dec0 Saarland -32 160 es51 Cataluntildea 09 25fi19 Laumlnsi-Suomi -54 297 fr10 Icircle de France 218 -307gr30 Attiki -17 234 fr51 Pays de la Loire -08 -67

fr52 Bretagne -23 12fr61 Aquitaine -14 -50fr62 Midi-Pyreacuteneacutees 15 -245fr71 Rhocircne-Alpes -61 -57fr82 Provence-Alpes-Cocircte dAzur -06 -18itc4 Lombardia -63 61itd3 Veneto -62 03itd5 Emilia-Romagna -16 16ite3 Marche 08 28ite4 Lazio -36 -47se23 Vaumlstsverige -26 -07

(4) (3)Employment Productivity Employment Productivity

de24 Oberfranken -116 186 fr24 Centre -106 06de71 Darmstadt -149 132 fr42 Alsace -154 61de73 Kassel -78 103 itc1 Piemonte -129 -48de92 Hannover -106 86 ite1 Toscana -98 -02dea1 Duumlsseldorf -126 177 se22 Sydsverige -91 62dea2 Koumlln -150 137 uke4 West Yorkshire -186 60dea4 Detmold -107 110 ukh1 East Anglia -137 71dea5 Arnsberg -129 122 uki2 Outer London -271 -67deb3 Rheinhessen-Pfalz -106 152 ukj3 Hampshire and Isle of Wight -81 51def0 Schleswig-Holstein -101 135 ukk1 Gloucestershire Wiltshire -169 62dk01 Hovedstaden -134 125 ukm2 Eastern Scotland -91 -116dk04 Midtjylland -146 118fi18 Etelauml-Suomi -86 339fr23 Haute-Normandie -171 77ie02 Southern and Eastern -78 426nl22 Gelderland -178 206nl32 Noord-Holland -138 180nl33 Zuid-Holland -129 296nl41 Noord-Brabant -173 269se11 Stockholm -136 363se12 Oumlstra Mellansverige -151 166ukd3 Greater Manchester -232 90ukd4 Lancashire -317 350ukf1 Derbyshire and Nottinghamshire -179 240ukf2 Leicestershire Rutland and Northants -233 191ukg3 West Midlands -313 180uki1 Inner London -142 102ukj1 Berkshire Bucks and Oxfordshire -246 208ukm3 South Western Scotland -207 128

Source Elaborated from Eurostat and national Statistical officesNote Shaded areas indicate regions with positive location effect

Em

ploy

men

t gro

wth

lt A

vera

ge v

alue

Productivity growth gt Average value Productivity growth lt Average value

Em

ploy

men

t gro

wth

gt A

vera

ge v

alue

Table 2 Employment and productivity growth rates in select regions manufacturing sector 2001-2005 (Averages of all regions +74 productivity -77 employment)

13

Delocation in the manufacturing sectors in the EU A regional overview

Finally the worst results can be seen in a fourth type of region with growth rates of

GVA and productivity below average and serious job destruction (Quadrant 3) This would

occur in regions with problems in restructuring their manufacturing sector or with low

levels of activity So together with those that are either restructuring via productivity or

via employment but without achieving a positive location effect they are clearly the

losers These include mainly British French Italian Belgian and Danish regions

To sum up these results confirm the direct relationship between growth in labour

efficiency and progress in the creation of productive industrial activity as pointed out

above and which can be more clearly by looking at Figure 1 In other words regions with

higher levels of productivity growth in general show a greater attractiveness for new

manufacturing firms

These differences in the behaviour of productivity and employment across regions

have led to remarkable disparities in rates of change of their industrial output during the

period Thus while a few increased their real manufacturing GVA by more than 20 in

the less dynamic ones the loss of industrial output is between a sixth and a third of the

initial value So significant changes had taken place in the relative position of some of

them (Figure 2)

Growth is led by a group of twenty regions located in Ireland Finland Greece

Germany Sweden Netherlands Spain and Austria with real GVA growth rates that are

more than five points above average As a result they increased their share in the industrial

output of the sample Looking at the complete list we see that the growth deviation from

the mean exhibits positive values in just over half of the regions examined among them

seven of the ten with the largest industrial scale in 2005 Thus the top ten regions have

increased their share from 32 to 33 percent in the last five years

Conversely among the less dynamic regions are found a great number of British and

some French and Italian ones which consequently have fallen back in the ranking of

NUTS 2 arranged by their contribution to aggregate manufacturing output

Between both groups there are approximately twenty regions mainly in Germany

with growth rates not far from average so their positions have hardly changed during the

last five years

14

Delocation in the manufacturing sectors in the EU A regional overview

Figure 2 Regional shares in the total sample manufacturing GVA(percentage at 2000 prices)

ITc4

UKf2

UKm3

Uke4

UKf1

UKk1

UKh1

AT22

ITe1

ES21

ES51

FR71

FR62

IT e4

DE24

DE26

SE23

FR61

FR51

FR82

SE22

ITe3

AT31

DK04

DE11

DE23

GR30

DEc0

ES30DEb3DEf0

DEa4

DE92

BE21

DE73

IE02

ITd3

FR52

DE91

DE13DE25

DE 12DE60

DE27

DE21

DE22

BE23DEa2

DE71

DE 14

SE12

ITd5

SE11

FI19

FR10

FR 24

ITe1

FI18

DK01 DEa5

FR4 2

DEa1NL32

NL3 3

FR2 3

NL22

NL 41

UKd3

UKj3

UKi1

UKm2

UKj1

UKd4

UKi2

Ukg3

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

Share in 2005

Source Elaborated from Eurostat and National Statistical offices

Rat

e of

cha

nge

200

1-20

05

70403530252015100500

Finally when the question is analyzed at Member State level (Table 1) it is possible

to check out the disparities among the above mentioned regions Thus while in some

15

Delocation in the manufacturing sectors in the EU A regional overview

countries such as Spain manufacturing growth has been clearly labour-intensive

something which can be extended to its regions as also happens in Greece and Austria

which both boast positive location effect and a job destruction rate far below the average

in others such as Holland Sweden and to a lesser extent Belgium Ireland and Finland

increasing of industrial activity is only due by advances in productivity so has been

accompanied by heavy job losses and these are patterns repeated in most of their regions

with noticeable relevance in the Netherlands As for the United Kingdom Belgium and

Denmark all share declining industrial production together with strong job cuts a pattern

that is replicated in many of their regions In this situation are the bulk of French and

Italian NUTS 2 just like the countries to which they belong although in this case job

destruction has been much less pronounced and below the group average Close to this

mean is Germany as well as most of their regions

Thus the analysis at Member States level also confirms the positive relationship

between the advance of industrial productivity and the attractiveness for new

manufacturing firms As demonstrated for their regions Ireland Finland Netherlands and

Greece following at quite distance from Germany are Member States showing a better

performance in productivity and with high location effects (see Figure 1)

In fact similarities between performances in the regions and in the Member States

they belong to suggest that the trend in regional industrial employment is to some extent

influenced by belonging to the same country in other words the locational attractions of

each area and its effort to improve productivity is partly due to national specificities or

idiosyncratic factors In fact as shown in Appendix 3 most European industrialized

regions follow the location patterns of the country they belong to reproducing its

productivity behaviour too In summary the evolution of these variables in each region

seems to have a clear `country effectacute that is different behaviours motivated by specific

factors andor differentiated industrial policies

For example in order to correctly interpret the results for Spanish regions it can not

be overlooked that during the period analyzed as in the whole country regions have

absorbed large numbers of immigrants which has boosted the proliferation of lower

productivity and lower wages in industrial activities However it may be thought that this

phenomenon has tended to conceal productivity increases in larger and better equipped

16

Delocation in the manufacturing sectors in the EU A regional overview

firms which could have been significant in the light of their results7 On the other hand

countries like Finland or Ireland with the lowest population increase have opted for high-

tech manufacturing sectors as a way to improve national competitiveness8 Nonetheless

this positive location effect has not reached in these countriesrsquo NUTS 2 enough magnitude

to offset completely the job losses arising from productivity improvements

On the other hand this `national effectacute has also played an important role in

explaining the economic outcomes of European regions in terms of per capita GDP in

accordance with the results obtained by Cuadrado-Roura (2001) using the data available

for 109 EU regions over the 1977-1994 period

Table 3 Export growth in key sectors by countries 2000-2005 (percentages)Impact ITon Total

Spain 65 91 72 158 38 111 04Germany 116 138 61 111 111 130 14United Kingdom 09 85 29 32 23 61 02France -09 102 25 94 118 78 -01Ireland -09 -102 01 25 -34 77 -03Finland 47 152 45 49 56 73 13Italy 40 88 41 90 56 89 02EU-15 70 141 59 91 67 111 11Source Elaborated from Comtrade

Shoes TotalIT Auto Textile Clothes

This importance of the `national effectacute should allow us to make a better approach to

the determinants of above-mentioned delocation patterns in future research One of such

determinants might be found in the different importance and performance by countries of

some manufacturing branches particularly affected by the increase in global competition as

their development has being receiving strong support in emerging countries from both

domestic policies and the location of big multinational companies This is the case of IT

manufacturing sectors or that of automobiles clothes textiles and shoes (Sachwald 2004)

Table 3 shows how the exports of these key sectors have evolved in some of the European

countries considered here It is amazing to see the poor results registered by IT sectors in

United Kingdom France and Ireland compared to the other countries mentioned Germany

in particular This evolution could be behind the stronger effects of delocation in France

and the United Kingdom although it must have produced similar results in the case of

Ireland Furthermore on such a basis what has happened in Italy would remain

7 For details see Myro et al (2008) 8 See Aacutelvarez Loacutepez et al (2007)

17

Delocation in the manufacturing sectors in the EU A regional overview

unexplained as the selected key sectors do not show a bad performance in terms of exports

in the latter country

One reason for such contradictory results lies in the possibility that production

instead of export reflects accurately what has been happening in every sector in terms of

activity However Aacutelvarez et al (2007) show very similar patterns in production and

export specialization from 1995 to 2005 except in the case of Ireland which thus appears

as a very particular one This is because IT specialization in production increases at the

same time as it decreases in exports Perhaps the strong growth in domestic demand

prevents IT exports from growing The opposite case could be that of Italy where slow

growth in domestic demand might have been pushing up IT export9

Anyway further developments on the determinants of delocation patterns require the

preparation of disaggregated data on production a task beyond the objectives of this paper

That could allow us to update the available manufacturing specialization analysis As the

two countries exhibiting higher delocation effects France and the United Kingdom

showed the lowest rates of specialization compared with the EU (Milderfart and Overman

2002 Combes and Overman 2003) something could have changed in the last few years

which would merit a fresh look based on more recent data

On the other hand it is now clear that insofar as delocation in manufacturing

activities has affected each country in a different way a relocation process inside the EU

and between countries has taken place This process deserved to be deeply analyzed

including a perspective by sectors updating what we know about changes in the

distribution of manufacturing industry in the EU and offering the possibility of exploring

whether the European Common Policy has influenced them in some way as was already

suggested by Milderfart-Knarvik and Overman (2002)

9 In Finland IT activities grows much more in production than in exports but in both aspects a positive growth is registered

18

Delocation in the manufacturing sectors in the EU A regional overview

19

Figure 3 Employment growth rates in selected regions 2001-2005 manufacturing and total economy (percentages)

Ukg3

UKi2

UKd4

UKj1

UKm2

UKi1 UKj3

UKd3

NL 41

NL22

FR23

NL33

NL32

DEa1

FR42

DEa5

DK01

FI18

ITe1

FR 24FR10

FI19

SE11

ITd5FR52

DE 14

DE71

DEa2

BE23

DE22

DE21

DE27

IC c4

DE60DE 12

DE25

DE13

DE9 1SE12

ITd3

IE02

DE73

BE21

DE92DEa4

DEf0

DEb3

ES30

DEc0

GR30

DE23

DE11

DK04

AT31

ITe3

SE22

FR82

FR51FR61

SE2 3

DE26

DE24

IT e4

FR62

FR71

ES51

ES2 1

ITe1

AT22

UKh1

UKk1

UKf1

Uke4

UKm3

UKf2

-20

-15

-10

-5

0

5

10

15

20

25

-40 -30 -20 -10 0 10 20 30

Manufacturing

Source Elaborated from Eurostat and National Statistical offices

Tota

l em

ploy

men

t

Delocation in the manufacturing sectors in the EU A regional overview

One last question that deserves consideration is the effect of industrial delocation

process on the regional economies Surprisingly manufacturing job losses have not been

translated into total employment which exhibit positive growth for most of the chosen

regions with the exceptions of some Germans and all the Netherlands (Figure 3) That

means the regions have been able to compensate the negative effect on the employment

from shrinking manufacturing allocating more resources inside the services sectors In fact

a substitution of manufacturing by services seems to have happened as are the region

showing more delocation in manufacturing those that increase in a higher measure the

employment in the service sectors that is the French British and Italian However the

relation is not accurate as there are not a strict quantitative correlation region by region

between rates of delocation and increase in the employment in the service activities

These good results in terms of general employment shown by the regions affected in

a higher measure by manufacturing delocation suggest the convenience of a more general

analysis of regional economic growth beyond the manufacturing sector as some regions

strongly shrinking manufacturing activities in favour of those of services might be leading

an interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

5- Final remarks

Comparative analysis of location patterns for a large sample of European regions -

characterized by having an upper-middle level income and a strong industrial sector-

during the last few years has revealed that only half of them have shown locational

attractions for manufacturing activities The regional patterns found clearly show the

influence of `national effectacute the regions showing location effects corresponding to most

of those located in Austria Germany the Netherlands Ireland Finland Sweden Spain and

Greece

On the contrary the delocation process has affected another half of the regions

included most of them French British and Italian Therefore delocation has contributed to

concentrating the European manufacturing industry in the areas with higher market

potential in the centre of continental Europe It has contributed furthermore to relocating

manufacturing activities inside every country something that requires further analysis

from future research

20

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

which disturbs the above-mentioned positive relationship suggesting the harsh process of

industrial restructuring involved

Figure 1 Location effect and productivity growth in select regions manufacturing sector 2001-2005 (percentages)

UKi2

EU 15

FR6 2

UKm2

ITc1

UKg3UKd 3

UKe4

UKk1FR71F R

FR10UKm3

UKj1UKf2

UKd 4

UKi1

D K

DK01

UKh1U K

FR2 3FR4 2

FR24ITe1

BE2 1ITe4FR51

FR61

ITITd 3

SE2 2FR8 2

SE23

NL2 2

UKf1DE2 4

NL3 2DEa1

DEf0

UKj3DE9 1 DE92

DEa2DE71DK0 4

S EDE25

DE6 0DE12

A T

DE14

ITe3

AT2 2

AT3 1ESD E

DE 11DE2 3

NL41N LDE2 7

DEc0

NL3 3

DE2 6DE2 1

ES3 0ES51

SE11

DE2 2

ES21

GR

FI19F I

FI18

IE

GR3 0

B E

DE73

DEb 3

IE0 2

ITd 5 DE13SE12

FR52ITc4BE2 3

DEa4DEa5

y = 07x - 59R2 = 06

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

40

-35 -30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30 35 40 45

Productivity grothw Source Elaborated from Eurostat and National Statistical offices

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

11

Delocation in the manufacturing sectors in the EU A regional overview

In contrast the productivity decline in half of the French and Italian regions albeit

with a positive impact on employment is largely offset by destruction of industrial

activity except in Icircle de France and Midi -Pyreacuteneacutees the two territories in which

productivity has a more negative trajectory

These results closely match those reported by Cuadrado Roura et al (2000) based on

the growth rates for regional productivity and employment during the period 1980-1993

grouping European regions in four typologies comparing the growth rates of these two

variables with their respective European averages

According to this approach and including the location effect in the analysis the

group of regions that are examined in this work would be categorized as follows (Table 2)

The first group includes areas with productivity and employment growth rates above

sample average and show positive location effect (Quadrant 1) So as has been underlined

in that work possibly this reflects the fact that the technological and organizational

changes introduced during the period would have been successful promoting the

diversification and specialization in activities with high growth levels and attracting new

investments which are less labour-intensive

In a second stage are those regions that as the above mentioned authors suggest

have implemented restructuring processes to eliminate the most inefficient outputs

achieving higher levels of productivity and in about half of them increased industrial

production at the expense of a notable decline in employment (Quadrant 4) They named

this typology `restructuring via productivityacute

With a few exceptions (those where the location effect is negative) the regions in

both groups are the winners of industrial dynamic during the period Most of them are

large industrial regions and are located in the traditional industrial growth centres As

demonstrated above they are mainly Finish German and Dutch as well as Greek and Irish

regions

In contrast another large group of regions due to different factors6 have opted for

more labour-intensive industrial models which in a few cases (especially in Spanish and

Austrian regions) have been accompanied by output increases so they can also be included

6 See Cuadrado et al (2000)

12

Delocation in the manufacturing sectors in the EU A regional overview

in the winners (Map 1) while in other the result has been a loss of productive activity

(Quadrant 2)

(1) (2)Employment Productivity Employment Productivity

de11 Stuttgart -48 125 at22 Steiermark 36 11de12 Karlsruhe -57 101 at31 Oberoumlsterreich 06 56de21 Oberbayern -37 179 be21 Prov Antwerpen -70 -05de22 Niederbayern -33 218 be23 Prov Oost-Vlaanderen -74 61de23 Oberpfalz -38 120 de13 Freiburg -45 45de25 Mittelfranken -53 80 de14 Tuumlbingen -23 72de26 Unterfranken -49 189 de91 Braunschweig -47 25de27 Schwaben -41 148 es21 Pais Vasco 74 13de60 Hamburg -60 97 es30 Comunidad de Madrid -31 63dec0 Saarland -32 160 es51 Cataluntildea 09 25fi19 Laumlnsi-Suomi -54 297 fr10 Icircle de France 218 -307gr30 Attiki -17 234 fr51 Pays de la Loire -08 -67

fr52 Bretagne -23 12fr61 Aquitaine -14 -50fr62 Midi-Pyreacuteneacutees 15 -245fr71 Rhocircne-Alpes -61 -57fr82 Provence-Alpes-Cocircte dAzur -06 -18itc4 Lombardia -63 61itd3 Veneto -62 03itd5 Emilia-Romagna -16 16ite3 Marche 08 28ite4 Lazio -36 -47se23 Vaumlstsverige -26 -07

(4) (3)Employment Productivity Employment Productivity

de24 Oberfranken -116 186 fr24 Centre -106 06de71 Darmstadt -149 132 fr42 Alsace -154 61de73 Kassel -78 103 itc1 Piemonte -129 -48de92 Hannover -106 86 ite1 Toscana -98 -02dea1 Duumlsseldorf -126 177 se22 Sydsverige -91 62dea2 Koumlln -150 137 uke4 West Yorkshire -186 60dea4 Detmold -107 110 ukh1 East Anglia -137 71dea5 Arnsberg -129 122 uki2 Outer London -271 -67deb3 Rheinhessen-Pfalz -106 152 ukj3 Hampshire and Isle of Wight -81 51def0 Schleswig-Holstein -101 135 ukk1 Gloucestershire Wiltshire -169 62dk01 Hovedstaden -134 125 ukm2 Eastern Scotland -91 -116dk04 Midtjylland -146 118fi18 Etelauml-Suomi -86 339fr23 Haute-Normandie -171 77ie02 Southern and Eastern -78 426nl22 Gelderland -178 206nl32 Noord-Holland -138 180nl33 Zuid-Holland -129 296nl41 Noord-Brabant -173 269se11 Stockholm -136 363se12 Oumlstra Mellansverige -151 166ukd3 Greater Manchester -232 90ukd4 Lancashire -317 350ukf1 Derbyshire and Nottinghamshire -179 240ukf2 Leicestershire Rutland and Northants -233 191ukg3 West Midlands -313 180uki1 Inner London -142 102ukj1 Berkshire Bucks and Oxfordshire -246 208ukm3 South Western Scotland -207 128

Source Elaborated from Eurostat and national Statistical officesNote Shaded areas indicate regions with positive location effect

Em

ploy

men

t gro

wth

lt A

vera

ge v

alue

Productivity growth gt Average value Productivity growth lt Average value

Em

ploy

men

t gro

wth

gt A

vera

ge v

alue

Table 2 Employment and productivity growth rates in select regions manufacturing sector 2001-2005 (Averages of all regions +74 productivity -77 employment)

13

Delocation in the manufacturing sectors in the EU A regional overview

Finally the worst results can be seen in a fourth type of region with growth rates of

GVA and productivity below average and serious job destruction (Quadrant 3) This would

occur in regions with problems in restructuring their manufacturing sector or with low

levels of activity So together with those that are either restructuring via productivity or

via employment but without achieving a positive location effect they are clearly the

losers These include mainly British French Italian Belgian and Danish regions

To sum up these results confirm the direct relationship between growth in labour

efficiency and progress in the creation of productive industrial activity as pointed out

above and which can be more clearly by looking at Figure 1 In other words regions with

higher levels of productivity growth in general show a greater attractiveness for new

manufacturing firms

These differences in the behaviour of productivity and employment across regions

have led to remarkable disparities in rates of change of their industrial output during the

period Thus while a few increased their real manufacturing GVA by more than 20 in

the less dynamic ones the loss of industrial output is between a sixth and a third of the

initial value So significant changes had taken place in the relative position of some of

them (Figure 2)

Growth is led by a group of twenty regions located in Ireland Finland Greece

Germany Sweden Netherlands Spain and Austria with real GVA growth rates that are

more than five points above average As a result they increased their share in the industrial

output of the sample Looking at the complete list we see that the growth deviation from

the mean exhibits positive values in just over half of the regions examined among them

seven of the ten with the largest industrial scale in 2005 Thus the top ten regions have

increased their share from 32 to 33 percent in the last five years

Conversely among the less dynamic regions are found a great number of British and

some French and Italian ones which consequently have fallen back in the ranking of

NUTS 2 arranged by their contribution to aggregate manufacturing output

Between both groups there are approximately twenty regions mainly in Germany

with growth rates not far from average so their positions have hardly changed during the

last five years

14

Delocation in the manufacturing sectors in the EU A regional overview

Figure 2 Regional shares in the total sample manufacturing GVA(percentage at 2000 prices)

ITc4

UKf2

UKm3

Uke4

UKf1

UKk1

UKh1

AT22

ITe1

ES21

ES51

FR71

FR62

IT e4

DE24

DE26

SE23

FR61

FR51

FR82

SE22

ITe3

AT31

DK04

DE11

DE23

GR30

DEc0

ES30DEb3DEf0

DEa4

DE92

BE21

DE73

IE02

ITd3

FR52

DE91

DE13DE25

DE 12DE60

DE27

DE21

DE22

BE23DEa2

DE71

DE 14

SE12

ITd5

SE11

FI19

FR10

FR 24

ITe1

FI18

DK01 DEa5

FR4 2

DEa1NL32

NL3 3

FR2 3

NL22

NL 41

UKd3

UKj3

UKi1

UKm2

UKj1

UKd4

UKi2

Ukg3

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

Share in 2005

Source Elaborated from Eurostat and National Statistical offices

Rat

e of

cha

nge

200

1-20

05

70403530252015100500

Finally when the question is analyzed at Member State level (Table 1) it is possible

to check out the disparities among the above mentioned regions Thus while in some

15

Delocation in the manufacturing sectors in the EU A regional overview

countries such as Spain manufacturing growth has been clearly labour-intensive

something which can be extended to its regions as also happens in Greece and Austria

which both boast positive location effect and a job destruction rate far below the average

in others such as Holland Sweden and to a lesser extent Belgium Ireland and Finland

increasing of industrial activity is only due by advances in productivity so has been

accompanied by heavy job losses and these are patterns repeated in most of their regions

with noticeable relevance in the Netherlands As for the United Kingdom Belgium and

Denmark all share declining industrial production together with strong job cuts a pattern

that is replicated in many of their regions In this situation are the bulk of French and

Italian NUTS 2 just like the countries to which they belong although in this case job

destruction has been much less pronounced and below the group average Close to this

mean is Germany as well as most of their regions

Thus the analysis at Member States level also confirms the positive relationship

between the advance of industrial productivity and the attractiveness for new

manufacturing firms As demonstrated for their regions Ireland Finland Netherlands and

Greece following at quite distance from Germany are Member States showing a better

performance in productivity and with high location effects (see Figure 1)

In fact similarities between performances in the regions and in the Member States

they belong to suggest that the trend in regional industrial employment is to some extent

influenced by belonging to the same country in other words the locational attractions of

each area and its effort to improve productivity is partly due to national specificities or

idiosyncratic factors In fact as shown in Appendix 3 most European industrialized

regions follow the location patterns of the country they belong to reproducing its

productivity behaviour too In summary the evolution of these variables in each region

seems to have a clear `country effectacute that is different behaviours motivated by specific

factors andor differentiated industrial policies

For example in order to correctly interpret the results for Spanish regions it can not

be overlooked that during the period analyzed as in the whole country regions have

absorbed large numbers of immigrants which has boosted the proliferation of lower

productivity and lower wages in industrial activities However it may be thought that this

phenomenon has tended to conceal productivity increases in larger and better equipped

16

Delocation in the manufacturing sectors in the EU A regional overview

firms which could have been significant in the light of their results7 On the other hand

countries like Finland or Ireland with the lowest population increase have opted for high-

tech manufacturing sectors as a way to improve national competitiveness8 Nonetheless

this positive location effect has not reached in these countriesrsquo NUTS 2 enough magnitude

to offset completely the job losses arising from productivity improvements

On the other hand this `national effectacute has also played an important role in

explaining the economic outcomes of European regions in terms of per capita GDP in

accordance with the results obtained by Cuadrado-Roura (2001) using the data available

for 109 EU regions over the 1977-1994 period

Table 3 Export growth in key sectors by countries 2000-2005 (percentages)Impact ITon Total

Spain 65 91 72 158 38 111 04Germany 116 138 61 111 111 130 14United Kingdom 09 85 29 32 23 61 02France -09 102 25 94 118 78 -01Ireland -09 -102 01 25 -34 77 -03Finland 47 152 45 49 56 73 13Italy 40 88 41 90 56 89 02EU-15 70 141 59 91 67 111 11Source Elaborated from Comtrade

Shoes TotalIT Auto Textile Clothes

This importance of the `national effectacute should allow us to make a better approach to

the determinants of above-mentioned delocation patterns in future research One of such

determinants might be found in the different importance and performance by countries of

some manufacturing branches particularly affected by the increase in global competition as

their development has being receiving strong support in emerging countries from both

domestic policies and the location of big multinational companies This is the case of IT

manufacturing sectors or that of automobiles clothes textiles and shoes (Sachwald 2004)

Table 3 shows how the exports of these key sectors have evolved in some of the European

countries considered here It is amazing to see the poor results registered by IT sectors in

United Kingdom France and Ireland compared to the other countries mentioned Germany

in particular This evolution could be behind the stronger effects of delocation in France

and the United Kingdom although it must have produced similar results in the case of

Ireland Furthermore on such a basis what has happened in Italy would remain

7 For details see Myro et al (2008) 8 See Aacutelvarez Loacutepez et al (2007)

17

Delocation in the manufacturing sectors in the EU A regional overview

unexplained as the selected key sectors do not show a bad performance in terms of exports

in the latter country

One reason for such contradictory results lies in the possibility that production

instead of export reflects accurately what has been happening in every sector in terms of

activity However Aacutelvarez et al (2007) show very similar patterns in production and

export specialization from 1995 to 2005 except in the case of Ireland which thus appears

as a very particular one This is because IT specialization in production increases at the

same time as it decreases in exports Perhaps the strong growth in domestic demand

prevents IT exports from growing The opposite case could be that of Italy where slow

growth in domestic demand might have been pushing up IT export9

Anyway further developments on the determinants of delocation patterns require the

preparation of disaggregated data on production a task beyond the objectives of this paper

That could allow us to update the available manufacturing specialization analysis As the

two countries exhibiting higher delocation effects France and the United Kingdom

showed the lowest rates of specialization compared with the EU (Milderfart and Overman

2002 Combes and Overman 2003) something could have changed in the last few years

which would merit a fresh look based on more recent data

On the other hand it is now clear that insofar as delocation in manufacturing

activities has affected each country in a different way a relocation process inside the EU

and between countries has taken place This process deserved to be deeply analyzed

including a perspective by sectors updating what we know about changes in the

distribution of manufacturing industry in the EU and offering the possibility of exploring

whether the European Common Policy has influenced them in some way as was already

suggested by Milderfart-Knarvik and Overman (2002)

9 In Finland IT activities grows much more in production than in exports but in both aspects a positive growth is registered

18

Delocation in the manufacturing sectors in the EU A regional overview

19

Figure 3 Employment growth rates in selected regions 2001-2005 manufacturing and total economy (percentages)

Ukg3

UKi2

UKd4

UKj1

UKm2

UKi1 UKj3

UKd3

NL 41

NL22

FR23

NL33

NL32

DEa1

FR42

DEa5

DK01

FI18

ITe1

FR 24FR10

FI19

SE11

ITd5FR52

DE 14

DE71

DEa2

BE23

DE22

DE21

DE27

IC c4

DE60DE 12

DE25

DE13

DE9 1SE12

ITd3

IE02

DE73

BE21

DE92DEa4

DEf0

DEb3

ES30

DEc0

GR30

DE23

DE11

DK04

AT31

ITe3

SE22

FR82

FR51FR61

SE2 3

DE26

DE24

IT e4

FR62

FR71

ES51

ES2 1

ITe1

AT22

UKh1

UKk1

UKf1

Uke4

UKm3

UKf2

-20

-15

-10

-5

0

5

10

15

20

25

-40 -30 -20 -10 0 10 20 30

Manufacturing

Source Elaborated from Eurostat and National Statistical offices

Tota

l em

ploy

men

t

Delocation in the manufacturing sectors in the EU A regional overview

One last question that deserves consideration is the effect of industrial delocation

process on the regional economies Surprisingly manufacturing job losses have not been

translated into total employment which exhibit positive growth for most of the chosen

regions with the exceptions of some Germans and all the Netherlands (Figure 3) That

means the regions have been able to compensate the negative effect on the employment

from shrinking manufacturing allocating more resources inside the services sectors In fact

a substitution of manufacturing by services seems to have happened as are the region

showing more delocation in manufacturing those that increase in a higher measure the

employment in the service sectors that is the French British and Italian However the

relation is not accurate as there are not a strict quantitative correlation region by region

between rates of delocation and increase in the employment in the service activities

These good results in terms of general employment shown by the regions affected in

a higher measure by manufacturing delocation suggest the convenience of a more general

analysis of regional economic growth beyond the manufacturing sector as some regions

strongly shrinking manufacturing activities in favour of those of services might be leading

an interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

5- Final remarks

Comparative analysis of location patterns for a large sample of European regions -

characterized by having an upper-middle level income and a strong industrial sector-

during the last few years has revealed that only half of them have shown locational

attractions for manufacturing activities The regional patterns found clearly show the

influence of `national effectacute the regions showing location effects corresponding to most

of those located in Austria Germany the Netherlands Ireland Finland Sweden Spain and

Greece

On the contrary the delocation process has affected another half of the regions

included most of them French British and Italian Therefore delocation has contributed to

concentrating the European manufacturing industry in the areas with higher market

potential in the centre of continental Europe It has contributed furthermore to relocating

manufacturing activities inside every country something that requires further analysis

from future research

20

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

In contrast the productivity decline in half of the French and Italian regions albeit

with a positive impact on employment is largely offset by destruction of industrial

activity except in Icircle de France and Midi -Pyreacuteneacutees the two territories in which

productivity has a more negative trajectory

These results closely match those reported by Cuadrado Roura et al (2000) based on

the growth rates for regional productivity and employment during the period 1980-1993

grouping European regions in four typologies comparing the growth rates of these two

variables with their respective European averages

According to this approach and including the location effect in the analysis the

group of regions that are examined in this work would be categorized as follows (Table 2)

The first group includes areas with productivity and employment growth rates above

sample average and show positive location effect (Quadrant 1) So as has been underlined

in that work possibly this reflects the fact that the technological and organizational

changes introduced during the period would have been successful promoting the

diversification and specialization in activities with high growth levels and attracting new

investments which are less labour-intensive

In a second stage are those regions that as the above mentioned authors suggest

have implemented restructuring processes to eliminate the most inefficient outputs

achieving higher levels of productivity and in about half of them increased industrial

production at the expense of a notable decline in employment (Quadrant 4) They named

this typology `restructuring via productivityacute

With a few exceptions (those where the location effect is negative) the regions in

both groups are the winners of industrial dynamic during the period Most of them are

large industrial regions and are located in the traditional industrial growth centres As

demonstrated above they are mainly Finish German and Dutch as well as Greek and Irish

regions

In contrast another large group of regions due to different factors6 have opted for

more labour-intensive industrial models which in a few cases (especially in Spanish and

Austrian regions) have been accompanied by output increases so they can also be included

6 See Cuadrado et al (2000)

12

Delocation in the manufacturing sectors in the EU A regional overview

in the winners (Map 1) while in other the result has been a loss of productive activity

(Quadrant 2)

(1) (2)Employment Productivity Employment Productivity

de11 Stuttgart -48 125 at22 Steiermark 36 11de12 Karlsruhe -57 101 at31 Oberoumlsterreich 06 56de21 Oberbayern -37 179 be21 Prov Antwerpen -70 -05de22 Niederbayern -33 218 be23 Prov Oost-Vlaanderen -74 61de23 Oberpfalz -38 120 de13 Freiburg -45 45de25 Mittelfranken -53 80 de14 Tuumlbingen -23 72de26 Unterfranken -49 189 de91 Braunschweig -47 25de27 Schwaben -41 148 es21 Pais Vasco 74 13de60 Hamburg -60 97 es30 Comunidad de Madrid -31 63dec0 Saarland -32 160 es51 Cataluntildea 09 25fi19 Laumlnsi-Suomi -54 297 fr10 Icircle de France 218 -307gr30 Attiki -17 234 fr51 Pays de la Loire -08 -67

fr52 Bretagne -23 12fr61 Aquitaine -14 -50fr62 Midi-Pyreacuteneacutees 15 -245fr71 Rhocircne-Alpes -61 -57fr82 Provence-Alpes-Cocircte dAzur -06 -18itc4 Lombardia -63 61itd3 Veneto -62 03itd5 Emilia-Romagna -16 16ite3 Marche 08 28ite4 Lazio -36 -47se23 Vaumlstsverige -26 -07

(4) (3)Employment Productivity Employment Productivity

de24 Oberfranken -116 186 fr24 Centre -106 06de71 Darmstadt -149 132 fr42 Alsace -154 61de73 Kassel -78 103 itc1 Piemonte -129 -48de92 Hannover -106 86 ite1 Toscana -98 -02dea1 Duumlsseldorf -126 177 se22 Sydsverige -91 62dea2 Koumlln -150 137 uke4 West Yorkshire -186 60dea4 Detmold -107 110 ukh1 East Anglia -137 71dea5 Arnsberg -129 122 uki2 Outer London -271 -67deb3 Rheinhessen-Pfalz -106 152 ukj3 Hampshire and Isle of Wight -81 51def0 Schleswig-Holstein -101 135 ukk1 Gloucestershire Wiltshire -169 62dk01 Hovedstaden -134 125 ukm2 Eastern Scotland -91 -116dk04 Midtjylland -146 118fi18 Etelauml-Suomi -86 339fr23 Haute-Normandie -171 77ie02 Southern and Eastern -78 426nl22 Gelderland -178 206nl32 Noord-Holland -138 180nl33 Zuid-Holland -129 296nl41 Noord-Brabant -173 269se11 Stockholm -136 363se12 Oumlstra Mellansverige -151 166ukd3 Greater Manchester -232 90ukd4 Lancashire -317 350ukf1 Derbyshire and Nottinghamshire -179 240ukf2 Leicestershire Rutland and Northants -233 191ukg3 West Midlands -313 180uki1 Inner London -142 102ukj1 Berkshire Bucks and Oxfordshire -246 208ukm3 South Western Scotland -207 128

Source Elaborated from Eurostat and national Statistical officesNote Shaded areas indicate regions with positive location effect

Em

ploy

men

t gro

wth

lt A

vera

ge v

alue

Productivity growth gt Average value Productivity growth lt Average value

Em

ploy

men

t gro

wth

gt A

vera

ge v

alue

Table 2 Employment and productivity growth rates in select regions manufacturing sector 2001-2005 (Averages of all regions +74 productivity -77 employment)

13

Delocation in the manufacturing sectors in the EU A regional overview

Finally the worst results can be seen in a fourth type of region with growth rates of

GVA and productivity below average and serious job destruction (Quadrant 3) This would

occur in regions with problems in restructuring their manufacturing sector or with low

levels of activity So together with those that are either restructuring via productivity or

via employment but without achieving a positive location effect they are clearly the

losers These include mainly British French Italian Belgian and Danish regions

To sum up these results confirm the direct relationship between growth in labour

efficiency and progress in the creation of productive industrial activity as pointed out

above and which can be more clearly by looking at Figure 1 In other words regions with

higher levels of productivity growth in general show a greater attractiveness for new

manufacturing firms

These differences in the behaviour of productivity and employment across regions

have led to remarkable disparities in rates of change of their industrial output during the

period Thus while a few increased their real manufacturing GVA by more than 20 in

the less dynamic ones the loss of industrial output is between a sixth and a third of the

initial value So significant changes had taken place in the relative position of some of

them (Figure 2)

Growth is led by a group of twenty regions located in Ireland Finland Greece

Germany Sweden Netherlands Spain and Austria with real GVA growth rates that are

more than five points above average As a result they increased their share in the industrial

output of the sample Looking at the complete list we see that the growth deviation from

the mean exhibits positive values in just over half of the regions examined among them

seven of the ten with the largest industrial scale in 2005 Thus the top ten regions have

increased their share from 32 to 33 percent in the last five years

Conversely among the less dynamic regions are found a great number of British and

some French and Italian ones which consequently have fallen back in the ranking of

NUTS 2 arranged by their contribution to aggregate manufacturing output

Between both groups there are approximately twenty regions mainly in Germany

with growth rates not far from average so their positions have hardly changed during the

last five years

14

Delocation in the manufacturing sectors in the EU A regional overview

Figure 2 Regional shares in the total sample manufacturing GVA(percentage at 2000 prices)

ITc4

UKf2

UKm3

Uke4

UKf1

UKk1

UKh1

AT22

ITe1

ES21

ES51

FR71

FR62

IT e4

DE24

DE26

SE23

FR61

FR51

FR82

SE22

ITe3

AT31

DK04

DE11

DE23

GR30

DEc0

ES30DEb3DEf0

DEa4

DE92

BE21

DE73

IE02

ITd3

FR52

DE91

DE13DE25

DE 12DE60

DE27

DE21

DE22

BE23DEa2

DE71

DE 14

SE12

ITd5

SE11

FI19

FR10

FR 24

ITe1

FI18

DK01 DEa5

FR4 2

DEa1NL32

NL3 3

FR2 3

NL22

NL 41

UKd3

UKj3

UKi1

UKm2

UKj1

UKd4

UKi2

Ukg3

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

Share in 2005

Source Elaborated from Eurostat and National Statistical offices

Rat

e of

cha

nge

200

1-20

05

70403530252015100500

Finally when the question is analyzed at Member State level (Table 1) it is possible

to check out the disparities among the above mentioned regions Thus while in some

15

Delocation in the manufacturing sectors in the EU A regional overview

countries such as Spain manufacturing growth has been clearly labour-intensive

something which can be extended to its regions as also happens in Greece and Austria

which both boast positive location effect and a job destruction rate far below the average

in others such as Holland Sweden and to a lesser extent Belgium Ireland and Finland

increasing of industrial activity is only due by advances in productivity so has been

accompanied by heavy job losses and these are patterns repeated in most of their regions

with noticeable relevance in the Netherlands As for the United Kingdom Belgium and

Denmark all share declining industrial production together with strong job cuts a pattern

that is replicated in many of their regions In this situation are the bulk of French and

Italian NUTS 2 just like the countries to which they belong although in this case job

destruction has been much less pronounced and below the group average Close to this

mean is Germany as well as most of their regions

Thus the analysis at Member States level also confirms the positive relationship

between the advance of industrial productivity and the attractiveness for new

manufacturing firms As demonstrated for their regions Ireland Finland Netherlands and

Greece following at quite distance from Germany are Member States showing a better

performance in productivity and with high location effects (see Figure 1)

In fact similarities between performances in the regions and in the Member States

they belong to suggest that the trend in regional industrial employment is to some extent

influenced by belonging to the same country in other words the locational attractions of

each area and its effort to improve productivity is partly due to national specificities or

idiosyncratic factors In fact as shown in Appendix 3 most European industrialized

regions follow the location patterns of the country they belong to reproducing its

productivity behaviour too In summary the evolution of these variables in each region

seems to have a clear `country effectacute that is different behaviours motivated by specific

factors andor differentiated industrial policies

For example in order to correctly interpret the results for Spanish regions it can not

be overlooked that during the period analyzed as in the whole country regions have

absorbed large numbers of immigrants which has boosted the proliferation of lower

productivity and lower wages in industrial activities However it may be thought that this

phenomenon has tended to conceal productivity increases in larger and better equipped

16

Delocation in the manufacturing sectors in the EU A regional overview

firms which could have been significant in the light of their results7 On the other hand

countries like Finland or Ireland with the lowest population increase have opted for high-

tech manufacturing sectors as a way to improve national competitiveness8 Nonetheless

this positive location effect has not reached in these countriesrsquo NUTS 2 enough magnitude

to offset completely the job losses arising from productivity improvements

On the other hand this `national effectacute has also played an important role in

explaining the economic outcomes of European regions in terms of per capita GDP in

accordance with the results obtained by Cuadrado-Roura (2001) using the data available

for 109 EU regions over the 1977-1994 period

Table 3 Export growth in key sectors by countries 2000-2005 (percentages)Impact ITon Total

Spain 65 91 72 158 38 111 04Germany 116 138 61 111 111 130 14United Kingdom 09 85 29 32 23 61 02France -09 102 25 94 118 78 -01Ireland -09 -102 01 25 -34 77 -03Finland 47 152 45 49 56 73 13Italy 40 88 41 90 56 89 02EU-15 70 141 59 91 67 111 11Source Elaborated from Comtrade

Shoes TotalIT Auto Textile Clothes

This importance of the `national effectacute should allow us to make a better approach to

the determinants of above-mentioned delocation patterns in future research One of such

determinants might be found in the different importance and performance by countries of

some manufacturing branches particularly affected by the increase in global competition as

their development has being receiving strong support in emerging countries from both

domestic policies and the location of big multinational companies This is the case of IT

manufacturing sectors or that of automobiles clothes textiles and shoes (Sachwald 2004)

Table 3 shows how the exports of these key sectors have evolved in some of the European

countries considered here It is amazing to see the poor results registered by IT sectors in

United Kingdom France and Ireland compared to the other countries mentioned Germany

in particular This evolution could be behind the stronger effects of delocation in France

and the United Kingdom although it must have produced similar results in the case of

Ireland Furthermore on such a basis what has happened in Italy would remain

7 For details see Myro et al (2008) 8 See Aacutelvarez Loacutepez et al (2007)

17

Delocation in the manufacturing sectors in the EU A regional overview

unexplained as the selected key sectors do not show a bad performance in terms of exports

in the latter country

One reason for such contradictory results lies in the possibility that production

instead of export reflects accurately what has been happening in every sector in terms of

activity However Aacutelvarez et al (2007) show very similar patterns in production and

export specialization from 1995 to 2005 except in the case of Ireland which thus appears

as a very particular one This is because IT specialization in production increases at the

same time as it decreases in exports Perhaps the strong growth in domestic demand

prevents IT exports from growing The opposite case could be that of Italy where slow

growth in domestic demand might have been pushing up IT export9

Anyway further developments on the determinants of delocation patterns require the

preparation of disaggregated data on production a task beyond the objectives of this paper

That could allow us to update the available manufacturing specialization analysis As the

two countries exhibiting higher delocation effects France and the United Kingdom

showed the lowest rates of specialization compared with the EU (Milderfart and Overman

2002 Combes and Overman 2003) something could have changed in the last few years

which would merit a fresh look based on more recent data

On the other hand it is now clear that insofar as delocation in manufacturing

activities has affected each country in a different way a relocation process inside the EU

and between countries has taken place This process deserved to be deeply analyzed

including a perspective by sectors updating what we know about changes in the

distribution of manufacturing industry in the EU and offering the possibility of exploring

whether the European Common Policy has influenced them in some way as was already

suggested by Milderfart-Knarvik and Overman (2002)

9 In Finland IT activities grows much more in production than in exports but in both aspects a positive growth is registered

18

Delocation in the manufacturing sectors in the EU A regional overview

19

Figure 3 Employment growth rates in selected regions 2001-2005 manufacturing and total economy (percentages)

Ukg3

UKi2

UKd4

UKj1

UKm2

UKi1 UKj3

UKd3

NL 41

NL22

FR23

NL33

NL32

DEa1

FR42

DEa5

DK01

FI18

ITe1

FR 24FR10

FI19

SE11

ITd5FR52

DE 14

DE71

DEa2

BE23

DE22

DE21

DE27

IC c4

DE60DE 12

DE25

DE13

DE9 1SE12

ITd3

IE02

DE73

BE21

DE92DEa4

DEf0

DEb3

ES30

DEc0

GR30

DE23

DE11

DK04

AT31

ITe3

SE22

FR82

FR51FR61

SE2 3

DE26

DE24

IT e4

FR62

FR71

ES51

ES2 1

ITe1

AT22

UKh1

UKk1

UKf1

Uke4

UKm3

UKf2

-20

-15

-10

-5

0

5

10

15

20

25

-40 -30 -20 -10 0 10 20 30

Manufacturing

Source Elaborated from Eurostat and National Statistical offices

Tota

l em

ploy

men

t

Delocation in the manufacturing sectors in the EU A regional overview

One last question that deserves consideration is the effect of industrial delocation

process on the regional economies Surprisingly manufacturing job losses have not been

translated into total employment which exhibit positive growth for most of the chosen

regions with the exceptions of some Germans and all the Netherlands (Figure 3) That

means the regions have been able to compensate the negative effect on the employment

from shrinking manufacturing allocating more resources inside the services sectors In fact

a substitution of manufacturing by services seems to have happened as are the region

showing more delocation in manufacturing those that increase in a higher measure the

employment in the service sectors that is the French British and Italian However the

relation is not accurate as there are not a strict quantitative correlation region by region

between rates of delocation and increase in the employment in the service activities

These good results in terms of general employment shown by the regions affected in

a higher measure by manufacturing delocation suggest the convenience of a more general

analysis of regional economic growth beyond the manufacturing sector as some regions

strongly shrinking manufacturing activities in favour of those of services might be leading

an interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

5- Final remarks

Comparative analysis of location patterns for a large sample of European regions -

characterized by having an upper-middle level income and a strong industrial sector-

during the last few years has revealed that only half of them have shown locational

attractions for manufacturing activities The regional patterns found clearly show the

influence of `national effectacute the regions showing location effects corresponding to most

of those located in Austria Germany the Netherlands Ireland Finland Sweden Spain and

Greece

On the contrary the delocation process has affected another half of the regions

included most of them French British and Italian Therefore delocation has contributed to

concentrating the European manufacturing industry in the areas with higher market

potential in the centre of continental Europe It has contributed furthermore to relocating

manufacturing activities inside every country something that requires further analysis

from future research

20

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

in the winners (Map 1) while in other the result has been a loss of productive activity

(Quadrant 2)

(1) (2)Employment Productivity Employment Productivity

de11 Stuttgart -48 125 at22 Steiermark 36 11de12 Karlsruhe -57 101 at31 Oberoumlsterreich 06 56de21 Oberbayern -37 179 be21 Prov Antwerpen -70 -05de22 Niederbayern -33 218 be23 Prov Oost-Vlaanderen -74 61de23 Oberpfalz -38 120 de13 Freiburg -45 45de25 Mittelfranken -53 80 de14 Tuumlbingen -23 72de26 Unterfranken -49 189 de91 Braunschweig -47 25de27 Schwaben -41 148 es21 Pais Vasco 74 13de60 Hamburg -60 97 es30 Comunidad de Madrid -31 63dec0 Saarland -32 160 es51 Cataluntildea 09 25fi19 Laumlnsi-Suomi -54 297 fr10 Icircle de France 218 -307gr30 Attiki -17 234 fr51 Pays de la Loire -08 -67

fr52 Bretagne -23 12fr61 Aquitaine -14 -50fr62 Midi-Pyreacuteneacutees 15 -245fr71 Rhocircne-Alpes -61 -57fr82 Provence-Alpes-Cocircte dAzur -06 -18itc4 Lombardia -63 61itd3 Veneto -62 03itd5 Emilia-Romagna -16 16ite3 Marche 08 28ite4 Lazio -36 -47se23 Vaumlstsverige -26 -07

(4) (3)Employment Productivity Employment Productivity

de24 Oberfranken -116 186 fr24 Centre -106 06de71 Darmstadt -149 132 fr42 Alsace -154 61de73 Kassel -78 103 itc1 Piemonte -129 -48de92 Hannover -106 86 ite1 Toscana -98 -02dea1 Duumlsseldorf -126 177 se22 Sydsverige -91 62dea2 Koumlln -150 137 uke4 West Yorkshire -186 60dea4 Detmold -107 110 ukh1 East Anglia -137 71dea5 Arnsberg -129 122 uki2 Outer London -271 -67deb3 Rheinhessen-Pfalz -106 152 ukj3 Hampshire and Isle of Wight -81 51def0 Schleswig-Holstein -101 135 ukk1 Gloucestershire Wiltshire -169 62dk01 Hovedstaden -134 125 ukm2 Eastern Scotland -91 -116dk04 Midtjylland -146 118fi18 Etelauml-Suomi -86 339fr23 Haute-Normandie -171 77ie02 Southern and Eastern -78 426nl22 Gelderland -178 206nl32 Noord-Holland -138 180nl33 Zuid-Holland -129 296nl41 Noord-Brabant -173 269se11 Stockholm -136 363se12 Oumlstra Mellansverige -151 166ukd3 Greater Manchester -232 90ukd4 Lancashire -317 350ukf1 Derbyshire and Nottinghamshire -179 240ukf2 Leicestershire Rutland and Northants -233 191ukg3 West Midlands -313 180uki1 Inner London -142 102ukj1 Berkshire Bucks and Oxfordshire -246 208ukm3 South Western Scotland -207 128

Source Elaborated from Eurostat and national Statistical officesNote Shaded areas indicate regions with positive location effect

Em

ploy

men

t gro

wth

lt A

vera

ge v

alue

Productivity growth gt Average value Productivity growth lt Average value

Em

ploy

men

t gro

wth

gt A

vera

ge v

alue

Table 2 Employment and productivity growth rates in select regions manufacturing sector 2001-2005 (Averages of all regions +74 productivity -77 employment)

13

Delocation in the manufacturing sectors in the EU A regional overview

Finally the worst results can be seen in a fourth type of region with growth rates of

GVA and productivity below average and serious job destruction (Quadrant 3) This would

occur in regions with problems in restructuring their manufacturing sector or with low

levels of activity So together with those that are either restructuring via productivity or

via employment but without achieving a positive location effect they are clearly the

losers These include mainly British French Italian Belgian and Danish regions

To sum up these results confirm the direct relationship between growth in labour

efficiency and progress in the creation of productive industrial activity as pointed out

above and which can be more clearly by looking at Figure 1 In other words regions with

higher levels of productivity growth in general show a greater attractiveness for new

manufacturing firms

These differences in the behaviour of productivity and employment across regions

have led to remarkable disparities in rates of change of their industrial output during the

period Thus while a few increased their real manufacturing GVA by more than 20 in

the less dynamic ones the loss of industrial output is between a sixth and a third of the

initial value So significant changes had taken place in the relative position of some of

them (Figure 2)

Growth is led by a group of twenty regions located in Ireland Finland Greece

Germany Sweden Netherlands Spain and Austria with real GVA growth rates that are

more than five points above average As a result they increased their share in the industrial

output of the sample Looking at the complete list we see that the growth deviation from

the mean exhibits positive values in just over half of the regions examined among them

seven of the ten with the largest industrial scale in 2005 Thus the top ten regions have

increased their share from 32 to 33 percent in the last five years

Conversely among the less dynamic regions are found a great number of British and

some French and Italian ones which consequently have fallen back in the ranking of

NUTS 2 arranged by their contribution to aggregate manufacturing output

Between both groups there are approximately twenty regions mainly in Germany

with growth rates not far from average so their positions have hardly changed during the

last five years

14

Delocation in the manufacturing sectors in the EU A regional overview

Figure 2 Regional shares in the total sample manufacturing GVA(percentage at 2000 prices)

ITc4

UKf2

UKm3

Uke4

UKf1

UKk1

UKh1

AT22

ITe1

ES21

ES51

FR71

FR62

IT e4

DE24

DE26

SE23

FR61

FR51

FR82

SE22

ITe3

AT31

DK04

DE11

DE23

GR30

DEc0

ES30DEb3DEf0

DEa4

DE92

BE21

DE73

IE02

ITd3

FR52

DE91

DE13DE25

DE 12DE60

DE27

DE21

DE22

BE23DEa2

DE71

DE 14

SE12

ITd5

SE11

FI19

FR10

FR 24

ITe1

FI18

DK01 DEa5

FR4 2

DEa1NL32

NL3 3

FR2 3

NL22

NL 41

UKd3

UKj3

UKi1

UKm2

UKj1

UKd4

UKi2

Ukg3

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

Share in 2005

Source Elaborated from Eurostat and National Statistical offices

Rat

e of

cha

nge

200

1-20

05

70403530252015100500

Finally when the question is analyzed at Member State level (Table 1) it is possible

to check out the disparities among the above mentioned regions Thus while in some

15

Delocation in the manufacturing sectors in the EU A regional overview

countries such as Spain manufacturing growth has been clearly labour-intensive

something which can be extended to its regions as also happens in Greece and Austria

which both boast positive location effect and a job destruction rate far below the average

in others such as Holland Sweden and to a lesser extent Belgium Ireland and Finland

increasing of industrial activity is only due by advances in productivity so has been

accompanied by heavy job losses and these are patterns repeated in most of their regions

with noticeable relevance in the Netherlands As for the United Kingdom Belgium and

Denmark all share declining industrial production together with strong job cuts a pattern

that is replicated in many of their regions In this situation are the bulk of French and

Italian NUTS 2 just like the countries to which they belong although in this case job

destruction has been much less pronounced and below the group average Close to this

mean is Germany as well as most of their regions

Thus the analysis at Member States level also confirms the positive relationship

between the advance of industrial productivity and the attractiveness for new

manufacturing firms As demonstrated for their regions Ireland Finland Netherlands and

Greece following at quite distance from Germany are Member States showing a better

performance in productivity and with high location effects (see Figure 1)

In fact similarities between performances in the regions and in the Member States

they belong to suggest that the trend in regional industrial employment is to some extent

influenced by belonging to the same country in other words the locational attractions of

each area and its effort to improve productivity is partly due to national specificities or

idiosyncratic factors In fact as shown in Appendix 3 most European industrialized

regions follow the location patterns of the country they belong to reproducing its

productivity behaviour too In summary the evolution of these variables in each region

seems to have a clear `country effectacute that is different behaviours motivated by specific

factors andor differentiated industrial policies

For example in order to correctly interpret the results for Spanish regions it can not

be overlooked that during the period analyzed as in the whole country regions have

absorbed large numbers of immigrants which has boosted the proliferation of lower

productivity and lower wages in industrial activities However it may be thought that this

phenomenon has tended to conceal productivity increases in larger and better equipped

16

Delocation in the manufacturing sectors in the EU A regional overview

firms which could have been significant in the light of their results7 On the other hand

countries like Finland or Ireland with the lowest population increase have opted for high-

tech manufacturing sectors as a way to improve national competitiveness8 Nonetheless

this positive location effect has not reached in these countriesrsquo NUTS 2 enough magnitude

to offset completely the job losses arising from productivity improvements

On the other hand this `national effectacute has also played an important role in

explaining the economic outcomes of European regions in terms of per capita GDP in

accordance with the results obtained by Cuadrado-Roura (2001) using the data available

for 109 EU regions over the 1977-1994 period

Table 3 Export growth in key sectors by countries 2000-2005 (percentages)Impact ITon Total

Spain 65 91 72 158 38 111 04Germany 116 138 61 111 111 130 14United Kingdom 09 85 29 32 23 61 02France -09 102 25 94 118 78 -01Ireland -09 -102 01 25 -34 77 -03Finland 47 152 45 49 56 73 13Italy 40 88 41 90 56 89 02EU-15 70 141 59 91 67 111 11Source Elaborated from Comtrade

Shoes TotalIT Auto Textile Clothes

This importance of the `national effectacute should allow us to make a better approach to

the determinants of above-mentioned delocation patterns in future research One of such

determinants might be found in the different importance and performance by countries of

some manufacturing branches particularly affected by the increase in global competition as

their development has being receiving strong support in emerging countries from both

domestic policies and the location of big multinational companies This is the case of IT

manufacturing sectors or that of automobiles clothes textiles and shoes (Sachwald 2004)

Table 3 shows how the exports of these key sectors have evolved in some of the European

countries considered here It is amazing to see the poor results registered by IT sectors in

United Kingdom France and Ireland compared to the other countries mentioned Germany

in particular This evolution could be behind the stronger effects of delocation in France

and the United Kingdom although it must have produced similar results in the case of

Ireland Furthermore on such a basis what has happened in Italy would remain

7 For details see Myro et al (2008) 8 See Aacutelvarez Loacutepez et al (2007)

17

Delocation in the manufacturing sectors in the EU A regional overview

unexplained as the selected key sectors do not show a bad performance in terms of exports

in the latter country

One reason for such contradictory results lies in the possibility that production

instead of export reflects accurately what has been happening in every sector in terms of

activity However Aacutelvarez et al (2007) show very similar patterns in production and

export specialization from 1995 to 2005 except in the case of Ireland which thus appears

as a very particular one This is because IT specialization in production increases at the

same time as it decreases in exports Perhaps the strong growth in domestic demand

prevents IT exports from growing The opposite case could be that of Italy where slow

growth in domestic demand might have been pushing up IT export9

Anyway further developments on the determinants of delocation patterns require the

preparation of disaggregated data on production a task beyond the objectives of this paper

That could allow us to update the available manufacturing specialization analysis As the

two countries exhibiting higher delocation effects France and the United Kingdom

showed the lowest rates of specialization compared with the EU (Milderfart and Overman

2002 Combes and Overman 2003) something could have changed in the last few years

which would merit a fresh look based on more recent data

On the other hand it is now clear that insofar as delocation in manufacturing

activities has affected each country in a different way a relocation process inside the EU

and between countries has taken place This process deserved to be deeply analyzed

including a perspective by sectors updating what we know about changes in the

distribution of manufacturing industry in the EU and offering the possibility of exploring

whether the European Common Policy has influenced them in some way as was already

suggested by Milderfart-Knarvik and Overman (2002)

9 In Finland IT activities grows much more in production than in exports but in both aspects a positive growth is registered

18

Delocation in the manufacturing sectors in the EU A regional overview

19

Figure 3 Employment growth rates in selected regions 2001-2005 manufacturing and total economy (percentages)

Ukg3

UKi2

UKd4

UKj1

UKm2

UKi1 UKj3

UKd3

NL 41

NL22

FR23

NL33

NL32

DEa1

FR42

DEa5

DK01

FI18

ITe1

FR 24FR10

FI19

SE11

ITd5FR52

DE 14

DE71

DEa2

BE23

DE22

DE21

DE27

IC c4

DE60DE 12

DE25

DE13

DE9 1SE12

ITd3

IE02

DE73

BE21

DE92DEa4

DEf0

DEb3

ES30

DEc0

GR30

DE23

DE11

DK04

AT31

ITe3

SE22

FR82

FR51FR61

SE2 3

DE26

DE24

IT e4

FR62

FR71

ES51

ES2 1

ITe1

AT22

UKh1

UKk1

UKf1

Uke4

UKm3

UKf2

-20

-15

-10

-5

0

5

10

15

20

25

-40 -30 -20 -10 0 10 20 30

Manufacturing

Source Elaborated from Eurostat and National Statistical offices

Tota

l em

ploy

men

t

Delocation in the manufacturing sectors in the EU A regional overview

One last question that deserves consideration is the effect of industrial delocation

process on the regional economies Surprisingly manufacturing job losses have not been

translated into total employment which exhibit positive growth for most of the chosen

regions with the exceptions of some Germans and all the Netherlands (Figure 3) That

means the regions have been able to compensate the negative effect on the employment

from shrinking manufacturing allocating more resources inside the services sectors In fact

a substitution of manufacturing by services seems to have happened as are the region

showing more delocation in manufacturing those that increase in a higher measure the

employment in the service sectors that is the French British and Italian However the

relation is not accurate as there are not a strict quantitative correlation region by region

between rates of delocation and increase in the employment in the service activities

These good results in terms of general employment shown by the regions affected in

a higher measure by manufacturing delocation suggest the convenience of a more general

analysis of regional economic growth beyond the manufacturing sector as some regions

strongly shrinking manufacturing activities in favour of those of services might be leading

an interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

5- Final remarks

Comparative analysis of location patterns for a large sample of European regions -

characterized by having an upper-middle level income and a strong industrial sector-

during the last few years has revealed that only half of them have shown locational

attractions for manufacturing activities The regional patterns found clearly show the

influence of `national effectacute the regions showing location effects corresponding to most

of those located in Austria Germany the Netherlands Ireland Finland Sweden Spain and

Greece

On the contrary the delocation process has affected another half of the regions

included most of them French British and Italian Therefore delocation has contributed to

concentrating the European manufacturing industry in the areas with higher market

potential in the centre of continental Europe It has contributed furthermore to relocating

manufacturing activities inside every country something that requires further analysis

from future research

20

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

Finally the worst results can be seen in a fourth type of region with growth rates of

GVA and productivity below average and serious job destruction (Quadrant 3) This would

occur in regions with problems in restructuring their manufacturing sector or with low

levels of activity So together with those that are either restructuring via productivity or

via employment but without achieving a positive location effect they are clearly the

losers These include mainly British French Italian Belgian and Danish regions

To sum up these results confirm the direct relationship between growth in labour

efficiency and progress in the creation of productive industrial activity as pointed out

above and which can be more clearly by looking at Figure 1 In other words regions with

higher levels of productivity growth in general show a greater attractiveness for new

manufacturing firms

These differences in the behaviour of productivity and employment across regions

have led to remarkable disparities in rates of change of their industrial output during the

period Thus while a few increased their real manufacturing GVA by more than 20 in

the less dynamic ones the loss of industrial output is between a sixth and a third of the

initial value So significant changes had taken place in the relative position of some of

them (Figure 2)

Growth is led by a group of twenty regions located in Ireland Finland Greece

Germany Sweden Netherlands Spain and Austria with real GVA growth rates that are

more than five points above average As a result they increased their share in the industrial

output of the sample Looking at the complete list we see that the growth deviation from

the mean exhibits positive values in just over half of the regions examined among them

seven of the ten with the largest industrial scale in 2005 Thus the top ten regions have

increased their share from 32 to 33 percent in the last five years

Conversely among the less dynamic regions are found a great number of British and

some French and Italian ones which consequently have fallen back in the ranking of

NUTS 2 arranged by their contribution to aggregate manufacturing output

Between both groups there are approximately twenty regions mainly in Germany

with growth rates not far from average so their positions have hardly changed during the

last five years

14

Delocation in the manufacturing sectors in the EU A regional overview

Figure 2 Regional shares in the total sample manufacturing GVA(percentage at 2000 prices)

ITc4

UKf2

UKm3

Uke4

UKf1

UKk1

UKh1

AT22

ITe1

ES21

ES51

FR71

FR62

IT e4

DE24

DE26

SE23

FR61

FR51

FR82

SE22

ITe3

AT31

DK04

DE11

DE23

GR30

DEc0

ES30DEb3DEf0

DEa4

DE92

BE21

DE73

IE02

ITd3

FR52

DE91

DE13DE25

DE 12DE60

DE27

DE21

DE22

BE23DEa2

DE71

DE 14

SE12

ITd5

SE11

FI19

FR10

FR 24

ITe1

FI18

DK01 DEa5

FR4 2

DEa1NL32

NL3 3

FR2 3

NL22

NL 41

UKd3

UKj3

UKi1

UKm2

UKj1

UKd4

UKi2

Ukg3

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

Share in 2005

Source Elaborated from Eurostat and National Statistical offices

Rat

e of

cha

nge

200

1-20

05

70403530252015100500

Finally when the question is analyzed at Member State level (Table 1) it is possible

to check out the disparities among the above mentioned regions Thus while in some

15

Delocation in the manufacturing sectors in the EU A regional overview

countries such as Spain manufacturing growth has been clearly labour-intensive

something which can be extended to its regions as also happens in Greece and Austria

which both boast positive location effect and a job destruction rate far below the average

in others such as Holland Sweden and to a lesser extent Belgium Ireland and Finland

increasing of industrial activity is only due by advances in productivity so has been

accompanied by heavy job losses and these are patterns repeated in most of their regions

with noticeable relevance in the Netherlands As for the United Kingdom Belgium and

Denmark all share declining industrial production together with strong job cuts a pattern

that is replicated in many of their regions In this situation are the bulk of French and

Italian NUTS 2 just like the countries to which they belong although in this case job

destruction has been much less pronounced and below the group average Close to this

mean is Germany as well as most of their regions

Thus the analysis at Member States level also confirms the positive relationship

between the advance of industrial productivity and the attractiveness for new

manufacturing firms As demonstrated for their regions Ireland Finland Netherlands and

Greece following at quite distance from Germany are Member States showing a better

performance in productivity and with high location effects (see Figure 1)

In fact similarities between performances in the regions and in the Member States

they belong to suggest that the trend in regional industrial employment is to some extent

influenced by belonging to the same country in other words the locational attractions of

each area and its effort to improve productivity is partly due to national specificities or

idiosyncratic factors In fact as shown in Appendix 3 most European industrialized

regions follow the location patterns of the country they belong to reproducing its

productivity behaviour too In summary the evolution of these variables in each region

seems to have a clear `country effectacute that is different behaviours motivated by specific

factors andor differentiated industrial policies

For example in order to correctly interpret the results for Spanish regions it can not

be overlooked that during the period analyzed as in the whole country regions have

absorbed large numbers of immigrants which has boosted the proliferation of lower

productivity and lower wages in industrial activities However it may be thought that this

phenomenon has tended to conceal productivity increases in larger and better equipped

16

Delocation in the manufacturing sectors in the EU A regional overview

firms which could have been significant in the light of their results7 On the other hand

countries like Finland or Ireland with the lowest population increase have opted for high-

tech manufacturing sectors as a way to improve national competitiveness8 Nonetheless

this positive location effect has not reached in these countriesrsquo NUTS 2 enough magnitude

to offset completely the job losses arising from productivity improvements

On the other hand this `national effectacute has also played an important role in

explaining the economic outcomes of European regions in terms of per capita GDP in

accordance with the results obtained by Cuadrado-Roura (2001) using the data available

for 109 EU regions over the 1977-1994 period

Table 3 Export growth in key sectors by countries 2000-2005 (percentages)Impact ITon Total

Spain 65 91 72 158 38 111 04Germany 116 138 61 111 111 130 14United Kingdom 09 85 29 32 23 61 02France -09 102 25 94 118 78 -01Ireland -09 -102 01 25 -34 77 -03Finland 47 152 45 49 56 73 13Italy 40 88 41 90 56 89 02EU-15 70 141 59 91 67 111 11Source Elaborated from Comtrade

Shoes TotalIT Auto Textile Clothes

This importance of the `national effectacute should allow us to make a better approach to

the determinants of above-mentioned delocation patterns in future research One of such

determinants might be found in the different importance and performance by countries of

some manufacturing branches particularly affected by the increase in global competition as

their development has being receiving strong support in emerging countries from both

domestic policies and the location of big multinational companies This is the case of IT

manufacturing sectors or that of automobiles clothes textiles and shoes (Sachwald 2004)

Table 3 shows how the exports of these key sectors have evolved in some of the European

countries considered here It is amazing to see the poor results registered by IT sectors in

United Kingdom France and Ireland compared to the other countries mentioned Germany

in particular This evolution could be behind the stronger effects of delocation in France

and the United Kingdom although it must have produced similar results in the case of

Ireland Furthermore on such a basis what has happened in Italy would remain

7 For details see Myro et al (2008) 8 See Aacutelvarez Loacutepez et al (2007)

17

Delocation in the manufacturing sectors in the EU A regional overview

unexplained as the selected key sectors do not show a bad performance in terms of exports

in the latter country

One reason for such contradictory results lies in the possibility that production

instead of export reflects accurately what has been happening in every sector in terms of

activity However Aacutelvarez et al (2007) show very similar patterns in production and

export specialization from 1995 to 2005 except in the case of Ireland which thus appears

as a very particular one This is because IT specialization in production increases at the

same time as it decreases in exports Perhaps the strong growth in domestic demand

prevents IT exports from growing The opposite case could be that of Italy where slow

growth in domestic demand might have been pushing up IT export9

Anyway further developments on the determinants of delocation patterns require the

preparation of disaggregated data on production a task beyond the objectives of this paper

That could allow us to update the available manufacturing specialization analysis As the

two countries exhibiting higher delocation effects France and the United Kingdom

showed the lowest rates of specialization compared with the EU (Milderfart and Overman

2002 Combes and Overman 2003) something could have changed in the last few years

which would merit a fresh look based on more recent data

On the other hand it is now clear that insofar as delocation in manufacturing

activities has affected each country in a different way a relocation process inside the EU

and between countries has taken place This process deserved to be deeply analyzed

including a perspective by sectors updating what we know about changes in the

distribution of manufacturing industry in the EU and offering the possibility of exploring

whether the European Common Policy has influenced them in some way as was already

suggested by Milderfart-Knarvik and Overman (2002)

9 In Finland IT activities grows much more in production than in exports but in both aspects a positive growth is registered

18

Delocation in the manufacturing sectors in the EU A regional overview

19

Figure 3 Employment growth rates in selected regions 2001-2005 manufacturing and total economy (percentages)

Ukg3

UKi2

UKd4

UKj1

UKm2

UKi1 UKj3

UKd3

NL 41

NL22

FR23

NL33

NL32

DEa1

FR42

DEa5

DK01

FI18

ITe1

FR 24FR10

FI19

SE11

ITd5FR52

DE 14

DE71

DEa2

BE23

DE22

DE21

DE27

IC c4

DE60DE 12

DE25

DE13

DE9 1SE12

ITd3

IE02

DE73

BE21

DE92DEa4

DEf0

DEb3

ES30

DEc0

GR30

DE23

DE11

DK04

AT31

ITe3

SE22

FR82

FR51FR61

SE2 3

DE26

DE24

IT e4

FR62

FR71

ES51

ES2 1

ITe1

AT22

UKh1

UKk1

UKf1

Uke4

UKm3

UKf2

-20

-15

-10

-5

0

5

10

15

20

25

-40 -30 -20 -10 0 10 20 30

Manufacturing

Source Elaborated from Eurostat and National Statistical offices

Tota

l em

ploy

men

t

Delocation in the manufacturing sectors in the EU A regional overview

One last question that deserves consideration is the effect of industrial delocation

process on the regional economies Surprisingly manufacturing job losses have not been

translated into total employment which exhibit positive growth for most of the chosen

regions with the exceptions of some Germans and all the Netherlands (Figure 3) That

means the regions have been able to compensate the negative effect on the employment

from shrinking manufacturing allocating more resources inside the services sectors In fact

a substitution of manufacturing by services seems to have happened as are the region

showing more delocation in manufacturing those that increase in a higher measure the

employment in the service sectors that is the French British and Italian However the

relation is not accurate as there are not a strict quantitative correlation region by region

between rates of delocation and increase in the employment in the service activities

These good results in terms of general employment shown by the regions affected in

a higher measure by manufacturing delocation suggest the convenience of a more general

analysis of regional economic growth beyond the manufacturing sector as some regions

strongly shrinking manufacturing activities in favour of those of services might be leading

an interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

5- Final remarks

Comparative analysis of location patterns for a large sample of European regions -

characterized by having an upper-middle level income and a strong industrial sector-

during the last few years has revealed that only half of them have shown locational

attractions for manufacturing activities The regional patterns found clearly show the

influence of `national effectacute the regions showing location effects corresponding to most

of those located in Austria Germany the Netherlands Ireland Finland Sweden Spain and

Greece

On the contrary the delocation process has affected another half of the regions

included most of them French British and Italian Therefore delocation has contributed to

concentrating the European manufacturing industry in the areas with higher market

potential in the centre of continental Europe It has contributed furthermore to relocating

manufacturing activities inside every country something that requires further analysis

from future research

20

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

Figure 2 Regional shares in the total sample manufacturing GVA(percentage at 2000 prices)

ITc4

UKf2

UKm3

Uke4

UKf1

UKk1

UKh1

AT22

ITe1

ES21

ES51

FR71

FR62

IT e4

DE24

DE26

SE23

FR61

FR51

FR82

SE22

ITe3

AT31

DK04

DE11

DE23

GR30

DEc0

ES30DEb3DEf0

DEa4

DE92

BE21

DE73

IE02

ITd3

FR52

DE91

DE13DE25

DE 12DE60

DE27

DE21

DE22

BE23DEa2

DE71

DE 14

SE12

ITd5

SE11

FI19

FR10

FR 24

ITe1

FI18

DK01 DEa5

FR4 2

DEa1NL32

NL3 3

FR2 3

NL22

NL 41

UKd3

UKj3

UKi1

UKm2

UKj1

UKd4

UKi2

Ukg3

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

Share in 2005

Source Elaborated from Eurostat and National Statistical offices

Rat

e of

cha

nge

200

1-20

05

70403530252015100500

Finally when the question is analyzed at Member State level (Table 1) it is possible

to check out the disparities among the above mentioned regions Thus while in some

15

Delocation in the manufacturing sectors in the EU A regional overview

countries such as Spain manufacturing growth has been clearly labour-intensive

something which can be extended to its regions as also happens in Greece and Austria

which both boast positive location effect and a job destruction rate far below the average

in others such as Holland Sweden and to a lesser extent Belgium Ireland and Finland

increasing of industrial activity is only due by advances in productivity so has been

accompanied by heavy job losses and these are patterns repeated in most of their regions

with noticeable relevance in the Netherlands As for the United Kingdom Belgium and

Denmark all share declining industrial production together with strong job cuts a pattern

that is replicated in many of their regions In this situation are the bulk of French and

Italian NUTS 2 just like the countries to which they belong although in this case job

destruction has been much less pronounced and below the group average Close to this

mean is Germany as well as most of their regions

Thus the analysis at Member States level also confirms the positive relationship

between the advance of industrial productivity and the attractiveness for new

manufacturing firms As demonstrated for their regions Ireland Finland Netherlands and

Greece following at quite distance from Germany are Member States showing a better

performance in productivity and with high location effects (see Figure 1)

In fact similarities between performances in the regions and in the Member States

they belong to suggest that the trend in regional industrial employment is to some extent

influenced by belonging to the same country in other words the locational attractions of

each area and its effort to improve productivity is partly due to national specificities or

idiosyncratic factors In fact as shown in Appendix 3 most European industrialized

regions follow the location patterns of the country they belong to reproducing its

productivity behaviour too In summary the evolution of these variables in each region

seems to have a clear `country effectacute that is different behaviours motivated by specific

factors andor differentiated industrial policies

For example in order to correctly interpret the results for Spanish regions it can not

be overlooked that during the period analyzed as in the whole country regions have

absorbed large numbers of immigrants which has boosted the proliferation of lower

productivity and lower wages in industrial activities However it may be thought that this

phenomenon has tended to conceal productivity increases in larger and better equipped

16

Delocation in the manufacturing sectors in the EU A regional overview

firms which could have been significant in the light of their results7 On the other hand

countries like Finland or Ireland with the lowest population increase have opted for high-

tech manufacturing sectors as a way to improve national competitiveness8 Nonetheless

this positive location effect has not reached in these countriesrsquo NUTS 2 enough magnitude

to offset completely the job losses arising from productivity improvements

On the other hand this `national effectacute has also played an important role in

explaining the economic outcomes of European regions in terms of per capita GDP in

accordance with the results obtained by Cuadrado-Roura (2001) using the data available

for 109 EU regions over the 1977-1994 period

Table 3 Export growth in key sectors by countries 2000-2005 (percentages)Impact ITon Total

Spain 65 91 72 158 38 111 04Germany 116 138 61 111 111 130 14United Kingdom 09 85 29 32 23 61 02France -09 102 25 94 118 78 -01Ireland -09 -102 01 25 -34 77 -03Finland 47 152 45 49 56 73 13Italy 40 88 41 90 56 89 02EU-15 70 141 59 91 67 111 11Source Elaborated from Comtrade

Shoes TotalIT Auto Textile Clothes

This importance of the `national effectacute should allow us to make a better approach to

the determinants of above-mentioned delocation patterns in future research One of such

determinants might be found in the different importance and performance by countries of

some manufacturing branches particularly affected by the increase in global competition as

their development has being receiving strong support in emerging countries from both

domestic policies and the location of big multinational companies This is the case of IT

manufacturing sectors or that of automobiles clothes textiles and shoes (Sachwald 2004)

Table 3 shows how the exports of these key sectors have evolved in some of the European

countries considered here It is amazing to see the poor results registered by IT sectors in

United Kingdom France and Ireland compared to the other countries mentioned Germany

in particular This evolution could be behind the stronger effects of delocation in France

and the United Kingdom although it must have produced similar results in the case of

Ireland Furthermore on such a basis what has happened in Italy would remain

7 For details see Myro et al (2008) 8 See Aacutelvarez Loacutepez et al (2007)

17

Delocation in the manufacturing sectors in the EU A regional overview

unexplained as the selected key sectors do not show a bad performance in terms of exports

in the latter country

One reason for such contradictory results lies in the possibility that production

instead of export reflects accurately what has been happening in every sector in terms of

activity However Aacutelvarez et al (2007) show very similar patterns in production and

export specialization from 1995 to 2005 except in the case of Ireland which thus appears

as a very particular one This is because IT specialization in production increases at the

same time as it decreases in exports Perhaps the strong growth in domestic demand

prevents IT exports from growing The opposite case could be that of Italy where slow

growth in domestic demand might have been pushing up IT export9

Anyway further developments on the determinants of delocation patterns require the

preparation of disaggregated data on production a task beyond the objectives of this paper

That could allow us to update the available manufacturing specialization analysis As the

two countries exhibiting higher delocation effects France and the United Kingdom

showed the lowest rates of specialization compared with the EU (Milderfart and Overman

2002 Combes and Overman 2003) something could have changed in the last few years

which would merit a fresh look based on more recent data

On the other hand it is now clear that insofar as delocation in manufacturing

activities has affected each country in a different way a relocation process inside the EU

and between countries has taken place This process deserved to be deeply analyzed

including a perspective by sectors updating what we know about changes in the

distribution of manufacturing industry in the EU and offering the possibility of exploring

whether the European Common Policy has influenced them in some way as was already

suggested by Milderfart-Knarvik and Overman (2002)

9 In Finland IT activities grows much more in production than in exports but in both aspects a positive growth is registered

18

Delocation in the manufacturing sectors in the EU A regional overview

19

Figure 3 Employment growth rates in selected regions 2001-2005 manufacturing and total economy (percentages)

Ukg3

UKi2

UKd4

UKj1

UKm2

UKi1 UKj3

UKd3

NL 41

NL22

FR23

NL33

NL32

DEa1

FR42

DEa5

DK01

FI18

ITe1

FR 24FR10

FI19

SE11

ITd5FR52

DE 14

DE71

DEa2

BE23

DE22

DE21

DE27

IC c4

DE60DE 12

DE25

DE13

DE9 1SE12

ITd3

IE02

DE73

BE21

DE92DEa4

DEf0

DEb3

ES30

DEc0

GR30

DE23

DE11

DK04

AT31

ITe3

SE22

FR82

FR51FR61

SE2 3

DE26

DE24

IT e4

FR62

FR71

ES51

ES2 1

ITe1

AT22

UKh1

UKk1

UKf1

Uke4

UKm3

UKf2

-20

-15

-10

-5

0

5

10

15

20

25

-40 -30 -20 -10 0 10 20 30

Manufacturing

Source Elaborated from Eurostat and National Statistical offices

Tota

l em

ploy

men

t

Delocation in the manufacturing sectors in the EU A regional overview

One last question that deserves consideration is the effect of industrial delocation

process on the regional economies Surprisingly manufacturing job losses have not been

translated into total employment which exhibit positive growth for most of the chosen

regions with the exceptions of some Germans and all the Netherlands (Figure 3) That

means the regions have been able to compensate the negative effect on the employment

from shrinking manufacturing allocating more resources inside the services sectors In fact

a substitution of manufacturing by services seems to have happened as are the region

showing more delocation in manufacturing those that increase in a higher measure the

employment in the service sectors that is the French British and Italian However the

relation is not accurate as there are not a strict quantitative correlation region by region

between rates of delocation and increase in the employment in the service activities

These good results in terms of general employment shown by the regions affected in

a higher measure by manufacturing delocation suggest the convenience of a more general

analysis of regional economic growth beyond the manufacturing sector as some regions

strongly shrinking manufacturing activities in favour of those of services might be leading

an interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

5- Final remarks

Comparative analysis of location patterns for a large sample of European regions -

characterized by having an upper-middle level income and a strong industrial sector-

during the last few years has revealed that only half of them have shown locational

attractions for manufacturing activities The regional patterns found clearly show the

influence of `national effectacute the regions showing location effects corresponding to most

of those located in Austria Germany the Netherlands Ireland Finland Sweden Spain and

Greece

On the contrary the delocation process has affected another half of the regions

included most of them French British and Italian Therefore delocation has contributed to

concentrating the European manufacturing industry in the areas with higher market

potential in the centre of continental Europe It has contributed furthermore to relocating

manufacturing activities inside every country something that requires further analysis

from future research

20

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

countries such as Spain manufacturing growth has been clearly labour-intensive

something which can be extended to its regions as also happens in Greece and Austria

which both boast positive location effect and a job destruction rate far below the average

in others such as Holland Sweden and to a lesser extent Belgium Ireland and Finland

increasing of industrial activity is only due by advances in productivity so has been

accompanied by heavy job losses and these are patterns repeated in most of their regions

with noticeable relevance in the Netherlands As for the United Kingdom Belgium and

Denmark all share declining industrial production together with strong job cuts a pattern

that is replicated in many of their regions In this situation are the bulk of French and

Italian NUTS 2 just like the countries to which they belong although in this case job

destruction has been much less pronounced and below the group average Close to this

mean is Germany as well as most of their regions

Thus the analysis at Member States level also confirms the positive relationship

between the advance of industrial productivity and the attractiveness for new

manufacturing firms As demonstrated for their regions Ireland Finland Netherlands and

Greece following at quite distance from Germany are Member States showing a better

performance in productivity and with high location effects (see Figure 1)

In fact similarities between performances in the regions and in the Member States

they belong to suggest that the trend in regional industrial employment is to some extent

influenced by belonging to the same country in other words the locational attractions of

each area and its effort to improve productivity is partly due to national specificities or

idiosyncratic factors In fact as shown in Appendix 3 most European industrialized

regions follow the location patterns of the country they belong to reproducing its

productivity behaviour too In summary the evolution of these variables in each region

seems to have a clear `country effectacute that is different behaviours motivated by specific

factors andor differentiated industrial policies

For example in order to correctly interpret the results for Spanish regions it can not

be overlooked that during the period analyzed as in the whole country regions have

absorbed large numbers of immigrants which has boosted the proliferation of lower

productivity and lower wages in industrial activities However it may be thought that this

phenomenon has tended to conceal productivity increases in larger and better equipped

16

Delocation in the manufacturing sectors in the EU A regional overview

firms which could have been significant in the light of their results7 On the other hand

countries like Finland or Ireland with the lowest population increase have opted for high-

tech manufacturing sectors as a way to improve national competitiveness8 Nonetheless

this positive location effect has not reached in these countriesrsquo NUTS 2 enough magnitude

to offset completely the job losses arising from productivity improvements

On the other hand this `national effectacute has also played an important role in

explaining the economic outcomes of European regions in terms of per capita GDP in

accordance with the results obtained by Cuadrado-Roura (2001) using the data available

for 109 EU regions over the 1977-1994 period

Table 3 Export growth in key sectors by countries 2000-2005 (percentages)Impact ITon Total

Spain 65 91 72 158 38 111 04Germany 116 138 61 111 111 130 14United Kingdom 09 85 29 32 23 61 02France -09 102 25 94 118 78 -01Ireland -09 -102 01 25 -34 77 -03Finland 47 152 45 49 56 73 13Italy 40 88 41 90 56 89 02EU-15 70 141 59 91 67 111 11Source Elaborated from Comtrade

Shoes TotalIT Auto Textile Clothes

This importance of the `national effectacute should allow us to make a better approach to

the determinants of above-mentioned delocation patterns in future research One of such

determinants might be found in the different importance and performance by countries of

some manufacturing branches particularly affected by the increase in global competition as

their development has being receiving strong support in emerging countries from both

domestic policies and the location of big multinational companies This is the case of IT

manufacturing sectors or that of automobiles clothes textiles and shoes (Sachwald 2004)

Table 3 shows how the exports of these key sectors have evolved in some of the European

countries considered here It is amazing to see the poor results registered by IT sectors in

United Kingdom France and Ireland compared to the other countries mentioned Germany

in particular This evolution could be behind the stronger effects of delocation in France

and the United Kingdom although it must have produced similar results in the case of

Ireland Furthermore on such a basis what has happened in Italy would remain

7 For details see Myro et al (2008) 8 See Aacutelvarez Loacutepez et al (2007)

17

Delocation in the manufacturing sectors in the EU A regional overview

unexplained as the selected key sectors do not show a bad performance in terms of exports

in the latter country

One reason for such contradictory results lies in the possibility that production

instead of export reflects accurately what has been happening in every sector in terms of

activity However Aacutelvarez et al (2007) show very similar patterns in production and

export specialization from 1995 to 2005 except in the case of Ireland which thus appears

as a very particular one This is because IT specialization in production increases at the

same time as it decreases in exports Perhaps the strong growth in domestic demand

prevents IT exports from growing The opposite case could be that of Italy where slow

growth in domestic demand might have been pushing up IT export9

Anyway further developments on the determinants of delocation patterns require the

preparation of disaggregated data on production a task beyond the objectives of this paper

That could allow us to update the available manufacturing specialization analysis As the

two countries exhibiting higher delocation effects France and the United Kingdom

showed the lowest rates of specialization compared with the EU (Milderfart and Overman

2002 Combes and Overman 2003) something could have changed in the last few years

which would merit a fresh look based on more recent data

On the other hand it is now clear that insofar as delocation in manufacturing

activities has affected each country in a different way a relocation process inside the EU

and between countries has taken place This process deserved to be deeply analyzed

including a perspective by sectors updating what we know about changes in the

distribution of manufacturing industry in the EU and offering the possibility of exploring

whether the European Common Policy has influenced them in some way as was already

suggested by Milderfart-Knarvik and Overman (2002)

9 In Finland IT activities grows much more in production than in exports but in both aspects a positive growth is registered

18

Delocation in the manufacturing sectors in the EU A regional overview

19

Figure 3 Employment growth rates in selected regions 2001-2005 manufacturing and total economy (percentages)

Ukg3

UKi2

UKd4

UKj1

UKm2

UKi1 UKj3

UKd3

NL 41

NL22

FR23

NL33

NL32

DEa1

FR42

DEa5

DK01

FI18

ITe1

FR 24FR10

FI19

SE11

ITd5FR52

DE 14

DE71

DEa2

BE23

DE22

DE21

DE27

IC c4

DE60DE 12

DE25

DE13

DE9 1SE12

ITd3

IE02

DE73

BE21

DE92DEa4

DEf0

DEb3

ES30

DEc0

GR30

DE23

DE11

DK04

AT31

ITe3

SE22

FR82

FR51FR61

SE2 3

DE26

DE24

IT e4

FR62

FR71

ES51

ES2 1

ITe1

AT22

UKh1

UKk1

UKf1

Uke4

UKm3

UKf2

-20

-15

-10

-5

0

5

10

15

20

25

-40 -30 -20 -10 0 10 20 30

Manufacturing

Source Elaborated from Eurostat and National Statistical offices

Tota

l em

ploy

men

t

Delocation in the manufacturing sectors in the EU A regional overview

One last question that deserves consideration is the effect of industrial delocation

process on the regional economies Surprisingly manufacturing job losses have not been

translated into total employment which exhibit positive growth for most of the chosen

regions with the exceptions of some Germans and all the Netherlands (Figure 3) That

means the regions have been able to compensate the negative effect on the employment

from shrinking manufacturing allocating more resources inside the services sectors In fact

a substitution of manufacturing by services seems to have happened as are the region

showing more delocation in manufacturing those that increase in a higher measure the

employment in the service sectors that is the French British and Italian However the

relation is not accurate as there are not a strict quantitative correlation region by region

between rates of delocation and increase in the employment in the service activities

These good results in terms of general employment shown by the regions affected in

a higher measure by manufacturing delocation suggest the convenience of a more general

analysis of regional economic growth beyond the manufacturing sector as some regions

strongly shrinking manufacturing activities in favour of those of services might be leading

an interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

5- Final remarks

Comparative analysis of location patterns for a large sample of European regions -

characterized by having an upper-middle level income and a strong industrial sector-

during the last few years has revealed that only half of them have shown locational

attractions for manufacturing activities The regional patterns found clearly show the

influence of `national effectacute the regions showing location effects corresponding to most

of those located in Austria Germany the Netherlands Ireland Finland Sweden Spain and

Greece

On the contrary the delocation process has affected another half of the regions

included most of them French British and Italian Therefore delocation has contributed to

concentrating the European manufacturing industry in the areas with higher market

potential in the centre of continental Europe It has contributed furthermore to relocating

manufacturing activities inside every country something that requires further analysis

from future research

20

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

firms which could have been significant in the light of their results7 On the other hand

countries like Finland or Ireland with the lowest population increase have opted for high-

tech manufacturing sectors as a way to improve national competitiveness8 Nonetheless

this positive location effect has not reached in these countriesrsquo NUTS 2 enough magnitude

to offset completely the job losses arising from productivity improvements

On the other hand this `national effectacute has also played an important role in

explaining the economic outcomes of European regions in terms of per capita GDP in

accordance with the results obtained by Cuadrado-Roura (2001) using the data available

for 109 EU regions over the 1977-1994 period

Table 3 Export growth in key sectors by countries 2000-2005 (percentages)Impact ITon Total

Spain 65 91 72 158 38 111 04Germany 116 138 61 111 111 130 14United Kingdom 09 85 29 32 23 61 02France -09 102 25 94 118 78 -01Ireland -09 -102 01 25 -34 77 -03Finland 47 152 45 49 56 73 13Italy 40 88 41 90 56 89 02EU-15 70 141 59 91 67 111 11Source Elaborated from Comtrade

Shoes TotalIT Auto Textile Clothes

This importance of the `national effectacute should allow us to make a better approach to

the determinants of above-mentioned delocation patterns in future research One of such

determinants might be found in the different importance and performance by countries of

some manufacturing branches particularly affected by the increase in global competition as

their development has being receiving strong support in emerging countries from both

domestic policies and the location of big multinational companies This is the case of IT

manufacturing sectors or that of automobiles clothes textiles and shoes (Sachwald 2004)

Table 3 shows how the exports of these key sectors have evolved in some of the European

countries considered here It is amazing to see the poor results registered by IT sectors in

United Kingdom France and Ireland compared to the other countries mentioned Germany

in particular This evolution could be behind the stronger effects of delocation in France

and the United Kingdom although it must have produced similar results in the case of

Ireland Furthermore on such a basis what has happened in Italy would remain

7 For details see Myro et al (2008) 8 See Aacutelvarez Loacutepez et al (2007)

17

Delocation in the manufacturing sectors in the EU A regional overview

unexplained as the selected key sectors do not show a bad performance in terms of exports

in the latter country

One reason for such contradictory results lies in the possibility that production

instead of export reflects accurately what has been happening in every sector in terms of

activity However Aacutelvarez et al (2007) show very similar patterns in production and

export specialization from 1995 to 2005 except in the case of Ireland which thus appears

as a very particular one This is because IT specialization in production increases at the

same time as it decreases in exports Perhaps the strong growth in domestic demand

prevents IT exports from growing The opposite case could be that of Italy where slow

growth in domestic demand might have been pushing up IT export9

Anyway further developments on the determinants of delocation patterns require the

preparation of disaggregated data on production a task beyond the objectives of this paper

That could allow us to update the available manufacturing specialization analysis As the

two countries exhibiting higher delocation effects France and the United Kingdom

showed the lowest rates of specialization compared with the EU (Milderfart and Overman

2002 Combes and Overman 2003) something could have changed in the last few years

which would merit a fresh look based on more recent data

On the other hand it is now clear that insofar as delocation in manufacturing

activities has affected each country in a different way a relocation process inside the EU

and between countries has taken place This process deserved to be deeply analyzed

including a perspective by sectors updating what we know about changes in the

distribution of manufacturing industry in the EU and offering the possibility of exploring

whether the European Common Policy has influenced them in some way as was already

suggested by Milderfart-Knarvik and Overman (2002)

9 In Finland IT activities grows much more in production than in exports but in both aspects a positive growth is registered

18

Delocation in the manufacturing sectors in the EU A regional overview

19

Figure 3 Employment growth rates in selected regions 2001-2005 manufacturing and total economy (percentages)

Ukg3

UKi2

UKd4

UKj1

UKm2

UKi1 UKj3

UKd3

NL 41

NL22

FR23

NL33

NL32

DEa1

FR42

DEa5

DK01

FI18

ITe1

FR 24FR10

FI19

SE11

ITd5FR52

DE 14

DE71

DEa2

BE23

DE22

DE21

DE27

IC c4

DE60DE 12

DE25

DE13

DE9 1SE12

ITd3

IE02

DE73

BE21

DE92DEa4

DEf0

DEb3

ES30

DEc0

GR30

DE23

DE11

DK04

AT31

ITe3

SE22

FR82

FR51FR61

SE2 3

DE26

DE24

IT e4

FR62

FR71

ES51

ES2 1

ITe1

AT22

UKh1

UKk1

UKf1

Uke4

UKm3

UKf2

-20

-15

-10

-5

0

5

10

15

20

25

-40 -30 -20 -10 0 10 20 30

Manufacturing

Source Elaborated from Eurostat and National Statistical offices

Tota

l em

ploy

men

t

Delocation in the manufacturing sectors in the EU A regional overview

One last question that deserves consideration is the effect of industrial delocation

process on the regional economies Surprisingly manufacturing job losses have not been

translated into total employment which exhibit positive growth for most of the chosen

regions with the exceptions of some Germans and all the Netherlands (Figure 3) That

means the regions have been able to compensate the negative effect on the employment

from shrinking manufacturing allocating more resources inside the services sectors In fact

a substitution of manufacturing by services seems to have happened as are the region

showing more delocation in manufacturing those that increase in a higher measure the

employment in the service sectors that is the French British and Italian However the

relation is not accurate as there are not a strict quantitative correlation region by region

between rates of delocation and increase in the employment in the service activities

These good results in terms of general employment shown by the regions affected in

a higher measure by manufacturing delocation suggest the convenience of a more general

analysis of regional economic growth beyond the manufacturing sector as some regions

strongly shrinking manufacturing activities in favour of those of services might be leading

an interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

5- Final remarks

Comparative analysis of location patterns for a large sample of European regions -

characterized by having an upper-middle level income and a strong industrial sector-

during the last few years has revealed that only half of them have shown locational

attractions for manufacturing activities The regional patterns found clearly show the

influence of `national effectacute the regions showing location effects corresponding to most

of those located in Austria Germany the Netherlands Ireland Finland Sweden Spain and

Greece

On the contrary the delocation process has affected another half of the regions

included most of them French British and Italian Therefore delocation has contributed to

concentrating the European manufacturing industry in the areas with higher market

potential in the centre of continental Europe It has contributed furthermore to relocating

manufacturing activities inside every country something that requires further analysis

from future research

20

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

unexplained as the selected key sectors do not show a bad performance in terms of exports

in the latter country

One reason for such contradictory results lies in the possibility that production

instead of export reflects accurately what has been happening in every sector in terms of

activity However Aacutelvarez et al (2007) show very similar patterns in production and

export specialization from 1995 to 2005 except in the case of Ireland which thus appears

as a very particular one This is because IT specialization in production increases at the

same time as it decreases in exports Perhaps the strong growth in domestic demand

prevents IT exports from growing The opposite case could be that of Italy where slow

growth in domestic demand might have been pushing up IT export9

Anyway further developments on the determinants of delocation patterns require the

preparation of disaggregated data on production a task beyond the objectives of this paper

That could allow us to update the available manufacturing specialization analysis As the

two countries exhibiting higher delocation effects France and the United Kingdom

showed the lowest rates of specialization compared with the EU (Milderfart and Overman

2002 Combes and Overman 2003) something could have changed in the last few years

which would merit a fresh look based on more recent data

On the other hand it is now clear that insofar as delocation in manufacturing

activities has affected each country in a different way a relocation process inside the EU

and between countries has taken place This process deserved to be deeply analyzed

including a perspective by sectors updating what we know about changes in the

distribution of manufacturing industry in the EU and offering the possibility of exploring

whether the European Common Policy has influenced them in some way as was already

suggested by Milderfart-Knarvik and Overman (2002)

9 In Finland IT activities grows much more in production than in exports but in both aspects a positive growth is registered

18

Delocation in the manufacturing sectors in the EU A regional overview

19

Figure 3 Employment growth rates in selected regions 2001-2005 manufacturing and total economy (percentages)

Ukg3

UKi2

UKd4

UKj1

UKm2

UKi1 UKj3

UKd3

NL 41

NL22

FR23

NL33

NL32

DEa1

FR42

DEa5

DK01

FI18

ITe1

FR 24FR10

FI19

SE11

ITd5FR52

DE 14

DE71

DEa2

BE23

DE22

DE21

DE27

IC c4

DE60DE 12

DE25

DE13

DE9 1SE12

ITd3

IE02

DE73

BE21

DE92DEa4

DEf0

DEb3

ES30

DEc0

GR30

DE23

DE11

DK04

AT31

ITe3

SE22

FR82

FR51FR61

SE2 3

DE26

DE24

IT e4

FR62

FR71

ES51

ES2 1

ITe1

AT22

UKh1

UKk1

UKf1

Uke4

UKm3

UKf2

-20

-15

-10

-5

0

5

10

15

20

25

-40 -30 -20 -10 0 10 20 30

Manufacturing

Source Elaborated from Eurostat and National Statistical offices

Tota

l em

ploy

men

t

Delocation in the manufacturing sectors in the EU A regional overview

One last question that deserves consideration is the effect of industrial delocation

process on the regional economies Surprisingly manufacturing job losses have not been

translated into total employment which exhibit positive growth for most of the chosen

regions with the exceptions of some Germans and all the Netherlands (Figure 3) That

means the regions have been able to compensate the negative effect on the employment

from shrinking manufacturing allocating more resources inside the services sectors In fact

a substitution of manufacturing by services seems to have happened as are the region

showing more delocation in manufacturing those that increase in a higher measure the

employment in the service sectors that is the French British and Italian However the

relation is not accurate as there are not a strict quantitative correlation region by region

between rates of delocation and increase in the employment in the service activities

These good results in terms of general employment shown by the regions affected in

a higher measure by manufacturing delocation suggest the convenience of a more general

analysis of regional economic growth beyond the manufacturing sector as some regions

strongly shrinking manufacturing activities in favour of those of services might be leading

an interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

5- Final remarks

Comparative analysis of location patterns for a large sample of European regions -

characterized by having an upper-middle level income and a strong industrial sector-

during the last few years has revealed that only half of them have shown locational

attractions for manufacturing activities The regional patterns found clearly show the

influence of `national effectacute the regions showing location effects corresponding to most

of those located in Austria Germany the Netherlands Ireland Finland Sweden Spain and

Greece

On the contrary the delocation process has affected another half of the regions

included most of them French British and Italian Therefore delocation has contributed to

concentrating the European manufacturing industry in the areas with higher market

potential in the centre of continental Europe It has contributed furthermore to relocating

manufacturing activities inside every country something that requires further analysis

from future research

20

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

19

Figure 3 Employment growth rates in selected regions 2001-2005 manufacturing and total economy (percentages)

Ukg3

UKi2

UKd4

UKj1

UKm2

UKi1 UKj3

UKd3

NL 41

NL22

FR23

NL33

NL32

DEa1

FR42

DEa5

DK01

FI18

ITe1

FR 24FR10

FI19

SE11

ITd5FR52

DE 14

DE71

DEa2

BE23

DE22

DE21

DE27

IC c4

DE60DE 12

DE25

DE13

DE9 1SE12

ITd3

IE02

DE73

BE21

DE92DEa4

DEf0

DEb3

ES30

DEc0

GR30

DE23

DE11

DK04

AT31

ITe3

SE22

FR82

FR51FR61

SE2 3

DE26

DE24

IT e4

FR62

FR71

ES51

ES2 1

ITe1

AT22

UKh1

UKk1

UKf1

Uke4

UKm3

UKf2

-20

-15

-10

-5

0

5

10

15

20

25

-40 -30 -20 -10 0 10 20 30

Manufacturing

Source Elaborated from Eurostat and National Statistical offices

Tota

l em

ploy

men

t

Delocation in the manufacturing sectors in the EU A regional overview

One last question that deserves consideration is the effect of industrial delocation

process on the regional economies Surprisingly manufacturing job losses have not been

translated into total employment which exhibit positive growth for most of the chosen

regions with the exceptions of some Germans and all the Netherlands (Figure 3) That

means the regions have been able to compensate the negative effect on the employment

from shrinking manufacturing allocating more resources inside the services sectors In fact

a substitution of manufacturing by services seems to have happened as are the region

showing more delocation in manufacturing those that increase in a higher measure the

employment in the service sectors that is the French British and Italian However the

relation is not accurate as there are not a strict quantitative correlation region by region

between rates of delocation and increase in the employment in the service activities

These good results in terms of general employment shown by the regions affected in

a higher measure by manufacturing delocation suggest the convenience of a more general

analysis of regional economic growth beyond the manufacturing sector as some regions

strongly shrinking manufacturing activities in favour of those of services might be leading

an interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

5- Final remarks

Comparative analysis of location patterns for a large sample of European regions -

characterized by having an upper-middle level income and a strong industrial sector-

during the last few years has revealed that only half of them have shown locational

attractions for manufacturing activities The regional patterns found clearly show the

influence of `national effectacute the regions showing location effects corresponding to most

of those located in Austria Germany the Netherlands Ireland Finland Sweden Spain and

Greece

On the contrary the delocation process has affected another half of the regions

included most of them French British and Italian Therefore delocation has contributed to

concentrating the European manufacturing industry in the areas with higher market

potential in the centre of continental Europe It has contributed furthermore to relocating

manufacturing activities inside every country something that requires further analysis

from future research

20

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

One last question that deserves consideration is the effect of industrial delocation

process on the regional economies Surprisingly manufacturing job losses have not been

translated into total employment which exhibit positive growth for most of the chosen

regions with the exceptions of some Germans and all the Netherlands (Figure 3) That

means the regions have been able to compensate the negative effect on the employment

from shrinking manufacturing allocating more resources inside the services sectors In fact

a substitution of manufacturing by services seems to have happened as are the region

showing more delocation in manufacturing those that increase in a higher measure the

employment in the service sectors that is the French British and Italian However the

relation is not accurate as there are not a strict quantitative correlation region by region

between rates of delocation and increase in the employment in the service activities

These good results in terms of general employment shown by the regions affected in

a higher measure by manufacturing delocation suggest the convenience of a more general

analysis of regional economic growth beyond the manufacturing sector as some regions

strongly shrinking manufacturing activities in favour of those of services might be leading

an interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

5- Final remarks

Comparative analysis of location patterns for a large sample of European regions -

characterized by having an upper-middle level income and a strong industrial sector-

during the last few years has revealed that only half of them have shown locational

attractions for manufacturing activities The regional patterns found clearly show the

influence of `national effectacute the regions showing location effects corresponding to most

of those located in Austria Germany the Netherlands Ireland Finland Sweden Spain and

Greece

On the contrary the delocation process has affected another half of the regions

included most of them French British and Italian Therefore delocation has contributed to

concentrating the European manufacturing industry in the areas with higher market

potential in the centre of continental Europe It has contributed furthermore to relocating

manufacturing activities inside every country something that requires further analysis

from future research

20

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

Attractiveness to location of manufacturing industries has been encouraged by

significant increases in labor productivity so in most of the regions showing positive

location patterns production increases have been compatibles with a decline in aggregate

manufacturing total employment More specifically in this situation are all regions in the

Netherlands Finland Greece and Ireland three quarters of the German ones two from the

United Kingdom two from Sweden and Madrid in Spain

On the contrary in the regions where delocation has taken place the increase in

labour productivity has been lower In fact thirteen of such regions show reductions in

labour productivity six of them placed in France three Italian the British Eastern Scotland

and Greater London Vaumlstsverige in Sweden and the Belgian Antwerp Besides the

positive impact of this rare performance on employment has been largely offset by

destruction of industrial activity except in the Icircle de France and Midi-Pyreacuteneacutees the two

territories in which productivity has a more negative trajectory

The positive link between location and labour productivity increase confirm the

result reached by other authors particularly Cuadrado Roura et al (2002) who referring

their analysis to previous years distinguish two groups of winners among the European

regions the first grouping regions with increased employment and labour productivity

perhaps reflecting the success of technological and organizational changes introduced

during the period promoting the diversification and specialization in activities with high

growth levels and attracting new investments less labour-intensive and a second one

formed by regions capable of implementing restructuring processes to eliminate the most

inefficient outputs achieve higher levels of productivity and increased industrial

production at the expense of a notable decline in employment

Nevertheless discrepancies in delocation patterns outlined above and different

achievements in terms of productivity improvement have not only been the result of

companiesacute response to intensified competition in each region but also of differential

features On the one hand is size which in larger regions like some of Italy Germany

France and Spain reduces the chances of scoring high growth rates On the other hand are

the national specificities such as the massive migrations received by the Spanish regions a

key to explain their model of industrial growth and the different public policies The

influence of the two latter factors appears remarkable given the similarities among the

location patterns of each of the regions examined and the Member States they belong to

21

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

Anyway the lack of data has prevented us from going further into a preliminary

explanatory analysis of the delocation patters found when looking at the different

manufacturing activities but the role played by the IT manufacturing sector and other

branches affected by intense international competition (automobile clothes and textiles

and shoes) might have been important

To finish just to point up that the effect of manufacturing delocation on aggregate

employment has been offset by growth of services sectors suggesting the convenience of a

more general analysis of regional economic growth beyond the manufacturing sector as

some regions showing greater change in employment between sectors might be leading an

interesting process of structural change that could light up one of the ways of facing the

increasing competition coming from emerging economies

22

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

References

Aacutelvarez Loacutepez ME R Myro and J Vega Crespo (2007) ldquoCambios Recientes en la

Especializacioacuten Inter-industrial de las Manufacturas Espantildeolasrdquo Papeles de Economiacutea

Espantildeola 112 2-21

Antragraves P and E Helpman (2004) ldquoGlobal Sourcingrdquo Journal of Political Economy

112(3) 552-580

Bjerring Olsen K (2006) Productivity Impacts of Offshoring and Outsourcing A Review

OECD Directorate for Science Technology and Industry

Baldwin RE and F Robert-Nicoud (2000) ldquoFree trade agreements without delocationrdquo

Canadian Journal of Economics 33(3) 766-786

Bhagwati J A Panagariya and TN Srinivasan (2004) ldquoThe Muddles over Outsourcingrdquo

Journal of Economic Perspectives 18(4) 93-114

Combes PP and HG Overman (2003) ldquoThe Spatial Distribution of Economic Activities

in the EUrdquo Center for Economic Performance London School of Economics and

Political Science

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2000) ldquoRegional

Productivity Patterns in Europe An Alternative Approachrdquo The Annals of Regional

Science 34(3) 365-384

Cuadrado Roura JR (2001) Regional convergence in the European Union From

hypothesis to the actual trends The Annals of Regional Science 35(3) 333-356

Cuadrado-Roura JR T Mancha-Navarro and R Garrido-Yserte (2002) ldquoRegional

Dynamics in the European Union Winners and Losersrdquo in Cuadrado-Roura JR and

M Parellada (eds) Regional Convergente in the European Union Springer-Verlag

Berlin 23-52

Fujita M P Krugman and A Venables (1999) The Spatial Economy Cambridge MA

The MIT Press

Grossman GM and E Rossi-Hansberg (2006) The Rise of Offshoring Itacutes Not Wine for

Cloth Anymore Princeton University unpublished

Helpman E D Marin and T Verdier (2008) The organization of firms in a global

economy Harvard University Press

Krugman P and AJ Venables (1990) ldquoIntegration and the Competitiveness of Peripheral

Industryrdquo in C Bliss and J Braga de Macedo (eds) Unity with Diversity in the

European Community Cambridge University Press

23

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

Krugman P (1999) Geography and Trade Cambridge MA The MIT Press

Midelfart-Knarvik KH and HG Overman (2002) Delocation and European Integration

is structural spending justified Economic Policy 17(35) 321-359

Myro R CM Fernaacutendez-Otheo L Labrador AB Baides Tudela ME Aacutelvarez Loacutepez

and Vega Crespo J (2008) Globalizacioacuten y Deslocalizacioacuten Importancia y Efectos

para la Industria Espantildeola Direccioacuten General de Poliacutetica de la PYME Ministerio de

Industria Turismo y Comercio (httpwwwipymeorgNRrdonlyresAD74121A-

DD3C-4564-92AD-17A2872508CA0GlobalizacionDeslocalizacionpdf)

Puga D (1999) The Rise and Fall of Regional Inequalities European Economic Review

43(2) 303-334

Puga D (2002) European Regional Policy in Light of Recent Location Theories Journal

of Economic Geography 2(4) 373-406

Sachwald F (2004) ldquoThe impact of EU enlargement on Firmsacute Strategies and the

Location of Production in Europerdquo Tokyo Club Research Meeting

24

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

25

Appendix 1 European regions NUTS2 in the sampleCode Region name Code Region name

NUTS2 and country to which it belong NUTS2 and country to which it belongat22 Steiermark gr30 Attikiat31 Oberoumlsterreich Greece

Austriaie02 Southern and Eastern

be21 Prov Antwerpen Irelandbe23 Prov Oost-Vlaanderen

Belgium itc1 Piemonteitc4 Lombardia

de11 Stuttgart itd3 Venetode12 Karlsruhe itd5 Emilia-Romagnade13 Freiburg ite1 Toscanade14 Tuumlbingen ite3 Marchede21 Oberbayern ite4 Laziode22 Niederbayern Italyde23 Oberpfalzde24 Oberfranken nl22 Gelderlandde25 Mittelfranken nl32 Noord-Hollandde26 Unterfranken nl33 Zuid-Hollandde27 Schwaben nl41 Noord-Brabantde60 Hamburg Netherlandsde71 Darmstadtde73 Kassel se11 Stockholmde91 Braunschweig se12 Oumlstra Mellansverigede92 Hannover se22 Sydsverigedea1 Duumlsseldorf se23 Vaumlstsverigedea2 Koumlln Swedendea4 Detmolddea5 Arnsberg ukd3 Greater Manchesterdeb3 Rheinhessen-Pfalz ukd4 Lancashiredec0 Saarland uke4 West Yorkshiredef0 Schleswig-Holstein ukf1 Derbyshire and Nottinghamshire

Germany ukf2 Leicestershire Rutland and Northantsukg3 West Midlands

dk01 Hovedstaden ukh1 East Angliadk04 Midtjylland uki1 Inner London

Denmark uki2 Outer Londonukj1 Berkshire Bucks and Oxfordshire

es21 Pais Vasco ukj3 Hampshire and Isle of Wightes30 Comunidad de Madrid ukk1 Gloucestershire Wiltshire and Bristoles51 Cataluntildea ukm2 Eastern Scotland

Spain ukm3 South Western ScotlandUnited Kingdom

fi18 Etelauml-Suomifi19 Laumlnsi-Suomi

Finland

fr10 Icircle de Francefr23 Haute-Normandiefr24 Centrefr42 Alsacefr51 Pays de la Loirefr52 Bretagnefr61 Aquitainefr62 Midi-Pyreacuteneacuteesfr71 Rhocircne-Alpesfr82 Provence-Alpes-Cocircte dAzur

France

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

26

Manufacturing GVA GDPpc( of total EU-15) (Index EU-15=100)

at22 Steiermark 040 9647at31 Oberoumlsterreich 063 10641be21 Prov Antwerpen 064 12780be23 Prov Oost-Vlaanderen 040 9589de11 Stuttgart 241 12296de12 Karlsruhe 132 11674de13 Freiburg 099 10018de14 Tuumlbingen 095 10780de21 Oberbayern 205 14684de22 Niederbayern 051 10090de23 Oberpfalz 046 10475de24 Oberfranken 047 9885de25 Mittelfranken 079 12022de26 Unterfranken 056 10283de27 Schwaben 084 10593de60 Hamburg 062 17922de71 Darmstadt 130 13985de73 Kassel 041 10006de91 Braunschweig 081 9675de92 Hannover 058 10000dea1 Duumlsseldorf 176 11397dea2 Koumlln 129 10387dea4 Detmold 085 9742dea5 Arnsberg 143 9395deb3 Rheinhessen-Pfalz 082 9505dec0 Saarland 040 9778def0 Schleswig-Holstein 061 9099dk01 Hovedstaden 043 18287dk04 Midtjylland 041 13581es21 Pais Vasco 070 9977es30 Comunidad de Madrid 093 10213es51 Cataluntildea 180 9310fi18 Etelauml-Suomi 123 13009fi19 Laumlnsi-Suomi 064 9954fr10 Icircle de France 221 15821fr23 Haute-Normandie 043 9333fr24 Centre 052 9211fr42 Alsace 048 9664fr51 Pays de la Loire 072 9353fr52 Bretagne 051 9124fr61 Aquitaine 044 9276fr62 Midi-Pyreacuteneacutees 040 9221fr71 Rhocircne-Alpes 140 10349fr82 Provence-Alpes-Cocircte dAzur 055 9715gr30 Attiki 039 9144ie02 Southern and Eastern 196 16068itc1 Piemonte 128 9973itc4 Lombardia 390 11863itd3 Veneto 176 10747itd5 Emilia-Romagna 164 11132ite1 Toscana 090 9929ite3 Marche 047 9078ite4 Lazio 062 11123nl22 Gelderland 043 9892nl32 Noord-Holland 051 13804nl33 Zuid-Holland 077 12003nl41 Noord-Brabant 098 11722se11 Stockholm 062 17037se12 Oumlstra Mellansverige 056 10433se22 Sydsverige 044 10865se23 Vaumlstsverige 074 11743ukd3 Greater Manchester 055 10374ukd4 Lancashire 046 9161uke4 West Yorkshire 054 10378ukf1 Derbyshire and Nottinghamshire 065 10283ukf2 Leicestershire Rutland and Northants 053 11179ukg3 West Midlands 071 10596ukh1 East Anglia 050 10169uki1 Inner London 067 28535uki2 Outer London 055 10231ukj1 Berkshire Bucks and Oxfordshire 058 15836ukj3 Hampshire and Isle of Wight 043 11334ukk1 Gloucestershire Wiltshire and Bristol 061 12628ukm2 Eastern Scotland 044 10947ukm3 South Western Scotland 053 10097

Source Elaborated from Eurostat and national Statistical offices

Appendix 2 GDP per Capita and regions share in EU-15 manufacturing GVA 2005

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 2 (Continued)

0

5

10

15

20

25

30

100 120 140 160 180 200 220 240 260 280

26

23

9

54

12

12

10 0 0 0 0 0 0 0 0

1

Reg iona l GD P per Ca pita (EU-1 5 =1 0 0 ) Histo g ra m

0

10

20

30

40

05 10 15 20 25 30 35 40

14

34

9

3

6

13 3

10 0 0 0 0

1

Industrial size of selected regions ( of EU-15 manufacturing GVA)Histogram

27

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

Appendix 3 Location effect and productivity growth rates in European manufacturing sector

(percentages)

Loca

tion

effe

ct (

of 2

000

empl

oym

ent)

Productivity grothw

AUSTRIA

AT

AT31AT221

-5

0

5

10

-5 0 5 10

GERMANY

DE91

DEa4

Dec0

EDf0

DE 24

DE71DE92

DE14

DE 26

DE73DE 60

DE 22

DE23

DE25

DE 13

DE a1

DEa2

DE27

DE 21

DE a5

DE12DE

DEb3

DE11

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

BELGIUM

BE21

BE

BE23

-10

-5

0

5

-5 0 5 10 15

DENMARK

DK04

DK

DK01

-10

-5

0

5

-5 0 5 10 15

SPAIN

ES

ES30

ES21

ES51

-5

0

5

10

-5 0 5 10

FRANCE

FR51

FR71

FR42

FR23

FR62

FR61

FR82FR52

FR24FR

FR10

-25

-20

-15

-10

-5

0

5

-35 -30 -25 -20 -15 -10 -5 0 5 10

HOLAND

NL33

NL22

NL41

NL32

NL

-5

0

5

10

15

20

-5 0 5 10 15 20 25 30

ITALY

ITe4

ITc1

ITe1

ITd3

ITd5

ITe3

ITc4

IT

-25

-20

-15

-10

-5

0

5

-10 -5 0 5 10

28

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29

Delocation in the manufacturing sectors in the EU A regional overview

Productivity grothw

Source Elaborated from Eurostat and National Statistical offices

Appendix 3 (continued)Lo

catio

n ef

fect

( o

f 200

0 em

ploy

men

t)FINLAND

FI FI19

FI18

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

GREECE

GR30GR

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30 35

IRELAND

IEIE02

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

UNITED KINDOGDOM

UKh1

UKm2

UKf1

UKj3

UKd3

UKf2

UKi2

UKk1

UKg3UKe4

UKm3

UKj1

UKd4

UKi1

UK

-35

-30

-25

-20

-15

-10

-5

0

5

10

-15 -10 -5 0 5 10 15 20 25 30 35 40

SWEDEN

SE11

SE12

SE23 SE22

SE

-5

0

5

10

15

20

25

30

35

-5 0 5 10 15 20 25 30 35 40 45

29