III Price Competitiveness - IMF eLibrary

III Price Competitiveness

Perhaps the element that is most commonly taken intoaccount in attempting to assess whether a country's ex-change rate is at an appropriate level is its price competi-tiveness. Such an assessment normally takes the form ofa straightforward comparison of past movements inprices at home and abroad, after adjustment for exchangerate changes. If the comparison indicates that the countryconsidered has experienced significant losses or gains inits price competitiveness in relation to a base periodwhen its position was considered adequate, this findingis viewed as establishing a presumption that the currentexchange rate is unsustainable.8 Sustainability can, ofcourse, be restored through a change in the nominalexchange rate, a change in the level of domestic currencyprices relative to those abroad, or some combination ofthe two. In many cases, a comparison of movements inproduction costs is made as a supplement, or an alterna-tive, to a comparison of movements in prices. Indicatorsof relative prices or costs adjusted for exchange ratechanges are here referred to as "real exchange rates."

The Fund's regular Article IV consultation reports onmember countries normally include a discussion of therelevant country's international competitiveness, and thisdiscussion is one of the bases for the section of the staffappraisal that deals with the exchange rate. In addition,following the review of surveillance procedures in March1983, the Fund decided to initiate, on an experimentalbasis, a system whereby the Executive Board would benotified regularly of all large changes in real effectiveexchange rates.

Outside the Fund, national authorities in virtually allindustrial countries also pay close attention to variousindicators of their cost and price competitiveness. Forexample, the recent report on intervention states that:

While a comparison of inflation rates—even in the longerrun—requires considerable judgment regarding thechoice of appropriate price indices and the base period,members agreed that it was a useful concept for assessingthe exchange rate.9

8Instead of analyzing changes in aggregate price indices relative to abase period, a more direct method of assessment would be to comparethe prices of specific goods and services in the home market with thoseprevailing abroad, after expressing both in a common currency. How-ever, the necessary data are not normally available for a sufficientlywide range of commodities.

9"Report of Working Group on Exchange Market Intervention,"p. 31.

In general, conclusions obtained from the applicationof this method are only considered as tentative. They areuseful to sound an alert but, because such indicators canbe quite distorted, they should not per se create alarm.Given the dearth of other suitable indicators, however,there are times when considerable practical importance isattached to the results of analyses of internationalcompetitiveness.

This section outlines the major assumptions thatunderlie the use of indicators of international competi-tiveness for the assessment of exchange rates and thendescribes the main statistical measures of competi-tiveness that are employed by the staff. The reliability ofthe preferred indicator—that based on relative nor-malized unit labor costs in manufacturing—is then con-sidered in detail. After a comment on competitivenessindicators based on consumer prices, the main conclu-sions of the section are briefly summarized.

Basic Assumptions of the CompetitivenessAnalysis

The use of indicators of international price competi-tiveness to assess the exchange rate is an application ofthe relative purchasing power parity (PPP) theory ofexchange rates. Without entering into the intricacies ofthe various versions of this theory, it is important to notethat it deals with currency relationships where: (1) coun-tries produce similar goods and there are no major im-pediments to free trade, (2) adjustment lags in the goodsmarkets are considerable because of information lags andother market inefficiencies, and (3) countries' overallprice levels in local currency tend to be determinedprimarily by domestic factors.

The first assumption is necessary in order for ex-change rates to converge toward their PPP level in thelonger run. If countries produced completely differentgoods, or if there were major impediments to free trade,there would be little or no tendency—even in the longerrun—for countries' price levels to converge when con-verted into the same numeraire currency. The secondassumption is required for the view that PPP holds in thelong run to be consistent with short-run divergencesbetween national price levels; if goods market pricesadjusted immediately there would be no need for the

6

©International Monetary Fund. Not for Redistribution

Assumptions of the Competitiveness Analysis

analysis of trends in competitiveness, since nominal ex-change rates would continuously move to offset inflationrate differentials. The third assumption is necessary forthe analysis to be relevant to the assessment of theexchange rate, rather than relative overall price levels inlocal currencies.

The empirical relevance of the second assumptionconcerning the existence of adjustment lags is widelyaccepted, since there is ample evidence that buyers in theinternational goods market do not immediately changetheir sources of supply in response to month-to-month(or even year-to-year) changes in relative prices. Morebasic issues arise in judging the empirical relevance ofthe first and third assumptions. On the whole, eventhough there is room for concern about restrictions thatimpede free trade in goods and services, industrial coun-tries tend to maintain a fairly liberal trade system withrelatively low tariffs, at least as far as nonagriculturalproducts are concerned.

There is much evidence, however, that the assumptionthat countries produce similar goods—that is, goods thatare close substitutes in demand—does not fit with condi-tions in the real world. In the first place, the degree ofspecialization in various categories of tradable goodsvaries significantly across countries. For example, Aus-tralia and Canada export relatively more raw and semi-processed materials than most other industrial countries.

More generally, specialization along the lines of com-parative advantage is the basic characteristic of free in-ternational trade. To illustrate this point, the structure ofproduction of the manufacturing sector—the dominantsegment of the tradable-goods producing sector in indus-trial countries—is presented in Table 1 for the threelargest industrial countries and for three smaller coun-tries. The table indicates significant cross-country differ-ences in the proportion of total value added that is con-tributed by the various industries. It is also relevant tonote that evidence at the industry level, such as thatgiven in Table 1, does not reflect the full extent ofstructural differences because much specialization occursat the level of individual products, rather than acrossbroad industrial categories. Furthermore, even in caseswhere the degree of specialization at the individual prod-uct level is similar in different countries, the specifictypes of commodities produced may still be greatly dif-ferentiated on account of differences in quality, deliverydates, after-sales service, and many other characteristics.

A second reason for differences in the composition ofoutput across countries is that there are many goods andservices whose use is country specific. For example,there is obviously little arbitrage between prices of newlyproduced houses in the United States and in France.Many types of services, including both government ser-vices and those provided by the private sector, are also

Table 1. Selected Countries:

Food, beverages,and tobacco

Textiles, weaving,apparel, and leather

Wood and wood products

Chemicals andrubber products

Petroleum refining

Nonmetallic minerals

Iron and steel

Nonferrous metals

Metal products

Nonelectricalmachinery

Electrical machinery

Precision instruments

Road vehicles

Aircraft

Ships

Miscellaneous

Shares

UnitedStates

13.1

7.4

12.7

11.1

2.5

3.7

5.3

2.1

6.6

8.6

11.1

2.5

7.4

2.9

1.6

1.6

of Value AddedFederal

Republic ofGermany

9.5

6.7

10.5

13.0

0.5

5.0

4.5

0.8

11.6

12.3

11.2

3.4

5.7

0.3

1.2

3.8

in Manufacturing1

Japan

10.5

6.7

10.1

9.7

4.7

4.2

3.9

1.6

6.4

11.7

10.5

1.8

6.0

—

4.1

3.1

Italy

9.3

15.1

11.0

11.6

1.2

6.7

4.9

1.1

7.2

8.3

6.6

2.4

4.8

1.2

1.4

1.2

Belgium

15.4

11.4

9.6

9.8

6.9

7.1

9.3

2.6

7.6

6.6

5.4

1.6

3.9

0.9

0.9

0.9

Sweden

10.6

7.0

22.2

7.8

0.8

4.8

5.2

1.0

3.6

17.8

6.8

1.2

7.2

0.3

2.9

0.6

Source: The input-output tables used were published by the statistical office of the European Communities or, in the case of non-EC countries, bynational statistical offices. The years of the tables are 1972 for the United States, 1975 for the Federal Republic of Germany, 1970 for Japan, 1975for Italy, 1970 for Belgium, and 1968 for Sweden.

1Because the level of disaggregation of the input-output tables varies across countries, not all categories of manufactures were available on aconsistent basis for all countries. For some countries, the disaggregation of metals into ferrous and nonferrous, of machinery into electrical,nonelectrical, and precision instruments, and of transport equipment into road vehicles, aircraft, and ships was estimated by the staff.

7


III • PRICE COMPETITIVENESS

country specific. It is possible to exclude most of these"nontradable" goods from comparisons of internationalprice levels, but this only reduces the problem. Forexample, if a country's tradable goods prices remain inline with those of similar goods in the world market, itwill not experience a decline in domestic and foreigndemands for these goods. However, if the ratio of non-tradable goods prices to tradable goods prices increases,it may still experience a shift in domestic productionfrom the tradable sector to the nontradable sector. Thus,an analysis that focuses on the relative prices of tradablegoods among countries—without taking into account rel-ative prices between tradable and nontradable goodswithin each country—may at times be misleading be-cause it ignores a major determinant of the sustainableexchange rate.

These observations are not intended to detract from therole played by relative prices in a country's foreign tradeperformance. Nevertheless, they do suggest that the PPPtheory has to be qualified in two respects. First, nomatter how much care is taken in the choice of the priceindices, the resulting indicator of competitiveness maystill fail to measure the true level of competitivenessbecause of composition effects; that is, because theweights given to the various items in the price indicesdiffer significantly across countries. This problem maybe quite significant because rates of growth of productiv-ity and, therefore, rates of change of prices, differ sub-stantially among sectors. For example, a country mayappear to be gaining competitiveness on the basis ofaggregate price indices when in fact the relative changein the indices may be caused by the country's heavyspecialization in an industry that has a relatively high rateof productivity growth and a declining price trend whencompared with other industries. Thus, whenever possi-ble, a study of competitiveness should include an exam-ination of whether the results are likely to be influencedby composition effects.

Second, the PPP theory also has to be qualified to takeinto account the fact that product differentiation amongcountries significantly limits the elasticities of substitu-tion among the goods that they supply, even in the longerrun. This has the major implication that any sustainedchanges in a country's endowment of economic re-sources, the pattern of demand for its products, or thedeterminants of its net capital flows may have to beoffset by a change in price competitiveness if the coun-try's overall payments balance is to remain in equilib-rium in the longer run. Thus, any assessment of anexchange rate requires both a consideration of changes incompetitiveness that have actually taken place and ananalysis of the change that would be called for to restorethe overall balance of payments to a sustainable level.10

There are also cases, particularly among the smallerindustrial countries, where the validity of the third as-sumption of the PPP approach—namely, that the overallprice level in local currency tends to be determinedmainly by domestic factors—is less tenable. As long asrigidities in the goods and labor markets are not tooextreme, overall price levels will normally remain con-sistent with domestic monetary developments in the lon-ger run. In this sense, domestic monetary policy can beviewed as the major determinant of a country's overallprice level. Nevertheless, large changes in nominal ex-change rates can create pressures on the monetary au-thorities that cannot always be fully resisted. Althoughthe strength of these pressures differs from country tocountry depending in part on the degree of openness ofthe economy and the pervasiveness of wage indexation,some allowance must normally be made for the feedbackeffect of exchange rate changes on costs and prices inlocal currency. For example, a change in the nominalexchange rate of 10 percent might induce a more or lessautomatic rise in domestic currency costs and prices of,say, 3 percent. Thus, a study that concluded that thenominal exchange rate was 10 percent too high on thebasis of current prices at home and abroad might implythat the nominal rate should be changed by 13 percent inorder for it to return to a sustainable level in real terms.

Some Measures of Competitiveness Usedby the Fund Staff

Most of the work done in the Fund on the measure-ment of competitiveness among industrial countries hasfocused on the five quarterly indicators now published inInternational Financial Statistics (IFS). All five indi-cators measure aspects of the international competi-tiveness of a country's manufacturing sector. The firstthree indices—those for relative export unit values,wholesale prices, and value added deflators—reflectprice competitiveness. The last two indices—those forrelative unit labor costs and normalized unit laborcosts—are intended to reflect cost competitiveness. Theweighting scheme for all of the indicators has been de-signed to take account not only of bilateral trade, but alsoof the degree to which various countries effectively com-pete because they tend to export the same types of prod-ucts to the same geographic markets.11

For the purpose of assessing the sustainability of ex-change rates, the first two indicators have importantlimitations that tend to cause them to underestimatechanges in international price competitiveness. For theindicator based on export unit values this problem arises

10This issue will be considered in more detail in Section IV, below.

11The weighting scheme is built up from disaggregated trade datafor manufactures and reflects both the relative importance of a coun-try's trading partners in its bilateral trade and the relative importanceof its competitors in third markets.

8


Reliability of the Normalized Unit Labor Cost Indicator

because, even though goods may be differentiated, inter-national competition places limits on how far an export-er's prices can diverge from those charged by competi-tors. The indicator also fails to reflect the prices of goodsthat are not exported anymore because their domesticproduction costs exceed the prices at which they wouldbe competitive in world markets. Similarly, prices forthe relatively homogeneous products that have a heavyweight in the wholesale price index tend to be similaracross countries after adjustment for exchange ratechanges, but this is hardly a sign that competitivenessamong countries is constant.12 In addition, bothwholesale and export price indicators are often stronglybiased as a result of composition effects because thestructure of exports and the weighting structure ofwholesale price indices differ substantially acrosscountries.

While the next indicator of price competitiveness—theindex based on value added deflators in manufacturing—is normally more relevant, it also has important disad-vantages. The main drawback of this indicator is thatactual data are published with a long lag and are avail-able quarterly for only a few countries. It is generallynecessary for the staff to extrapolate for five or sixquarters beyond the most recent benchmark year on thebasis of wholesale prices for manufactures adjusted toexclude the influence of changes in raw materials prices.This is a severe limitation since, as explained above,movements in wholesale price indices can often be amisleading indicator of changes in competitiveness.

Given the problems with measures of price competi-tiveness, more reliance is generally placed on indices oflabor cost developments in the manufacturing sector.However, the fourth indicator—that based on currentlabor costs—is very sensitive to cyclical changes in pro-ductivity over periods of several years. For this reason,the indicator based on normalized unit labor costs (i.e.,adjusted for cyclical variations in productivity) has gen-erally been regarded as the most useful in assessingexchange rates, and it was recently selected to implementthe experimental system of information notices for the 13member countries for which the indicator is available.13

Thus, it is particularly useful to discuss the main char-acteristics of this indicator at some length and to analyzeits past behavior.

The series on normalized unit labor costs is calculatedby dividing an index of actual hourly compensation perworker by an index of output per manhour in the manu-facturing sector adjusted so as to eliminate the estimated

12The wholesale price index is also subject to problems of doublecounting, since the prices of certain intermediate commodities may becounted both separately and as part of final output.

13The data are published for the following member countries: theUnited States, Canada, Japan, France, Federal Republic of Germany,Italy, Norway, Sweden, Denmark, Austria, Belgium, the Nether-lands, and the United Kingdom; they are also available forSwitzerland.

effect of the cyclical swings in economic activity onproductivity.14 The indicator for each country representsthe ratio of that country's index of normalized unit laborcosts to a weighted geometric average of correspondingindices for the other 13 industrial countries included inthe analysis, after all of the national indices are ex-pressed in terms of a common currency.

To the extent that the price structure in a country'sfactor markets tends to fluctuate less over time than theprice structure of its goods markets, the normalized unitlabor cost indicator benefits from its focus on factorprices rather than goods prices; in particular, there is lessrisk of a composition bias. Moreover, factor prices in thetradable goods sector—particularly money wage rates—are less subject than goods prices to direct foreign influ-ences. Price indices for factors of production in thetradable goods sector can, therefore, be used directly.There is less need to estimate the relative long-run evolu-tion of the tradable versus the nontradable sectors, or tomake allowance for the distorting effects of competitors'prices on domestic prices.

However, these conceptual advantages are relativerather than absolute; that is, they are significant mainlyin relation to the major disadvantages of alternative indi-cators. In particular, the risk of serious composition bias,while smaller than for alternative indicators, is not neg-ligible. Rates of growth of productivity may differwidely among industries and, as indicated above, suchcross-country differences may reflect primarily their dif-fering compositions of total manufacturing output.Moreover, since the only factor of production included inthis indicator is labor, it may be biased because of cross-country differences in longer-run rates of change incapital/labor ratios. To eliminate this source of bias itwould be necessary to adjust the changes in productivity(in terms of output per manhour) for the effects ofchanges in capital/labor ratios; this is a difficult task thatso far has not been attempted by the staff.

Reliability of the Normalized Unit LaborCost Indicator

The practical reliability of the normalized unit laborcost indicator can be assessed by (i) considering itshistorical performance, (ii) analyzing whether merchan-dise trade flows are, in fact, responsive to movements inthis particular indicator of international competitiveness,and (iii) determining whether developments in nor-malized unit labor costs in local currencies are largelyindependent of nominal exchange rate movements.

14A technical note on the indicator is provided in each issue of IFS,listed in the Contents as "Cost and Price Comparisons forManufacturing."

9



Historical Performance of the Normalized UnitLabor Cost Indicator

As an aid to analyzing the historical behavior of theindicator of relative normalized unit labor costs, Chart 2presents quarterly data for 12 member countries.15 Theperiod covered in the chart runs from the beginning of1963 to the first quarter of 1983. On the strict assump-tions of the PPP theory, one would expect the data for thereal effective exchange rate based on normalized unitlabor costs to exhibit relatively short-lived fluctuationsaround a more or less stable long-run trend. Further-more, one would also expect fluctuations around thetrend to be closely associated with overshooting of thenominal exchange rate resulting from short-run devel-opments in financial markets. Even if these assumptionswere not fully satisfied, however, international pricecomparisons could still be useful so long as the timingand magnitude of shifts in cost and price competitivenesswere clearly associated with recognized major structuralchanges in the goods and capital markets.

The information in Chart 2 suggests that, on the abovecriteria, the record of the normalized unit labor costindicator of the real exchange rate during the past 20years has been mixed. First, there are five countries—theUnited States, Japan, Canada, Italy, and the Nether-lands—for which the level of the normalized unit laborcost indicator in the first quarter of 1983 differed by lessthan 15 percent from the level that had prevailed in thefirst quarter of 1963. However, for these countries thedifference between the maximum and minimum levels ofthe real effective exchange rate during the 20-year periodwas more than 30 percent, and in the case of the UnitedStates the average level of the real effective exchangerate during the period 1973-82 was 35 percent lowerthan it had been during the preceding decade. In addi-tion, while marked movements in the real effective ex-change rate were associated with similar movements innominal effective exchange rates, such movementstended to persist rather than being reversed in a fewyears. Examples are the appreciation of the Dutch guil-der between 1963 and 1974, and the depreciation of theU.S. dollar between 1969 and 1973.

Second, in two cases—Norway between 1969 and1977 and the United Kingdom between 1976 and 1981—the sharp appreciation in the real effective exchange ratewas associated with a major structural change, namely,the discovery and exploitation of North Sea oil reserves.Even in these cases, however, it is noteworthy that thechanges in real effective exchange rates associated withthe structural shifts were very large, raising the question

15As noted above, the indicators of the cost and price competi-tiveness for the manufacturing sector are available in IFS for 14countries. For purposes of exposition, however, only 12 of thesecountries are included in the panels of Chart 2. The omitted countriesare Austria and Switzerland.

of whether the movements that actually occurred were anappropriate response to the change in comparative ad-vantage, or whether they partly represented overshootingof the nominal effective exchange rate. The competi-tiveness indicators, taken by themselves, provide onlylimited information on this issue. Furthermore, while theshifts that occurred in Norway and the United Kingdomwere particularly large ones, most other industrial econo-mies also experienced significant structural changes dur-ing this period. The magnitudes of the changes in the realeffective exchange rates of Norway and the United King-dom suggest that even the effects of smaller structuralchanges on the sustainable exchange rate cannot neces-sarily be neglected in making assessments of interna-tional price competitiveness.

Third, there are several cases where the indicator de-picted in Chart 2 did not suggest a worsening of interna-tional competitiveness, but the countries concerned ex-perienced growing external imbalances. For example,the current account and overall payments positions ofBelgium and Sweden began to weaken from the mid-1970s, even though their international competitivenessas measured by the relative unit labor cost indicatorshowed little deterioration during this period. Further-more, despite the substantial apparent improvement in itscompetitiveness according to the conventional measuresduring the period 1979-81, Belgium failed to show asignificant recovery of export volumes, manufacturingsector employment, or current account balance. In Swe-den, the combined effects of the 1981 and 1982 devalua-tions led in 1983 to improved international competi-tiveness, a gain in market shares for manufacturedexports, and an improvement in the profitability ofSwedish industry. Nevertheless, the experience of thesetwo countries suggests the need for caution in interpret-ing statistical indicators of competitiveness.

The country cases cited above suggest that there is avariety of situations where standard indicators such asrelative normalized unit labor costs appear to provideonly limited information about changes in a country'sinternational competitiveness or in the sustainable levelof its exchange rate. For this reason, the Fund, in analyz-ing exchange rate sustainability on the basis of competi-tiveness indicators, has on occasion extended its analysisto incorporate other types of evidence including trends inmanufacturing sector profitability, export market shares,and capital formation in the sectors exposed to foreigncompetition. Such supplementary indicators suggest thepossibility that during the later years of the 1970s thesecountries experienced a significant deterioration in theiroverall international competitiveness, even though theirindices of relative normalized unit labor costs (Chart 2)were showing an improvement. Two likely causes ofsuch a development are a loss of comparative advantagein traditional export industries (such as shipbuilding,machine tools, and autos (Sweden), and textiles and steel

10



Chart 2. Selected Industrial Countries: Relative Normalized Unit Labor Costs in Manufacturing, Adjustedfor Exchange Rate Changes, First Quarter 1963-Third Quarter 1983(Period average 1963-82 - 100)

Mean 1963 -72period Mean 1973 -82 period

UNITED STATES FRANCE

FEDERAL REPUBLIC OF GERMANY UNITED KINGDOM

JAPAN CANADA

(Belgium)); and/or a process of "marginalization" inwhich the squeeze on profits caused by domestic costincreases in excess of the rise in world market pricesforced the least viable firms in the tradable goods sectorout of business.

For example, during the mid-1970s Belgium experi-enced strong wage-push pressures that sharply reducedthe rate of growth of output in the manufacturing sectorand led to the fastest rate of decline in manufacturingsector employment among the industrial countries. But,

because marginal firms were being eliminated at a rela-tively rapid rate, the measured growth of labor produc-tivity in the Belgian manufacturing sector accelerated,and standard indicators of the real effective exchangerate actually showed a gain in Belgium's internationalcompetitiveness after 1977. Similar developments mayhave affected the indices of competitiveness for suchcountries as Denmark, the Netherlands, and Sweden.Such explanations provide an ex post rationale for thelack of correlation between the apparent improvement in

11


Ill • PRICE COMPETITIVENESS

Chart 2 {concluded). Selected Industrial Countries: Relative Normalized Unit Labor Costs in Manufacturing,Adjusted for Exchange Rate Changes, First Quarter 1963-Third Quarter 1983(Period average 1963-82 = 100)

Mean 1963 -72 period Mean 1973 -82 period

BELGIUM NETHERLANDS

DENMARK NORWAY

ITALY SWEDEN

relative unit labor costs and current account devel-opments, but they give little indication of the size of thechange in the real exchange rate index that would haveproduced a sustainable overall payments position.

Fourth, in some cases the timing of the change in theindicator depicted in Chart 2 differed substantially fromthe timing of major structural changes in the goods andcapital markets. For example, between 1969 and 1974the real effective exchange rate of the U.S. dollar depre-ciated by 29 percent, that for the Federal Republic of

Germany appreciated by 32 percent, and that for Japanappreciated by 15 percent. These large changes in theindicators of international competitiveness occurred dur-ing the transition from the Bretton Woods regime offixed exchange rates to one of managed flexibility, butthey were not associated with any specific developmentsin the structure of goods or capital markets. It is widelyassumed that during the period prior to 1970-73 therewas a steady deterioration in the competitiveness of theUnited States relative to Germany and Japan which ac-

12



counted for the large change that occurred subsequently.However, the empirical indicator of cost and price com-petitiveness, which remained relatively stable for allthree of these countries during the period from 1963 to1969 (Chart 2), gave little prior warning about either thesize of this disequilibrium or the need for a change in realexchange rates. Even after the fact, the timing of ex-change rate movements can only be attributed to thechange that occurred in the exchange rate regime, sug-gesting that during this period national authorities wereable, through their policy actions, to maintain exchangerates at unsustainable levels for substantial periods oftime. While such leeway has doubtless been reduced tothe extent that capital has become increasingly mobilesince the early 1970s, the possibility remains that com-petitiveness indices may give only limited warning ofsuch misalignments.

Fifth, since standard indicators of a country's interna-tional competitiveness are based on weighted averages ofbilateral real exchange rates, the issue arises as towhether the weights used are an accurate reflection of therelative importance of partner countries. If this were notthe case, changes in bilateral real exchange rates betweencountries that are close trading partners could lead tosubstantial longer-term changes in the structure of trade,even if—because of offsetting movements in bilateralrates—standard measures of overall competitiveness ap-peared stable. This issue is important because actualmovements in bilateral rates have been considerable.Chart 3 presents data on the real bilateral exchange ratesbetween the deutsche mark and the currencies of five ofits partner countries (France, Italy, Netherlands, Bel-gium, and Denmark) in the European Monetary System(EMS). It is apparent from these data that even in thecase of closely integrated economies that belong to aregional exchange arrangement, movements in real bilat-eral exchange rates were rather large, both prior to thesetting up of the EMS and subsequently.

Responsiveness of Trade Flows to Changes inthe Normalized Unit Labor Cost Indicator

An alternative way of assessing the reliability of thenormalized unit labor cost indicator is to see how well itperforms as an explanatory variable in an econometricmodel of foreign trade flows. To be a highly reliableguide for assessing the sustainability of the exchangerate, the indicator should consistently show a strongeffect on foreign trade flows, especially in the long run.This linkage should be reflected in relatively large andstatistically robust estimates of the long-run elasticitiesof import and export volumes with respect to theindicator.

As an experiment, the equations for the volumes ofexports and imports of manufactures in the Fund's worldtrade model16 were re-estimated using the normalizedunit labor cost indicator as the variable reflecting eachcountry's international competitiveness. The results arepresented in Table 2. The estimated long-run elasticitiesvary considerably across countries, but, on the whole,they cannot be described as large. For example, con-sidering the case of a country where imports and exportsare initially equal and where the estimated elasticities are0.5 in absolute terms for both imports and exports, a10 percent increase in relative normalized unit laborcosts would lead to a trade balance deficit in constantprices amounting to 10 percent of the initial flow ofexports or imports. Allowing for a rise in the terms oftrade of say 5 percent,17 the deficit in nominal termswould be only 5 percent of the initial flow of exports orimports. Moreover, the standard errors of the estimatedelasticities are rather high, indicating that the magnitudeof the response of foreign trade flows to the standardmeasure of cost competitiveness remains uncertain.

The evidence in Table 2 must be viewed as quitetentative. In particular, it is extremely difficult to distin-guish among trend developments, structural shifts, andthe longer-run effects of lagged competitiveness changesin the econometric determination of foreign trade flows.Thus, the relatively low estimated elasticities with re-spect to the chosen competitiveness indicator may wellbe the result of the necessarily simplistic specification ofthe trade equations.18 Nevertheless, the evidence doessuggest that the responsiveness of trade flows to thisindicator, while sizable, is not so large and so preciselyknown that the indicator, taken by itself, can always berelied upon as a fully satisfactory measure of the sustain-able exchange rate.

Feedback Effects of Exchange Rate Changes onNormalized Unit Labor Costs

A further issue that arises in using indicators of inter-national competitiveness is that of the extent to which amajor change in the nominal effective exchange rate willcreate feedback effects that alter the levels of prices or

16See Michael C. Deppler and Duncan M. Ripley, "The WorldTrade Model: Merchandise Trade," Staff Papers, International Mone-tary Fund (Washington), Vol. 25 (March 1978), pp. 147-206.

17See Jacques R. Artus, "The Behavior of Export Prices for Manu-factures," Staff Papers, International Monetary Fund (Washington),Vol. 21 (November 1974), pp. 583-604.

18In addition to specification problems related to the trade equa-tions, the low estimated elasticities are likely to reflect the existence ofmeasurement errors in the normalized unit labor cost indicator, as wellas the fact that this indicator is far from being a perfect proxy for thecompetitiveness of exportables.

13



Chart 3. Selected Members of EMS: Bilateral Relative Normalized Unit Labor Costs in Manufacturing,Adjusted for Exchange Rate Changes, First Quarter 1963-Third Quarter 1983(Period average 1963-82 = 100)

DEUTSCHE MARKIN TERMS OF FRENCH FRANC

DEUTSCHE MARKIN TERMS OF BELGIAN FRANC

DEUTSCHE MARKIN TERMS OF ITALIAN LIRA

DEUTSCHE MARKIN TERMS OF DANISH KRONE

DEUTSCHE MARKIN TERMS OF NETHERLANDS GUILDER

unit labor costs in a direction that tends to restore pur-chasing power parity. Chart 4 provides evidence on thisissue for periods corresponding to relatively largechanges in the nominal effective exchange rates of threemajor industrial countries (the United States, Japan, andthe Federal Republic of Germany) and three of thesmaller countries (Belgium, Norway, and Sweden). Inthe charts, the dashed line is the nominal effective ex-change rate (foreign currency per unit of domestic cur-rency), and the solid line is the ratio of partner countryunit labor costs expressed in local currency to those inthe country under consideration. If there is a significant

feedback effect from exchange rates on domestic laborcosts, then one would expect to see an appreciation of thenominal effective exchange rate (a rise in the dashedline) followed by a rise in the solid line, reflecting aslower rate of increase in domestic-currency unit laborcosts relative to those prevailing abroad. The six panelsin Chart 4, however, provide little evidence that thefeedback effects of exchange rate changes on domesticunit labor costs are significant for either the larger or thesmaller countries considered.

Note on Consumer Price Indicators

Even though, as noted above, most of the work donein the Fund on the measurement of competitivenessamong industrial countries has focused on the five indi-cators for manufacturing published in IFS—particularlyon the indicator based on normalized unit labor costs—there are a number of countries where considerations ofdata availability have so far precluded the calculation ofsuch indicators. For these countries, as well as for devel-

14



Table 2. Fourteen Industrial Countries: Elasticities of Imports and Exportsof Manufactures With Respect to Relative Normalized Unit Labor Costs1

Austria

Belgium

Canada

Denmark

France

Germany, FederalRepublic of

Italy

Japan

Netherlands

Norway

Sweden

Switzerland

United Kingdom

United States

1Elasticities are defined such that their expected signs are negative. The impact elasticity gives the response of import or export volume in the firstsix months; the long-run elasticity gives the total response, which is generally assumed to occur within 2-3 years. Standard errors are in parentheses.

ImpactElasticity

-0.09(0.22)

—

-0.12(0.19)

-0.17(0.09)

-0.23(0.12)

-0.08(0.18)

-0.58(0.30)

-0.06(0.14)

-0.18(0.44)

-0.11(0.30)

-0.33(0.24)

—

-0.26(0.12)

—

Import Volume

Long-RunElasticity

-0.72(0.33)

—

-0.12(0.19)

-0.17(0.09)

-0.60(0.21)

-0.49(0.18)

-0.58(0.30)

-1.27(0.32)

-0.20(0.55)

-0.47(0.59)

-0.45(0.28)

-1.10(0.25)

-0.78(0.22)

-1.03(0.23)

ImpactElasticity

-0.33(0.16)

-0.06(0.17)

-0.33(0.08)

-0.12(0.17)

-0.13(0.07)

—

—

-0.35(0.10)

—

-0.23(0.21)

-0.42(0.11)

—

—

-0.26(0.17)

Export Volume

Long-RunElasticity

-0.56(0.24)

-0.52(0.29)

-1.14(0.26)

-0.12(0.17)

-0.48(0.14)

-0.36(0.13)

-0.51(0.40)

-1.30(0.28)

-0.07(0.40)

-1.25(0.70)

-0.82(0.17)

-0.18(0.18)

-0.16(0.18)

-0.93(0.19)

oping countries, the measurement of competitiveness isoften based on consumer price indices, as for example inthe system of information notices for real exchange ratemovements that is being undertaken on an experimentalbasis by the Fund. For this reason, it is useful to comparethe indicator based on consumer prices with that basedon normalized unit labor costs in cases where the twoindicators can be calculated. To illustrate the results ofsuch a comparison, the left-hand panels of Chart 5 dis-play the real bilateral exchange rates between Japan andthe United States, the Federal Republic of Germany, andthe United Kingdom, respectively, based on both con-sumer price indices (dotted line) and indices of nor-malized unit labor costs (solid line). For all three cases,the two indices essentially show a similar pattern ofvariation throughout the period chosen. Nevertheless,owing to their differing longer-term trend movements,the levels of the two indicators diverged markedly by theend of the period of the two decades shown in the chart.For example, compared with the first quarter 1963 baseperiod, the cumulative net appreciation of Japan's real

exchange rate with the United States in the first quarterof 1983 was 71 percent, when measured in terms of theindicator based on consumer prices, but only 29 percentwhen measured in terms of normalized unit labor costs.Japan's real exchange rate with the Federal Republic ofGermany experienced a net appreciation of nearly64 percent in terms of consumer prices over the sameperiod, whereas the normalized unit labor cost indicator,by contrast, indicated a net depreciation of 6 percent.Similarly, using the former indicator, Japan's real ex-change rate with the United Kingdom experienced a netappreciation of 66 percent, while the latter measureshowed a depreciation of 8 percent against sterling.

Given Japan's trade performance as compared with theother three countries cited in Chart 5, the indications ofcumulative changes in Japan's international competi-tiveness provided by the normalized unit labor cost indi-cator seem much more plausible than those based onconsumer prices. This conclusion is not surprising, sincelong-run increases in the relative prices of nontradedgoods were a major source of trend movements in Ja-

15



Chart 4. Selected Industrial Countries: Nominal Effective Exchange Rates and Relative Normalized UnitLabor Costs, in Local Currencies(First quarter of sample period = 100)

Nominal effective exchange rate1 Relative normalized unit labor cost, in local currencies2

UNITED STATES BELGIUM

JAPAN NORWAY

FEDERAL REPUBLIC OF GERMANY SWEDEN

foreign currency per unit of domestic currency.2Ratio of partners' to own labor costs.

pan's consumer price index during the period consid-ered. These observations suggest that although relativenormalized unit labor costs and consumer prices maygive similar indications of short run developments, theformer indicator is superior for the analysis of trends ininternational competitiveness over periods spanning sev-eral years.

A further problem that arises in using competitiveness

indicators based on consumer prices, illustrated in theright-hand panel of Chart 5, is the fact that consumerprices appear to be more sensitive than unit labor costindicators to the feedback effects of nominal exchangerate changes, particularly because a significant portion ofthe consumption basket consists of imports. The right-hand panels of Chart 5 are the same as those presented inChart 4, except that a dotted line has been added to

16



Chart 5. Selected Industrial Countries: Indicators of Competitiveness Based on Consumer Prices

Bilateral relative normalized unit labor costs,adjusted for exchange rate changes

Bilateral relative consumer price indices,adjusted for exchange rate changes

Nominal effective exchange rate1

Relative normalized unit labor costs, in local currencies2

Relative consumer price indices, in local currencies2

JAPAN VIS-A-VISUNITED STATES

BELGIUM

JAPAN VIS-A-VISFEDERAL REPUBLICOF GERMANY

NORWAY

JAPAN VIS-A-VISUNITED KINGDOM

SWEDEN

'Foreign currency per unit of domestic currency.2Ratio of partners' to own labor costs or consumer prices.

represent the ratio of partner-country to domestic con-sumer prices, with each measured in local currency. Inall three cases, there appears to be some evidence of afeedback from exchange rate changes to domestic con-sumer prices. For example, the February 1982 devalua-

tion in Belgium appears to have had a stronger feedbackeffect on domestic consumer prices than on unit laborcosts in local currency, owing to the large weight ofimported goods in the consumer price index and theimplementation of a strict incomes policy.

17



Summary

The discussion of this section suggests the need forcaution in using standard indices of international com-petitiveness to assess the longer-run sustainability of acountry's exchange rate and external position, particu-larly at times when major changes are occurring inunderlying real sector conditions. There can often be

cases where a given exchange rate is no longer sustain-able, despite the apparent stability of conventional mea-sures of the real effective exchange rate, and the oppositecase is no less common. In such circumstances, evidencebased on indices of competitiveness must be supple-mented by other information on the economy, such asdata on the return on capital in the tradable goods sector,or on the growth of the nontradables sector (particularlypublic services) relative to the traded sector over time.

18


III Price Competitiveness - IMF eLibrary

Documents

Transcript of III Price Competitiveness - IMF eLibrary