FDI Attractiveness of Indian States and Location Choice of ...

GLAMS Research Publications

Technical Papers

• N Viswanadham and Roshan Gaonkar, A Conceptual and Analytical Frameworkfor the Management of Risk in Supply Chains, to appear in IEEE TASE, 2007.

• D Garg, Y Narahari, and N Viswanadham, Achieving sharp deliveries in Sup-ply Chains through Variance Pool Allocation, European Journal of OperationsResearch, May 2006.

• N Viswanadham and Kannan Balaji, Foreign Direct Investment or Outsourcing :A Tax Integrated Supply Chain Decision Model, to appear in E-business and E-commerce, Kluwer, Eds: Christopher Tang, Wei Kwok Kee and CP Teo, 2006.

• Sunil Chopra, Usha Mohan and Gilles Reinhardt, The Importance of DecouplingRecurrent and Disruption Risks in a Supply Chain, to appear in Naval LogisticsResearch, 2006.

• N Viswanadham and S Kameshwaran, Location Selection in Global Supply Chains,2006.

• Milind Sohoni, Sunil Chopra, Usha Mohan and M Nuri Sendil, Impact of Stair-Step Incentives and Dealer Structures on a Manufacturers Sales Variance, 2005.

• Ram Bala, Pricing of Software Services, 2005.

• Ram Bala, Pricing and Market Segmentation for Software Upgrades, Ram Bala,2005.

• Ram Bala, Renting of Software Services under Competition, 2004.

• Milind Sohoni, Ellis L Johnson and T Glenn Bailey, Operational Airline ReserveCrew Planning, 2004.

Books

• Achieving Rural and Global Supply Chain Excellence: The Indian Way, Eds: NViswanadham, December 2006.

Thought Leadership Papers

• N Viswanadham and Roshan Gaonkar, E-Logistics - Trends and Opportunities,TDB Tradelink Web-site, Jan 2001.

• N Viswanadham, Supply Chain Automation: The Past, Present and Future, June2001.

• N Viswanadham and Roshan Gaonkar, Foundations of E-Supply Chains.

Achieving Rural & GlobalSupply Chain Excellence

The Indian Way

Edited by

N ViswanadhamExecutive Director, GLAMSIndian School of BusinessHyderabad 500032

Center for Global Logistics and Manufacturing StrategiesIndian School of BusinessGachibowli, Hyderabad 500032

Disclaimer: The contributors of this book have used their best efforts inpreparing this manuscript. These efforts include the research and devel-opment of the theories and methods to determine their usefulness. Thecontributors or the ISB makes no warranty of any kind, expressed or im-plied, with regard to the documentation contained in this book.

Copyright c! 2006 by ISB

All rights reserved. No part of this publication may be reproduced, stored ina retrieval system or transmitted in any form or by any means, mechanical,photocopying, recording, or otherwise, without the prior written permissionof GLAMS, ISB, Gachibowli, Hyderabad 500032

Printed in India. Not for sale.

Released byHonourable Prime Minister of India

Dr Manmohan Singh,during the inaugural session of the

Global Logistics Summit 2006December 5-6, 2006

Indian School of BusinessGachibowli, Hyderabad 500032

Contents

List of Figures viii

List of Tables ix

Foreword 1M Rammohan Rao

Introduction 3N Viswanadham

1 Can India be the Food Basket for the World? 9N Viswanadham

1.1 Introduction 91.2 Current State of Indian Agricultural Industry 101.3 The Indian Food Processing Industry 111.4 The Food Supply Chain 121.5 Food Supply Chain Cluster 131.6 Government Initiatives to Promote Food Exports 171.7 Private Sector Initiatives 191.8 Opportunities for Improving Food Supply Chain 191.9 Opportunities for Research in Food Supply Chain 201.10 Conclusions 21

References 22

2 Dynamics of Retail in India: Case Studies of Andhra Pradesh andPunjab 23

N Viswanadham and Navolina Patnaik

2.1 Introduction 232.2 Food and Retail in India 252.3 Food and Retail in Andhra Pradesh 30

xi

xii Contents

2.4 Food and Retail in Punjab 362.5 Conclusions 43

References 46

3 Impact of Avian Influenza in the Indian Poultry Industry: A SupplyChain Risk Perspective 49

N Viswanadham, Usha Mohan, and Prachi Trikha

3.1 Introduction 493.2 Literature Review 503.3 Avian Influenza: Background and Analysis 523.4 Supply Chain Risk Mitigation Strategies 593.5 Poultry Supply Chain in India 603.6 Risk Management in Supply Chains 643.7 Conclusions and Discussions 68

References 69

4 Rural Business Transformation - Empowering Villages using Kisan-bandhu 71

N Viswanadham and D Ramakrishna

4.1 Introduction 714.2 Private and Public Sector Initiatives in India 734.3 Rural Business Transformation (RBT) 764.4 Village as a Business System 804.5 Rural Supply Chain Transformation 834.6 Rural Retailing 874.7 Conclusion 89

References 90

5 Design of Special Economic Zones as Economic Engines of Growth 93N Viswanadham

5.1 Introduction 935.2 Background on SEZs 945.3 SEZ Policy 2005 975.4 Economic and Technical Issues concerning SEZs 1015.5 Performance of the Indian SEZs 1035.6 Market Attractiveness of Indian States 1055.7 Design of SEZ 1075.8 Selection of Location 1105.9 Conclusions 114

Contents xiii

References 114

6 FDI Attractiveness of Indian States and Location Choice of MNCs 117N Viswanadham and S Kameshwaran

6.1 Introduction 1176.2 Location Choice Problem 1196.3 Related Literature 1246.4 Methodology 1286.5 Analysis: PERC based Hierarchical Structuring 1316.6 Synthesis: Analytic Hierarchy Process 1396.7 Location of a Biotech R&D Centre 1406.8 Measuring FDI and Market Attractiveness 1456.9 Final Notes and Proposals 146

References 148

7 Design of Competitive Indian Construction Supply Chain Networks 153N Viswanadham and Vinit Kumar

7.1 Introduction 1537.2 Construction Supply Chains: Characteristics 1557.3 Indian Construction Supply Chains: Status, Issues & Scope 1577.4 Indian Construction Supply Chains: Strategies 1687.5 Conclusions 184

Appendix: Cases of Government Infrastructure Projects & Causes ofDelays 185Appendix: Indian Construction System 187References 189

Index 193

xiv Contents

6 FDI Attractiveness of IndianStates and Location Choice ofMNCs

N Viswandham and S Kameshwaran

6.1 Introduction

A global supply chain spans several countries and regions of the globe.Trade liberalization (European Union, NAFTA) and information technologyhave accelerated the growth of global supply chains, whereby a firm can in-vest and trade across national borders. It is now a competitive requirementthat firms invest all over the globe to access markets, technology, and talent.Firms could trade across national borders either by intra-firm-trade usingforeign direct investment (FDI) or arms-length-trade (foreign outsourcing).International trade and FDI have been among the fastest growing economicactivities around the world. FDI is the movement of capital across nationalfrontiers in a manner that grants the investor control over the acquired asset.FDI includes corporate activities such as building plants or subsidiaries inforeign countries, and buying controlling stakes or shares in foreign compa-nies.

FDI stocks now constitutes over 20% of global GDP, which in effect hascontributed to the economic growth of developing countries. Firms locatedin industrialised countries pursue vertical disintegration of their productionprocesses by outsourcing some stages in foreign countries where economicconditions are more advantageous. For example, Intel Corporation assem-bles most of its microchips in wholly-owned subsidiaries in China, CostaRica, Malaysia, and Philippines. With proven results that FDI by multi-national corporations (MNCs) increase employment, exports, revenue, andknowledge spill over to host country’s private/public sectors, many govern-ments have introduced various forms of investment incentives, to encourageforeign owned companies to invest in their jurisdiction.

Pre 1990, India allowed only up to 40% FDI, selectively in few sectors.The FDI policy reforms that started in 1991 has resulted in tremendous in-

117

118 FDI Attractiveness of Indian States and Location Choice of MNCs

crease of FDI flows into the country. Currently FDI up to 100% is permittedin all sectors except a few. The total inflow of FDI has increased almost1321% from USD 264.1 million in 1991 to USD 3.8 billion in 2004 [17]. Thesectors attracting investments are diverse including software, transporta-tion, food processing, chemicals, and metallurgical industries. The top tencountries investing in India include Mauritius, Japan, and South Korea inaddition to USA, Germany, France, and Switzerland.

One of the strategic decisions faced by an MNC is the location decisionof the subsidiary within India. Usually the location choice problem of MNCis only considered at a national level in the literature. This national levelchoice is made in tandem with other organizational decisions like mode ofFDI (joint venture, acquisition, greenfield), investment budget, level of activ-ities at the new subsidiary, etc. Choosing a nation for FDI is hence relateddirectly with many other firm-level decisions. In this work, we considerthe location choice at the subnational level, with focus on location choice inIndia.

An MNC with greenfield FDI has to set up a new subsidiary in In-dia. The diversities in India pose great challenges and also opportunitiesin choosing an optimal location. We consider in this work a single plantlocation. Given the type of the business undertaken by an MNC, it is rel-atively easy to list the potential locations in India. However, choosing theoptimal one taking into account the objectives of the firm and the locationcharacteristics is a non-trivial problem. One can obtain substantial amountof information regarding the locations from specialized sources or the sta-tistical data provided by research agencies and management consultants.However, the challenge is in using the above fragmented data and informa-tion about the locations in a rational way to evaluate and rank the locations.

On the other hand, the state governments and the economic developmentagencies (EDAs) take measures to improve their respective locations in or-der to attract FDI. In order to measure the FDI attractiveness and compareit with other competing locations, it is necessary to identify the locationfactors that influence FDI and synthesize the information to quantify the at-tractiveness. Thus the complementing entities viz. the MNC and the EDA,have a common problem to solve: evaluating a finite number of locations withrespect to a multitude of factors. We develop a decision support framework,which helps in using the vast and diverse information about locations in arational and a systematic way.

Specifically, we are concerned with ranking of N alternate locationsbased on M location attributes. This is a multi-attribute decision analysis prob-lem. Our interest is not about location selection for particular industry likeautomobiles or for a specific kind of facility like warehouse. In contrast, wedevelop a generic framework that can be used by both the MNC and the

Location Choice Problem 119

EDAs for evaluating locations. It is obvious that this problem is unique inthe requirements of the investing firm, the intended business activity, andthe industry. However, they all share certain generic characteristics and inthis work we identify and formalize these characteristics. We develop a hi-erarchical structuring, based on the PERC model [42], to homogeneouslycluster the M location attributes under appropriate criteria. Following thisanalysis stage is the synthesis, where the weights for the attributes and thecriteria are first elicited from the decision maker using the above hierar-chical clusters. This is then combined with the information about the Nlocations for the M attributes to arrive at the final ranking. For this stage,we recommend the use of analytic hierarchy process (AHP), though in prin-ciple other multicriteria decision techniques like multicriteria utility theory(MAUT) can be used. We illustrate the applicability of the framework for alocation decision of a biotech R&D facility in India.

6.2 Location Choice Problem

India is administratively divided into 28 states and 7 union territories. Dueto the diversities in geography, climate, culture, language, tradition, re-sources, and political governance, economic performance of Indian statesvary drastically. For example, the gap in per capita incomes is lot widerbetween the states Maharashtra and Uttar Pradesh, or Punjab and Mad-hya Pradesh, than it is between India and China [46]. The investment cli-mate variations within India is very significant: almost all the FDI wereattracted by the following high performing states: Maharasthra, Delhi, Gu-jarat, Andhra Pradesh, Karnataka, Punjab, Tamil Nadu, and Haryana.

With such diversities in India, it is mandatory for both business andpublic policy decision-makers to consider the location choice problem at amore explicitly local level. The development of areas of free trade and freefactor mobility, such as EU and NAFTA, have led to researchers take intoaccount factors pertaining to the level of the region or city rather than atthe level of the nation. A decision-making process for plant location in EU[50] consisted of selection of country, followed by the selection of the siteswithin the country. Different factors were considered for the above twodecisions. Hence, it is important to identify the differences in the locationchoice problem at the national and the subnational level.


6.2.1 Location Choice: National vs Subnational

An MNC faces two kinds of location decisions in FDI. The first locationdecision of an MNC is to select a nation for FDI. This decision is a complexone as it subsumes within it the following decisions [35]:

• Mode of entry: whether the FDI is implemented through a greenfieldinvestment or an acquisition or joint venture.

• Industry of entry: whether the FDI occurs in the main line of businessof the parent MNC or represents a diversification away from this busi-ness.

Once the country for FDI is determined, the firm has to choose the lo-cation within the country for its intended subsidiary. If the mode of entryis through acquisition, one of the deciding factors is the location of the firmbeing acquired. However, for greenfield investments and joint ventures (insome cases), the MNC faces the subnational location choice problem. Thisproblem inherits many of its characteristics from the national level problem.Based on the nature of the firm, the location choice problem is defined dif-ferently. In this work, we are concerned with a single-plant firm: an MNCthat wishes to locate a single plant or business activity. The location choiceproblem is to find a single location from a given set of potential locations.

6.2.2 Dynamics of the Industrial Location Decision

Consider the example of an MNC interested in setting up a drugs and phar-maceutical plant in India. The industry is clustered in the following fourcities: Ahmedabad, Mumbai, Delhi, and Hyderabad1. The MNC is inter-ested in locating its subsidiary in one of the above four locations. Thisindustrial location decision making is a highly complex process with mul-tifaceted characteristics including tangible and intangible elements that arevery difficult to measure and evaluate [21]. The basic steps involved toarrive at the best possible location recommendation are [32]:

1. The basic requirements of the location project are first identified, usu-ally by using an all-purpose location questionnaire. From this, criticaland desirable factors for the locations are determined. Critical factorsare mandatory factors that play a prohibitive role in identifying lo-cations. For example, if seaport is a mandatory factor then locationswithout seaports should be eliminated for consideration.

1http://www.nifindia.org/bd/list industrial clusters


2. A list of N alternate locations are shortlisted that satisfy the mandatorycritical factors. Matching algorithms are sometimes used to determinethese locations by estimating matching scores between the locationsand the desirable location factors.

3. Based on the intended investment and the nature of the investing firm,M location attributes are identified. These are the location factors overwhich the N locations are to be evaluated.

4. Information about the N locations for each of the M attributes areobtained using public databases, private investigations, and personalmeetings with the local authorities. This includes quantitative infor-mation like economic cost analysis of non-recurring and recurringcosts, prediction of future sales, return of investment, production ef-ficiency, government incentives, etc. If required, negotiation with thegovernment authorities are carried out for special incentives and sub-sidies. Also qualitative information like living conditions and politicalclimate are also determined.

5. The N locations are ranked with respect to the M attributes. This is amulticriteria decision analysis problem.

The first two steps are relatively easier. They are both dependent on theindustry level characteristics and firm level characteristics. There are somemandatory requirements for any business investment. For example, thenew manufacturing plant should have easy access to a port on the easterncoast is a mandatory requirement. This may be because the new plant willhave foreign trade with Singapore and Malaysia. With this constraint, onecan list the possible locations. With several such mandatory requirementsor constraints, identifying the possible locations is a constraint satisfactionproblem. If there is only one location that satisfies all the constraints, thenthe location choice is solved. If there is more than one location, then theoptimal has to be chosen.

The negotiation with the government authorities for incentives and sub-sidies often play a major role in location decision. The Tamil Nadu gov-ernment’s hospitality was one of the influencing factors for Ford’s choice ofChennai for its integrated manufacturing facility (in addition to other fac-tors like skilled workforce, ports, electricity, water, etc) 2. It has been pointedout that the location of FDI is a two player game between the MNC and thehost government. The host government attempts to elicit desired behaviorfrom the MNC using direct (through legislative and executive controls) and

2http://www.chennaibest.com/cityresources/Automotive/fordinterview.asp


Table 6.1 Factors influencing location decisions

Industry Inputs Industrial electricity charge; industrial water charge; avail-ability of land; land cost; labour wage; overstaffing rate;number of white-collar workers; number of blue-collarworkers; educational & training institutes; collaborationwith local universities;

Agglomeration &Network Economies

Localization economies measure; economic diversity:Chinitz-Jacobs diversity measure and Herfindhal mea-sure; location quotient;

CommunicationTechnologies

Number of days to get connections for various technolo-gies: telephone, wireless, internet; bandwidth; networkreadiness index; mail & postal;

Transport Distance to nearby sea port, airport, railway station; trans-portation costs for various modes; domestic and interna-tional connectivity;

Laws & Regulations Difficulty of interface with various government depart-ments: labour, customs and excise, income tax, pollu-tion control, electricity board, water board; corruptionlevel; amount of time spend with government officials;frequency of government official visits;

Economic & Finan-cial

Corporate income tax; imports tax; exports tax; financialincentives; availability of funds and loans; GDP; growthrate; buying power;

Risks Political stability; intellectual property protection; friend-liness of the government; conflicts with the neighboringgovernments; communal disputes;

Living Conditions Consumer price index; crime rate; real estate prices; num-ber of hospitals;

Third Party Services Legal; advertisement; logistics;


indirect (through incentives) stimuli [29]. The literature on the interactionbetween the host government and the MNC is comparatively smaller, sincein the international business intellectual tradition, the firm is always as-sumed to choose amongst several alternative locations, greatly reducing thebargaining power and role of the host government [35]. The multi-lateralinteraction of the MNC with several host governments simultaneously canalso be modeled as a auction game with competitive bidding [10]. In thiswork, our focus is on the evaluation and ranking of the locations, which isthe integral component of the location decision.

The main theme of this work is the last step in the above process, whichis to evaluate and rank the locations. If there is only one location factor, thenthe problem is trivial. With more than one location factor, the problem is amulti-attribute decision analysis problem [51]: evaluating the locations withrespect to multiple number of attributes or factors. For this evaluation, thedecision maker requires the information regarding the locations for each ofthe factors. If the location factor is number of research organizations, thenthe decision maker should know the above for each of the locations. Withthis information, one can determine how good or bad a location is with re-spect to each of the attributes. If there is a location which is superior to theother locations in all the attributes, it is the optimal choice. However, suchlocations rarely exist in reality. Usually, a location is better with respect tosome set of attributes and some other location is better with respect to a dif-ferent set of attributes. Hence the problem is to consolidate and synthesizeall the information to rank the locations with respect to all the attributes.

As mentioned earlier, this problem is also encountered in measuring theFDI or market attractiveness of locations. In the industrial location choiceproblem, the location attributes depend on the type of investing firm whereasin measuring the FDI attractiveness they depend on the industry. For R&D,number of research institutions and IT connectivity are important location fac-tors. On the other hand, for an manufacturing industry, land cost, labourwage, electricity charge, water availability are the factors. Depending on thefirm to locate or the industry, the number of factors can vary from 10 to100. A list of location factors considered in several studies [50, 17, 46] istabulated in Table 6.1 under different categories. The list is not exhaustiveand some of the factors are interrelated. However, it provides a higher levelview of the factors that can influence the location choice decision.

With respect to the drug and pharmaceutical MNC example, the evalu-ation and ranking of the locations can be explained as follows. Firstly, theMNC is assumed to possess the following information:

• Potential locations: Ahmedabad, Mumbai, Delhi, and Hyderabad.


Location Choice Multi-attribute Analysis

Locations AlternativesLocations attributes Attributes

Firm level characteristics Weights for the attributesLocation information Attribute values

Table 6.2 Mapping of the location choice problem as a multi-attribute analysis prob-lem

• Firm level characteristics: Imports, exports, local market, size of the firm,etc.

• Location factors: Transportation costs, labour costs, availability of sup-pliers, ports and logistics network, corporate tax, corruption, electric-ity, etc.

• Location information: Information about the four locations with respectto the above location attributes. The information can be obtainedthrough individual investigation and/or third party databases (Worldbank 3, Indian Investment Centre 4).

Given the above information, one has to evaluate a location based onmultitude of information like transportation costs, tax, etc. The evaluationis achieved through the use of scores and weights. The scores are the at-tributes values for the locations. The weights reflect the relative preferenceof the firm for the attributes. The relative preference or importance of theattributes is based on the firm level characteristics. For example, if the firmhas considerable exports and imports, then presence of a port in the locationis relatively more important. Each location is evaluated by combining thescores using appropriate weights. The mapping of evaluation of the locationsas the multi-attribute decision problem is summarized in Table 6.2. In thenext section, we review the related techniques and models for the locationchoice problem in the literature.

6.3 Related Literature

The industrial location choice problem is addressed by location theory. Thelocation theory is concerned with the geographic location of an economic ac-

3http://www.worldbank.org/sar4http://iic.nic.in/

Related Literature 125

tivity and hence it has become an integral part of economic geography, regionalscience, and spatial economics. The MNC location behavior, in particular, hasbeen studied by international business and management science community.This problem has been studied with various assumptions and views by theabove disciplines. It has been noted that the above disciplines in isolationare unsuitable to explain the current MNCs location behavior [31]. How-ever, the above disciplines, though varying in their perspectives, have iden-tified several significant location determinants and decision insights, whichare used to develop our decision support framework. We briefly reviewthe related literature by broadly classifying them based on the conceptualframeworks.

6.3.1 Location-Production Models

These are the earliest and simplest analytical models and referred com-monly as Weber’s and Moses models. These classical and neoclassical mi-croeconomic models analyze the production behavior of an individual styl-ized firm in relation to the spatial economic costs. These costs include locallabour prices, land costs, transportation costs, and telecommunication costs.The objective is to locate a plant in the plane by minimizing the weightedsum of Euclidean distances from that plant to a finite number of sites cor-responding to the markets where the plant purchases its inputs and sellsits outputs. Some of these models also consider the production as a de-cision variable, which often depends on the location. Interested readersare referred to [30] for an excellent review of the range of microeconomiclocation-production models. In our work, we consider only the locationdecision and production levels are not variables in the model. The com-mon feature of the above models is the consideration of the firm or plant inisolation without any competition from other firms.

6.3.2 Agglomeration Economies

The inclusion of other firms from the same or related industry in the analy-sis, brings out a new set of factors for location decision. The regional sciencecommunity use the term economies of agglomeration to describe the benefitsthat firms obtain when locating near each other. It is related to the idea ofeconomies of scale and network effects, in that the more related firms thatare clustered together, the lower the production cost and the greater themarket. The production costs are lower due to availability of specialized re-sources, such as competing suppliers, skilled labour, and infrastructure. Onthe demand side, the informational externalities from other firms and thereduction in consumer search costs are beneficial for total market demand.


Several studies show that the agglomeration economies are dominant fac-tors in the location choice of MNCs for FDI [22, 8, 6, 4, 20].

This agglomeration phenomena, from management science literature, isexplained using the clusters [38]. Clusters are geographic concentrations ofinterconnected companies and institutions in a particular field. The linkedindustries and institutions can consist of suppliers to universities to govern-ment agencies. Clusters promote both competition and cooperation. For afirm, location in clusters is a source of competitive advantage. The otherrelated model is core periphery model that explains why certain regions orcities attract more industries than the others. Such clustering of industriesare explained through cumulative causation or multiplier effect. New economicgeography [19] uses general equilibrium models to explain the location ofindustrial and economic activities. All these models reinforce that the pres-ence of other firms in the same or related industries is an important locationdecision factor.

6.3.3 International Business Literature

The Dunning’s eclectic OLI (ownership, location, and internalization) [13,14] framework is the widely accepted model for the study of MNCs. TheOLI framework suggests that a firm will prefer FDI to trade and becomean MNC if the following three conditions are satisfied. First, the firm mustpossess ownership advantages not available to other firms in terms of su-perior technology, firm size, brand name, etc. Second, the foreign marketshould offer location specific advantages like market size, cheap resources,and infrastructure. Finally, there should be internalization advantages, whicheliminates the transaction and coordinations costs associated with marketinteraction and internalizes these activities by bringing them inside the hi-erarchy of the firm. The framework is also used in the analysis of locationdecision [34] and mode of entry decision [7]. Accordingly, the location deci-sion is contingent on ownership and internalization factors. This frameworkis more useful for location choice at the national level, which also takes intoaccount the mode of entry.

6.3.4 Multiattribute Decision Models

Except for the location-production models, the above are not prescriptivedecision models that can aid an decision maker (DM) to select a location. Alinear additive MAUT based location evaluation was used to locate a manu-facturing facility in [26]. AHP has been used extensively for a wide varietyof location problems: generic plant location [47, 49]; sure service terminallocation [23]; landfill siting problem [16]; location of international consoli-

Related Literature 127

dation terminals [33]; overseas plant location [50]; global facility location-allocation problem [3]; industrial plant location [1]; and international loca-tion decisions [2]. All the above models first created a hierarchical structureof criteria and used the multicriteria technique (AHP/MAUT) for locationevaluation.

6.3.5 Investment Climate, FDI Attractiveness, and MarketAttractiveness

As noted earlier, the difference between the location choice and measuringmarket attractiveness is that the former is firm specific, while the latter is usu-ally industry specific. Investment climate studies are conducted at nationallevels, comparing different nations, and also at regional and city levels, usu-ally for a specific industry [46]. These studies basically combine differentparameter values in measuring the attractiveness of a location. The prob-lem with these aggregate measures is the implicit assumption that locationcharacteristics have same advantages or disadvantages for all MNCs in anindustry. However, there are evidences that the factors affecting locationchoices are not identical across MNCs and do not exist in isolation from thecharacteristics of the investing firms [36]. For example, good port and cus-toms infrastructure can be a major advantage to firms engaged in exportingand have only more limited and indirect effects on other firms. The rank-ings and measures of investment climate are helpful for public policy [45],for identifying the potential locations, and location determinants. However,for the final choice of the location, the firm level characteristics need to betaken into account, as the costs and benefits of a location depends on them.

6.3.6 Industry Best Practices

Location consultants aid the firms in the complex decision making of se-lecting a location for expansion or investment or relocation. They usuallysupport the whole location selection process (all the steps outlined in Sec-tion 6.2.2), starting from the initial search of locations till the negotiationson investment subsidies and agreements on land and/or buildings. Thebook by Marcel de Meirlier [32] is an excellent source on industrial loca-tion selection process with real world cases and experiences. It should benoted that the location consultants do a lot more than the evaluation of lo-cations. However, the author who has worked on more than 1500 locationcases states that this step (step 5 in Section 6.2.2) is the most difficult andleast understood part of location studies [32]. The author uses a multicrite-ria model with assigning of weights directly in a one to ten scale and manyof the location consultants use this simple linear additive technique. IBM


Plant Location International5 uses a hierarchy of criteria clustered as qualityfactors and cost factors. The weights are directly assigned based on experi-ence. Many popular location consultants listed by Global Direct InvestmentSolutions6 use the technique of directly assigning and combine them in alinear additive fashion to get a final rank.

Buck Consultants International7 (BCI) developed a cost-quality matrixto compare different locations. The matrix has a vertical axis showing allcosts for the next 5 years (labour costs, transport costs, occupancy costs,etc. minus investment subsidies/grants), and a horizontal axis showingthe quality of the investment climate in a weighted form (labour qualityand flexibility, labour regulations, technological expertise available, logisticsqualities of the location, availability of suppliers, etc.). An optimal locationis the one with low costs and high quality business environment.

DealTek8 provides a web based decision support software called as DEALSto search and rank locations in US. The software combines the economic anddemographic data with the user inputs to rank the locations. The user in-puts include the usual project specific information and financial projections.The innovative feature of the software is that it allows the user to inputpossible business scenarios, under varying assumptions on the economyconditions, like Optimistic, Most Likely, and Pessimistic. The locations arefirst shortlisted using user selected criteria from a set of pre-defined criterialike labor cost and labor availability. Then these locations are ranked takinginto account the business scenarios, financial projections, etc. The softwarealso includes what-if sensitivity analysis module, which allows the user tovary the parameter values to see their effect on the rankings.

The above brief survey is intended to provide an overview of the locationselection problem studied from different perspectives. For a more exposi-tory review, see [48] and references therein. In the following we describeour proposed two stage analysis-synthesis decision framework.

6.4 Methodology

Let there be N locations (alternatives) and M location factors (attributes).Let the required information about these N locations with respect to Mfactors be known. Now to evaluate the locations, two numerical entities arerequired:

5http://www.ibm.com/bcs/pli6http://www.gdi-solutions.com/profsvcs/lists/location consultants.htm7http://www.bciglobal.com8http://www.dealtek.com

Methodology 129

Location Factors (Attributes)Att1 · · · Attj · · · AttM Total Scoring

(Weights) w1 · · · wj · · · wM for the Location(Scores)

Loc1 s1,1 · · · s1,j · · · s1,M {WT}{S1}

Locations...

.... . .

.... . .

...(Alternatives) Loci si,1 · · · si,j · · · si,M {WT}{Si}

......

. . ....

. . ....

LocN sN,1 · · · sN,j · · · sN,M {WT}{SN}

Fig. 6.1 Multiattribute Evaluation with Weights and Scores

• Weights: W = {w1, . . . , wj, . . . , wM}.

• Scores: Si = {si1, . . . , sij, . . . , siM}, i = 1, . . . , N.

The relative importance of the location attributes (as perceived by theDM based on the firm’s objectives and constraints) are denoted using theweights W. The scores, on the other hand, rates the performance of a lo-cation for each of the attributes. The total score for a location i is thenobtained by combining the weights W and the scores Si in some mathemat-ically acceptable way, which depends on how the weights and scores arerepresented and obtained. Figure 6.1 illustrates the multicriteria analysisusing scores and weights. In the figure, the total scoring function {WT}{Si}for location i is left unspecified and it can be any mathematically acceptablecombination. The various multicriteria evaluation techniques differ by theway they estimate the weights and scores, and by the way the combine them.They can be formalized in two steps as arranging the criteria into a hier-archy (analysis) and then measuring how well the alternatives perform oneach criterion (synthesis) [37].

6.4.1 Analysis

The analysis starts with the identification of criteria and attributes. We usethe word criteria to refer to objectives or directions along which the DMseeks better performance from the alternatives. The performance is mea-sured in terms of attributes. For example, the criterion economic factors canbe measured using attributes income tax, property tax, and sales tax. Thereis a considerable interplay in the identification of criteria and attributes.This complex creative process is achieved through hierarchically structur-ing homogeneous clusters of criteria and attributes [28, 27, 39]. The result-ing multilevel hierarchy is shown in Figure 6.2. The fundamental criteria


Fundamental

Criterion K

Fundamental

Criterion k

Fundamental

Criterion 1

Goal Sub-Criterion

Attribute i

Attribute j

. . .

. . .

. . .

. . .

. . .

.

.

.

.

.

.

Fig. 6.2 Hierarchical Structuring of Criteria and Attributes

are next to the overall goal and can be further divided into sub-criteria,sub-sub-criteria, and so on, till they cannot be further subdivided. This lastlevel contains measurable attributes, which in our case are the M locationattributes. These are measurable in the sense that numerical scores can begiven to the locations for each of these attributes. The hierarchy impliesa one way dependence relationship from a parent node to its child node.There are models that can incorporate other kinds of dependence like feed-back, by creating a network of criteria and attribute nodes [27, 40].

6.4.2 Synthesis

Given the hierarchy, the next step is the application of a multicriteria de-cision analysis technique, which determines the weights W, scores S, andcombine them to give a final ranking. The hierarchical clustering of criteriaand attributes is used to elicit the weights W from the decision maker. Tobe precise, the weights W are derived for all the criteria and sub-criteria, inaddition to the attributes. The final score for a location is obtained throughmultilinear superposition of weights and scores. First the linear combina-tion of scores with the corresponding weights is obtained for attributes thathave the same parent criterion. This becomes the score for that criterion.This is repeated for all attributes and all criteria at each level, till the goalnode is reached, which gives the final consolidated score for the location.

Analysis: PERC based Hierarchical Structuring 131

Two commonly used multi-attribute decision analysis techniques [5] forproblems involving few alternatives and many attributes are: analytic hierar-chy process (AHP) [39] and multi-attribute utility theory (MAUT) [28, 15]. Thetwo approaches differ both in theory and in practice in the assessment ofcriteria weights and scores. Unlike AHP, there are many ways by whichthe weights and scores can be obtained in MAUT. MAUT allows the DMto directly state the W (values) or estimate it as a utility function identi-fied through risk lotteries. AHP uses pairewise comparisons of hierarchicalfactors to derive W as ratio-scale measures. There have been long standingdebates between these two schools of thought over the mathematical authen-ticity of the techniques involved. Nevertheless, both have been successfullyemployed in practice. An insightful comparison of both the techniques ispresented in [5]. For a comprehensive study of different multicriteria tech-niques the reader is referred to [37].

We propose the use of AHP for the multi-attribute evaluation of loca-tions. AHP has been used to resolve multi-attribute decisions in many hun-dreds of diverse applications by cognizant decision makers. The applica-tions include business and R&D decisions by Xerox, engineering decisionsby NASA, resource allocation decisions by DoE, strategic planning by DoD,etc [18, 41]. It has also been proposed as a decision methodology for locationchoice problem [50] and for evaluating market attractiveness [41, Chapter 9].

6.5 Analysis: PERC based Hierarchical Structuring

The analysis stage consists of identifying fundamental criteria and sub-criteria therein, to homogeneously cluster the locations factors or attributes.This would, to large extent, depend on the industry and the nature of in-vestment. Our aim is to develop a generic analysis framework, which wouldhelp in identifying and grouping the attributes for location selection. We usethe PERC model [42] to identify the activities and factors along four dimen-sions that affect the business operations and output. Using the PERC model,the DM can structure the list of seemingly unrelated and incomparable lo-cation factors hierarchically with suitable sub-categories as shown in Figure6.3.

6.5.1 PERC Model

There is little work in any area of multiple criteria decision making to adviseon how hierarchies should be constructed and what makes a good hierar-chical representation [5]. However, there are broad guidelines on the hierar-chy development process and the properties that a hierarchical structuring


Fundamental Criteria Sub-criteria

Location of Japanese firms in European Commission [50]1. Labour labour force, costs, union;2. Markets product market, raw materials;3. Transport airways, railways, seaport, roadways;4. Financial inducement tax, country risk, loan availability;5. Living conditions firms from host country, educational facilities, crimes,

consumer price index;6. Environment for oper-ations

electricity rate, water charge, sewage facilities, rulesand regulations;

International location decision for manufacturing plants [2]1. Cost direct costs, indirect costs;2. Quality of products labour, infrastructure;3. Time to markets markets, suppliers, macro-environment;

Facility location selection [49]1. Market growth potential, proximity to market, raw materials;2. Transportation land, water, air;3. Labour cost, availability of skilled and semi-skilled labour;4. Community housing, educational, business climate;

Locating global R&D operations [25]1. Demand factors proximity to the final market, growth potential, re-

sponse to local variations2. Supply factors local scientific talent, local technology, know-how3. General competitivefactors

competitive environment

Benchmarking of European locations by IBM PLI [24]1. Cost property costs, labour costs;2. Quality staff availability, language skills, labour laws, inter-

national accessibility, attractivesness for internationalstaff;

Market attractiveness of developing countries [41, Chapter 9]1. Political factors turmoil, strategic relevance;2. Economic-financialfactors

risk of direct investment, GDP, inflation rate, growthrate of GDP;

Investment climate of India for manufacturing industry [46]1. Business environment regulation, corruption, infrastructure, factor markets;2. Agglomerationeconomies

own industry concentration, economic diversity, spa-tial distribution;

Table 6.3 Fundamental Criteria and Sub-Criteria of Various Hierarchical Structuring.


Fig. 6.3 PERC based hierarchical structuring

should posses [28, 27]. In general, this phase is entirely under the control ofthe DM. There are many different ways one can cluster and form a hierarchyof criteria and it is not possible to claim or prove the betterness of one overother. Table 6.3 presents different hierarchical clustering proposed in liter-ature and being used in practice. The list is not exhaustive and it includeslocation selection problems studied from varying perspectives: locating inforeign countries [50, 2]; locating manufacturing industries [49, 1, 2]; locat-ing R&D facilities [25]; industry location consultants [24]; measuring marketattractiveness of countries [41] and locations within a country [46]. There aretwo things that are evident from the Table 6.3. Firstly, the hierarchical struc-ture depends on the particular industry and firm, and for the same type,one can arrive at different structures. The second observation is that thereare some common features in these seemingly different location problems.To complement and extend these efforts, we propose here an hierarchicalstructuring based on [42] the PERC model, which is an abstract higher levelmodel that consists of the following four fundamental criteria:


Fundamental Criteria Sub-criteria

1. Product/Process valuechain

suppliers, markets and demand, agglomerationeconomies and clusters, knowledge sharing and col-laboration;

2. Economic integration economic policies, trade facilitations, laws & regula-tions, financial inducements and incentives, politicalfactors, living conditions;

3. Resources and man-agement

human resources (skilled workforce), financial re-sources (loans, investors), utilities and industry inputs(land, water, power, educational and training insti-tutes), management services (legal, marketing, busi-ness consulting, financial planning);

4. Connecting technolo-gies

transport (rail, road, air, sea), information and com-munication technologies (Internet, wireless, landline,data);

Table 6.4 Fundamental Criteria and Sub-Criteria of PERC Model.

• Product/Process Value Chain

• Economic Integration

• Resources and Management

• Connecting Technologies

The above are the fundamental criteria in the hierarchical structuring.The sub-criteria under each of them are shown in Table 6.4. There wouldbe uncertainties in any kind of business investment. Those that could beleveraged for growth are opportunities and those that would affect the firmnegatively are risks. Their certain counterparts are benefits and costs, re-spectively. To enhance the understanding the four fundamental criteria, thebenefits, opportunities, costs, and risks associated with them are listed inTable 6.5. They are explained in detail in the following.

6.5.2 Product/Process Value Chain

The investment is intended for some business process like manufacturinga product or providing a service. This criterion is about the value chaindimension of the intended subsidiary. It is not about the entire global sup-ply or value chain, but the part confined to the location. It is concernedabout the forward and backward linkages, supply and demand (marketconditions), agglomeration economies (competing and complementing busi-nesses) and business process innovation (adapting to local markets, creating


Benefits Opportunities Costs Risks

- Local Demand - Market growth - Production cost - Deviations in de-mand and supply

- Own-industryconcentration

- Product/processinnovation

- Operating cost - Local competition

- Inter-industryconcentration

- Spin-offs/ins - Ignorance of do-main knowledge

- Special economiczones and technol-ogy parks

- Knowledge shar-ing and collabora-tion

- Direct and indirectcosts

- Knowledge spill-over and IP rightsviolation

- Incentives: taxes,utilities, exports,imports

- Improving publicfacilities

- Taxes: income,land, utilities, ex-ports, imports

- Exposure of com-pany informationwhile applying forincentives

- Flexible labourlaws

- Delays due to reg-ulations

- Anti-dumping,Breach of promisesby government

- Transparent regu-lations

- Corruption - Political instability

- Living conditions - Cost of living - Crime and terror-ism, Bankruptcy

- Investors and loanavailability

- Developing inte-grated services

- Cost of utilities:land, power, water

- Sub-optimal qual-ity

- Institutions: ed-ucational, training,research

- Customized train-ing

- Cost of raw mate-rials

- Unskilled labour

- Value-added ser-vices

- Services cost - Exposure of busi-ness process- Labour strikes

- Transport infras-tructure: airports,seaports, railways,roadways

- Developing publicfacilities

- Freight costs - Disruption of con-nectivity due to nat-ural calamities

- Custom clearancedelays

- Security in globalsourcing

- Network connec-tivity

- Bandwidth cost - Network reliabil-ity and security

Table 6.5 Benefits, Opportunities, Costs, and Risks.


new business opportunities). The above aspects of the location are clusteredunder this criterion.

The important sub-criteria considered are the supply-demand and theagglomeration. The traditional supply-demand factors are enhanced inglobal supply chains in terms of cheap supplies, local demand, and po-tential for market growth. However, deviations and disruptions in demandand supply can result is costly discrepancies elsewhere in the global supplychain. Similarly, the stronger agglomeration economies and cluster effectsprovide many benefits and opportunities and also pose major risks. Thepresence of related businesses in the location reduces supply costs and pro-vides huge demands. It also enables knowledge sharing and collaborationto make the business process efficient but knowledge spill-over could resultin IP violations. Ignorance of domain knowledge will be an added advan-tage of the locals with added expertise due to knowledge spill-over. Theagglomeration, on the other hand, can help in business process innovationsby leveraging the global expertise to meet local demands. The DM, hencehave to take into account the above conflicting factors to arrive at an optimallocation decision.

6.5.3 Economic Integration

The economic and political factors play an important role in FDI and MNClocation choice. The interaction between the investing firm and the host gov-ernment during the location decision has been modeled using game theoryin the international business discipline [29, 10]. The firm is always assumedto choose amongst several alternative locations, greatly reducing the bar-gaining power and role of the host government [35]. However, the economicand political profile of the government plays a significant role in the locationdecision.

This criterion include taxes (income, sales, trade, import, export), regu-latory framework (labor, environmental, legal), trade agreements, govern-ment incentives and subsidies, political stability, and living conditions. Theobvious benefits of incentives, subsidies, and trade agreements also comewith a great pool of risks like anti-dumping, voluntary export restrictions,and breach of promises. Once the investments are made, the bargainingpowers of the firms are lost and are dependent on the functioning of thegovernment. The exposure of the firm’s information while applying for in-centives is another risk encountered commonly in practice [32]. The indirectinfluences of the government like terrorism and crime are other sources ofrisk.

The regulatory framework is another major sub-criterion that is govern-ment dependent. Many studies [45, 17] based on surveys have indicated


that rules and regulations (labor laws, licensing, environmental) are seenas hinderances by firms. The bureaucratic framework results in unproduc-tive delays and non-transparent functioning leads to corruption. Finally, theintangible attributes living conditions and public attitude also have subtleeffects on location selection.

The role of economic integration in location selection is evident at thenational level, while choosing countries for investment. They also play animportant role at the sub-national level for countries like India with federalsystem that is co-administered with several regional or state governments.Incentives, subsidies, and regulatory framework are generally the dominat-ing factors while choosing a location within India.

6.5.4 Resources and Management

The third criterion covers the resources and the management of resources.The resources include human (skilled and unskilled), natural (raw materials,land, coast line), utilities (water, electricity), and also financial (loans, banks,venture capitalists). Management of resources is an important sub-criterionthat is overlooked. Many global business operations largely depend on re-source management skills like global sourcing, global marketing, researchand training institutions, legal services, human resource training, and finan-cial planning. The resource management complements with the resourcesand sometimes even substitutes when the resources are not available like inthe case of Singapore.

6.5.5 Connecting Technologies

The final criteria is about how a firm connects to the external world usingthe transport and network infrastructure. The inbound and outbound flowof materials, manpower, information, and data are considered in this cri-terion. The obvious attributes include availability of sea ports, air posts,railways, road ways, freight forward costs, lead time, network readiness, ITconnectivity, mobile networks, postal and courier system, IT enabled ser-vices, etc. The other network components due to globalization are customsclearance and quality tracking systems. International logistics flows are sub-stantially more complex with more documentation like commercial invoicesand customs paperwork. Hence, locations that employ automated tradedocumentation are advantageous.

The following characteristics of the hierarchy using the PERC model areworth noting.


• The four fundamental criteria are integral to many global investments.Indeed, they can be interpreted as four forces whose interplay affectthe evolution of a global supply chain. Their relative importance, how-ever, depends on the industry and the investing firm. For example, inlocating a large manufacturing plant, all the four criteria have almostequal significance, whereas in locating a call center, network connec-tivity and resources play a dominant role. Thus its widespread appli-cability lies in its genericness.

• The model is complete in the sense of covering all aspects and takesan end-to-end view of a global investment.

• It is a top-down approach by starting with the fundamental criteriafirst and then identifying the suitable sub-criteria and the attributes. Itis not based on the importance of the attributes, perceived a priori bythe DM. It is also not classified by the tangible or intangible nature ofthe attributes. It can aid the practitioners in identifying and groupingthe attributes for new-age business processes, which are not yet wellstudied in the literature.

• The number of fundamental criteria in the model is optimal. As ob-served earlier, the main reason for an hierarchical structuring is toelicit the preferences of the DM for the attributes in terms of weights.It has been long observed in cognitive science that the comparativecapability of human brain is limited to five distinct entities simultane-ously. Thus too many fundamental criteria is undesirable and too fewwould result clustering of non-homogeneous attributes.

• The model assumes that information about a location for the locationfactors is already known. For example, it is usual in production in-vestments to forecast and project sales for period of five years. Similarto this are the risk evaluations. It is assumed that such calculations arealready available to the DM.

• The model only suggests what can be included under a criterion.Some of the attributes in Table 6.5 are interrelated. For example, thebenefit high labor availability is a mathematical inverse of labour cost.They can also be indirectly related like high taxes and poor infrastruc-ture. The DM should make sure that such related attributes are notincluded to avoid double counting. The lack of a step-by-step cook-book procedure is an obvious outcome of its genericness, though theTables 6.4 and 6.5 can be used as look-up tables by the practitioners.

Synthesis: Analytic Hierarchy Process 139

Scale Definition Explanation

1 Equal importance Two factors contribute equally to the ob-jective

3 Moderate importanceof one over another

Experience and judgment strongly favorone factor over another

5 Essential or strongimportance

Experience and judgement strongly favorone factor over another

7 Very strong impor-tance

An activity is strongly favored and itsdominance demonstrated in practice

9 Extreme importance The evidence favoring one factor over an-other is of tile highest possible order ofaffirmation

2,4,6,8 Intermediate valuesbetween the twoadjacent judgments

When compromise is needed

Reciprocals If factor F1 has one of the above numbers assigned to it when com-pared with factor F2, then F2 has the reciprocal value when com-pared with F1

Table 6.6 The fundamental scale used in AHP for pairwise comparison

6.6 Synthesis: Analytic Hierarchy Process

The second phase in the multi-criteria evaluation is to determine the scoresfor the locations for each of the attributes and weights for the attributes andcriteria that reflect their relative significance. This multidimensional scaleof measurements are then combined into a unidimensional scale of ranks.Out of the two commonly used methodologies, MAUT is preferred moreby location consultants (Section 6.3.6), whereas AHP is advocated more byacademicians (Section 6.3.4). To be more precise, location consultants usevalue function by directly assigning values to the criteria in some intervalscale (say 1 to 10) and not utility function, which is elicited through risk lot-teries. This direct assignment of values is subject to high degree of humansubjective errors. On the other hand, use of utility functions, though math-ematically infallible, is found to be difficult to understand and implementin practice for DMs. AHP is widely accepted as the best trade-off betweenmathematical accuracy and implementation in practice. It is also best suitedto handle both tangible and intangible attributes along with objective andperception data, which is the norm in the location selection problem.

Given the problem in a hierarchical structure, AHP determines the scoresand weights using pairwise comparisons of sibling nodes under each parent


node. AHP ingeniously derives the weights and scores using pairwise com-parison. Actually, it does not distinguish between scores and weights andthey are called as priorities. The priorities are numbers on ratio scales.

The priorities are obtained for each set of siblings separately. The siblingsare compared pairwise with respect to their parent and a numeric value isgiven, which represents the ratio of preference between the two factors. Amatrix of pairwise comparisons is constructed by reference to the semanticscale and 1-9 numeric scale, shown in Table 6.6. Let A denote the matrixwith scales assigned through pairwise comparisons. A is a positive, ratiomatrix with aij = 1/aji. The priorities of the factors w is the normalizedeigenvector associated with the largest eigenvalue of A. It is obtained bysolving the following set of linear equations:

Aw = !maxw (6.1)

The largest eigenvalue !max will be greater than or equal to the size ofmatrix A. If it is equal to the size of A, then the judgement made frompairwise comparison is consistent. However, such consistency is rarely areality in real world and AHP allows for such inconsistencies. Because eachpairwise comparison is already a ratio, the resulting priorities will be ratio-scale measures as well. Once the priorities are obtained for all the nodes,the location is evaluated by multi-linear superposition.

6.7 Location of a Biotech R&D Centre

In this section, we illustrate the applicability of our proposed decision frame-work in modeling and solving the location selection problem faced by aglobal biotech firm. The problem is to locate a R&D facility in India. Thefollowing is not an in depth analysis, which is beyond the scope of thiswork. Further, the evaluation of locations in reality is ultimately made withsubjective judgement (though objective and perception data are used) by theinvesting firm based on its objectives and characteristics [46].

India has long enjoyed a reputation as a destination for IT and busi-ness process outsourcing. Now, the country is fast emerging as a majorcenter for cutting-edge research and development (R&D) projects for globalmultinationals. According to an UNCTAD survey, India has emerged as thethird most attractive prospective destination for setting up research centresby the world’s largest corporations looking to expand their R&D activitiesworldwide during 2005-09 [44]. The list of multinationals with R&D cen-tres in India includes General Electric, Microsoft, IBM, Cisco, Intel, GeneralMotors, Astra Zeneca, Motorola, and Texas Instruments. The best-knownIndian R&D companies are in pharmaceuticals: Ranbaxy, Dr. Reddy’s Labs,

Location of a Biotech R&D Centre 141

Sun Pharma, Biocon, Reliance Life Sciences, and Shanta Biotech. Thoughmany of these are Indian companies, there are new companies like Divi’slabs, Vimta labs, and Matrix labs, which do contract R&D for multinationals.

Given the natural resources and the skilled workforce, India has identi-fied its potential in biotechnology nearly two decades ago. Department ofBiotechnology9 was setup under the Ministry of Science and Technology in1986. With biotechnology industry registering over 35% growth in the lastfew years, this industry is seen as one of the key drivers that will contributeto the socio-economic growth. So several states are making conscious ef-forts to create a conducive environment to attract entrepreneurs to set uptheir units and leverage on the vast talent pool and rich biodiversities inthe respective states. With the success in IT using the IT parks, the centraland state governments are keen to replicate it in biotechnology. As a result,India will have at least 20 biotech parks in the next few years.

6.7.1 Location Determinants

The literature identifies several distinct categories of factors that influencethe location of foreign R&D centre. The factors can be broadly categorizedas [25]: demand factors, supply factors, and general competitive factors. A dif-ferent categorization was used by [11]: economic environment, institutional en-vironment, science and telecommunications environment, and MNC competitivefactors.

The above categorization were used, in general, to explain the R&D ex-pansion of a MNC in foreign countries, along with the location decision.Table 6.7 presents the location factors in terms of importance for greenfieldR&D according to Buck Consultants International, a location consultancyfirm [9]. Firms choose for locations where highly qualified labour is avail-able at reasonable costs, where universities and technological institutes con-duct state-of-the-art research, and where travel to foreign destinations iseasy.

The above factors are for national level and some of them are irrelevant tothe subnational location choice. In particular, corporate tax and income tax arecentrally administered, and do not vary across locations. For our particularcase of locating in a biotech park, many more factors become redundant.Most of the regulatory frameworks are from the central government andthe state government usually provides a single point interface, if located ina biotech park. The pollution control and treating of toxic and chemicalwastes are taken care at the special facilities in the park. Likewise, the

9http://www.dbtindia.nic.in


Crucial Very Important Important Less Important

availability andcosts of highlyqualified labour;

investment andtechnologygrants;

quality of lo-cal suppliers,customers andcompetitors;

quality and costsof telecommu-nication andenergy;

availability andquality of uni-versities andtechnologicalinstitutes;

availabilityof technol-ogy/scienceparks;

corporate taxregime, incometax regime;

macro-economicprofile and politi-cal stability;

proximity andquality of inter-national airport;

quality of life,costs of living,and internationalschools;

regulatory frame-work;

Table 6.7 Location factors for greenfield R&D

Educational & re-search institutes;

International air-port connectivity;

Cost of living; Quality of life;

Own industryconcentration;

Inter-industryconcentration;

Collaborationwith universitiesand researchcenters;

Investment andtechnologicalgrants;

Incentives andsubsidies;

Legal and valueadded services;

Network connec-tivity;

Political support

Table 6.8 Location factors for subnational location choice of a Biotech R&D

infrastructure needed for conducting research are also available at all parksand hence need not be considered. We identified 12 factors that are relevantto our subnational location choice of an R&D centre in a biotech park, whichare shown in Table 6.8.

6.7.2 Hierarchical Structuring using the PERC Model

The hierarchical structuring of the location parameters under the four di-mensions is shown in Figure 6.4. The factors related to the value chain are:qwn-industry concentration, inter-industry concentration, and collaboration withregional universities and research centers.

Location of a Biotech R&D Centre 143

Fig. 6.4 Hierarchical structuring of the R&D location factors

They capture the demand factors, supply factors, agglomeration economies,and cluster effects. The own-industry concentration reveals the cluster effectwhich is positively correlated with the suppliers, market demand, comple-menting firms, and competing firms. Biotech is a multidisciplinary industrywith linkages to chemistry, biology, and IT. The inter-industry concentrationis used to model these linkages. The above factors also capture the knowl-edge spill over across the firms and the industries. Collaboration with thelocal universities and research institutions is a very crucial activity in thevalue chain of and R&D industry and it is positively correlated with thegrowth of the organization.

Economic integration comprises of the following economic, political, andsocial factors: incentives and subsidies, investment and technological grants, po-litical support, cost of living, and quality of life. The cost of living, quality oflife, and political support influence the employees’ living conditions in thecity and hence affect the MNC’s ability to employ highly skilled workforce.The government incentives and the availability of venture capital invest-


ments are very important for the establishment and the operations of thesubsidiary.

Resources is the crucial factor for a service industry like R&D. The fol-lowing two factors are considered, respectively, under Resources & Manage-ment: educational and research institutes and legal and value added services. Theeducational and research institutes provide the required workforce for thecentre. Further, an R&D centre will need several value added and man-agement services like legal consulting (for patenting), financial planning,human resources management, etc.

Finally, the Connecting Technologies consists of the following obvious fac-tors: international airport connectivity and network readiness. The networkreadiness refers to the connectivity with respect to IT, ITES, and telecom-munication, including wired and mobile communication.

Following observations about the PERC based hierarchical structuringcan be immediately made by referring to Figure 6.4:

1. Firstly, from the practical perspective, there is a quantum leap in theunderstanding and representation of the problem from the list of fac-tors in tables 6.7 and 6.8 to Figure 6.4. The PERC based structuringprovides the decision maker with a holistic view of how informationabout various location factors will be synthesized to evaluate a loca-tion.

2. Secondly, from the theoretical perspective, the PERC structuring en-ables the decision maker to compare and judge the factors with re-spect to a common feature or objective. This makes the methodologytheoretically, and as well as practically, implementable.

The model could be further expanded by including more factors at thelast level of hierarchy, thus making it more fine. The own industry concen-tration is a measure of localization economies of agglomeration economies.Biotechnology industry can be further subdivided as bio pharma, bio indus-trial, bio services, bio agri, bioinformatics, and bio suppliers. Similarly, one cansplit inter-industry concentration into IT, auto, chemicals, leather, etc.

The incentives and subsidies offered by the state governments includewaiver on sales tax, tax on capital investment, tax on captive energy, stampduties, and registration fees for locating in the biotech park. The total sub-sidy, thus, should be calculated using judicious combination of all these val-ues. The influence of the educational and research institutes can be further sub-divided by taking into account the number of graduate and post-graduatepass-outs, Ph.Ds, post doctorates, publications, and patents. Services likelegal and business consulting are very important for start-ups and R&D es-tablishments for filing patents and like. The overall contribution of value

Measuring FDI and Market Attractiveness 145

added services can be obtained from combining individual availability of ser-vices. For the sake of simplicity, we have confined our model to just twolevels of hierarchies as shown in the Figure 6.4.

Given the hierarchical structuring, the next step is to use AHP to evaluatethe scores for the locations. For a detailed analysis of applying AHP toderive scores and weights, and synthesizing them to evaluate the locationsfor biotech R&D, see [43].

6.8 Measuring FDI and Market Attractiveness

The evaluation of location from an MNC’s perspective for its location choiceproblem was dealt in detail in the previous section. In a similar way, onecan measure investment climate and market attractiveness using the PERCmodel and AHP.

6.8.1 Investment Climate

Private firms invest in new ideas and new facilities that strengthen the foun-dation of economic growth. They include farmers and micro-entrepreneursto MNCs and their investment in a region is mainly determined by theinvestment climate. Investment climate reflects the location-specific factorsthat shape the opportunities and incentives for firms to invest productively,create jobs, and expand. More specifically it is the policy, institutional, andbehavioral environment, both present and expected, that influences the returns, andrisks, associated with investment. Three main features of the investment cli-mate are:

• Macro-economic factors (including political stability)

• Governance

• Infrastructure

Our proposed methodology can be used to measure investment climate.The PERC model, in addition to the above features, includes the Prod-uct/Process value chain and Resources & Management dimensions. This isessential for measuring the investment climate for specific industries likeautomobiles. The P dimension provides the forward and backward link-ages like supply, demand, and markets, which are essential in measuringindustry specific investment climates. The location of biotech firm in theprevious section is firm-specific. If the attributes and weights are derivedwith industry-specific characteristics, then the methodology can be used toevaluate investment climates at international, national, regional or city level.


6.8.2 Market Attractiveness of Tier II Cities

With the sustained FDI inflows in various sectors, the tier I cities like Ban-galore, Chennai, Hyderabad, Mumbai, and Delhi are saturated and the in-frastructure is overwhelmed. The city limits have expanded considerablyand many development agencies are targeting tier II cities for development.The proposed decision framework can be applied to evaluate and rank thepotential tier II cities. The evaluation can also be used to measure the timeand cost involved in preparing the cities to be future centers of economic ac-tivities. The planning commission and economic development agencies canleverage the proposed methodology to systematically identify, plan, and de-velop potential tier II locations.

6.9 Final Notes and Proposals

Evaluating and ranking locations for investment based on a multitude offactors was the central theme of this article. A decision framework wasproposed for location evaluation from the perspective of an MNC that issearching for a location to invest in its subsidiary. An example of an MNClocating its Biotech R&D facility in India was described in detail. The lo-cation evaluation was solved from the perspective of an investing MNC.However, this problem is encountered as the core or sub-problems in manyscenarios. We discuss here other related decision problem that can be solvedusing the proposed approach.

6.9.1 Location of Retail Stores

India’s retail sector is wearing new clothes and with a three-year com-pounded annual growth rate of 46.64%, retail is the fastest growing sec-tor in the Indian economy. Traditional markets are making way for newformats such as departmental stores, hypermarkets, supermarkets and spe-cialty stores. Chapter 2 in this volume provides a detailed exposition on thedynamics of retail in India.

The process of determining optimal locations for new retail stores, fran-chise locations, restaurants, and financial institutions is a related but differ-ent location selection problem. It includes profiling customers, understand-ing competitors, and analyzing demographic, census, and market data todetermine how one can reach and service both new and existing customersmost profitably. Specifically, following issues need to be addressed:

• Where are the customers, best prospects, current stores, and competi-tors?

Final Notes and Proposals 147

• How many suitable markets exist in a given geographic area for theproducts/services being offered?

• How many locations are necessary to optimally cover a given marketarea and which locations offer the greatest potential?

The decision problem is similar to the traditional facility location prob-lem, but with a large number of tangible and intangible factors. The pro-posed decision framework could be used adapted to first list down the pos-sible locations. However, one has to take into account the interaction effectsof locations and create an optimization model to optimally choose the set oflocations for the stores.

6.9.2 Design of SEZ

India over the past decade has progressively opened up its economy to effec-tively face new challenges and opportunities of the 21st Century. To com-pete in the global market, the Government of India has liberalised exportpolicies and licensing of technology and implemented tax reforms provid-ing various incentives. The tangible fruition of these efforts is the introduc-tion of special economic zones (SEZs) in the Exports-Imports policy of March2000 (refer Chapter 5 of this volume for more details).

SEZs are geographical regions that have economic laws different from acountry’s typical economic laws. The goal is usually an increase in FDI inthe country. Traditionally SEZs are created as open markets within an econ-omy that is dominated by distortionary trade, macro and exchange regula-tion and other regulatory governmental controls. SEZs are believed to createa conducive environment to promote investment and exports.

There are many design issues regarding the SEZs (refer Chapter 5 ofthis volume for more details). First is the location of SEZ or rather whichlocation is to be declared as an SEZ. This decision problem, faced by thegovernment, should take into account location and region specific factorslike availability of labor, utilities, resources, connection to highways andports, infrastructure, incentives, and governance. Usually the incentive andconcession packages are uniform for all the SEZs in the country, but thereare certain regulatory framework that depends on the state governmentslike the labour laws. One can easily see that above factors can be analyzedusing the PERC model and can be used to evaluate and rank locations thatcan be potential SEZs. Moreover, this can aid the governments in planningand developing the locations of interest to be conducive to be an SEZ.

The second decision problem faced by the government is to decide onwhich project(s) to accept for developing an SEZ. More specifically, an op-timal portfolio of projects is to be decided based on the SEZ specific char-


acteristics. For example, the proposed industry segments in Navi MumbaiSEZ consists of 18 industries including biotech, gems and jewellery, ITeE,and leather products. Here again, the PERC model can be used to analysisthe location specific factors and industry needs, and AHP can be used toselect a portfolio of projects.

The third is the design of SEZ from the developer perspective to deter-mine the composition of companies in the SEZ. This includes determiningthe companies that care of logistics, MRO, information and communicationinfrastructure, venture capital and investments, human resource training,global marketing, etc. One can see that the PERC model can be applica-ble in this scenario, probably in combination with AHP and optimizationtechniques.

References

1. Petroni Alberto. The logistics of industrial location decisions: An ap-plication of the analytic hierarchy process methodology. InternationalJournal of Logistics: Research and Applications, 3(3):273–289, 2000.

2. W. Atthirawong. A Framework for International Location Decision-Makingin Manufacturing using the Analytical Hierarchy Process Approach. PhDthesis, School of Mechanical, Materials, Manufacturing Engineering andManagement, University of Nottingham, UK, 2002.

3. M. A. Badri. Combining the analytic hierarchic process and goal pro-gramming for global facility location-allocation problem. InternationalJournal of Production Economics, 62(3):237–248, 1999.

4. Salvador Barrios, Holger Gorg, and Eric Strobl. Multinationals’ locationchoice, agglomeration economies, and public incentives. InternationalRegional Science Review, 29:81–107, 2006.

5. Valerie Belton. A comparison of the analytic hierarchy process and asimple multi-attribute value function. European journal of Operational Re-search, 26:7–21, 1986.

6. Fabienne Boudier-Bensebaa. Agglomeration economies and locationchoice: Foreign direct investment in Hungary. Economics of Transition,13(4):605–628, 2005.

7. Lance Eliot Brouthers, Keith D. Brouthers, and Steve Werner. Is Dun-ning’s eclectic framework descriptive of normative? Journal of Interna-tional Business Studies, 30(4):831–844, 1999.


8. John A. Cantwell and Lucia Piscitello. The recent location of foreignR&D activities by large MNCs in the European regions: The role ofdifferent sources of spillovers. In ERSA Congress, Finland, 2003.

9. Maarten Cornet and Marieke Rensman. The location of R&D in theNetherlands: Trends, determinants and policy. CPB Documents 14, CPBNetherlands Bureau for Economic Policy Analysis, 2001. available athttp://ideas.repec.org/p/cpb/docmnt/14.html.

10. P. Dicken. Seducing foreign investors - the competitive bidding strate-gies of local and regional agencies in the UK. In M. Hebbert and J. C.Hansen, editors, Unfamiliar territory: The re-shaping of European geography.Aldershot, Hants: Gower, 1990.

11. J. P. Doh, G. K. Jones, H. J. Teegen, and R. Mudambi. Foreign re-search and development and host country environment: An empiricalexamination of US international R&D. Management International Review,45(2):121–154, 2005.

12. Zvi Drezner and Horst W. Hamacher, editors. Facility Location: Applica-tions and theory. Springer-Verlag, Berlin, 2002.

13. J. H. Dunning. Trade, location of economic activity and the MNE:a search for an eclectic approach. In B. Ohlin, P.O. Hesselborn, andP.M. Wijkman, editors, The International Allocation of economic Activity.Macmillan, London, 1977.

14. J. H. Dunning. Explaining International Production. Unwin Hyman, Lon-don, 1988.

15. W. Edwards and J. R. Newman. Multiattribute Evaluation. Sage BeverlyHills, CA, London, 1982.

16. E. Erkut and S. R. Morgan. Locating obnoxious facilities in the publicsector: An application of the analytic hierarchy process to municipallandfill siting decisions. Socio-Economic Planning Science, 25(2):89–102,1991.

17. FICCI. FICCI handbook on foreign direct investment. Report, Federa-tion of Indian Chambers of Commerce and Industry, FICCI, New Delhi,October 2005.

18. Ernest H. Forman and Saul I. Gass. The analytic hierarchy process - anexposition. Operations Research, 49(4):469–486, 2001.

19. M. Fujita, P. Krugman, and A. Venables. The Spatial Economy: Cities,Regions, and International Trade. MIT Press, Cambridge, MA, 1999.


20. Rachel Griffith, Michael P. Devereux, and Helen Simpson. Firm locationdecisions, regional grants, and agglomeration externalities. AIM Re-search Working Paper Series 038-May-2006, Advance Institute of Man-agement Research, May 2006.

21. Roger Hayter. The Dynamics of Industrial Location: The Factory, the Firmand the Production System. John Wiley and Sons, Chichester, 1997.

22. Keith Head, John Ries, and Deborah Swenson. Agglomeration benefitsand location choice: Evidence from Japanese manufacturing investmentin the United States. Journal of International Economics, 38:223–247, May1995.

23. G. G. Hedge and P. R. Tadikamalla. Site selection for a ‘sure serviceterminal. European Journal of Operational Research, 48:77–80, 1990.

24. IBM. Investment strategies and location benchmarking study. Locationbenchmarking analysis, Oxford Intelligence, February 2004.

25. G. Jones and H. Davis. National culture and innovation: implicationsfor locating global R&D operations. Management International Review,40(1):11–39, 2000.

26. C. Jungthirapanich and C. O. Benjamin. A knowledge-based decisionsupport system for locating a manufacturing facility. IIE Transactions,27:789–799, 1995.

27. R. Keeney. Value-Focused Thinking: A Path to Creative Decisionmaking.Harvard University Press, Cambridge, Massachusetts, 1992.

28. R. Keeney and H. Raiffa. Decisions with Multiple Objectives: Preference andValue Tradeoffs. Wiley, New York, 1976.

29. D. W. Loree and S. E. Guisinger. Policy and non-policy determinantsof US equity foreign direct investment. Journal of International BusinessStudies, 26(2):281–299, 1995.

30. Philip McCann. Classical and neoclassical location-production models.In Philip McCann, editor, Industrial Location Economics. Edward Elgar,Cheltenham, UK, 2002.

31. Philip McCann and Ram Mudambi. The location behavior of the multi-national enterprise: Some analytical issues. Growth and Change, 35(4):491–524, 2004.

32. Marcel de Meirler. Location Location Location: A Plant Location and SiteSelection Guide. Wiley, New York, 2006.


33. H. Min. Location analysis of international consolidation terminals usingthe analytic hierarchy process. Journal of Business Logistics, 15(2):25–44,1994.

34. Ram Mudambi. The MNE investment location decision: Some empiricalevidence. Managerial and Decision Economics, 16:249–257, 1995.

35. Ram Mudambi. The location decision of the multinational firm: A sur-vey. In Philip McCann, editor, Industrial Location Economics. EdwardElgar, Cheltenham, UK, 2002.

36. L. Nachum and C. Wymbs. Firm-specific attributes and MNE locationchoices: Financial and professional service FDI to New York and London. ESRCCentre for Business Research - Working Papers wp223, ESRC Centre forBusiness Research, March 2002.

37. David L. Olson. Decision Aids for Selection Problems. Springer, New York,1996.

38. Michael E. Porter. Clusters and the new economics of competition. Har-vard Business Review, pages 77–90, November-December 1998.

39. Thomas L. Saaty. The analytic hierarchy process. McGraw-Hill, New York,1980.

40. Thomas L. Saaty. The Analytic Network Process: Decision Making withDependence and Feedback. RWS Publications, Pittsburgh, PA, 1996.

41. Thomas L. Saaty and Luis G. Vargas. Models, methods, concepts andapplications of the analytic hierarchy process. Kluwer Internation, Mas-sachusetts, USA, 2001.

42. N. Viswanadham. Global manufacturing and service strategies. Term 5Course Lecture Notes, Indian School of Business, 2006.

43. N. Viswanadham and S. Kameshwaran. A decision framework for the loca-tion choice of MNCs in India. GLAMS Research Report, Indian School ofBusiness, 2006.

44. UNCTAD. Transnational corporations and internalization of R&D. Worldinvestment report, UNCTAD, 2005.

45. World Bank. A better investment for everyone. World developmentreport, World Bank, 2005.

46. World Bank. India: Investment climate and manufacturing industry.Investment climate assessment report, World Bank and International Fi-nance Corporation, November 2005.

152

47. J. A. Wu and N. L. Wu. Analysing multi-dimensional attribute for thesingle plant location problem via adaptation of the analytic hierarchyprocess. International Journal of Operations and Production Management,4(3):13–21, 1984.

48. C. H. Yang. Identifying and testing the decision making factors related tokey industries choice of location. PhD thesis, Faculty of Commerce andManagement, Griffith University, Brisbane, 2004.

49. Jiaquin Yang and Huei Lee. An AHP decision model for facility locationselection. Facilities, 15(9/10):241–254, 1997.

50. Shigeru Yurimoto and Tadayuki Masui. Design of a decision supportsystem for overseas plant location in the EC. International Journal of Pro-duction Economics, 41:411–418, 1995.

51. M. Zeleny. Multiple criteria decision making. McGraw-Hill, New York,1982.

FDI Attractiveness of Indian States and Location Choice of ...

Documents

Transcript of FDI Attractiveness of Indian States and Location Choice of ...