German Multinational Firms in India

Table of Contents

List of Figures

List of Tables

List of Abbreviations

List of Symbols in the main text

1. Introduction
1.1. Background and Problem Description
1.2. Project Organization

2. Theoretical Foundation
2.1. Definitions
2.1.1. Foreign Direct Investment
2.1.2. Multinational Firms
2.2. Analytical Framework
2.2.1. Theoretical Background
2.2.2. The Eclectic (OLI) Paradigm of International Production
2.2.3. The Knowledge-Capital Model
2.3. Determinants of Multinational Firm Decisions
2.3.1. Firm-Level Determinants
2.3.2. Sector-Level Determinants
2.4. Host Country Determinants
2.4.1. Three Types of Location Determinants
2.4.2. Policy Determinants
2.4.3. Economic Determinants
2.4.4. Business Facilitation
2.5. Interrelations of Determinants and FDI Types
2.6. Market Entry Strategies

3. India
3.1. Economic Reforms
3.2. Economic Development
3.3. FDI in India
3.3.1. FDI Regulation
3.3.2. FDI Growth
3.3.3. Evidence on FDI Determinants in India
3.4. Investment Climate India
3.4.1. FDI Rankings
3.4.2. Challenges
3.4.3. Opportunities

4. German Multinationals in India
4.1. Data and Procedure
4.2. German FDI in India – Stylized Facts
4.2.1. FDI Stock
4.2.2. Sectoral and Regional Distribution
4.3. Firm-Level Descriptive Evidence
4.3.1. FDI Types
4.3.2. Agglomeration
4.3.3. Firm-Level Characteristics
4.3.4. Market Entry and Ownership
4.4. Firm-Level Statistical Analysis
4.4.1. Sector-wise Testing for FDI Types
4.4.2. Baseline Model
4.4.3. Extended Baseline Model
4.5. Profitability
4.5.1. Stylized Facts
4.5.2. Baseline Model

5. Management Surveys: Factors for Success
5.1. Strategic Aspects
5.2. Operational Aspects

6. Conclusion
6.1. Economic Policy Implications
6.2. Corporate Strategy Implications

Annex A – The Knowledge-Capital Model

Annex B – Additional Tables and Figures

Annex C – Econometric Issues

References

List of Figures

Figure 1: Real GDP growth rate India (1980-2007)

Figure 2: FDI stock and inflows as GDP-percentage – India (1989-2007)

Figure 3: German FDI stock in India (1989-2007)

Figure 4: Sector-wise stock share of German FDI in India in percent (1989-2007)

Figure 5: Agglomeration of German affiliates in India relative to full dataset (2007)

Figure 6: Profitability and sales growth (2002-2007)

Figure 7: Profitability in quartiles India versus USA (1989-2007)

Figure 8: ROA growth since initial investment of German MNFs in six countries

List of Tables

Table 1: FDI determinants - theoretical predictions

Table 2: FDI inflows by sector in India

Table 3: Average firm- & plant-level size-ratio for selected sectors (2002-2007)

Table 4: Sector-wise comparison: parent firm versus affiliate (1989-2007)

Table 5: Parent firm degree of internationality (2002-2007)

Table 6: Route of market entry (2005-2007)

Table 7: Pooled-OLS Baseline Model Regression and Extension

Table 8: Detailed Sector Groupings

Table 9: Size of German parent firms in India by employees (2002-2007)

Table 10: Definition and sources of explanatory variables

Table 11: STATA do files

Table 12: Sector-wise regression results

Table 13: Profitability regression results

List of Abbreviations

illustration not visible in this excerpt

List of Symbols in the main text

1. Introduction

1.1. Background and Problem Description

Global foreign direct investment (FDI) has risen sharply in the 1990s. At the same time and even more strikingly FDI in emerging market economies has spurred dramatically.^[2] Arising at the end of the 20th century and only known from 19th century trading companies as the East India Company, FDI in developing and emerging markets represents today a new phenomenon. Traditionally, literature has mostly focused on FDI between industrialized countries, and publications on FDI in emerging markets remain relatively sparse.^[3] Yet, the rise of large emerging market countries such as China and India has attracted interest of an increasing number of multinational firms (MNFs). The combination of high real economic growth and new FDI regulation in these countries creates a specific environment where MNFs have to adjust corporate strategies for successful market penetration. Further, government’s perception of FDI has changed towards increasingly attracting FDI.^[4][1]

The present diploma thesis attempts to contribute to the broad issue of understanding why firms become multinational and why they invest in specific countries. Determinants affecting corporate strategies of MNFs and the influence of FDI related economic policies will be specifically analyzed for the case of German MNFs in India for two reasons:

Firstly, India is of particular interest as its economy has been opened up remarkably for FDI as a consequence of the economic reform process in the early 1990s. FDI in India, one of the Asian emerging market countries, may therefore suitably represent recent FDI trends to developing and emerging market countries. FDI analysis in these countries often concentrates on China. Given that India is most prominently referred to the Chinese case in terms of economic perspective and potential FDI growth, the analysis allows to draw attention on a second large and prospective FDI player in the region.

Secondly, a “newly” available firm-level dataset from the Deutsche Bundesbank provides detailed data on German FDI in India. The dataset allows to jointly analyze determinants of MNF decisions and host country location factors.

The following thesis attempts to provide a (1) profound theoretical background, (2) empirical validation of theoretical predictions using firm-level data and management survey results and (3) implications on corporate strategy and economic policy.

1.2. Project Organization

Section 2 provides the theoretical foundation by outlining an explicit analytical framework to clearly separate determinants. Determinants of MNFs and host countries are specifically discussed and interrelated, before analyzing the effect of MNF determinants on market entry strategies.

Section 3 introduces the Indian case. Based on a brief overview of economic reforms and growth, consequences of the latter on FDI regulation and subsequent FDI growth are presented. Finally, specific host country determinants are discussed from the point of view of MNFs.

Section 4 provides the empirical foundation of the thesis. Based on general analysis of FDI patterns and descriptive firm-level evidence, regressions to test determinants as suggested by theory follow. Further, the same determinants are considered to explain profitability of FDI projects.

In section 5 findings from management surveys with respect to MNF strategies and factors for success are presented, thereby confronting econometric findings from previous sections with survey results.

Section 6, finally, proposes a conclusion, presenting implications of the thesis results on corporate strategies of MNFs in India and on FDI related economic policy issues in India.

2. Theoretical Foundation

2.1. Definitions

2.1.1. Foreign Direct Investment

FDI as defined by the OECD Benchmark Definition is characterized by a

“…lasting interest by a resident entity in one economy (‘‘direct investor’’) in an entity resident in an economy other than that of the investor (‘‘direct investment enterprise’’). The lasting interest implies the existence of a long-term relationship between the direct investor and the enterprise and a significant degree of influence on the management of the enterprise.” (OECD, 1996)

In the following, the direct investor is referred to as the parent firm or MNF. On the other hand, the direct investment enterprise will be denoted as affiliate.^[5]

FDI comprises the establishment of new affiliates, investment in existing foreign enterprises and in established affiliates. Furthermore, direct investment does not only include transactions between parent firm and affiliates but also between affiliate and third tier enterprises. The latter type of investment is referred to as indirect investment. Invested capital includes equity capital, reinvested earnings and interfirm debt transactions.

FDI has to be distinguished from portfolio investment. While FDI refers to a significant degree of influence and physical expansion in another country, portfolio investment is limited to the buying and selling of minority stakes below 10 percent of the ordinary shares or voting power. Further, investment of 10 percent or more is typically declared a direct investment.^[6] Therefore, FDI is considered an “active” investment and portfolio investments, where no direct influence is exerted, displays a “passive” investment. Portfolio investment primarily aims at short-term financial returns, including easy divestment and no significant control on the management of the firm; FDI, however, is intended to raise profitability through the long-term injection of resources or management skills into the firm. The resources and factor inputs transferred may be tangible or an intangible combination of technology, management skills, production processes and financial resources.^[7]

The investing entity may be an individual, a multinational firm, a financial institution or a government. In principal MNFs can be considered to be the major source of FDI, generating 95 percent of world FDI flows.^[8] Data on FDI are typically reported as flows in the capital account of national balance of payments statistics. In addition, stock data are available, indicating cumulated flows or as reported from balance sheet data of affiliates.

2.1.2. Multinational Firms

Multinational firms operate in more than one country by means of FDI, thereby controlling and managing production establishments at different geographical locations. The two terms MNF and FDI are employed more or less interchangeably.^[9] MNFs comprise the parent firm, based in the home (source) country, and foreign affiliates in host countries.^[10]

Depending on the degree of ownership, the foreign firms may range from affiliates (associates) to subsidiaries to fully owned subsidiaries. According to the OECD and the IMF^[11] …

“…the foreign firm can be defined as a subsidiary if the foreign investor controls more than 50 percent of the shareholder’s voting power or has the right to appoint or remove a majority of the member of this enterprise’s administrative, management or supervisory body. Otherwise it can be defined as an associate enterprise if the foreign investors own between 10 and 50 percent of the voting shares.” (Barba-Navaretti et al., 2004)

There are different types of MNFs. While some control multidomestic foreign subsidiaries, producing for the respective local market only, others have affiliates to keep a regionally or globally coordinated network of asset-creating and asset-exploiting activities to increase general production efficiency. Caves (1996) distinguishes MNFs relative to output into three categories: (1) horizontally integrated firms, (2) vertically integrated firms and (3) diversified firms. Horizontally integrated firms produce outputs broadly similar to production in plants established in different countries to serve local markets. Vertically integrated firms produce outputs in plants located in host countries to serve as inputs in other plants. If outputs cannot be classified as one of the former two, MNFs are subsumed under diversified firms.^[12] Hence, horizontally integrated MNFs undertake horizontal FDI (HFDI) referring to the duplication of a subset of firm activities to serve local markets. Vertical FDI (VFDI), in addition, describes the fragmentation of the value-added chain to produce intermediate products in a host country carried out by vertically integrated MNFs in order to re-export intermediate products. HFDI is referred to as “market-seeking” FDI and VFDI as “efficiency-seeking” FDI. When undertaking FDI, MNFs have two possible entry routes: greenfield (setting up new production facilities) and brownfield (international cross-border mergers and acquisitions [M&A]).

As types of MNFs may vary substantially in practice, Dunning and Lundan (2008) specify MNFs further according to distinctive features such as ownership, product diversification, age and geographical control. MNFs may be (1) publicly or privately owned, (2) large diversified global corporations or single-product firms and (3) just recently established or may be in existence for several decades already. (4) Most often MNFs are controlled in the home country but are internationally managed and owned.^[13]

2.2. Analytical Framework

2.2.1. Theoretical Background

In this section an introduction to the theoretical background is given. Development of theories explaining FDI and MNFs is outlined. The background delineates theories of the analytical framework presented in subsequent sections. The framework builds on the

“eclectic (OLI) paradigm of international production” proposed by Dunning (1980) which will then be extended by the “knowledge-capital model” by Markusen (2002). Hence, the analysis of MNFs will be conducted from the perspective of international business and microeconomic trade theory. Approaches from macroeconomics are not considered. Theory of location as broader part of trade theory is only considered casually. However, it is not part of the theoretical framework. Based on the framework detailed investigation of determinants of MNF decisions and host country location will follow in sections 2.3 and 2.4.

In the broadest sense, FDI and MNFs are part of international integration and as a consequence they can be related to international economics. In more detail, international integration is composed of trade in terms of international movements of goods and services on the one hand and of international factor movements on the other. Furthermore, international factor movements comprise international labor and capital movements. Finally, FDI is part of international capital movements. This includes not only FDI, but also international borrowing and lending. However, the present analysis deals only with FDI. Therefore portfolio investments, referring to financial international integration are not considered.

Despite the delimitation from portfolio investments, historically both have been explained by the same traditional trade theories. In accordance with macroeconomic approaches, the early explanation of international economics consisted of simply regarding MNFs as “arbitrager of equity capital” (Caves, 1996). Fundamentally, this approach is associated with Heckscher-Ohlin trade theory, which is based on Ricardo’s principal “comparative advantages” approach of classic trade theory. This approach states that countries specialize in production where labor productivity is highest and thus opportunity costs low, thereby determining trade patterns.^[14] Heckscher-Ohlin build on this and explain trade patterns by the relative abundance of factors of production. According to the factor-proportions theory “factors are expensive where they are scarce and cheap where they are abundant” (Reinert and Ramkishen, 2009). Consequently, the assumption is that capital is directed to countries where marginal productivity of capital is highest, yielding high rates of return. Hence, capital is supposed to flow to capital scarce countries. In addition, capital is considered to be a homogenous factor. The “capital-arbitrage hypothesis” was the dominating theory of FDI until the 1970s.^[15] Nevertheless it has not been confirmed empirically and it remains fairly limited in its predictions.^[16]

Hymer (1976) was the first to reject the “capital-arbitrage theory”, thereby refusing the assumptions of traditional trade theory. Traditional trade theory assumes perfect competitive markets, constant returns to scale and absence of transport costs. At this point of time, trade theory completely ignored the individual firm. Hymer’s proposition of monopolistic compensatory advantages is a first step towards the consideration of the microeconomics of the firm. However, not all assumptions are rejected as for instance the assumption of profit maximizing firms. In other word’s, Hymer

“laid the foundations for a microeconomic explanation of the MNE by pointing out that they are not randomly distributed among industries and that competitive conditions, in particular product markets, clearly influence foreign investment” (Caves, 1996).

Hymer’s approach was developed further in international business literature, respectively by Dunning’s ownership-location-internalization (OLI) approach which was later extended to the Eclectic (OLI) Paradigm of International Production. Both theoretical concepts are discussed in greater depth in section 2.2.2. Dunning’s approach introduces individual firms to the analysis and thereby includes the theory of the firm. This entails, in particular, the inclusion of theories from industrial organization based on transaction cost theory of Coase (1936).

By the 1980s, microeconomic approaches have been incorporated into international trade theory. In addition, trade theory shifted away from the macroeconomic approach by distinguishing FDI and portfolio investments.^[17] In particular industrial organization aspects were integrated into international trade theory at that time. Prior to this point in time there existed two separate strands of literature. On the one hand trade theory providing general-equilibrium models based on the mentioned assumption of perfect competition and constant returns to scale. And partial-equilibrium approaches considering industrial organization effects of trade existed on the other hand.^[18] Since the 1980s international trade theory combines both strands adding aspects of increasing returns to scale, imperfect competition and product differentiation to traditional trade-theory. Analogous to Dunning theories combine the assumption of “comparative advantages” between locations with advantages arising at the firm level. Succeeding Dunning, general-equilibrium models of microeconomic based trade theory have been introduced. Important micro-based trade theories have been proposed by Helpman (1984) and Markusen (1984). In the “factor-proportion model” by Helpman, firms may fragment their activities in different production stages and locate activities across different countries according to factor price differentials. The model does not consider trade costs and MNFs cannot arise between similar countries. Therefore the model only explains VFDI. In contrast, Markusen’s approach which has later been developed to the “proximity-concentration” model^[19] explains HFDI. Under the assumption of trade costs, firms may opt between exporting or producing the same good in the host country to realize a better market access. Both models have been extended and refined over the years and were integrated by Markusen (2002) in the “knowledge-capital model”. The model builds on Dunning’s OLI approach and extends the framework towards a differentiation of horizontal and vertical FDI. As mentioned before, both models constitute the analytical framework of the subsequent sections.

2.2.2. The Eclectic (OLI) Paradigm of International Production

According to Dunning’s OLI framework three conditions have to be fulfilled for firms to become MNFs. Firms must have an ownership-specific advantage (O), a location-specific advantage (L) and an internalization advantage (I).

Under the assumptions of perfect competitive markets implying the assumptions of atomistic competition, transparency and full information, absence of entry barriers, homogeneity of products and rationality of agents, there is no possibility in classic trade theory for foreign firms to serve host country markets more efficiently than their indigenous counterparts localized in the host country.^[20] Contrary to the formulated assumptions, however, indigenous firms are in effect in a more favorable competitive position in terms of information about their country, the national economy, language, law and politics. Foreign firms are therefore disadvantaged and have to incur this lack of information as costs, which may be considerable and increasing with geographical and cultural distance. However, these costs represent fixed costs that occur only once at market entry. Yet, MNFs can also be confronted with variable permanent costs arising from local discrimination by national governments, consumers and suppliers. Nevertheless, foreign firms are found to find it profitable to engage in international markets even though they have to face the described disadvantages. How can this be explained? Hymer argues that in order to be able to serve foreign markets better than indigenous firms, foreign firms have to possess compensatory advantages. Compensatory advantages must be monopolistic considering that local firms would otherwise acquire the advantage and consequently predominate the market. Hence, in order to explain the existence of MNFs and FDI in general, an imperfect market setting has to be assumed.

What kind of advantages must MNFs possess to outweigh the competitive disadvantages compared to indigenous firms? The exclusively owned advantages have to be in the form of innovatory, cost, financial or marketing advantages.^[21] Under the neoclassical assumption of profit-maximizing firms, the advantages arise from minimizing costs of production and associated logistical activities of the firm. Therefore, MNFs must own specific proprietary assets to compensate for the disadvantages faced in host country markets. Yet, this contradicts traditional assumptions of homogeneous capital endowment. Thus, ownership specific advantages refer to the ownership of proprietary assets that yield some market power in host country markets.^[22] Proprietary assets are typically intangible “…such as a superior product, production process, patent, trade-mark…” (Reinert and Ramkishen, 2009). Unique characteristics of proprietary assets will be discussed in detail in section 2.3.1.

But the ownership of proprietary assets alone is not adequate assuming that firms may simply choose to serve host countries by means of export. Thus, location-specific advantages must exist to determine where FDI is directed. Obviously, some countries attract substantially higher FDI than others. This suggests that countries, as well as firms, possess specific competitive advantages over other countries. Although monopolistic with respect to the country level, location factors are external to the firm and they are typically equally available to all firms accessing the market. Location specific advantages create incentives to invest, if production in the host country is more profitable than producing in the home or a third country.^[23] In contrast to firm-level specific factors, location factors are the only variables host country governments can influence directly.^[24]

A third condition builds on transaction cost theory approaches. Dunning and Lundan (2008) argue that firms must have an internalization advantage from applying proprietary assets within the MNF, instead of simply licensing them or contracting with arm’s length firms resident in the host country. Hence, ownership and location advantages alone do not explain the existence of MNFs.^[25] This would only apply under the assumption of perfect and complete markets.^[26] Yet, in an imperfect market setting, arm’s length transfers between firms are prone to market failures, fostering the existence of multiplant firms. In this case transfers of proprietary assets are linked to transaction costs (search and information costs, bargaining and decision costs, policing and enforcement costs). Thus, when transacting externally (outside the firm), costs arise from imperfect information, contractual incompleteness and the quest to protect proprietary assets, which require the firm to fully control the asset. In addition, internalization may also be driven by scale economies. A more profound elaboration of internalization is given in section 2.3.1.

2.2.3. The Knowledge-Capital Model

Based on Dunning’s ILO framework, Markusen (2002) integrated Helpman’s “factor-proportion model” (1984) and the earlier “proximity-concentration model” (1998) into the “knowledge-capital model”, which explains both horizontal and vertical MNFs in a general-equilibrium context. As indicated, HFDI refers to “market-seeking” motives where outputs are produced and sold in the host country market. VFDI, in contrast, describes “efficiency-seeking” motives where investment is driven by cost-side considerations.

The knowledge-capital model^[27] (hereafter “the model”) intersects with the OLI framework with respect to ownership and proprietary assets that are referred to as knowledge capital in the model. The model assumes three main features about knowledge capital: (1) fragmentation, (2) skilled labor intensity and (3) jointness. Geographical fragmentation of services from knowledge-based assets (e.g. skilled employees) is possible at little extra costs once the asset is produced. Thus, the cost difference of supplying the asset to a single plant at home versus servicing a single foreign plant is small. Furthermore, knowledge-based activities carried out in headquarters are skilled-labor intensive in contrast to final production. These characteristics motivate vertical multinationals to fragment production and to place single-plants and headquarters in different countries according to factor prices and market size.^[28] Jointness on the other hand motivates HFDI. To a certain extent knowledge-based services (e.g. blueprints) have joint-input characteristics, so the costs of setting up a second single plant is much lower than setting up a second firm with a local plant. In other words the costs of servicing additional production facilities is low. As fragmentation refers to the transfer of technology it is only available at one plant at a time and therefore nonjoint. Jointness, however, allows to serve several plants with the specific asset at the same time.

In the following, the basic assumption of the model will be outlined. The model considers two countries (i and j), two factors of production (skilled labor S and unskilled labor L) and two goods. Y is the output of a homogenous good produced with constant returns to scale by competitive firms. Good Y is unskilled labor intensive and used as numéraire in the model. Good X (skilled labor intensive) is produced by imperfectly competitive Cournot firms under increasing returns to scale. Firms may opt for HFDI (horizontal firms), VFDI (vertical firms) or export (domestic firms). HFDI entails maintaining two plants in each country and locating headquarters in the home country. VFDI implies headquarter activities in the home country and a single plant in the host country with the option to re-export to the home country. Finally, firms may simply choose to export to host countries keeping headquarters and a single plant in the home country. In other words there exist three kinds of type- k firms: h (horizontal), d (domestic) and v (vertical). Considering that both countries may be in the role of a home country there exist six firm types. Horizontal firms have higher fixed costs due to fragmentation and setting up a second plant. As vertical firms run a single foreign plant, they only face the relatively lower costs of fragmentation. Unlike the former two, domestic firms only incorporate standard fixed costs (which all firms have to incorporate) like skilled labor headquarter fixed costs and unskilled labor fixed costs of production. However, considering the jointness property, total fixed costs of a horizontal firm are lower than fixed costs of two domestic firms.

For the sake of simplicity, only predictions of the model will be presented in the following rather than extensively discussing the model from a formal point of view in this section (for a more rigorous formal presentation, see annex A). After determining optimal Cournot output, the model derives free entry conditions that determine the equilibrium number of respective firm types. These conditions contain markup revenues on the LHS and fixed costs on the RHS which, if subtracted from each other, yield the following firm profits:

Abbildung in dieser Leseprobe nicht enthalten

Note that Abbildung in dieser Leseprobe nicht enthaltenas domestic and vertical firms have to incorporate transport costs from export, their markup revenues are therefore lower.^[29] Anyhow, both are positive. Abbildung in dieser Leseprobe nicht enthaltendisplays income of country i, Abbildung in dieser Leseprobe nicht enthaltenrepresents fixed costs of a type- k firm headquartered in country i and Abbildung in dieser Leseprobe nicht enthaltenand Abbildung in dieser Leseprobe nicht enthaltenare factor prices of unskilled and skilled labor in country i. As mentioned before fixed costs increase with additional plants (supplementary unskilled labor fixed costs) and with foreign investment (skilled labor fixed costs). Similar interpretations apply for inequalities in country j.

Ceteris paribus predictions can be deduced when changing one variable keeping all other endogenous variables constant. Considering the profits of firm types, the following predictions can be derived:

1. Given that horizontal MNFs face no transport costs, they have the highest markup revenues. But they also have the highest fixed costs. Horizontal multinationals dominate if world income, trade costs (transport costs) and similarity in terms of size and factor costs are high. A change in world income: Abbildung in dieser Leseprobe nicht enthaltenyields highest profits for horizontal MNFs: Abbildung in dieser Leseprobe nicht enthalten. This is because horizontal MNFs have no transport costs to incur, hence relative profits will also be higher with increasing trade costs.
2. Nevertheless, vertical and domestic firms dominate if world income and trade costs are low and if countries are asymmetric in size and factor prices. A change in distribution of world income Abbildung in dieser Leseprobe nicht enthaltenfavors domestic and vertical firms selling in country i, since they are confronted with higher income but lower fixed costs compared to firms in country j which fall behind. Horizontal MNFs profits do not change: Abbildung in dieser Leseprobe nicht enthalten. In addition, a change in factor prices Abbildung in dieser Leseprobe nicht enthaltenincurs gains for firms in country j. Markup revenues of domestic firms and vertical MNFs in j rise as margins are higher. Both also benefit from lower fixed costs just as horizontal MNFs in country j. The other way around, firms in country i “lose” as markups decrease and fixed costs rise: Abbildung in dieser Leseprobe nicht enthalten.
3. So far, results do not alter much between domestic and vertical firms. Although one may remark that if factor prices are equal domestic firms will dominate vertical MNFs due to lower fixed costs. In addition, as indicated under the assumption of a large host country market (e.g. j), domestic firms in j and vertical MNFs based in i dominate their respective counterparts. That is to say in the model VFDI is strictly one way as (unlike single-plant products) only knowledge capital is exported. In this case domestic firms would also dominate vertical MNFs if factor prices would be lower in the country providing the larger market. If, however, the small country is skilled-labor-abundant, then the vertical MNF will dominate by fully leveraging on the possibility to arbitrage between factor costs. “Hence, vertical MNFs are most likely if one country is small and skilled-labor-abundant” (Markusen, 2002). In other words, this scenario assumes Abbildung in dieser Leseprobe nicht enthaltenand Abbildung in dieser Leseprobe nicht enthalten. Secondly, vertical MNFs may dominate even if both countries are of similar size but with differing factor costs. If country i is relatively skilled-labor-abundant compared to country j, then vertical MNFs of country i may serve country j with higher markup revenues than domestic firms of country j due to lower fixed costs. Scenario 2 assumes Abbildung in dieser Leseprobe nicht enthaltenand Abbildung in dieser Leseprobe nicht enthalten. In both scenarios profits derived from VFDI are relatively larger Abbildung in dieser Leseprobe nicht enthalten.

To sum up theoretical predictions for MNFs: HFDI will prevail if countries are similar in size and in factor endowments and if trade costs are high. On the other hand, vertically integrated MNFs are dominant if the home country is small with relative abundance of skilled-labor and if trade costs are low.

2.3. Determinants of Multinational Firm Decisions

2.3.1. Firm-Level Determinants

As indicated in the theoretical framework, an important condition for MNFs to evolve is the ownership of proprietary assets which are exclusively owned by MNF and which motivate to internalize such assets. Analogous to knowledge capital they have unique characteristics and features that create competitive advantages and that allow to apply the assets internationally at low costs. First of all the asset’s productivity may be higher than that of comparable assets owned by competing firms and assets may rely on specific knowledge for cheaper and better production at given input prices. In addition, they may represent advantages in the promotion of products, resulting in higher revenue productivity as consumers are willing to pay premiums. But most important, proprietary assets are mobile between national markets and the lifespan is relatively longer than the investment horizon^[30]. Based on the assumption of fragmentation and in particular of jointness formulated with respect to knowledge capital, proprietary assets are limitless in a sense that they have a public goods like character of non-rivalry within the firm. The marginal cost of deploying proprietary assets tend to zero suggesting that they are subject to substantial firm-level economies of scale. This applies particularly to intangible firm assets such as scientific knowledge, entrepreneurial skills or the firm’s organizational system. Yet, the public goods character can also be applied to certain tangible assets, as for instance headquarter buildings, which are only required once independent of the firm’s output level.^[31] Proprietary assets allow to explain differences between firms that become multinational and firms that remain domestic. MNFs are larger in size (measured by domestic market sales) and have higher market shares as they exploit full domestic potential before going abroad. They have higher advertising and R&D expenditures and higher product diversity.^[32]

With regard to the assumption of imperfectly competitive markets as presented in the framework, there exist multiple incentives for firms that motivate internalization of proprietary assets instead of licensing to an outside firm at arm’s length. Based on the assumptions of fragmentation and jointness proposed in the knowledge capital model, incentives to internalize generally arise in the context of information asymmetries. Different incentives may be of particular importance for specific types of FDI. However, it is important to mention that most of the incentives, arising from transaction costs theory may not be specific to the international context. Incentives related to jointness that are especially of high relevance for HFDI are presented in the following^[33]:

(1) The inability to make a convincing disclosure about proprietary assets entails the risk of loosing the firm’s asset when licensed to a third party. Thus, internal transfer is preferred to protect the intangible asset.
(2) Licensing involves the risk of not meeting quality standards implying negative repercussions on the parent firm’s reputation. A licensee may be subject to free riding on the established reputation and a firm’s quality in order to realize short run profits on the cost savings, thereby negatively impacting reputation in the long run.
(3) Accounting for potential buyer’s and licensee’s uncertainty about the asset’s true value, it is difficult for the MNF to convince buyers or licensees of the fair value, without disclosing the secrets themselves.
(4) Failed recognition of product potential by unaffiliated firms might make an external transfer impossible. As the potential is not fully exploited, revenues will not be maximized.
(5) Local firms have much better information about market potentials than the MNF. Hence, MNFs cannot be sure whether the licensee uses the informational advantage to extract rents from the MNF by withholding the information that market potential is actually much higher.

In addition, motives to internalize may be related to the assumption of fragmentation. This refers to transaction costs originating foremostly from technically new or complex projects that rely on proprietary assets. Hence, the following incentives may apply to both types of FDI^[34]:

(1) A novel and complex proprietary asset might include tacit knowledge embodied in the skill of personnel that can solely be transferred by employees of the parent firm, who are familiar with the technology. This would require training of foreign technical and managerial personnel which might be more cost intensive than setting up an own subsidiary.^[35]
(2) Newness and complexity of technology create significant uncertainty with respect to transfer potential and start-up costs. These uncertainties hamper the establishment of complete licensing contracts.

The following two motives originate from hold-up problems. The hold-up problem refers to the situation where distribution of profits depends on the relative bargaining power of two parties. Uncertainty about losing bargaining power impedes first-best solutions.^[36]

(3) Firms relying on natural resources base their capacity planning on assumptions about future prices and availability of raw material. Suppliers of raw material, however, hold the better information about prices and supply hence they may overstate availability to the firm. The higher the capacity the firm builds the higher it will accept to pay later on.

(4) In contrast, if the supplier provides a good requiring specific investment upfront the bargaining position may be weakened. Fearing this, the supplier will save costs by making suboptimal investments upfront.

The more complex and technologically sophisticated the proprietary assets are the stronger motivations to internalize will be. From an organizational behavior perspective another motive to invest may simply rely on the management’s personal goals, attitudes and ambitions. If bonuses are paid according to the volume of sales, managers will have motives to enlarge the firm through FDI.^[37]

2.3.2. Sector-Level Determinants

Motives to undertake FDI may differ among sectors, depending on the relative importance of firm-level and plant-level economies of scale. Economies of scale occurring on the firm level and affecting the entire MNF are referred to as firm-level economies of scale. They are closely related to the assumption of the public goods like character of proprietary assets. In contrast there exist plant-level economies of scale occurring on the level of individual plants. Generally, FDI is more likely to occur in sectors where large firm-level economies of scale are observed, as fragmentation becomes relatively cheap. High plant-level economies, on the other hand, result in efficiency losses when splitting similar activities geographically and thus making FDI only attractive, if entire stages of the value-added chain are transferred to plants in foreign locations.^[38] High firm-level economies of scale motivate FDI in general. HFDI, however, is hampered if economies of scale at the plant-level are important, thus it is expected to occur only in sectors with low plant-level economies of scale.^[39]

Based on approaches from oligopolistic theory and location theory there are motives stemming from effects of sectoral agglomeration to undertake FDI. The analysis encompasses two perspectives: (1) herd behavior or follow-the-leader behavior and (2) external economies.

(1) Herd behavior refers to some sort of oligopolistic reaction. Oligopolists, being risk minimizers, seek to avert destructive competition. FDI motivation exists in a loose-knit oligopoly. In this case, due to a lack of sufficient contractual consensus to coordinate activities, simple imitative behavior prevails and oligopolists would therefore follow each other into new (foreign) markets to assure commercial interests.^[40] Knickerbocker (1973) argued that imitation motivates foreign investment in the following way:

“Rival A establishes a subsidiary in France. Rivals B and C recognize that this investment might knock out their export business in France and give A first-mover advantage if the investment should prove successful. Still worse, A might discover some competitive asset in France that it could repatriate to torment B and C on their native soil. These considerations dispose B and C to imitate A and found their own subsidiaries in France. Their combined expansions of capacity of course should cause excess capacity and/ or depress prices in the French market – a deterrent. On the other hand if investments turn out badly for all parties, they do share some oligopolistic understanding and hence excess profits (worldwide) that will make the losses bearable.” (Knickerbocker in Caves, 1996)

Hence, in order to protect market share, firms may be motivated to enter a foreign market via FDI.^[41]

(2) Agglomeration effects may also be stimulated by external economies. These exist when a firm’s cost efficiency improves as a function of an external factor, such as sector size. Geographic agglomerations on a sector level may provide positive external effects through better access to suppliers and production factors as well as improved infrastructure or technology spillovers.^[42] This may also apply when suppliers or customers have established foreign production facilities requirering the MNF to follow in order to retain business. Specific external economies in the context of MNFs may also originate from large expatriate communities in the host country^[43].

A third motivation related to the sector level is to reduce competition. Reducing competition implies increasing market power and thus higher sales and profits. Hymer (1976) postulates that the control of foreign firms aims at removing competition between the parent firm and the foreign firm in the host country market:

“It frequently happens that enterprises in different countries compete with each other because they sell in the same market or because some of the firms sell to other firms. […] One form of collusion is to have the various enterprises owned and controlled by one firm. This is one motivation for firms to control enterprises in foreign countries.” (Hymer, 1976, pg. 25)

Motives for collusion may be (1) generating synergies with local production units to improve market power, (2) improving distribution channels, (3) greater product ranges by acquiring competitors with complementary products, (4) access to host government contracts through mergers with local firms, (5) establishing collaborative alliances in order to prevent competitors from merging.^[44]

2.4. Host Country Determinants

2.4.1. Three Types of Location Determinants

Firm and sector-level determinants explain why firms become multinational, however, they do not determine where FDI is undertaken. As indicated, country-level factors are decisive for determining whether production in a host country is more profitable than serving the market via exports. Following the categorization proposed by the World Investment Report (WIR) 1998^[45] three types of location factors may be distinguished: (1) policy determinants, (2) economic determinants and (3) business facilitation. Type one and two assemble long-term determinants that, if at all, can only be influenced by host governments via long-term policies. The third type, on the other hand, represents short-term incentives. Types are ranked according to relevance, assuming that MNFs will primarily examine fundamental factors like country risk; even if economic prospects are promising, FDI may not happen due to unfavorable political conditions. This is illustrated by Schneider and Frey:

“A country in which there is political unrest or in which there is threat of having the investment nationalized (without adequate compensation) is more of a risk and therefore ceteris paribus less attractive to invest in than a country offering political stability and a guarantee of property rights.” (Schneider & Frey, 1985)

Clearly, this may be in particular important with respect to FDI in developing countries where political stability for instance might be a more critical factor than in developed countries. However, if basic prerequisites are fulfilled, FDI decisions will depend on economic factors. Considering increased competition for FDI, pro-active investment incentives (business facilitation) gain importance.

[...]

^[1] For model notation see pp. 63

^[2] UNCATD (2005), p.3

^[3] Wezel (2003), abstract

^[4] Henley (2004), p.1

^[5] A more detailed discussion of direct investment enterprise nomenclature follows in section 2.1.2.

^[6] IMF (2003), pp.162

^[7] Buckley & Clegg (1991), p.250

^[8] Accolley (2003), p.5

^[9] Reinert & Ramkishen (2009), p.444

^[10] Barba-Navaretti et al. (2004), p.300

^[11] OECD (1996); IMF (1993) in Barba-Navaretti et al. (2004), p.1

^[12] Caves (1996), pp.2

^[13] Dunning & Lundan (2008), pp.6

^[14] Krugman & Obstfeld (2003), pp.10

^[15] Caves (1996), p. 26

^[16] Hymer (1976), pp. 11

^[17] Reinert & Ramkishen (2009), p.444

^[18] Markusen (2002), p.1

^[19] Markusen & Venables (1998)

^[20] Hymer (1976), p.81

^[21] Hymer (1976), p.84

^[22] Dunning & Lundan (2008), pp.98

^[23] Dunning & Lundan (2008), pp.100

^[24] UNCTAD (1998), p.92

^[25] Caves (1996), pp.32

^[26] Dunning & Lundan (2008), p.108

^[27] Markusen (2002), Chp.5, 7, 8

^[28] In the model headquarter location decisions are based solely on factor price differentials.

^[29] A more detailed composition of markups is provided in annex A

^[30] Caves (1996), pp.2

^[31] Barba-Navaretti et al. (2004), pp.28

^[32] Caves (1996), p.59

^[33] Barba-Navaretti et al. (2004), Edwards (2002), p.35; pp.119; Markusen et al. (1995), pp.404; Kumar (1994), pp.64

^[34] Barba-Navaretti et al. (2004), pp.36, 102; Caves (1996), p.14; Markusen et al. (1995), pp.404; Kumar (1994), pp.64

^[35] Barba-Navaretti et al. (2004), pp.115

^[36] Schiller (1994), p.7

^[37] Buckley & Clegg (1991), p.252

^[38] Barba-Navaretti et al. (2004), pp.28

^[39] Barba-Navaretti et al. (2004), pp.30

^[40] Buckley & Casson (2002), pp.78

^[41] Caves (1996), pp. 86

^[42] Krugman & Obstfeld (2003), pp.147

^[43] Dunning & Lundan (2008), pp.87

^[44] Dunning & Lundan (2008), pp.73

^[45] UNCATD (1998), p.91