FOREIGN DIRECT INVESTMENT AND THE GROWTH OF AFRICAN ECONOMIES:

(1)

FOREIGN DIRECT INVESTMENT AND THE GROWTH OF AFRICAN ECONOMIES:

Determinants, the Role Played by Local Financial Markets and Labor Skills.

Master s Thesis

MSc. INTERNATIONAL ECONOMICS AND BUSINESS

Justice Joel Kamote (1474049)

UNIVERSITY OF GRONINGEN GRONINGEN, THE NETHERLANDS

AUGUST 2006

(2)

Acknowledgements

I would like to thank Dr. Desislava Dikova, my supervisor, for our fruitful discussions and her excellent supervision. But most of all, I am grateful for her patience in answering all of my questions.

My special thanks go to the Eric Bleumink Fund of the University of Groningen for their financial support that gave me an opportunity to study in The Netherlands especially at this university. Equally important, I thank Mrs. M.V. Rothengatter-Lake of the admissions office for her support during and after the application processes to this University.

Naturally, none of these could have been possible without the help and support of my family; my wife Tumaini, my daughters; Julieth and Jacqueline; and my parents. A call back home during my stay here was a source of new pace, new zeal and new vigor.

Finally, I would also like to thank the following friends who made my stay in Groningen a wonderful adventure; Zeng (David), Migeyo, Winneke, Pasha, Masako, and Bore.

Groningen, August 4, 2006

Justice Joel Kamote

(3)

Abstract

In this thesis, we examine the extent of institutional reforms in improving country attractiveness for the location of foreign direct investments (FDI) and whether host country s labor skills and financial markets act as the best conduits for the FDI spillovers to influence economic growth. Our empirical analysis, using cross-country data from 1985-2003, shows that several formal institutions are found to play significant role in influencing FDI: price liberalization, business climate, development of financial sector and corruption. We also found that the absorptive capacity of local labor force in transmitting FDI spillovers to influence growth to be significant. This finding support the policy motive for undertaking institutional measures such as developing better education institutions and technical training programs that improve the host country s stock of human capital. We did not however find any support for the role played by local financial institutions in transmitting FDI externalities to influence growth, a finding that suggest that local financial institutions in Africa are incapable of influencing FDI spillovers on growth. This conclusion calls for African countries to take deliberate policy measures to liberalize and develop the financial sector.

Keywords: Foreign Direct Investment, Africa, Institutional Reforms, Growth, Financial Markets, Labor Skills.

(4)

Chapter 1 1 Introduction

1.1 Background

The study of foreign direct investment (FDI) in the world economy has received a particular attention in the last 15 years of the 20^th century. This follows an ever increasing flow of FDI far outpacing the traditionally experienced flows of trade and income during the same period. While world real GDP experienced an increase of 2.5 per cent per year in the period 1985 to 1999 and world exports rose by 5.6 percent in the same period, the world flows of FDI raised by 17.7 percent, Navaretti and Venables (2004 p. 3).

Comparing to pre 1985 figures, world real GDP grew by 3.1 per cent per year, world exports grew by 5.2 per cent and FDI grew by 4.2 percent per year¹. It is due to this significant role played by FDI in the global economy that shifts the focus of scholars and business community from global trade to FDI.

FDI is considered as an important factor that determines the growth of economies as it contributes to capital stock accumulation and brings with it new technologies from home to host countries of FDI. The growth miracle experienced by East Asian countries in recent decades is largely explained by the role played by FDI, Gary Dean, (2000). This significant importance of FDI in promoting economic growth is not taken for granted however. Outward-oriented policies adopted by respective country governments together with other country determinants such as natural resource endowments (OECD 2002), institutional development (Bevan et al. 2004) play a significant role in determining FDI inflows to host countries. Thus the attractiveness of a country plays a significant role in determining the flows of world FDI.

(7)

(2003). Theory suggests that inward FDI bring capital and technologies that the recipient country lacks, and these, combined with others factors available in the host economy bring about economic growth². Indeed FDI has the potential to generate employment, increase productivity, transfer skills and knowledge, enhance exports and contribute to long-run economic growth of the recipient country.

In this connection, the role played by financial markets is to make it possible for the spillovers and linkages associated with FDI to materialize. Alfaro et. al., (2003) points out that though significant amount of FDI arrives through mergers and acquisitions, it is not just the easy availability of loans but also well-functioning stock markets that matter.

Well-functioning stock markets enlarge the scale of resources for entrepreneurs, and therefore play an important role in creating linkages between domestic and foreign investors. Highly developed financial institutions may act as a decisive factor in determining the interaction between foreign investors and local market beyond hiring labor, Alfaro et. al. (2003).

The link between FDI and growth through technological progress is gauged by knowledge and skills transmitted through training of recipient country labor force.

Borensztein (1998) point out that FDI act as medium for the implementation of new technologies packed together with the inflows of FDI bundles. The interaction of host country labor force or local firms with foreign-owned projects determines the extent of technology transfer from FDI to host country. Thus institutional reforms that improve the quality of the host country s stock of human capital are likely to influence the impact of FDI on economic growth. Borensztein (1998) analyze the effects of FDI flows from industrial countries to 69 developing countries over the last two decades and find that FDI is an important vehicle for transferring technology and that it has a growth- enhancing effect; however, this finding holds only if the recipient country has a minimum threshold stock of human capital ranging between 0.76 and 1 years of post primary schooling.

2 See section two for the presentation of the theory

(8)

The World Bank s 2001 publication on global development finance summarizes the importance of absorptive capacities and success of FDI in influencing growth.

Absorptive capacities include macroeconomic management (captured by inflation and trade openness), infrastructure (telephone lines and paved roads), and human capital (share of labor force with secondary education and percentage of population with access to sanitation). Though the publication do not point financial markets, but the empirical evidence on FDI and growth, with financial market interacting with FDI have been studied to a greater extent and reached positive conclusions, King and Levine (1993a,1993b), Beck, Levine and Loayza (2000), and Levine, Loayza and Beck (2000).

All these studies suggest that financial systems are important for productivity and development. Developed financial institutions plays a crucial role in allocating both domestic and foreign financial resources to the most productive investment projects, and hence influence growth, Bevan et. al., (2004), Alfaro et. al. (2003).

Africa is lagging behind in attracting foreign investment though the amount of inward FDI has been increasing in recent years (See table 1). An overview of FDI for development given by OECD (2002) shows that the entire continent (except South Africa) received FDI flows worth US$ 8.2 billion in year 2000, an amount equal to inward FDI attracted by Finland alone in the same year. This figure account for about 0.6 per cent of total world FDI flows, OECD (2002).

Insert Table 1 about here

The low levels of FDI to Africa can partly be explained by the late adoption of outward-

(9)

factors believed to hold back FDI flows to Africa include the perceived poor sustainability of national economic policies and poor quality of public services, D. Dollar and W. Eastely, (1999). Other studies stress macroeconomic policy failures as deflecting FDI flows from the region, UNCTAD, (2005). These include irresponsible fiscal and monetary policies that generate unsustainable budget deficits and inflationary pressures, which in turn raise local production costs, generating exchange rate instability, and making the region too risky for foreign investors, (Reinhart and Rogoff, 2002).

1.2 Institutional Reforms and Location of Foreign Direct Investments

Douglas C. North, (1991) define institutions as humanly devised constraints that structure political, economic and social interactions. While human create institutions, institutions shape the environment where individuals interact. Foreign investors, just as other businessmen, view institutions as assets that directly affect their businesses. Institutions form part of created assets , Bevan et. al. (2004) of countries, and arguably become increasingly significant relative to more conventional natural resources , like raw materials (United Nations, 2002). North, (1990) point out that economic institutions create the incentives faced by domestic economic actors as well as foreign investors. This provides the motivation for including institutional reforms into the study of foreign investments, (Bevan et. al., (2004); government policy (Gomes-Casseres, 1991), intellectual property rights protection (Oxley, 1999) or political risk (Henisz, 2000) especially for the choice of entry modes.

Moreover, institutions reduce transaction costs, North (1991). Foreign investors are concerned with institutional framework in the host country of their investments not only for entry decisions (Bevan et. al., 2004) but also the constraints associated with the complexities of the local institutions (North, 1991). Because foreign investors do not operate in a separate world, transaction costs associated with their interaction with other economic actors become an important component of their businesses. Wealth- maximizing individuals will usually find it worthwhile to cooperate with other players

(10)

when they play repeated game, they possess complete information about other player s past performance and when there are small numbers of players, North (1991). Such game theoretic context becomes very important when the motives of foreign investors go beyond sourcing inputs to incorporate the markets for their final produce. In this spirit, well developed institutions become relevant in minimizing transaction costs.

An interesting work by Bevan et. al. (2004) explores the impact of institutional development in transition economies on the location choice of foreign direct investments.

The authors show that several formal institutions such as private ownership, banking sector reform, foreign exchange and trade liberalization and legal development to have significant impact of FDI inflows. Transition economies provide a clear picture on how institutional reforms contribute to the increase in FDI flows, Bevan et al. (2004). For instance the shift from planned economic to liberal market policies (e.g. Russia) and the increasing role played by private sector in these economies provide grounds for favoring institutional reforms policies.

African economies have not isolated themselves in the move towards institutional reforms in the post World War II, though their adoption came a bit late. Such reforms involved fundamental policy and institutional transformations that shifts the role played by states to market which is important especially in societies that lack strong entrepreneurial capacities, ESRF, (2003).

1.3 Purpose

Different to other studies on the subject in Africa, we address to what extent the institutional reforms have improved the attractiveness of African countries to contribute to inward flows of investment, and whether the inflows of FDI contributed to economic

(11)

Specifically, we investigate two main questions:

First, to determine whether institutional reforms have improved the level of attractiveness of a target country and investigate how the variation of this attractiveness would affect FDI inflows,

Second, to determine how FDI inflows affect GDP growth through financial markets developments and skill levels.

Our contribution to the existing literature is two-folds: first we consider the role of institutional development in improving the attractiveness of a country for the location of foreign direct investments. In this respect, instead of pooling together the factors that define the attractiveness of a country for the location of FDI, as most of previous studies have done, we stress the role of the quality of institutions in defining the attractive of a country. Second, we examine how the quality of institutions, particularly the financial institutions and institutions that help improve the quality of the stock of human capital, in enhancing the effect of FDI on economic growth. In particular we investigate how the extent of cross-country variation in quality of institutions influences the inflows of FDI and its impact on growth.

Following this introduction, the thesis continues as follows: chapter two presents a review of literature, theory and methodologies used in previous studies for the determinants of FDI and the relationship between FDI and growth. It then followed by our conceptual arguments and hypotheses. Chapter four presents our methodological approach. Chapter five presents some specification tests for the models. And chapter six presents findings from our empirical analysis. Chapter seven presents discussion and policy implications of our findings and, finally, our paper ends with conclusion in chapter eight.

(12)

Chapter 2 2 Literature Review

2.1 Kojima s Macroeconomic Theory of FDI

Kiyoshi Kojima, (1973, 1978, 1982, and 1990) describes a theory of trade-oriented FDI, taking into account the principle of comparative advantage. The theory is essentially an extension of the neo-classical factor endowments tradition in that it explains trade in intermediate products. The basic tenet is that FDI should originate in the home country s comparatively disadvantaged industry, which is prospectively a comparatively advantage industry in the host country Kojima (1982).

Kojima and Ozawa (1984) and Phongpaichit (1990: 5) point out that, the Japanese approach to FDI differs in an interesting way from that of mainstream economics in that Japanese approach was concerned with changing comparative advantage, Kojima (1973, 1975, 1985). Kojima s hypothesis conveys two types of FDI: trade oriented and anti-trade oriented (Kojima 1973, 1975, 1985, OECD 1987: 36), depending on two issues; first whether FDI works in harmony with comparative advantage, and second, whether the flows of FDI work against comparative advantages, H.C. Blompvist, (2004).

In the trade-oriented case, the motivation to undertake FDI is to exploit location advantages in order to take advantages of international markets. Hence, the home country is the one with a comparative disadvantage and the foreign investment undertaken tends to generate trade and enhance industrial restructuring in both home and host countries as the share of industries with comparative advantage (or at least potential comparative advantage) will grow, compared to other industries (Daguila and Nguyen 1994, Kojima 1973, Lizando 1991, Mortimore 1993). In this approach, the distribution of value-added

(13)

In the anti-trade case, FDI are considered to be in industries where the home country has a comparative advantage. In such case investors inspire to protect the oligopolistic advantage and a crucial objective is to preserve the investor s market share, Peltola, (1994). FDI will substitute domestic production for trade and in the process it is likely to crowd out local production in the host country, H.C. Blompvist, (2004). If inward FDI is technologically advanced than that in the host country, this type of investment will have limited linkages with local economy in the host country.

Kojima s hypothesis was that Japanese FDI is at large trade-oriented, while American investment is anti-trade oriented. So long as the later case does not comply with comparative advantage, Kojima concludes that it is welfare reducing in both home and host countries³.

Kojima and Ozawa (1984, 1985) give a formal presentation of Kojima s ideas. They point out that if the scope of international trade is generated by a situation in international immobile factors of production and mobile goods. (If capital is mobile, the relative capital/labor abundance, and hence, the relative factor prices are equalized. In that case, no foundation for tradition trade remains). Their idea is that a third factor of production, in this case industry-specific entrepreneurial endowments , may render trade more advantageous than in the traditional Heckscher-Ohlin framework, if the other factors are immobile and if there exists differences in the relative endowments of the third factor, capital in this case, between the two countries. Thus, Kojima recommend that a country s industries with comparative disadvantages could improve profitability by relocating to a country where the same industries have comparative advantages, Kojima (1979). The relocation of resources tends to broaden the basis for trade and enhance welfare in both home and host countries.

Kojima describes two recommendations in this view when distinguishing between inward and outward FDI. The first is that outbound FDI should be undertaken by countries that produce intermediate products that require resources and capabilities in which the home

3 See Kojima and Ozawa 1984 for a discussion of welfare effects of different types of FDI.

(14)

country has a comparative advantage, but that create value-added activities that require resources and capabilities of which that country is disadvantaged. The second is that, inbound FDI should import intermediate products that require resources and capabilities of which the recipient country is disadvantaged, but the use of which requires resources and capabilities in which it has a comparative advantage.

These two recommendations suggest that the FDI inflows to African countries are primarily targeted to natural resource-sectors such as mining, fishing, tourism and in those activities that require the relatively abundant and cheap factor input, labor.

Existence of natural resources by itself cannot maximize the inflows and benefits of FDI to the recipient country, but rather institutions that improve the attractiveness of a country and those reduce the costs associated with FDI projects can maximize both the inflows and the growth impact of FDI, Alfaro et. el. (2003).

2.2 Review of Methodological Approaches to the Study of FDI and its Impact on Growth

This part surveys the recent methodological approaches used by a number of scholars to analyze the determinants of FDI on one hand and the impact of FDI on growth on the other. We first present a brief assessment of studies on the determinants of FDI to developing countries, and Africa in particular. We then proceed with a review of the techniques employed to assess the impact of FDI on growth in part (2.2). We focus on those studies on developing countries and Africa in particular for two main reasons; first, the determinants of FDI inflows to Africa are quite different from those of other regions⁴; and second, there is perceived political grouping of the regions of the world based on the

(15)

interests of the big economic powers⁵. This kind of classification shows clearly that there might be determinants of FDI flows other than economic and natural resource endowments that account for the ever increasing flows of investments to the rest of the world which Sub-Saharan Africa does not possess. We think isolating the studies conducted for Africa will help us assess and choose the appropriate methodological approach that best fit our research problem. We will however refer to studies conducted in transition economies particularly on the role of institutions and FDI on growth.

2.2.1 Methodology review: determinants of FDI

Four studies: Morisset (2000), Aseidu (2002), Beatrice Mkenda and Adolf Mkenda (2004) and Addison and Heshimati (2003) employ different econometric techniques and different dataset to analyze the variation in country attractiveness in attracting FDI flows.

Their findings differ to some extent partly due to different estimation techniques and datasets. While all studies were conducted for a set of developing countries, the findings turn to be quite different when the countries analyzed are all in Africa, Sub-Saharan Africa or all developing countries.

These four studies find that openness to trade, market size (proxied either by GDP growth or total population), natural resources and skills levels to be statistically significant in attracting FDI. In addition they find that size of government has the expected sign though in most cases it is found to be insignificant. On the other hand, democracy, infrastructure, and degree of industrialization are insignificant though this finding is not consistent in all studies.

5 In a Reflections Speech at the University of Dar es Salaam, Tanzania, Mwalimu Julius K. Nyerere (1997) challenges African Leaders on the perceived classification of flows of investment from developed countries to developing countries. He classifies the world into three big economic powers and four Mexican economies that are to a greater extent influenced by the distance from the big economic powers: US and Canada invest in Mexico to avoid Mexican migrants to North America; Japan invests in East Asian to avoid East Asians to migrate to Japan; and Europe has two Mexicans, East Europe and North Africa.

(16)

Jennifer Tobin and Susan Rose (2003), take a different approach to assess whether Bilateral Investment Treaties (BITs) fuel investment flows from developed countries to developing countries and whether those treaties have any impact on business environment for local private investments. The authors find that GDP per capita, market size and political risk to be significant determinants of FDI flows. More importantly, when they include BITs into the model the rest of explanatory variables remain significant as in the first estimation, with BITs having a positive impact on FDI.

Beven et. al. (2004) use a cross-section analysis of a panel of source and host countries (in Eastern European transition economies) of FDI to assess the impact of institutional reforms on FDI flows from source to host countries. The authors find that development of formal institutions plays a significant role in determining the flows of FDI. The authors point out that investors view institutions as created assets that form part of country assets that influence their decisions to locate in a particular country.

In this thesis, we borrow the approach used by Bevan et. al. (2004) to assess the role played by local institutions in influencing FDI flows. We think these institutional reform measures help improve the country attractiveness for attracting FDI inflows (See Bevan et. al. 2004).

2.2.2 Methodology review: FDI and growth

Economic scholars have extensively debated the relationship between FDI flows and economic growth. de Mello (1997) presents two channels through which FDI may be growth enhancing⁶. The long-established view suggest that FDI flows contribute to economic growth by increasing the country s stock of capital, while in most recent years,

(17)

a bundle embody advanced technology, and that through training and skill acquisition of recipient country s labor force as well as introduction of alternative management practices and better business organization, the effects of FDI on growth through technology progress will accelerate economic growth. Townsend (2003) uses endogenous growth model to investigate how technological progress transmitted through labor skills acts as an important determinant of growth. Both these channels of knowledge transfers augment the existing stock of capital, and together they enhance economic growth.

A second group of scholars focus on the role played by financial instituions in their interaction with FDI to influence economic growth. Alfaro et al (2003), use a set of financial variables in growth regressions to analyze their interaction with FDI in enhancing economic growth. After controlling for endogeneity of FDI and legal origins of financial systems, they find that having well-developed financial market, FDI role in promoting growth is much significant. Consistent findings are obtained by Hermes and Lensink (2003) who find that the host country s financial system is a precondition for FDI to have impact on economic growth. The authors use a simple model of technological change⁷to analyze the impact of the interaction between financial system and FDI on economic growth. They find that out of the 67 countries analyzed 37 (most of them in Latin America and Asia), have sufficiently developed financial institutions to support FDI inflows to have impact on economic growth.

The third study that takes into account the role played by development of financial institutions is by Carkovik and Levine (2002) who investigates whether the growth-effect of FDI depends on the level of educational attainment of the host country, level of economic development of the host country, the level of financial development of the host country and trade openness. They find that the exogenous component of FDI does not exert a consistent positive impact of FDI on growth. They find positive and significant impact of FDI on growth after including initial income per capita and average years of schooling as well as when controlling for government size and inflation. Their OLS regression results suggest FDI has a positive growth effect in sufficiently financially

7 See Barro and Sala-i-Martin (1995)

(18)

developed countries, which confirms earlier findings. However they fail to establish from their findings, at what levels of financial development can exert force of FDI to enhance growth.

Other studies that take into account the role played by financial markets include Levine and Zervos (1998) who analyze at country level, the role played by different financial institutions, and find that while stock markets and banks provide different services; both stock market liquidity and banking development positively predict growth, capital accumulation and productivity improvements. Rajan and Zingales (1998) take a different root; studying at industry level, find that the state of financial development reduces the cost of external finance to firms, thereby promoting growth. Wurgler (2000) combine industry level and country level data and find that even if financial development does not lead to higher levels of investment; it allocates the available resources better and hence promotes growth.

Thus to a greater extent, the studies reviewed above give a picture of the role played by financial institutions in transmitting positive FDI externalities to influence economic growth.

The findings presented above must be viewed skeptically; however, some of the previous studies do not control for country-specific effects and the use of lagged dependent variables as regressors in the growth models. These shortcomings may potentially bias the estimated coefficients as well as the standard errors. Our study takes care of country- specific effects in the assessment of country attractiveness in influencing FDI flows.

While we employ endogenous growth models in the growth regressions, the interaction between FDI and levels of financial market development and skill levels is also used to examine whether these channels enhance FDI to fuel economic growth.

(19)

2.3 Scope and Boundaries of our Study

In this thesis, we examine to what extent institutional reforms have improved country attractiveness to influence FDI inflows for our first model, and the impact of FDI on growth through technological progress and financial institutions as transmission mechanisms. The choice of the estimation techniques we employ is partly based on the underlying assumption we make regarding the nature of our sample countries and their integration in the global economy.

Moreover, we confine ourselves in studying a subset of African countries which we believe may be in the same state of development with respect to economic and institutional reforms and therefore may possess similar economic environments. The contradicting findings on the subject in previous studies might have been contributed by analyzing countries with different state of affairs, an aspect that may lead to obtaining opposing effects on their estimation. Townsend (2003) explain how analysing a dataset with countries that differ to a greater extent in levels of development may affect the estimated results and may as well lead to obtaining opposing effects in the hypothesized relationships.

Different from previous studies of FDI on Africa, this paper seeks to integrate the impact of institutional reforms on attractiveness of a country in influencing FDI flows and the impact of FDI on growth through technological progress and financial system development. In the former case, most of previous studies use pure cross-sectional data to determine the extent to which differences in host country characteristics explain the variation in FDI across countries. The analysis in this case attempt to answer the question: assume country J and country K differ in variable y by one unit. What is the expected difference in FDI/GDP between countries J and K? This thesis takes a different approach. First, it investigates to what extent the institutional reforms have improved country attractiveness to influence FDI inflows and second, to what extent the inflows of FDI exert shock on economic growth. The relevant question in this case is: suppose country J increases variable y by one unit. What is the expected change in country J s

(20)

FDI/GDP? Fixed effects framework becomes relevant in this case, since it allows us to assess changes within different cross-sectional units over time.

On the later case, scholars have employed a number of approaches to the study of the relationship between FDI and growth. The earliest approach was that FDI flows contribute to growth through increasing country s stock of capital. The second was that FDI act as a channel through which international technology transfer takes place in this case level of labor force skills plays an important role, and lastly, the role played by financial system using a series of financial market development indicators in their interaction with FDI to influence growth. In this paper, we use the endogenous growth model, borrowed directly from Borensztein et. al. (1998); also used by Isaac Townsend (2003). The authors use black market premium as one of their regressors as well as initial GDP per capita. In our case, we use inflation to capture the state of price liberalization in a country. While these authors use Seemingly Unrelated Regression estimation technique, this thesis employs fixed effects technique. The reason for diverging from their estimation technique is well defended by our sample selection and the results obtained in our specification tests as discussed in chapter five and presented in section 6.1.

To investigate the role played by financial institutions, we use a battery of financial market development indicators in the growth regression. These variables were also used by Alfaro et. al. (2003). We use the same financial indicators used by the authors and we add another variable, Financial System Deposits, which we believe it indicate a proportionately large size of the financial sector.

(21)

Chapter 3 3 Conceptual Arguments and Hypotheses

Institutions, FDI and Growth

The significance of institutions for economic growth has long been stressed in the literature since the times of Adam Smith. In the more recent years we acknowledge the contribution of scholars such as David Landes, (1998) and Nobel Prize Winner Douglas North in 1993 for their tremendous work on this line.

Much of the recent work has been motivated by the existing cross-country differences in income growth, and most of recent scholars emphasize that the cross-country differences in income growth can be explained by institutions; their origins and their level of development, World Economic Outlook, (2003). The report highlights that much of the observed income differences among countries are closely correlated with indicators of institutional quality.

What do the existing literature tells us about institutions and growth? The answer to this question can give us a clear picture of the role played by institutions on economic growth.

The World Economic Outlook report, (2003), gives two approaches to answering this question. The best approach to this question, according to the report is to refer to two main definitions of institutions as informed in the literature. The first and most cited definition is that of Douglas North, (1990), who defines institutions as both formal and informal constraints of political, economic and social interactions. From this viewpoint, good institutions are viewed as devices that establish an incentive structure that reduces uncertainty and promotes efficiency-and hence contributes to stronger economic performance, World Economic Outlook, (2003).

The other definition looks at institutions as particular entities, procedural apparatus, and regulatory frameworks that affect performance of economic actors primarily through fostering better policy choices, World Economic Outlook, (2003). Such devices include central bank independence and balanced budget amendments; the existence and design of

(22)

international trade agreements; and regulations governing the functioning of product, labor and financial markets.

The two definitions above show clearly how institutions become important for economic development and growth. Our approach in this thesis has been to incorporate elements of each of the two definitions. We consider existence of both informal and formal constraints on business operations to have significant impact on the decisions of foreign investors to locate in a particular country. On the other hand, the existence of formal institutions such as financial markets and central bank independence to have particular importance in policy choices, World Economic Outlook, (2003) and well as allocating the available financial resources most efficiently (Alfaro et al. 2003), and hence contribute to growth. Our findings, particularly for the role played by local labor force skills in interaction with direct investment lend support to the argument that institutional setup that foster better policy choices that improve the quality of local labor skills may lead to economic growth. In a summary assessment reported in the World Economic Outlook, (2003), points out that different measures of institutional development such as quality of governance, including the degree of corruption; political rights; public sector efficiency; and regulation systems, and also the extent of legal protection of private property as well as the level of institutional and other limits placed on political leaders affect growth.

While better institutions affect growth, they also affect the location decisions of foreign investors and hence affect the economic performance of the recipient countries of FDI.

Bevan et. al., (2004) find that institutional setup in the Eastern European transition economies play a crucial role in influencing the inflows of FDI to the region.

Institutions as created assets for the location of FDI in the host country

(23)

business operations signal a poor state of business environment. Thus, we therefore propose:

Hypothesis 1a: Countries with high quality business climate receive more FDI flows.

Developed financial infrastructure and capital markets reduce the transaction costs associated with local financial services such as sourcing credit and settlement of payments. In addition, well developed financial infrastructure guarantees access to complementary local funds, which can reduce foreign investors exposure to exchange rate risk (Bevan et. al. 2004). Moreover, local economic agents can as well get access to bank credit which increases the demand for input and consumer goods, Bevan et. al.

(2004).

Alfaro et. al. (2003) point out that well developed and well functioning stock markets not only increase the spectrum of sources of finance but also to efficiently allocate resources to the most economic sectors and hence accelerates economic growth. Thus, financial institution reforms increase business opportunities for both foreign investors as well as domestic local investors. Thus, we therefore propose:

Hypothesis 1b: Countries with more developed financial market infrastructure receive more FDI flows.

Foreign investors have to adapt to a variety of regulatory regimes applied in different countries. Such regulatory regimes, which comprise of establishing market institutions, competition policy and liberalization of prices, have greater impact on the way economic actors interact among themselves and with the government, and most importantly affect the nature of competition in the local and at the international markets. Thus, liberalization of markets and development of regulation and competition policy becomes a central aspect of deciding where to invest, Bevan et. al. (2004).

(24)

In this paper, we consider the importance of price liberalization in the decision of foreign investors on their location of their activities. Price liberalization was among the reforms packages adopted by African countries (Jason A. Lovelace, 2005), which involved abolition of government control on prices to let the market mechanism do the job.

Foreign investors prefer to operate in competitive domestic markets, Bevan et. al. (2004).

Thus the incentive for foreign investors to operate in liberalized markets, as informed in the literature (c.f Bevan et. al. 2004), is that, price liberalization creates new business opportunities for foreign investors, while at the same time the abolition of exchange rate restrictions and multiple exchange rates allows repatriation of profits and reduces transaction costs. A more liberalized market guarantees stability in prices.

Thus, we suggest the following hypothesis regarding price liberalization:

Hypothesis 1c: Countries with less inflation receive more FDI flows.

Quality of institutions in the host country of foreign investors is an important aspect that influences the decisions of foreign investors. The concerns about perceptions and assessments of public institutions in affecting both growth and the location of FDI is important especially about how well they function and what their impact on the behavior of private sector. An empirical work conducted for the World Economic Outlook report of 2003 finds that three measures of institutions are important in attracting FDI inflows and influence growth. These measures include: quality of governance, including the degree of corruption, political rights, public sector efficiency and regulatory burdens; the extent of legal protection of private property and how well such laws are enforced; and lastly, the level of institutional and other limits placed on political leaders. The report further points out that the perception of the political, economic and policy climate embodied in the institutional measures are likely to be of key importance in shaping

(25)

Existing institutional framework shapes the legal framework, property right protection, and the way settlement of business disputes is conducted, Bevan et. al. (2004) and more generally they reduce transaction costs (North, 1991). Weak institutions indicate poor coordination of businesses and existence of corruption. In this paper, we use perception of corruption to proxy for the quality of existing institutions in the host country of foreign investments. Perception of corruption is not however the aggregate measure of governance, but it can as well be used to explain the quality of institutions in a particular country.

We hypothesize the following regarding quality of institutions, location decisions of foreign investors and the inflows of FDI to African countries:

Hypothesis 1d: Countries with less corruption receive more FDI flows.

The existing empirical literature shows that there is strong correlation between institutional development and growth. The World Economic Outlook, (2003), provides a brief summary of empirical work that proved the existence of strong correlations between institutional quality and growth. According to the report, high income countries tend to have relatively strong and developed institutions as compared to low income countries.

And among the developing regions, the differences in per capita income levels and institutional quality behave in a cyclical way and are not consistent, with Sub Saharan Africa being the region with the most weakest institutions and highest growth volatility, World Economic Outlook, (2003). Such variations in institutional quality and their impact on economic development can be explained by a variety of factors⁸. Such disparity in institutional setup has been found to explain much of the institutional differences and hence economic performance, World Economic Outlook, (2003).

8 See Acemoglu, Johnson and Robinson, (2001a, 2002), for an explanation of the role of history and geography, particularly the motives of European colonization in shaping institutions and growth

(26)

In the recent years, economic scholars have developed a substantial empirical analysis on the roles of institutions and macroeconomic policies taking into account the impact of each of the institutional factors on per capita GDP, economic growth, and growth volatility. In an attempt to examine how institutions affect the level of GDP per capita, a background work for the World Economic Outlook, (2003), finds that each of the institutional measures has a statistically significant impact on GDP per capita. Similar findings are obtained by Hall and Jones, (1999), Acemoglu, Johnson and Robinson, (2001a), Easterly and Levine, (2003), and Rodrik, Subramanian and Trebbi, (2002).

In the analysis conducted for the World Outlook report, (2003), using a standard growth model, the researchers finds that the aggregate governance measure of institutions is itself capable of explaining nearly three-fourths of the cross-country difference in income per capita.

The second question investigated and reported in the report is how institutions affect GDP growth. Using a standard growth model to capture the effects of institutions and policies on cross-country variations in GDP growth the World Outlook report, (2003), points out that institutions have a strong and significant impact on GDP growth.

The existing literature informs us that while FDI can transmit greater knowledge spillovers, (de Mello, 1997), a host country s capacity to take advantage of these externalities might be limited to local conditions. More specifically, the literature has dwelt on FDI s contribution on capital accumulation, technological innovation and nature and development of institutions in enhancing the spillover effects of FDI on growth. In this thesis, apart from examining the role of institutions in shaping the attractiveness of a country for the location of foreign investment, this thesis further assess the role of financial institutions and labor skills as mechanisms through which FDI associated

(27)

In the neoclassical theory view, FDI influences income growth by increasing the amount of capital per person. This however holds in the short-run due to diminishing returns to capital. It is only in recent years when endogenous growth theorists (e.g., Romer, 1986, and Lucus, 1988), developed growth models in which FDI stimulate long-run growth through variables such as human capital and research and development (R&D). These authors suggest that, through technology transfer to their affiliates, and technology spillovers to unaffiliated firms in the host country, multinational enterprises (MNEs) can speed up the development of new intermediate product varieties, raise labor productivity, raise product quality, facilitate international collaboration in R&D and introduce new forms of management.

Except for a limited number of studies at micro-level, majority of studies at macro-level (e.g., Balasubramanyam et. al., 1996, Keller, 1996, and OECD, 2002), find that FDI contribute to total factor productivity and income growth in the recipient country of FDI beyond what domestic investment would generate.

In a study that analyze the effects of FDI flows from industrial countries to 69 developing countries, Borensztein et. al. (1998) find the positive technological externalities associated with FDI become important when the level of human capital in the host country ranges between 0.76 and 1 year of post primary schooling. Put together, the above findings suggest that policies that promote host country technological capabilities such as well developed education institutions, technical training, and R&D increase the aggregate rate of technology transfer from FDI and are important conditions for positive externalities of FDI on growth. Thus we propose:

Hypothesis 2a: The joint effect of FDI inflows and host country skills of labor force will have a positive effect on GDP growth.

Developed financial institutions are believed to play a significant role in influencing the positive externalities associated with FDI on growth. The literature puts the argument that, well-functioning financial markets, by means of lowering transaction costs, ensures

(28)

that capital is allocated to those investments that yield the highest returns, and therefore enhances growth rates, Alfaro et. al. (2003). Thus we propose:

Hypothesis 2b: The joint effect of FDI inflows and financial market development will have a positive effect on GDP growth.

(29)

Chapter 4 4 Methodology

4.1 Sample and Data

4.1.1 Sample

To select the countries we need for our analysis we use three criteria based on our research objectives: First, we consider all countries that have operated under political stability during the sample period. We obtain such information from The World Fact book. The reason for this criterion is that an unattractive investment environment is expected to affect FDI inflows and therefore have impact on economic growth. Second, we consider in our sample those countries that apart from their liberalization efforts under Structural Adjustment Programs (SAPs) during the 1980s, established measures to attract FDI inflows. Such measures include putting in place one-stop centre for investors, adopted deliberate policies to attract foreign investors, opening up capital accounts, liberalizing trade policies, and instituting market-oriented policies by lifting controls on the exchange rate and consumer commodities (Dunning 2001).

In addition, the SAPs involved the withdrawal of government from production activities giving room for private sector participation. Such measure ensures foreign investors a substantial share in the local markets. Third we eliminate those countries with inconsistent data in our sample period and for a number of variables. Exceptions are made however, where a country meets conditions one and two but with some missing data. We remain with a sample of 14 countries. The sampled countries, with their respective regions in brackets are provided in appendix A.1.1.

Our sample period starts from 1985 to 2003. We think this is appropriate period of assessing the efforts to attract FDI as well as its impact on growth as many African countries adopted market liberalization policies starting the mid 1980s.

(30)

4.1.2 Data sources and definitions

This section describes the data used in the empirical analysis, particularly the measures of FDI, economic growth, skill levels, financial development indicators and the institutional reforms indicators that define the attractiveness of a country for attracting FDI inflows.

However before defining our data, we introduce the models we estimate in the following paragraph.

We estimate two models separately each for one objective. We estimate the first model using fixed effects technique, with FDI as dependent variable and our institutional reform indicators as explanatory variables together with a number of controls. The second model is estimated twice; first we estimate an endogenous growth model in which technological progress act as a channel through which FDI spillovers takes place. Second, we estimate the same model using financial market development indicators as mechanism through which FDI spillovers do materialize, (Alfaro 2003). Thus while the first growth model investigate how skill levels matters in transferring technology from FDI to influence economic growth, the second growth model uses financial market indicators to investigate FDI externalities through financial markets, The relationship of the dependent and independent variables (with their expected signs) are given in the tables below:

Model 1: FDI as a dependent variable

Dependent variable Independent variable Expected sign

FDI Open (Openness to trade) +

Domestic credit to private sector (%GDP) +

Inflation -

Corruption perception Index -

(31)

Model 2a: Growth, Foreign Direct Investment and Skill levels

Dependent Variable Independent variable Expected sign

per capita GDP growth Life schooling +

Life schooling*FDI +

FDI¹ +

1 This variable enters as a ratio of GDP

Model 2b: Growth, Foreign Direct Investment and Finance

Dependent Variable Independent variable Expected sign

per capita GDP growth FINANCE +

FINANCE*FDI +

FDI¹ +

1 This variable enters as a ratio of GDP.

Note: We have six indicators of financial market development each enter the model independently

There are several sources for the data on FDI and growth. We use two main sources of data; the World Bank s World Development Indicators (WDI CD-ROM 2005) and the United Nations Centre for Trade and Development (UNCTAD) database. We believe these sources provide the most accurate data because compilation of the data is done by country statistical agencies in collaboration with World Bank and/or International Monetary Fund (IMF). UNCTAD database provides us with data on FDI and GDP.

According to the Balance of Payment Manual Fifth Edition, IMF and the Detailed Benchmark definition of Foreign Direct Investment, FDI refers to an investment made to acquire lasting interest in enterprises operating outside of the country of the investor.

Data on Corruption Perception Index comes from the Internet Center for Corruption Research (a joint initiative of The University of Passau and Transparency International).

Financial development indicators come from World Bank Financial Structure Database.

We obtain data on education from UNESCO Institute for Statistics, and the rest of the data comes from WDI CD-ROM 2005. Below we provide definitions of the data and their relationship. A Detailed description of all the data and their sources is included in the Data Appendix.

(32)

Our first model investigates how institutional reforms have improved the attractiveness of a country to influence the inflows of FDI.

Four institution reform indicators are included in our first regression as independent variables. First, Domestic Credit to Private Sector (henceforth, DCPS) Is credit by financial intermediaries to the private sector as a share of GDP (Beck, Levine, Loayza, 2000). This captures the level of development of local financial institutions and is expected to be positively related to FDI inflows (Alfaro et. El, 2003, Carkovik and Levine 2002). Second, Openness to Trade (henceforth, Open) - is measured as the sum of exports and imports as a ratio of GDP. This variable measures the extent and quality of the business climate of a particular country. Investors are concerned of various issues that hinder the day to day business conduct such as tax system complications that pose problems in processing imports or exports. Thus we use openness to trade to capture the quality of business climate in a given country. However, this variable is imperfect measure of business climate, due to limitations of appropriate data and its availability for the countries we study for our sample period, we opt to use this variable.

Our third variable is Corruption Perception Index (henceforth, Corrupindex) This variable indicates the quality of institutions in a country, as it is one of good governance indicators. A country with high quality of institutions implies that there are well-defined institutions for citizen and business society in particular to interact with. As such we expect countries with high quality institutions to attract more FDI. We use corruption perception index to capture the country s quality of institutions. And finally, Inflation (henceforth, CPI) - is the average growth rate of consumer price index at 2000 constant prices. This variable indicates the extent to which a given country has undertaken appropriate measures to liberalize prices. We use inflation to capture the extent of liberalization of prices/or macroeconomic stability of a given economy. We predict an

(33)

high population and higher incomes to warrant returns on their investments. We use population as a proxy for market size. We predict a positive relationship between population and FDI inflow, (see Beatrice Mkenda and Adolf Mkenda (2004)). Life Schooling (henceforth, lschol) Is the average number of years spent from Primary to tertiary levels. Foreign investors are attracted to countries or regions with well trained labor force that have the necessary skills to perform MNEs activities. Consistent with other studies, we use average number of years of schooling from primary to tertiary education to capture the extent and quality of skill acquisition by the labor force. We expect a positive relationship between skill levels and FDI inflows. Telephone Mainlines (henceforth, TEL) is the average number of telephone mainline subscribers per a 1000 people. We use telephone mainline subscribers to proxy for service infrastructure such as paved roads and Internet users. These are public good in nature but they affect the daily conduct of business. Thus we expect countries with well developed service infrastructure to attract huge amount of foreign investments.

The last control variable we use is Industry Value Added as a Percentage of GDP (henceforth, INDVAD). Foreign investments depend very much on the level of development in both the upstream and downstream industries for sourcing inputs and marketing their products in a given economy. Thus the level of industrialization of a given country is likely to influence FDI flows to a particular country. We use industry- value added as a percentage of GDP to measure the degree of industrialization a country has achieved. We expect a positive relationship between this variable and FDI inflows.

The dependent variable, Foreign Direct Investment (henceforth, FDI) is the net inflows of investment to acquire a lasting management interest (10 percent or more of voting stock) in an enterprise operating in an economy other than that of investor. It is the sum of equity capital, reinvestment earnings, other long-term capital, and short-term capital as shown in the balance of payments statistics. Gross FDI figures is the sum of the absolute value of inflows and outflows accounted in the balance of payments financial accounts.

In our model we use the inflows to the economy as this reflect the net FDI flow to a particular country.

(34)

Our second set of variables is those included in our first growth model. This model investigates the role played by skill levels of the recipient country labor force in which technological spillovers from FDI-based projects are transmitted to influence growth.

The dependent variable is per capital GDP growth and our independent variable is Life Schooling (henceforth, lschol) - is the average number of years of education from primary to tertiary level. The quality and level of skills of the recipient country s labor force indicate the ability of local workers and firms to systematically adapt to new and advanced technologies used by foreign investors (in the host country of FDI), either through training or other spillover effects to increase productivity and hence economic growth. Hence we expect, countries with well trained labor force to realize FDI-led growth.

We also include a number of controls that we believe play an important role in influencing growth. These include: quality of local institutions, government consumption, openness to trade and macroeconomic stability.

The third set of data we use are those included in our second growth model. These are the financial market development indicators. These variables are expected to influence the effect of FDI on growth. These data were constructed by King and Levine (1993a), Levine and Zervos (1998), and Levine et al. (2000) and also used by Alfaro et al.

2003).These financial market series cover the stock market data and lending in the economy. The data can be grouped into two categories: those relating to the stock market (or equity markets) and those relating to banking sector (or formerly credit markets). The data we use are available from the World Bank New Database on Financial Development and Structure Database (1960-2004)⁹.

(35)

Six financial variables are included in our growth regression, each at a time, First Liquid liabilities of the Financial System (henceforth, LLY): equals the sum of currency plus demand and interest-bearing liabilities of banks and nonfinancial intermediaries divided by GDP. It is the broadest measure of financial intermediation and includes three types of financial institutions: the central bank, deposit money banks and other financial institutions. Hence, LLY provides a measure for the overall size of the financial sector without distinguishing between different financial institutions. Alfaro et al. (2003) have used this measure. Second, Private Sector Credit (henceforth, PRIVCR): is the value of credits by financial intermediaries to the private sector divided by GDP. Third, Commercial-Central Bank Assets (henceforth BTOT): equals the ratio of commercial bank assets divided by commercial bank plus central bank assets. It measures the degree to which commercial bank versus central bank allocates society s savings in productive investments. King and Levine (1993a), Levine et al. (2000) and Alfaro et al. (2003) have used this measure. BTOT provides a relative importance of different financial institutions and sectors relative to each other. Fourth, Bank Credit (henceforth, BANKCR): is the credit by deposit money banks (excluding non-bank credits) to the private sector divided by GDP. Fifth, Financial System Deposits (FDGDP): equal demand, time and saving deposits in deposit money banks and other financial institutions as a share of GDP. It reflects the real deposits of the financial system as is deflated using end-of period as well as average annual CPI.

Data on stock market include variables introduced by Levine and Zervos (1998). These data was also used by Alfaro et al. (2003) and is available in the World Bank Financial Structure Database. In this study we use Capitalization (henceforth, SCAPT): equals the average of listed domestic shares on domestic exchanges in a year as a share of the size of the economy (the GDP). This variable captures the relative size of the domestic stock market.

(36)

The dependent variable is as in the first growth model, is the annual growth rate of real per capita GDP. We obtain this data series from World Bank Development Indicators (WDI CD-ROM 2005).

We also include a number of controls; Inflation (henceforth CPI): equals the annual change of Consumer Price Index (at 2000 constant prices), is used as a proxy for macroeconomic stability. Others include government consumption, openness to trade and skill levels.

Both our growth models use FDI as the ratio of FDI to GDP as an independent variable as well as an interaction term. The ratio of FDI to GDP takes into account the relative proportions of inflows of FDI adjusted to the size of the economy. Previous studies that use the same variable include Townsend (2003), Alfaro et al. (2003), Carkovic and Levi ne, (2002), and Assanie and Singleton, (2002). De Mello, (1999) also points out that FDI flows are sensitive to cross-country differences.

4.2 Design

4.2.1 Analytical framework

A general implication of our research objectives is that given the institutional indicators that define the attractiveness of a country, we expect to see an increasing inflow of FDI and these flows of FDI have impact on the economic growth. Thus, we construct the following diagram to represent the interrelation between FDI determinants, the FDI flow and GDP growth.

(37)

Fig 1: Analytical framework

4.3 Econometric Framework

This section describes the econometric methods we use to analyze the determinants of FDI inflows in our first specification and the effects of FDI inflows on growth in the second. First we use the Fixed Effects Model to analyze how institutional reforms have improved country attractiveness to influence FDI inflows to Africa and then use two growth regression models to study the effects of FDI on growth through technology progress and the financial markets channels. As with model one, the diagnosis results presented in section 6.1 below guide us to use fixed effects technique for the growth models

4.3.1. Country attractiveness and FDI inflows

The selection of the fixed effects model was based on the specification test as presented in section 6.1. The test fails to reject the null hypothesis that the cross-section fixed effects are redundant, and thus we estimate our first model using fixed effects technique.

The cross-country, cross-time panel estimation with fixed effects uses annual dataset for fourteen countries over the sample period, 1985-2003, such that there are nineteen

FDI Inflow Institutional

Reform Indicators that define Country Attractiveness

GDP Growth

Other Growth factors

FOREIGN DIRECT INVESTMENT AND THE GROWTH OF AFRICAN ECONOMIES: