On the Drivers of FDI and Portfolio Investment: A Simultaneous Equations Approach


    This paper assesses the drivers of Foreign Direct Investment (FDI) and portfolio investment using simultaneous equations, which control for the correlation between the two components of private capital flows. The results show that in addition to infrastructure and financial development, capital control and economic growth significantly explain private capital flows in developing countries. The paper also highlights that the impact of financial development on portfolio investment is nonlinear. Indeed, lax monetary policy and excessive credit provision could weaken the financial system and significantly reduce portfolio investment flows in the long run.


    Foreign direct investment , portfolio investment , financial development , non linearity , simultaneous equations , three stage least squares

  • 1. Introduction

    According to the neoclassical economic theory – assuming free capital markets and diminishing returns – capital should flow from capital-abundant countries (developed countries) to capital-scarce countries (developing countries) leading to the equalization of marginal returns to capital. In reality, this theoretical prediction is not observed, leading to an important paradox in international macroeconomics: the ‘Lucas paradox’. Private capital flows are important in financing development, especially in the context of insufficient and unstable aid, which makes it crucial to understand why the neoclassical theory is not observed. Why does capital not flow to developing countries where their marginal return is higher? Answering this question requires the study of the determinants of private capital flows. For foreign private capital, we consider net flows of FDI, portfolio investments and debts.

    Following the Asian crisis, a number of studies on the determinants of private capital flows emerged. These studies were generally based on an approach that distinguishes between external determinants (exogenous to the economy receiving capital, or ‘push factors’) and internal determinants1 (under the recipient economy’s control, or ‘pull factors’). The analysis of external factors explains how the economic conditions of capital-exporting countries (developed countries) influences capital inflows in developing countries. These external factors reflect the opportunity cost of investment in these countries. The international interest rate and world growth rates, generally approximated by those of the United States, are the most influential factors. Low profit in developed countries is a significant cause of capital flows to developing countries where profits’ prospects can be more promising. One of the first investigations of private capital flows determinants was made by Calvo et al. (1996). Using a sample of 10 Latin American countries over the period 1988–1991, they find that capital flows are mainly influenced by the external factors, namely the growth rate and the interest rate of developed countries. Many authors showed the importance of the external factors (international interest rate and international growth rate) in determining private capital flows (Calvo et al., 1996; Fernandez-Arias, 1996; Montiel&Reinhart, 1999; Kim, 2000; Ying&Kim, 2001; Ferrucci et al., 2004). A larger number of studies revealed the dominant role of internal factors (macroeconomic conditions of the recipient country) in the explanation of private capital inflows (Root & Ahmed, 1979; Schneider & Frey, 1985; Fernandez-Aria, 1996; Ahn et al., 1998; Gastanaga et al., 1998; Asiedu, 2002). Internal factors are the macroeconomic conditions of the recipient country that influence private capital flows to this country. A stable macroeconomic environment is favorable to investment decisions, creation of value added, and productivity. Internal factors include economic growth rate, inflation, trade openness, education, and political stability, which can be influenced by national-level policies. Studies that are more recent use the ‘Lucas paradox’ to explain the determinants of private capital flows.2 Following Lucas, these studies differentiate the determinants of capital flows into economic fundamentals with the ability to affect the production structure (education, institutions, and so forth) and capital market imperfections (mainly informational asymmetry). Alfaro et al. (2006, 2008), through a cross-sectional study, find that the ‘Lucas paradox’ is explained by the quality of institutions, education, inflation, and financial development. According to Reinhart and Rogoff (2004), the ‘Lucas paradox’ exists because of political risk and credit market imperfections. Reinhart and Rogoff (2004) argue that the reduction of credit market imperfections through better institutions would allow externalities, in particular those related to the human capital, to play a more significant role. Recent studies also illustrated the importance of business environment for private capital flows (Martin & Rose-Innes, 2004; Asiedu, 2006; Naudé & Krugell, 2007; Bénassy-Quéré et al., 2007; Kinda, 2010).

    All of these studies lead to different conclusions about the factors which significantly influence private capital inflows to a country. Another crucial element to attracting FDI is building industrial capacity. This includes developing infrastructure and human capital; strengthening institutional capabilities and economic openness; and promoting sound macroeconomic policies (low inflation, strong and sustainable economic growth). The purpose of this study is to extend the ‘Lucas paradox’ approach (which considers only the economic fundamentals3 and capital market imperfections), by integrating external factors from the traditional approach (‘push-pull factors’). Emphasis will be given to physical infrastructure and financial development that have received insufficient attention in the literature (especially for financial development) given the importance of their contribution for countries attractiveness to private capital flows. In addition, the paper controls for non-traditional variables such as capital control and banking crises. We will analyze aggregated private capital flows and their components. Breaking-up aggregate private capital flows allows the differentiation between short-term and long-term flows, which can have some common determinants while other factors are specific to certain flows. Contrary to past studies, this paper takes into account the relationship between different components of private capital and non-linear effects of financial development.

    The rest of the paper is organized into two main sections: the first section analyzes the theoretical relation between private capital flows, infrastructure, and financial development and describes a simple framework based on the ‘Lucas paradox’ approach. The second part of the paper is devoted to the empirical analysis of the determinants of private capital flows followed by robustness checks. The last part concludes.

    1Studies also focus on contagion during episodes of surges in private capital flows between large countries and their smaller neighbors who benefit from externalities resulting from the high attractiveness of the large countries (Calvo et al., 1996; Hernandez et al., 2001). Competition between countries of the same area for better attractiveness to private capital flows could also happen (Kang et al., 2003).   2A very recent approach, applied to emerging countries, consists of the estimation of a model of supply and demand of capital flows. Then, using the maximum likelihood method, this approach estimates the probability of disequilibrium between supply and demand of capital (Mody & Taylor, 2004).   3The economic fundamentals include industrial capacity main determinants.

    2. Infrastructure and Private Capital Flows

    Infrastructure availability is one of the key elements needed to run efficient business. A large number of studies (The World Bank, 1994; Temple, 1999; Demurger, 2001; Willoughby, 2003) highlight the role of infrastructure (telecommunications, electricity, etc) for economic growth and development. Beyond its direct effect on economic growth, infrastructure also affects growth by increasing private investment.4 A better availability of infrastructure increases the output of private investment by reducing transactions costs and enabling firms to get closer to their customers and suppliers, making it possible for the firms to increase their potential markets and thus their opportunities for profit. Well-developed telecommunications infrastructure, for example, can help firms to access financial resources through financial markets. Firms that do not have access to modern telecommunication services, reliable provision of electricity, or developed road systems invest less and have fewer productive investments (regardless of whether they are local or foreign).When the provision of well-functioning infrastructure fails, firms are sometimes forced to pay the costs of providing infrastructure themselves, such as electricity through power generating units, in order to continue their activities. This type of provision is generally more costly than traditional infrastructure provision. In addition to these high costs of provision, firms also support other costs due to damages caused by power outages.

    The determinants of FDI may vary according to their type. FDI in manufacturing, services or in oil, gas and mineral extraction may have different determinants. Moreover, variables such as infrastructure, education or inflation may have different effects depending on the destination of FDI.

    In previous studies, the importance of physical infrastructure in determining the attractiveness of foreign private capital essentially focused on FDI. Loree and Guisinger (1995) find that countries with developed infrastructure (measured by a multidimensional index of infrastructure) receive more FDI from the United States. Wheeler and Mody (1992) and Mody and Srinivasan (1998) find similar results. Kumar (2002), with a sample of 66 countries over 1982–1994, finds that the development of infrastructure, measured by a composite index, has a positive effect on FDI inflows. Ngowi (2001), Asiedu (2002) using a sample of African countries, and Jenkins and Thomas (2002) using a sample of Southern African countries, obtain similar results. The limited resources of the public sector in developing countries, coupled with profitable opportunities in some infrastructure projects (electricity, telecommunications, etc), lead to the provision of infrastructure by the private sector. Given the high cost of infrastructure investments, private corporations carrying out this type of investment are generally foreign. Sader (2000) finds that between 1990 and 1998, 17% of FDI flows received by developing countries were directed to infrastructure projects. According to Ramamurti and Doh (2004), FDI financing infrastructure represents one third of capital inflows to developing countries in the beginning of the 1990s.

    4See Blejer and Khan (1984), Greene and Villanueva (1991), Serven and Solimano (1993).

    3. Financial Development and Private Capital Flows

    Financial development may increase private investments due to better access of firms to capital.5 With the emergence of financial intermediaries, financial development reduces transactions costs through lower informational asymmetry and better risk management and coverage.The reduction of informational asymmetry through financial intermediaries has a considerable effect on foreign capital and investments. In fact, in addition to the informational asymmetry supported by the local entrepreneurs, the distance between foreign investors and local markets generally increases this already existing information asymmetry. Foreign investors know neither the opportunities nor the risks of the local market as well as local investors do. Financial intermediaries can provide information about local market risks, providing more credibility to potential profit in the country. This stimulates the entry of new investors, in particular foreign investors, in the local market. Huang (2006), focuses only on domestic investment, but suggests an empirical model for the importance of financial development on investment. Using a sample of 43 developing countries over 1970–1998, he finds that financial development significantly and positively affects private investment. The author also concludes that private investment has a positive and significant effect on financial development. A developed financial sector also facilitates interactions between foreign and local firms and their suppliers and clients. The importance of financial intermediaries could also vary according to the type of private flows. Indeed, even if financial development significantly explains countries’ attractiveness to FDI and debts, financial intermediaries’ contribution for portfolio investments is more significant. Portfolio investments generally require a relatively developed financial sector and possibly the pre-existence of a stock market. Financial development, itself, can imply the entry of newbanks or newactors in the local market.The process of financial liberalization with bank privatization implies acquisitions in the form of FDI or portfolio investment, increasing of foreign private capital inflows. The importance of financial development for FDI could, however, be reduced with the entry of multinational banks which tend to follow their corporate clients.

    As mentioned by Levine (1997), studies on financial development and investments generally do not distinguish domestic investments from foreign investments. Focusing only on foreign capital, this study enriches the scarce literature on this topic. To the best of our knowledge, very few studies deal specifically with the effect of financial development on private capital flows, precisely FDI. Hausmann and Fernandez-Arias (2000) find that countries with the least developed capital markets tend to have more FDI inflows. According to the authors, FDI can be alternative financing for the firms that do not have access to capital markets. However, using a sample of 81 foreign firms based in Southern African countries, Jenkins and Thomas (2002) show that the better financial sector development in South Africa helps the country to attract relatively more FDI than other African countries. Montiel (2006), in a theoretical analysis, argues that Africa does not attract enough foreign private capital to finance sectors with high potential profits because of Africa’s human capital weakness, lack of infrastructure, and bad institutional quality. Montiel (2006) underlines that when African countries are relativelywell endowed in these factors; financial underdevelopment explains their low attractiveness to foreign capital.

    5See Levine (1997, 2003) for a review of the theoretical and empirical literature.

    4. The Theoretical Framework

    The ‘Lucas paradox’ is derived froma simple neoclassical growth model assuming a common technology to all economies. Let us consider a Cobb-Douglas production function with constant return to scales, representing a small open economy in which the production (Y) is obtained from the combination of capital (K) and labor (L).


    A is the productivity factor and reflects the technological level that can be the stock of human capital (Lucas, 1990). Assuming a common technological level in all economies and perfect capital mobility, capital will flow from most endowed economies (in capital) to the least endowed countries because of the property of diminishing returns.Thatwould lead to a convergence and equality of the interest rates. Considering two economies i and j, the interest rate rt would be defined as follows:


    However, the prediction of interest convergence is not observed, leading to the ‘Lucas paradox’. According to Lucas, this paradox is mainly due to capital market imperfections