Rethinking Foreign Investment for Sustainable Development

Lessons from Latin America

Wimbledon Publishing Company

Contents

FDI AND SUSTAINABLE DEVELOPMENT IN THE AMERICAS

Kevin P. Gallagher, Daniel Chudnovsky and Roberto Porzecanski

Since the early 1980s nations in Latin America have been implementing a cluster of deep reforms to their economies. Referred to in the United States as the Washington Consensus and in Latin America as "neoliberalism," the reforms include a package of economic policies intended to promote economic development by opening national economies to global market forces. Over the last twenty-five years, governments throughout Latin America have reduced tariffs and other protectionist measures, eliminated barriers to foreign investment, restored "fiscal discipline" by reducing government spending and promoted the export sector of the economy (Williamson 1990).

Now, after 25 years of free-market reforms, many citizens in the hemisphere — and some governments — are questioning the wisdom of the Washington Consensus. Indeed, between October 2005 and December 2006, sixteen Latin American nations held either presidential or congressional elections. Nearly all of these contests have been referred to as referendum on the reforms. In many of the region's most significant economies — Argentina, Bolivia, Brazil, Chile, Uruguay, and Venezuela, candidates critical of the Washington Consensus prevailed. In other nations the outcome of the vote was so close that right-leaning governments at the very least have no mandate to deepen existing reforms.

This sea change in Latin American democracy has been portrayed in the Western press as an irrational resurgence by protectionists. However, a closer look at the record of the Washington Consensus shows that the concerns of citizens and governments can be justified. Indeed, the region has not experienced the economic growth that was promised would come as a result of the reforms. Economic growth has occurred at an annual rate of less than two percent between 1980 and 2005, compared to a rate of 5.5 percent between 1960 and 1980. Growth was faster during the 1990s than in the 1980s, but it still did not compare to the period previous to the reforms. Chile is the one exception, where growth rates almost doubled over the past twenty years compared to the 1960 to 1980 period. In addition, a debate has arisen over the extent to which Chile deviated significantly from Washington Consensus policies to achieve that growth.

The promise, among others, of following these policies is that FDI by multinational corporations will flow to your country and be a source of dynamic growth. Beyond boosting income and employment, the hope was that manufacturing FDI would bring knowledge spillovers that would build the skill and technological capacities of local firms, catalyzing broad-based economic growth, and environmental spillovers that would mitigate the domestic ecological impacts of industrial transformation.

This book evaluates the extent to which FDI fostered sustainable development in the Americas. Drawing on case studies from across the region — Argentina, Brazil, Bolivia, Chile, Costa Rica, Ecuador, Mexico, Uruguay, and Venezuela — the authors in this volume specifically look at how foreign investment during the reform period has affected economic growth, environmental policy and performance, and the countries' political economy. The authors have each authored numerous studies on the performance of FDI in their countries and region. Their chapters in this book synthesize that work and the work of others, most often for the first time in the English language. Hence, these chapters should not be seen as original research studies, but as synthesizing assessments of the situation in specific regions, written by leading in-country experts in the field. By and large, and consistent with the broader literature on the subject, the authors in this volume find that investment regime liberalization-led FDI has been a limited success at best in the Latin American case (for an exhaustive review of the literature, see Gallagher and Porzecanski (2007). In summary, we find that:

1. FDI was concentrated in a smaller handful of countries in the region.

2. FDI was attracted by traditional determinants, not whether a nation has a regional or bi-lateral trade and/or investment treaty (RBTIA) or if it can serve as a pollution haven for foreign firms.

3. When FDI did come, foreign firms tend to have higher levels of productivity and higher wages than were likely to increase trade in the region.

4. FDI fell far short of generating "spillovers" and backward linkages that help countries develop, and in many cases wiped out locally competing firms thereby "crowding out" domestic investment.

5. The environmental performance of foreign firms was mixed, in some cases leading to upgrading of environmental performance, and in others performing the same or worse than domestic counterparts.

It should be said up front that although this volume is highly critical of the performance of FDI under the Washington Consensus, the findings here should not be in any way be interpreted as recommending that FDI is not beneficial for sustainable development. Indeed, for economies to develop in a sustainable manner all forms of investment are crucial. The findings in this volume suggest that LAC (Latin America and the Caribbean) is simply not rising to the challenge to make FDI work for sustainable development. In order to rise to this challenge, the authors in this volume suggest three lessons for LAC and other developing countries seeking to successfully place FDI as part of a comprehensive development strategy:

1. FDI is not an end, but a means to sustainable development. A liberalization-led strategy will not automatically attract FDI nor generate economic growth in an environmentally sustainable manner when FDI does come.

2. FDI policy needs to be conducted in parallel with significant and targeted domestic policies that upgrade the capabilities of national firms and provide a benchmark of environmental protection.

3. International agreements, whether at the World Trade Organization (WTO) or at the level of RBTIAs, need to leave developing nations the "policy space" to pursue the domestic policies necessary to foster development through FDI.

This chapter is organized as follows: The first section provides an overview of FDI trends in LAC and the literature that examines the determinants of FDI in the region under the Washington Consensus. Section 2 outlines the theoretical framework regarding FDI and sustainable development and briefly reviews the literature on that subject. Section 3 consists of an outline of the volume as a whole and of an overview of the major findings. Finally, Section 4 suggests some policy recommendations to rectify the problems identified in the volume and suggests avenues for future research in this area.

FDI Trends and Determinants

LAC has been one of the largest recipients of FDI since the early 1990s. This section provides an overview of those trends and discusses the literature on the determinants of that FDI. By and large, the key factors, among others, that attracted FDI to LAC during the period were the ability to get market access to LAC markets, the ability to serve as an export platform to other markets, political and macroeconomic stability. Interestingly the literature finds no relationship between whether a nation has signed a regional or bilateral trade or investment agreement (RBTIA) with an investing nation. What's more, there is no evidence that suggests that LAC is a "pollution haven," where foreign firms move to the region to avoid relatively stronger environmental standards in the developed world.

There is no question that the region experienced an unprecedented amount of FDI since the reform period began. For some countries it has been truly impressive. Figure 1.1 exhibits annual FDI flows to LAC from 1980 to 2006.

The 1990s was a period of unprecedented increases in the level of FDI in the world economy as a whole, reaching $1.6 trillion in the year 2000. However, the lion's share of that investment — 70 percent of all FDI — stayed in developed countries. Of the FDI that did accrue to the developing world during the 1990s, almost 80 percent of it flowed to just 10 countries. Five of those countries (Brazil, Mexico, Argentina, Bermuda, and Chile) are in LAC. Even among the ten countries that benefited most heavily from FDI, the distribution was skewed; China, Brazil and Mexico received 58 percent of all FDI that flowed to the developing world in the 1990s (UNCTAD 2002).

Although the LAC region received a great deal of FDI, these flows were highly concentrated in just a handful of countries. Table 1.1 lists the top 31 nations that received the highest amount of annual FDI flows during the period 1990 to 2006. Brazil, Mexico, Argentina, Chile and Colombia top the list. Indeed, these nations received 81 percent of all FDI in the region. Investment in the top fifteen countries accounts for almost 97 percent of all FDI.

What factors led to the upsurge in FDI into the region (and the lack of flows in some countries)? The vast majority of studies on the determinants of FDI in LAC are econometric in nature. In other words, using FDI flows (or FDI/per capita) as a dependent variable, analysts statistically examine the extent to which other factors independently affect the level of such flows. There is unanimity among these studies that large and growing economies with low levels of inflation and debt (i.e., macroeconomic stability) are the key determinants of FDI in the region. There is also a consensus that weak environmental standards (in and of themselves) do not significantly attract FDI in the region. The jury is still out on the question of whether new treaties for trade and investment have independently led to attracting FDI. The following is an exhaustive guide the literature on these subjects.

In most cases, these studies have taken the form of cross-sectional analyses of total FDI flows for different groups of Latin American countries in the 1990s (and in some cases, longer periods). While each study has different model specifications, as they attempt to explain the relationship between FDI and one additional determinant, they generally share several core control variables. These studies find FDI to be positively and significantly correlated with the market size of the receiving economy, and negatively but significantly correlated with the level of inflation and/or the level of external debt in the receiving country. Both variables are generally used to proxy macroeconomic stability (Nunnenkamp 2000; Arbelaez, Daniels, et al. 2002; Chudnovsky, Lopez, et al. 2002; Bengoa Calvo and Sanchez-Robles 2003; Bittencourt and Domingo 2004; Tuman and Emmert 2004; Gallagher and Birch 2005). These variables are found to be particularly important in the case of "market seeking" FDI — that is, FDI aimed at exploiting the domestic market (Chudnovsky and Lopez 2000). Agosin and Machado note, "in spite of the talk about the internationalization of production and the increasing global market orientation of MNEs [MNCs], looking at the broad picture, the size of domestic markets still seems to matter most to foreign investors" (Agosin and Machado 2006). In addition, the existence of resources (either natural of human) has also been generally found to be an important determinant of investment, particularly in the cases where FDI is "resource seeking" or "export oriented" (Chudnovsky and Lopez 2000).

In part because of the region's history and risk of expropriation (as mentioned above), it has been argued that the adoption of RBTIAs — which protect investors from, among other things, expropriation — should therefore be expected to promote foreign direct investment. In recent years, a literature has developed that attempts to test for the existence of a causal link between the adoption of investment treaties and an increase in the inflows of FDI. While most studies look at the issue at the developing country level, a recent study has focused on the case of Latin America. After conducting a cross-sectional data analysis for 133 countries between 1993 and 1995, UNCTAD finds that the impact of bilateral investment treaties (BITs) on FDI is small and secondary to the effects of other determinants, especially market size. UNCTAD's finding is shared by Hallward-Dreimeier, who looks at data from 20 OECD countries flowing to 31 developing countries from 1980 to 2000 (though unfortunately, it is unclear how many Latin American countries are included in the sample). Work by Tobin and Rose-Ackerman, who examine FDI for 63 countries (20 of which are from Latin America) from 1975 to 2000, also supports this conclusion (UNCTAD 1998; Hallward-Dreimeier 2003; Tobin and Rose-Ackerman 2004).

Some studies, however, have found a positive association between the adoption of BITs and foreign direct investment flows. Neumayer and Spess look at 119 developing countries (29 of which are in Latin America) between 1970 and 2001. They use as an independent variable the number of BITs a developing country has signed with OECD countries, weighted by the world share of outward FDI flow that the OECD country accounts for. They find that developing countries that sign more BITs with developed countries receive more FDI inflows (Neumayer and Spess 2005). This conclusion is shared by Egger and Pfaffermayr, who looking at the issue from the supply side; the authors examine a sample of 19 high-income-source countries and more than 50 host countries, 8 of which are from Latin America, and find that BITs exert a positive and significant effect on real stocks of outward FDI, with a lower bound of 15 percent (Egger and Pfaffermayr 2004). Finally, Salacuse and Sullivan look at 33 developing countries, eight of which are from Latin America, and find that the presence of a BIT with the United States has a large, positive and significant association with a country's overall FDI inflows. However, they find that this is not the case for BITs with other OECD countries (which have a weak positive, but not statistically significant, effect), nor for BITs with other developing countries (which have weak negative, but not statistically significant, effect) (Salacuse and Sullivan 2005). As pointed out above, however, none of these studies focuses exclusively on Latin America.

The only study to have undertaken a region-specific study of this kind is that of Gallagher and Birch, who find very limited evidence that BITs in general attract additional FDI and strong evidence that an investment agreement with the United States will not lead to additional FDI (Gallagher and Birch 2005). This second finding is consistent with the conclusions of Tobin and Rose-Ackerman, but contradictory to that of Salacuse and Sullivan, although both studies conduct non-region specific analyses (Tobin and Rose-Ackerman 2004; Salacuse and Sullivan 2005).

There are also a significant number of studies that have focused on the impact that processes of regional integration might have had on FDI flows to Latin America. In a study conducted by panel data analysis that looks at bilateral FDI flows (not exclusively for Latin America, but with the aim of assessing the possible FDI impact of the FTAA), Daude, Levy Yeyati et al. argue that sharing membership in a regional integration agreement with a source country increases the likelihood of receiving FDI from that country by 27 percent, although these gains are very unlikely to be distributed evenly between members (Daude, Levy Yeyati et al. 2003). Aguilar and Vallejo, on the other hand, disagree with this finding, arguing that the FDI effects of a preferential trade agreement are ambiguous and depend on which effect (investment creation, investment diversion) prevails. They argue that the results obtained by Daude, Levy Yeyati et al. may be a consequence of the fact that these authors fail in their regressions to control for institutional and infrastructure quality. This critique is shared by Chudnovsky and Lopez and Bittecourt and Domingo (Chudnovsky and Lopez 2001; Aguilar and Vallejo 2002; Chudnovsky, Lopez et al. 2002; Bittencourt and Domingo 2004).