4

Finance

Introduction

Global financial flows are an important resource for developing countries. These capital flows augment domestic savings and can contribute to investment, growth, financial sector development, technology transfer, and poverty reduction. These possibilities are reflected in the long-standing view in international economics that capital flows are beneficial in almost all circumstances. This view is backed by evidence suggesting that the growth gains from capital flows are of the same order of magnitude as the growth gains from trade.1 However, there is also evidence that capital flows may entail potential costs that are both larger than in the case of trade and tend to be disproportionately carried by poor people.2 Additionally, it has become clear that not all capital flows are the same in their benefit and cost characteristics. For these reasons, a careful assessment of the impact of capital flows on global poverty does not lend itself to across-the-board statements. Rather, the cost and benefit characteristics of distinct types of capital flows must be considered in some detail.

Capital Flows and Balance of Payments

From a macroeconomic standpoint, capital flows are activities that influence the “capital account” of countries’ balance of payments involving the exchange of assets, whereas trade activities influence the “current account” of the balance of payments where no assets are exchanged. Both of these components of the balance of payments accounts record transactions between each country and the rest of the world. The assets exchanged on the capital account of the balance of payments consist of various financial objects with monetary values that can change over time in the portfolios of both individuals and firms.3

Capital Flows: Classifications

There are a number of legitimate ways to classify capital flows and various subcomponents of the capital account. In this chapter, we will distinguish among four types:

• foreign direct investment
• equity portfolio investment
• bond finance
• commercial bank lending.

Foreign Direct Investment

Foreign direct investment (FDI) is the acquisition of shares by a firm in a foreign-based enterprise that exceeds a threshold of 10 percent, implying managerial participation in the foreign enterprise.4 FDI is one means of effecting services trade, such as when a bank uses a foreign subsidiary to provide financial services abroad, but is also important in the production of merchandise. Under the right conditions, FDI can generate direct and indirect increases in employment, promote competition, improve the training of host-country workers, and transfer technology from developed to developing countries. It may also subject workers to unsafe working conditions, compromise the natural environment, and increase the dominance of foreign culture over host-country cultures.5

Equity Portfolio Investment

Equity portfolio investment is similar to FDI in that it involves the ownership of shares in foreign countries. It differs from FDI in that the share holdings are too small to imply managerial participation over the foreign enterprise. It is thus indirect investment, rather than direct investment. Because equity portfolio investment is undertaken for portfolio reasons rather than managerial reasons, the behavior of investors can be quite different than with FDI. To generalize, equity portfolio investment tends to be motivated by a shorter time horizon than FDI’s horizon and is subject to the portfolio considerations of investors.

Bond Finance

Bond finance or debt issuance is a second kind of portfolio activity. In a bond finance transaction, the government or firms in developing countries issue bonds to foreign investors. These bonds can be issued in either the domestic currency or in foreign currencies, and can involve different kinds of default risks. Bond portfolio investment has in common with equity portfolio investment that both are held along with equities in international portfolios. Portfolio considerations and their relatively short-term characteristics are therefore important to both. Indeed, they are often combined in simple balance of payments accounts under the heading of “portfolio investment.”

Key Terms and Concepts

absorptive capacity

backward links

balance of payments

bond finance

capital account

commercial bank lending

contagion

equity portfolio investment

flight capital

foreign direct investment (FDI)

market failure

microfinance institutions (MFIs)

technology transfer

Commercial Bank Lending

Commercial bank lending is another form of debt. Unlike bond finance, it does not involve a tradable asset.6 Commercial bank loans can be short-term or long-term loans, can be made with fixed or flexible interest rates, and can take the form of inter-bank loans. A single bank or a syndicate of banks can be involved in any particular loan package. As we discuss in this chapter, commercial bank lending can potentially play a role in financial crises.

Impact of Capital Flows

Although FDI can affect poor people directly by generating employment and transferring technology, much of the potential impact of other capital flows on poverty is indirect, taking place through the broad process of financial development. For this reason, before taking up the individual categories of capital flows and their potential impacts on poor people in the rest of this chapter, we consider both the recent trends in capital flows to the developing world and the overall process of financial development.

Recent Trends

To understand the role of capital flows in the economies of developing countries, as well as their implications for poor people, we need to look at their evolution over time. In chapter 2, we noted the volatile nature of debt and equity relative to FDI. Here we distinguish among four types of capital flows: FDI, equity portfolio investment, bond portfolio investment, and commercial bank lending.

Capital Flows to Low-Income Countries

When we consider these flows for the low-income countries (figure 4.1), a number of important observations can be made.

Figure 4.1 Net Private Capital Flows to Low-Income Countries, 1970–2004

Source: World Bank 2006a.

Commercial Bank Lending as Primary Source

First, until the 1990s, commercial bank lending was the primary source of foreign capital for the low-income countries, although it declined significantly after the 1982 debt crisis. On a net basis, bank lending remained positive during the 1990s but then became negative in 1998 after the Asian financial crisis (box 4.1). This reflected the drastic decline in access to commercial bank funds suffered by the low-income countries and their continued payments of old commercial bank debt. This net outflow continued until 2003, when commercial bank lending again became positive, but it reappeared in 2004.

Box 4.1 Financial Crises

Balance of payments, debt, and financial crises have plagued the developing world with detrimental impacts for poor people. One notable crisis of the 1980s was that of Mexico. The decade of the 1980s began with a significant increase in real interest rates and a significant decline in non-oil commodity prices. These increased borrowing costs and reduced export revenues for many developing countries, including Mexico. In August 1982, in the face of capital flight, the Mexican government announced that it would stop servicing its foreign currency debt. Subsequently, both Argentina and Brazil entered into similar debt and balance of payments crises.

Despite the efforts of the International Monetary Fund to effectively address these crises, international commercial banks began to withdraw credit from many of the developing countries of the world, and the debt crisis became global. Within a few years of the outbreak of these crises, the phenomenon of net capital outflows appeared in which the capital account payments of debtor countries exceeded their capital account receipts. Poverty increased substantially, and much of the developing world, particularly Latin America and Africa, entered what came to be known as the lost decade.

Mexico underwent a second crisis in late 1994 and early 1995, and this was soon followed by the “Asian crisis.” Beginning in July 1997, crises struck Thailand, Indonesia, the Republic of Korea, and Malaysia, and in August 1998, a crisis also hit Russia. In each of these cases, sharp depreciations of the currencies resulted. Subsequent crises hit Brazil in 1999 and Argentina in 2001, bringing the crisis process back to Latin America again.

Eichengreen (1999) makes a distinction between “low-tech” debt and balance of payments crises such as those of the 1980s and “high-tech” financial crises of the 1990s. According to Eichengreen, the following features distinguish the latter, more recent crises: 1. Financial firms have significant exposures in real estate and equities; 2. Capital accounts are liberalized to allow firms (including banks) to take on short-term foreign debt, including debt denominated in foreign currencies; 3. Banks are less than fully regulated and supervised as the countries involved liberalize financial markets and capital accounts; and 4. Firms (including banks) do not adequately hedge their foreign exchange exposures, resulting in vulnerable financial positions. Policy makers need to be aware of each of these potential causes to better mitigate the risks poor people suffer when hit by the effects of crises.

Source: Authors.

Increase in Significance of FDI

Second, although FDI comprised a significant portion of total capital inflows to the low-income countries in the 1970s, it became even more significant in the 1990s, far exceeding commercial bank lending. Despite its stagnation after the 1997 Asian crisis, FDI remains the most important source of foreign capital flows for the poorest nations of the world, increasing in the 2001 to 2004 period. As we will see in this chapter, however, these FDI flows are both small relative to the total population of the low-income countries and very unevenly distributed among them.

Unreliability of Portfolio Investment

Third, portfolio investment in the form of bonds and equities has been a significant although somewhat fickle source of resources for the low-income countries. These investments provided substantial positive net flows for a few years in the 1990s, and took on a negative value after the Asian crisis. As noted by Prasad and others (2003), “FDI flows are the least volatile of the different categories of private capital flows to developing countries…. Portfolio flows tend to be far more volatile and prone to abrupt reversals than FDI” (p. 16). This was indeed the case for the low-income countries after the Asian crisis, but their net portfolio equity inflows recovered substantially in 2003 and 2004.

Capital Flows to Middle-Income Countries

Capital flows to the middle-income countries (figure 4.2) have behaved somewhat differently than those to the low-income countries. Again, in the early years, commercial bank lending was the most important source of foreign capital. This type of lending was reduced after the 1982 Mexican crisis, and FDI began to replace it in the late 1980s. As with the low-income countries, despite some stagnation after the 1997 Asian crisis, FDI currently dwarfs all other sources of capital flows to the middle-income countries. Commercial bank lending and portfolio bond and equity flows were important during the 1990s.7 Portfolio bond flows held up better after the Asian crisis than they did in the low-income countries, and these flows recovered substantially beginning in 2003; in the low-income countries equity investment is currently more important. However, on a net basis for the middle-income countries, currently none of these three, nondirect capital flows is large compared with FDI.8

Figure 4.2 Net Private Capital Flows to Middle-Income Countries, 1970–2004

Source: World Bank 2006a.

Reasons for Increase in Capital Flows

What explains the significant increase in capital flows to the developing world beginning in the 1990s? Analysts such as Calvo, Leiderman, and Reinhart (1996) and de la Torre and Schmukler (2004) typically divide explanations into internal or “pull” factors and external or “push” factors. Internal factors included improved relationships between developing countries and their creditors, the pursuit of more sound fiscal and monetary policies, capital account liberalization, and the privatization of state-owned assets.9 External factors included the decline in world interest rates that made assets in developing countries relatively more attractive for portfolio investments, recessions in major developed countries, and a growing integration of world capital markets. This last external factor involved the increased participation of global financial firms in developing-country markets and the increased role of institutional investors. As such, trade in financial services supported expansions in capital flows, with the trade and capital flow dimensions of globalization interacting to support the expansion in private capital flows. All of these factors changed the global capital flow regime in significant ways.

Capital Flows as a Percentage of GDP

It is useful to consider the recent history of total capital flows to the developing countries as a percentage of GDP for both low- and middle-income countries (figure 4.3). Until the 1990s, these ranged between 1 and 2 percent of GDP. During the 1990s, for the reasons just described, these values increased to a maximum of just over 2 percent and 5 percent of GDP for the low- and middle-income countries, respectively. The flows fell precipitously as a percentage of GDP for the middle-income countries after the 1997 Asian crisis, however. Private capital flows are therefore relatively small as a percentage of developing country GDP and somewhat volatile. Over the last three decades, capital flows have increased slightly as a percentage of GDP for the developing world.

Figure 4.3 Net Private Capital Flows to Low- and Middle-Income Countries as a Percentage of GDP, 1970–2004

Source: World Bank 2006a.

Summary of Evolution of Capital Flows

To summarize, before the late 1980s, commercial bank lending was an important source of foreign capital. Since then, however, FDI has become the most important capital flow for both the low- and middle-income countries. To a significant extent, this reflected mergers and acquisitions in response to the privatization that began in the late 1980s. FDI has become somewhat stagnant for both low- and middle-income countries in recent years. However, as noted by Mody (2004), “during a period when portfolio flows boomed and then crashed, FDI remained a resilient form of external finance” (p. 1218). Commercial bank lending and portfolio bond and equity flows experienced something of a renaissance during the early 1990s, but they are not currently significant sources of capital for the developing world. When these flows have been significant, they have been more volatile than FDI flows.10

Financial Development

A basic, theoretical insight in the field of international finance suggests that flows of capital from developed to developing countries can improve welfare for the countries’ populations.11 Developing countries can receive net inflows of capital and invest it at relatively high rates of return, the capital being supplied from developed countries where rates of return are relatively low. This reflects the fact that, at early stages of development, the need for capital is high, while domestic saving is low. As development proceeds, the need for capital slowly declines and domestic saving slowly increases. This theoretical framework is highly idealized, however, and a number of intervening factors inhibit capital flows from developed to developing countries. These include political risk, default risk, differences in levels of human capital and technology, and differences in institutional quality.12 In addition, as first pointed out by Hymer (1976), rate of return analysis is wholly insufficient to explain FDI flows.13

The Developing World as Exporter of Capital

Since 2000, the idealized flow of funds has been turned on its head. Due in large part to a current account deficit in the United States that has exceeded 5 percent of its GDP, and reflecting the high level of savings of central banks in developing countries, the developing world is now an exporter of capital to the developed world rather than an importer. Chinese official reserves alone exceed US$850 billion. For rich countries as a whole, for example, global capital imports exceeded US$300 billion in 2004. The United States actually imported approximately US$650 billion in 2004, with capital exports of Japan, the European Union, and other high-income countries making up the $350 billion difference.14 Most of these developing-country capital exporters are in East Asia and the Middle East. What is worrying about this situation is that U.S. capital imports failed to decline as much as would normally be expected in the 2001–2 recession. This indicates that the United States is structurally rather than cyclically claiming the bulk of the world’s savings, the opposite of what we would desire from a development or poverty-alleviation standpoint. Estimates and forecasts of the Institute of International Finance (2005) indicate that this situation will persist. Addressing fiscal imbalances in the United States is, therefore, of key importance to poverty-alleviating capital flows.

Financial Markets: Global and Domestic Capital Flows

Financial markets, both global and domestic, are an important component of economic development and poverty alleviation. Capital flows can support savings mobilization and deployment, financial sector development, and technology transfer. They also have the potential to manage various types of risk and to monitor the performance of firms’ managers. Empirically, there is some evidence that financial sector development helps to explain economic growth.15 It appears that this growth effect occurs primarily through increasing productivity and reducing external finance costs rather than through increases in savings. As the World Bank (2001) notes, “Rigorous and diverse econometric evidence shows that the contribution of finance to long-term growth is achieved by improving the economy’s total factor productivity, rather than on the rate of capital accumulation” (p. 6).16 The growth effects of financial sector development appear to be important for both banking and equity markets.17

Impact of Global Capital Flows

Global capital flows are only one aspect of financial sector development, and the flows themselves are not always taken into account in growth investigations. There is evidence that not all types of capital flows have the same impact on growth. For example, capital inflows in the form of equity portfolio investment appear to be more beneficial than both bond finance and commercial bank lending.18 More generally, there is widespread agreement that financial globalization via capital flows can make effective financial development more challenging by increasing both benefits and risks.19 General, across-the-board statements about the role of capital flows in financial development are thus difficult to support.

The Importance of Domestic Savings

It is also important to view capital flows in the context of domestic capital mobilization. As we saw in figures 4.1 and 4.2, the most important type of capital inflow into developing countries is FDI. However, on average, FDI is but a small portion of total domestic investment (and savings) in low- and middle-income countries. For these countries in 2002, it was less than 6 percent and 11 percent of total domestic investment, respectively. Therefore, increases in the capital flow dimension of economic globalization have in no way lessened the importance of domestic investment, domestic savings, and the domestic environment influencing them.

Flight Capital

Residents of low- and middle-income countries hold a great deal of their wealth in the form of flight capital, which are assets held abroad because of poor domestic investment opportunities and high domestic risks. Estimates of the magnitude of flight capital vary and, because such flows are not always officially recorded, must be regarded with some caution. They suggest, however, that approximately 40 percent of the private wealth of Africa and the Middle East is held by its residents outside of these regions, and that the figure for Latin America is approximately 10 percent.20 Consequently, any significant improvement in the investment and financial climates of these developing regions could result in a substantial repatriation of resources. Not all resources for development and poverty alleviation need to come from foreign sources—it would be possible for repatriated wealth to provide a substantial portion of a country’s resources.

Financial Market Failures

The financial markets involved in equity portfolio investment, bond finance, and commercial bank lending are characterized by a number of imperfections, which economists call market failures. In normal circumstances, these imperfections contribute to a certain amount of market volatility. Under certain circumstances that are not fully understood, they can lead to full-blown financial crises of the kind experienced in Mexico (1994–5), Asia (1997), Russia (1998), Brazil (1999), and Argentina (2001) (see box 4.1).21 It is important to understand these market imperfections to appreciate the effects of these three types of capital flows on poor people.

Asymmetry of Information in Financial Markets

Financial markets do not operate with full information. By their very nature (which involves the exchange of assets), financial markets have an important intertemporal component, and no market participant possesses perfect information about the future. Consequently, financial markets inherently involve an intertemporal “leap in the dark” of one sort or another. There is often an asymmetry in the information available to borrowers and lenders in which borrowers have more information about their creditworthiness than lenders.22 Such asymmetries in information can lead to market failure in which changes in expectations cause swift changes in behavior, despite the lack of change in fundamental economic conditions. This can be pernicious because lender confidence consequently tends to be “pro-cyclical,” remaining strong in business upturns but suddenly evaporating during downturns of various sorts. Attempts to overcome certain market failures in microfinance to better provide financial services to poor people are described in box 4.2.

Box 4.2 Targeting Poor People: Commercial Microfinance

From the point of view of alleviating poverty, some of the informational imperfections of financial markets make it difficult for commercial financial corporations to assess the creditworthiness of poor borrowers, including poor entrepreneurs. These imperfections or barriers include physical remoteness, the lack of tangible assets to serve as collateral, the lack of property rights, and the cost of contracting. All of these factors tend to exclude poor people from basic financial services.

What are now called microfinance institutions (MFIs) have evolved over the last few decades to fill these gaps in commercial finance. An MFI is an organization that provides financial services of any kind to the poor. At present, they provide such services to poor individuals, including entrepreneurs. MFIs overcome financial market imperfections through group lending practices in which a borrower’s associates become co-signers to the loan. Along with group lending, MFIs use a number of other mechanisms to facilitate effective credit provision. These include creative incentives such as progressive lending, tailored repayment schedules, collateral substitutes, and a focus on women who typically have significantly better repayment rates. In addition, MFIs are beginning to offer noncredit financial services such as savings arrangements to poor people.

The question on the minds of many in the MFI community is whether these institutions will be able to move in large numbers in the direction of commercialization. The steps in this process include an increasingly business-oriented approach, achieving operational and financial self-sufficiency, the increased use of commercial funding sources, and operating as for-profit institutions.

The difficult challenge in making the transition to increased commercialization is not losing sight of the central mission of serving poor populations. There is thus a tension between commercialization and “mission drift.” This has been a central problem for MFIs pursuing commercial status, such as BURO Tangail (BT) in Bangladesh, and Bank Rakyat’s Micro-business Division in Indonesia. Evidence presented in Morduch (1999) suggests that most MFIs will have difficulty in making this transition, but those that do will have contributed significantly to the design of poverty-alleviating finance, no small achievement.

Despite some skepticism, steps have indeed been taken in the direction of commercial microfinance. In 2001, a number of aid organizations and MFIs launched AFRICAP, a commercial MFI facility in Africa. AFRICAP is a for-profit equity investment company incorporated in Mauritius and operating out of Dakar, Senegal. It invests in leading African MFIs and is capitalized at US$13 million. Commercial banks are also beginning to investigate microfinance. For example, in 2004, Deutsche Bank launched a commercial microfinance facility based on its previous experience with MFIs. The facility will serve MFIs in Africa, South Asia, and Latin America using local commercial banks as intermediaries. This lending facility is funded at a level of US$60 million. Time will tell whether these initiatives prove to be successful. Given the importance of the credit constraint for small firms and entrepreneurs, and the central role in job creation and poverty reduction, it is essential that policy makers at the national and global levels focus their attention on developing vibrant microfinance institutions.

Source: Charitonenko and Campion undated; Morduch 1999; and World Bank 2001.

Credit Rating Agencies

Credit rating agencies provide information to potential investors in capital markets. Their ratings are closely scrutinized by investors, and achieving “investment-grade” status is an important milestone for governments, public utilities, and corporate entities seeking to raise money in international markets. Extending the reach of credit rating agencies to include a growing number of developing countries has helped to widen information about these markets.

Credit rating agencies play a vital role in overcoming information asymmetries in financial markets, but the fact that they only partially extend across middle-income countries and are virtually absent from low-income countries means that their limitations need to be recognized. Their ratings also are not predictive in nature: countries and companies that have performed well in ratings have succumbed to crises and failure. Credit rating agencies enhance information flows, and this may even at times exacerbate herding in markets. They perform a vital function, but are no panacea for asymmetry of information in the markets.

Role of Government Failure in Financial Market Failure

The market failure of financial markets is compounded by government failure in that necessary attempts to regulate financial markets can at times make matters worse. Like other financial market participants, governments also suffer from imperfect information, and their attempts to offer support in times of crisis can provide an incentive for excessively risky behavior in financial markets, something that economists call “moral hazard.” Both market failure and government failure characterize markets in equities, bonds, and bank lending, and they can complicate hoped-for effects of poverty alleviation. For example, market failure and government failure in financial markets combine in certain circumstances to generate a behavior known as contagion. This is where problems with regard to one financial instrument or country spread to other financial instruments or countries. The key contributing factor is attempts by market participants to maintain liquidity.

Role of Foreign Short-Term Lending in Financial Market Failure

Imperfections in financial markets appear to be particularly problematic when commercial banks in developing countries are given access to short-term, foreign lending sources.23 The resulting problems have three causes:

• First, systems of financial intermediation in developing countries tend to rely heavily on the banking sector, because other types of financial intermediation are typically underdeveloped.
• Second, developing countries have been encouraged to liberalize domestic financial markets, sometimes before systems of prudential bank regulation and management are put in place.
• Third, developing countries have sometimes prematurely liberalized their capital accounts, on which most of the private capital flows examined in this chapter take place.24

Consequently, and as will be discussed below, care must be taken in managing evolving banking systems and their access to international capital flows.

Impact of Market Failure on Poor People

What is the implication of market failure for poor people? Financial crises are devastating to poor people and should therefore be avoided if at all possible. Poor people are particularly vulnerable to crises because they do not have the savings or safety nets to protect themselves from the income losses that are an inherent part of these events. It is common to consider the costs of crises in terms of the percentage of GDP lost in a particular country or region. For example, Eichengreen (2004) estimates the cost of an average crisis to be approximately 9 percent of GDP for the country in question. In the case of the Asian crisis, Dobson and Hufbauer (2001) estimated the cost to the region at up to 1.5 percent of GDP, but World Bank estimates suggest that it involved 20 million persons falling back into poverty and 1 million children being withdrawn from school.25 Even this estimate may understate the impact. More recent estimates by Suryahadi, Sumarto, and Pritchett (2003), suggest that, in Indonesia alone and at the peak of the increase in poverty, approximately 35 million persons were pushed into absolute poverty. During the Argentine crisis of 2001, close to one-fourth of the population became extremely poor, while one-half of the population fell below the national poverty line.26 Changes in poverty of these magnitudes matter a great deal.27 As recognized by Eichengreen (2004),

To remind oneself of the immediacy of these effects, it is only necessary to observe that Indonesia and Argentina experienced larger falls in output and real incomes than that suffered by the United States in the Great Depression, an event that produced a revolution in social and economic policy. This is another way of saying that the social impact of financial crises can be enormous. (pp. 9–10)

Financial Sector Reforms

Given these potentially severe poverty effects, caution is warranted. Although there is not complete agreement among those who have examined these issues, there is some evidence that the sequence and timing of financial sector reforms can mitigate financial turmoil and, thereby, prevent negative effects on poor people. As mentioned above, the liberalization of financial markets can strengthen the development process in the long run. However, financial liberalization without the proper surveillance capability may destabilize local financial sectors, real economies, and domestic political environments.28 There are many examples where excessive liquidity associated with booms and market overconfidence were followed by excessive pessimism and capital flow reversals. In all of these outcomes, poor people suffer the most. Careful financial sector development should therefore be combined with carefully-targeted safety nets to protect poor people.29

Foreign Direct Investment

Foreign direct investment can have positive effects on poverty by creating employment, improving technology and human capital, and promoting competition. While much of FDI contributes in this way, at times it may adversely affect certain dimensions of poverty through unsafe working conditions and environmental destruction. Nevertheless, if we were to identify the most promising category of capital flows from the point of view of poverty alleviation, FDI would be it.

FDI Recipient Countries

Global flows of FDI are highly concentrated, with the low-income countries being dramatically uninvolved as FDI recipients (figure 4.4). In 2004, for example, the low-income countries accounted for 37 percent of global population, only 9 percent of global GDP, and a mere 3 percent of global FDI. The middle-income countries are much more active as FDI recipients. In 2004, these countries accounted for 47 percent of global population, 36 percent of global GDP, and 31 percent of global FDI. The bulk of global FDI (66 percent in 2004) goes to the high-income countries of the world. As Dobson and Hufbauer (2001) put it, “the vast bulk of FDI represents investment made by one rich country in another rich country” (p. 33).

Figure 4.4 Global Shares of Population, PPP GDP, and FDI, 2004

Source: World Bank, World Development Indicators 2006a.

The lack of involvement on the part of low-income countries as FDI hosts is a major impediment to poverty alleviation. Even the fact that the low-income countries received only 2 percent of FDI in 2004 vastly understates the problem, because this 2 percent of the total flows is highly concentrated among these poor countries. Just two countries—India and Vietnam—receive nearly one-half of this 2 percent, for example. For most of the low-income countries, access to FDI has remained elusive. Growth is essentially driven by domestic investment to an even greater extent than when FDI is present. Foreign aid in support of growth is important to these countries, a fact that we will return to in chapter 5.

Multinational Enterprises

Many developing countries lack access to the technologies available in developed countries. Since multinational enterprises (MNEs) account for approximately three-fourths of worldwide civilian research and development, hosting MNEs from developed countries is, potentially, one important way to gain access to that technology. There are two problems with this idealized scenario, however. First, MNEs will employ the technology that most suits their strategic needs, not necessarily the development needs of host countries. For example, MNEs can employ processes that are much more capital intensive than host-country employment considerations may want.30 Second, there is a strong tendency for MNEs to conduct their research and development in their home bases rather than in host countries.31 Consequently, there are limits to the transfer of new technologies to host countries; as we show below, there is considerable scope for enhancing the benefits of foreign investment.

Potential Benefits of MNEs to Host Country: Technology Transfer

Despite these limitations, there is evidence that, in some important cases, MNEs do transfer technology and establish significant relationships with host-country suppliers by what economists call backward links.Moran (2001), for example, reviewed the evidence in the automotive, computer, and electronics sectors. He summarizes his conclusion as follows:32

Foreign investors whose local operations comprise an integral part of the parent’s global or regional sourcing network introduce state-of-the-art technology and business practices into the host economy both via the investment that the parent makes in the performance of its own subsidiary and via the supervision that the parent and subsidiary exercise over the performance of local suppliers. (p. 24)

Benefits from Local Source Inputs

If the foreign MNE begins to source inputs locally rather than importing them, the host country can gain a number of important benefits:

• Employment can increase because the sourced inputs represent new production.
• Production technologies can be better adapted to local conditions because suppliers are more likely to employ labor-intensive processes.
• The MNE can transfer state-of-the-art business practices and technologies to the local suppliers.
• It is possible that the local suppliers can coalesce into a spatial cluster that supports innovation and upgrading.33

The policies required to support such links are considered in box 4.3.

Box 4.3 Creating Links

The World Trade Organization includes an Agreement on Trade-Related Investment Measures (TRIMs) that prohibits domestic content, trade balancing, foreign exchange balancing, and domestic sales requirements placed on the MNEs hosted by WTO members. Indeed, many international economic policy experts are now calling for policies that would go beyond TRIMs to require the abandonment of all policies that discriminate between domestic and foreign firms. In this changed policy context, how can developing countries use policies to obtain the most benefits from FDI?

New thinking suggests that local content requirements should be replaced by efforts to support local suppliers in their efforts to secure contracts with foreign MNEs. The backward linkage promotion process involves many players, including the government, foreign MNEs, local suppliers, professional organizations, commercial organizations, and academic institutions. The key role of the government is that of coordinator, attempting to bridge the “information gaps” among the players. The government can do this in a number of different ways:

• First, in the realm of information, attempts can be made to provide a matching service between MNEs and local suppliers. This can be done by inviting the relevant players to link-promotion forums.
• Second, in the realm of technology, efforts can be made to provide support in standards formation, materials testing, and patent registration. Providing support in these areas has been part of the function of the Singapore Institute of Standards and Industrial Research. In addition, foreign MNEs can be invited to be involved in programs designed to upgrade local suppliers’ technological capabilities.
• Third, in human resources development, efforts can be made to provide technical training and managerial training.
• Finally, in the area of finance, obstacles to access on the part of small firms can be removed. This has been one of the functions of the Korean Technology Banking Corporation, for example.

Efforts in these and other areas typically must be coordinated by a lead agency. In the cases of Costa Rica, Ireland, and Singapore, the Costa Rican Investment Board, an Irish National Linkage Program, and the Singapore Economic Development Board have played this role. Other developing countries can learn from these experiences.

Source: Battat, Frank, and Shen 1996; UNCTAD 2001; and Reinert 2005.

Taking Japanese MNEs in Thailand as an example of creating links, Moran (2001) reports

As for the impact of foreign investors on local Thai firms, the Japanese assemblers took an active role in organizing “cooperation clubs” of the kind that were characteristic of supplier relations in the home country to assist with quality control, cost reduction, scheduling and delivery, and product improvement. Within the first 10 years after the turn toward offshore sourcing by the Japanese parents, some 150 local firms qualified for original equipment manufacturer (OEM) status…. An additional 200 to 250 Thai firms received replacement equipments manufacturer (REM) certification. These suppliers, like the foreign affiliates themselves, were able to capture economies of scale, and to use different and more sophisticated production techniques, than local firms elsewhere in Asia. (pp. 17–18)

Benefits from Spillover Effects

Another avenue through which MNEs can positively affect host economies is through “spillovers” to other sectors of these economies. For example, FDI in the automotive sector might benefit production in the machine tools sector. Indeed, there is a presumption in much of the literature on FDI that MNEs provide positive spillovers in the form of upgrading technology to domestic firms in the host country. This line of thinking goes back to Caves (1974) who tested this possibility for Canada and Australia. The evidence to date suggests that such spillovers do occur in some circumstances and can be significant.34 However, in the words of Blomström and Sjöholm (1999), they are not “guaranteed, automatic, or free.” For example, Haddad and Harrison (1993) and Kokko, Tansini, and Zejan (1996) failed to find such effects for Morocco and Uruguay, respectively. Aitken and Harrison (1999) also failed to find such positive spillovers for the case of Venezuela. Indeed, their evidence suggests the presence of negative spillovers due to market-stealing effects. Blomström and Sjöholm (1999) find positive spillovers in the case of Indonesia, attributing them to the increased competition that FDI brings. Haskel, Pereira, and Slaughter (2002) found positive spillovers in the United Kingdom, but such evidence might not apply to the developing country context.

Factors that Determine Spillover Effects

What determines whether positive technology spillovers will occur? Many factors are involved, and these include host country policies, MNE behavior, and industry characteristics.

Learning and Education

One key factor is the capacity of local firms to absorb foreign technologies. Blomström and Kokko (2003) suggest that learning is a key capacity that is responsive to various host country policies, and evidence presented in Tsang, Nguyen, and Erramilli (2004) in the case of Vietnam supports this view. Because learning is so important, a lack of human capital in the form of skills and education tends to prevent the generation of positive spillovers. For example, Kokko (1994) found evidence of these learning barriers in Mexico in that spillovers were negatively related to the productivity gaps between MNEs and domestic firms. Additionally, Kokko and Blomström (1995) found evidence that the technology transfers of U.S. MNEs have been positively affected by levels of education in host countries.

Wages

There is some evidence that MNEs offer higher wages than domestic firms. This is the conclusion of te Velde and Morrissey (2003) based on evidence from five African countries. This effect is more predominant for skilled than unskilled workers. In the long run, wages depend on education and training levels, and there is some evidence that MNEs will engage in important training activities. This appears to be more likely when MNEs are large, operate in competitive environments, and are export oriented.35 As with the wage effects, however, training is more likely to be directed toward skilled than toward unskilled workers.36 In a way similar to that of international trade (discussed in chapter 3), then, FDI can have differential effects that are positive for skilled workers but that exclude unskilled workers. This can result in what te Velde (2001) refers to as the “low-income low-skill trap.”

Avoiding the Low-Income, Low-Skill Trap

Basic education and skills development have a major role in making the most of FDI for poverty alleviation.37 Even Singapore, the preeminent example of free market development, has intervened in labor markets to avoid any possible low-income low-skill trap. As described by te Velde (2001),

The Skills Development Fund (SDF) in Singapore is an example of how MNEs (and other firms) can be engaged in more training. The Productivity and Standards Board (PSB), responsible for the SDF, imposes a 1 percent levy on the payroll of employers for every worker earning less than a pre-determined amount. This levy is distributed to firms that send their low-paying employees to approved training courses. This has had a significant impact on skill-upgrading in Singapore. (p. 24)

This is just one example of efforts to make the most of FDI to alleviate poverty. Generalizing such processes throughout the developing world requires new policy efforts along the lines of those outlined in box 4.3 for building a synergistic relationship between skills development and FDI.

Costs of Hosting FDI

Hosting FDI is not without its potential costs.38 Concern has been raised about the practice of transfer pricing. Transfer pricing involves the manipulation of the prices of intrafirm trade by MNEs to reduce their global tax payments. In the case of the United States, with more resources to martial against this practice than any other country, it has been estimated that annual losses in tax revenue are on the order of US$50 billion.39 The solution to the transfer-pricing problem is multifaceted and not straightforward, but it is clear that a multilateral approach is the preferred solution.40 Such options include forming international guidelines and codes of conduct, using international standards of invoicing and customs procedures, harmonizing global tax systems, negotiating and concluding international conventions, and establishing international arbitration procedures. However, to make these options work, resources would need to be provided for many developing countries for them to effectively combat transfer-pricing abuses.

FDI and Poverty Alleviation: An Assessment

FDI is perhaps the most important capital flow from the point of view of poverty alleviation. FDI can be a means of employment generation, especially when it takes place in labor-intensive sectors. It can also be a means of technology and management transfer, especially where effective backward links have been established. This gain, and its potential to help poor people, involves learning processes; these, in turn, require that minimal thresholds of human capital be met. Without advances in education, training, and health, few long-term gains from FDI take place. Advances in the investment climate and the environment for doing business, both for foreign and domestic investors, are also vital. Policy makers have a role to play in facilitating investment by combating corruption, streamlining procedures, and investing in the physical and human capacities that are the foundation for not only foreign but more importantly for domestic investment (See World bank 2005e).

For FDI to have a greater role in poverty reduction, it is also necessary to address the following issues. First, it is highly concentrated among developed countries and just a handful of developing countries. Second, for extractive industries (such as mining and petroleum), steps must be taken to ensure that the FDI does indeed contribute to poverty alleviation, especially in the context of weak governance mechanisms. Third, transfer pricing abuses may rob developing countries of the tax revenues they desperately need to make the very investments required for FDI to contribute positively to poverty alleviation. We return to some of these issues in chapter 8.

Equity Portfolio Investment

Evidence suggests that financial development has a positive impact on growth, but not all types of financial activity and capital inflows have the same effects. In particular, capital inflows in the form of equity portfolio investment might be more beneficial than either bond finance or commercial bank lending. Reisen and Soto (2001) have examined the impact of all four capital inflows considered in this chapter on growth for a sample of 44 countries. They found that FDI did indeed have a positive impact on economic growth. The most positive growth impact, however, came from equity portfolio flows. Bond finance, considered below, did not have any impact on growth; commercial bank lending, also considered below, had a negative impact. These results suggest that equity inflows, along with FDI, could play an especially positive role in growth, development, and poverty alleviation.

Reasons for High Equity Portfolio Impact

Why can equity portfolio investment play a positive role in growth and development, at least under some circumstances? Rousseau and Wachtel (2000) summarize research on this question with four possibilities:

• Equity portfolio inflows are an important source of funds for developing countries.
• The development of equity markets helps to provide an exit mechanism for venture capitalists, and this increases entrepreneurial activity.
• Portfolio inflows assist developing countries to move from short-term finance to longer-term finance. They also help to finance investment in projects that have economies of scale.
• The development of equity markets provides an informational mechanism evaluating the performance of domestic firms and can help provide incentives to managers to perform well.

Some evidence suggests that institutional investors managing equity flows are less likely than banks to engage in herd and contagion behavior, thus making volatility less of an issue.41 However, under some circumstances, these herd and contagion behaviors do indeed appear, especially for nonresident, foreign investors who are at an informational disadvantage.42 The degree of recent volatility in net inward portfolio equity flows to developing countries can be seen in figure 4.5. The Asian crisis of 1997 had a significant, detrimental effect on these inflows, as did the 2001–2 recession (see box 4.1). Recovery began in 2003, supporting significant gains in emerging-market stock indices, and continued through 2004 despite equity market corrections. Estimates and forecasts from the Institute of International Finance (2005) indicate that portfolio equity investment will maintain itself at least through 2005. More long term, according to the World Bank (2004b), prospects for equity finance are positive, although risks remain. As noted by World Bank (2005a), the increases in equity flows during 2003 and 2004 were highly concentrated regionally in East Asia, South Asia, and South Africa. The Latin America and Caribbean region actually experienced a loss.

Figure 4.5 Net Inward Portfolio Equity Flows to Developing Countries, 1995–2004

Source: World Bank 2006a.

Role of Equity Market Prices

The evidence on the role of global integration on equity market prices, as opposed to flows, is more mixed. For example, Asian equity markets recovered values fairly quickly after the 1997 crisis. More to the point, as noted by the World Bank (2001), “There is no clear theoretical presumption as to whether local stock prices will be more or less volatile after integration into the world market. Integration should insulate the prices from shocks that affect the non-market wealth or savings behavior of local investors, but could expose them more to fluctuations in world asset prices and to shifts in external investor preferences” (p. 173). There appears to be no clear evidence that one or the other of these two influences dominates in practice.43

Features of Developing-Country Equity Markets

In general, equity markets are underdeveloped in much of the developing world. For example, nearly the entire net portfolio equity inflows into Sub-Saharan Africa are accounted for by one country alone: South Africa. The World Bank (2004b) summarized the features of developing-country equity markets as follows:

Market capitalization as a share of GDP in low-income countries is about one-sixth of that in high-income countries. Even in the middle-income countries, the share is only about one-third of that in industrial countries. Stock exchanges in developing countries also tend to lag technologically behind developed markets. Technology plays a major role in the trading, clearance, and settlement processes; problems in those areas can discourage sophisticated investors. Institutions that supervise and support the operation of the stock exchange also tend to be weaker in developing countries. (p. 95)

The example of the Nairobi Stock Exchange is taken up in box 4.4. Elsewhere in Africa, the Johannesburg stock market has a capitalization that far exceeds that of the rest of Africa. In many respects, it has become a model for emerging markets, but it too has suffered from the migration of some of the most global listings to London and New York.

Box 4.4 The Nairobi Stock Exchange

Although equity exchange in Kenya has a history going back to the 1920s under British colonial rule, the Nairobi Stock Exchange (NSE) came into being only in 1954. From its inception until 1991, it was a voluntary association registered under the Kenyan Societies Act. At about the same time, the NSE came under the regulation of the Capital Markets Authority (CMA). In 1995, exchange controls that limited foreign participation in the NSE were removed.

As with most other African stock exchanges, the NSE has some characteristics of a frontier market. Its market capitalization peaked at approximately US$2 billion in 1998, and its value traded peaked at only approximately US$100 million in 1997. The number of companies listed in 2002 was only 50, down from 57 a decade earlier. That said, there are institutional changes that have strengthened the exchange, such as the implementation of an electronic central depository system, the adoption of international accounting standards, and the establishment of a Capital Markets Appeals Tribunal. All of these are valuable changes, but the exchange has some distance to travel in fulfilling development promises.

Source: Ngugi, Murinde, and Green 2002; Nairobi Stock Exchange, Market Intelligence (Kenya) accessible at http://www.mi.co.ke/economy_and_markets/markets/nse/rise_growth_of_nse.asp.

Development of Equity Markets

The development of equity markets in low- and middle-income countries is more complex than it might first appear. This is because of the recent trends in the globalization of financial services. Observers have pointed to a set of domestic factors as particularly important in equity market development. These factors include sound macroeconomic policies, minimal degrees of technology, legal systems that protect shareholders, and open financial markets. However, as pointed out by Claessens, Klingebiel, and Schmukler (2002), these are precisely the factors that tend to promote the “migration” of equity exchange out of developing countries to the major exchanges in financial capitals of developed countries. This migration process complicates standard notions of equity market development. Steil (2001) has argued that the way forward is to link local markets with global markets. However, medium-size firms with local information needs might still benefit from some kind of domestic equity market. This is an area that urgently requires the development of novel approaches.

The Promise of Portfolio Equity for Poverty Alleviation

Capital flows in the form of portfolio equity hold out some promise for poverty alleviation. Along with FDI, these indirect equity investments appear to have a positive impact on growth through a variety of mechanisms. Contagion and herd behavior are less prevalent than they are with commercial bank lending, and flows are expected to hold relatively steady in the near future. However, the underdeveloped nature of equity markets in most developing countries and some degree of market volatility are two obstacles associated with global equity flows. Developing equity financing to ameliorate these obstacles is a long-term priority for poor people, as these indirect investments can help to offset some of the problems with other sources of flows. This development needs to proceed in an open fashion that does not favor a narrow, investing elite with inside knowledge, but rather offers an open system for all medium-size and large firms.

Debt: Bond Finance and Commercial Bank Lending

In the financial world, there are significant differences between portfolio equity investment and debt. This shows up in the fact that, in bankruptcy, debt is given priority over equity. This tends to support the preference for debt over equity in markets, a preference that might well be misplaced from the perspective of development and poverty alleviation. In this section, we consider two types of debt: bond finance and commercial bank lending. The main difference between these two forms of debt is that bonds are in the form of tradable assets. This provides more flexibility to lenders than bank lending.

Bond Finance

Net debt flows to the developing world have evolved markedly in recent years (figure 4.6). Bond finance fell gradually between 1998 and 2001, and the 2003 to 2004 period showed substantial recovery in net flows to a level above the 1997 to 1998 period. That said, the World Bank (2004b, 2005a) identifies a number of risks with regard to bond investment. These include interest rate increases in high-income countries, which are already beginning at the time of this writing and continued fiscal imbalances in some developed countries such as the United States.44 The World Bank (2005a) warns that “The risk of an abrupt increase in U.S. interest rates remains a serious concern. Large, sudden movements in long-term rates, in particular, could provoke a sharp widening of emerging market bond spreads” (p. 21). This potential increase in bond spreads could raise the cost of capital via bond finance for the developing world. At the time of writing, however, these spreads are at a historical low, reflecting the strength of emerging markets’ macroeconomic performance, as well as the low yields offered by alternative investment classes. The heightened appetite for emerging markets has seen a growing number of developing countries accessing significant volumes of finance from the bond markets. Traditional issuers, such as Mexico, have been able to extend the term of their bond issuance up to 30 years. At the same time, corporate borrowing has grown significantly and in 2004 accounted for about half of total emerging market bond issuance. The long-dated sovereign bonds better match the assets and liabilities of governments and build the benchmark yield curve (which plot the yields of different maturities) Increased issuance in local currencies–such as Panda (Chinese Renminbi), Peso (Mexico), Real (Brazil), and Rand (South Africa) bonds–reduces the currency mismatch, which has been a major cause of past financial instability. Meanwhile, greater corporate access has provided a new and often lower cost of finance for emerging businesses. As the need for sovereign borrowing by middle-income countries has declined—not least in those Asian economies with large foreign exchange surpluses, corporate borrowing has provided a welcome alternative for investors seeking relatively high yields. Longer-dated offerings by their governments have enabled corporate issues in certain emerging markets to price their own longer-dated offerings by reference to these government benchmarks. The challenge for policy makers is to sustain and deepen this access and to broaden this access from the handful of countries with investment-grade that has been able to benefit from this widening access to capital markets. Enthusiasm for this asset class, however, has in the past proved short-lived. Prudence and a balanced portfolio approach thus remain necessary.

Figure 4.6 Net Inward Debt Flows to Developing Countries, 1997–2004

Source: World Bank 2006a.

Commercial Bank Lending

Commercial bank lending fell precipitously in 1999 on a net basis and did not recover until 2004 (figure 4.6). Commercial bank lending on a gross basis is a different story (figure 4.7). As noted by the World Bank (2005a), despite declines in net lending, commercial bank lending is still used by a wide variety of developing countries: “Twice as many countries tapped this segment of the debt markets in 2004 than the bond financing segment” (p. 20). The private or corporate sector is emerging as an increasingly important borrower in this regard.

Figure 4.7 Gross Inward Bank Lending, 1996–2004

Source: World Bank 2005a, 2006b.

Above, we briefly analyzed some features of financial markets that give them some important, “imperfect” characteristics. As mentioned, commercial bank lending appears to be particularly prone to these imperfections. For example, Dobson and Hufbauer (2001) note that “Bank lending may be more prone to run than portfolio capital, because banks themselves are highly leveraged, and they are relying on the borrower’s balance sheet to ensure repayment” (p. 47). The World Bank (2001) also notes that “Incentives are key to limiting undue risk-taking and fraudulent behavior in the management and supervision of financial intermediaries—especially banks that are prone to costly failure” (p. 3).

Before the Asian crisis, such an assessment might have been seen as exaggerated. Indeed, the precrisis data examined by Sarno and Taylor (1999) provided a relatively sanguine conclusion about commercial bank lending. Subsequent events, however, showed otherwise. Much of the debt involved in the Asian crisis was composed of short-term, interbank loans, and, as we saw in figures 4.1 and 4.2, net commercial bank lending flows quickly became negative for both the low- and middle-income countries after 1997. Indeed, Goldstein, Kaminsky, and Reinhart (2000) include short-term capital flows (most of which are interbank commercial lending) as a significant predictor of future financial crises based on a broad sample of countries.

Supporting the Development of Banking Sectors

What can be done to support the safe development of banking sectors in developing countries? Some of the necessary steps can be thought of in terms of information, institutions, and incentives. For information, it is important for banks to embrace internationally sanctioned accounting and auditing procedures and to make the results of these assessments available to the public. For institutions or the rules of the “banking game,” risk management practices (both credit and currency) must be sufficiently stringent and prudential regulation systems must be well developed. For currency risk, the World Bank (2004b) notes that “particular care should be taken to ensure that foreign-currency liabilities are appropriately hedged” (p. 30).45 These information and institutional safeguards are no small task, and they inevitably cannot be achieved in the short term.46 Consequently, these safeguards should be buttressed with incentive measures in the form of market-friendly taxes on banking capital inflows. For example, Eichengreen (1999) argues that “banks borrowing abroad should be required to put up additional noninterest-bearing reserves with the central bank” (p. 117). Such taxes on short-term capital inflows have been applied by Chile and others to prevent destabilizing episodes of overborrowing.47

Debt Flows Compared with Equity Flows

Debt flows in the form of bond finance and commercial bank lending appear to have different properties than equity flows in the form of FDI and portfolio equity investment. They are more prone to the imperfect behaviors that characterize financial markets and their positive growth effects do not seem to be as large as those associated with equity flows. Consequently, debt finance must be used cautiously and should be hedged against exchange rate risks.48

Summary

The different forms of capital flows are best seen as complements, not substitutes. The capital flows with the greatest potential contribution to poverty alleviation are both FDI and equity investment. Equity-related finance brings with it the natural benefits of risk-sharing, and is far less subject to the sudden stops and reversals of debt flows. In the case of FDI, this is because investors have a tendency to reinvest a portion of retained earnings. Also, FDI capital stock depreciates, and new inflows are needed to sustain the existing capital stock. Finally, the benefits of FDI go beyond those relating to narrow financial issues: new ideas, technologies, and improvements in skills and training are all potential and important spillovers.49

Long-Term Trend of Financial Development

The long-term trend of financial development is probably toward a mix of all four capital flows described in this chapter. The positions of developing countries with regard to capital flows can be generalized as in figure 4.8. Here, for simplicity, our four types of capital flows are represented as the four corners of a diamond, with the relative strength of any particular flow indicated by proximity to a corner of the diamond. In the short term, most developing countries have no choice about their position in the diamond because they are constrained by the availability and cost of different capital flows. In the medium term, however, their actions can yield some influence over the availability and cost of capital flows: their choices can expand.

Figure 4.8 Composition of Financial Development

Source: Authors.

For example, there is a group of low-income countries who find themselves at approximately point “1” in the diagram, relying primarily on the commercial bank lending form of capital flows.50 Another set of low-income countries find themselves at approximately point “2,” with a mix of commercial bank lending and FDI, the latter probably concentrated in petroleum or minerals. Many middle-income countries find themselves at approximately point “3,” with the addition of some bond finance and equity portfolio investment. As financial development proceeds, there will be a move to somewhere in the vicinity of point “4” in which there is a broad mix of all four capital flows. Maintaining this position in a stable way would require that the financial development of the country be designed to mitigate the imperfections discussed in this chapter. Capital inflows have a vital role to play. As noted by Fernandez-Arias and Montiel (1996), “the possibility that capital inflows may be welfare reducing does not mean that they are invariably harmful” (p. 57, emphasis added).

Absorptive Capacity Requirements

If there is any convergence in the emerging literature on capital flows in the developing world, it concerns absorptive capacity, which acts as a set of necessary conditions for potential poverty-alleviating effects. That is, for capital flows to positively help poor people, a number of things must be true:

• First, human capital must be developed. Without it, the hoped-for positive spillovers from FDI will not emerge.
• Second, the domestic financial markets must be “deep” enough to support liquidity. Without liquidity, volatility will be a problem.
• Third, systems of oversight and regulation of domestic financial markets must be developed enough to prevent excessive volatility and crises.
• Fourth, levels of corruption should be low and strenuously combatted.51

Each of these conditions takes some time to fulfill. Until they are achieved, caution is warranted. Thus, we sound a note of warning in providing an overall assessment of global capital flows and their relationship to poverty. As Hanson, Honohan, and Majnoni (2003) concluded, “the globalization of finance is not an unmixed blessing, but it appears to be inexorable” (p. 25). Capital flows are an “inexorable” aspect of financial globalization, which have potential benefits and costs that are significant. Because poor people are particularly vulnerable to the potential costs, any errors in managing capital flows should be on the side of caution.

Impact of Fiscal Imbalance in the United States on Poverty Alleviation

Finally, as discussed in the beginning of this chapter, in significant measure because of unaddressed fiscal imbalances in the United States, aggregate capital flows are currently from developing to developed regions of the world. The United States continues to claim the bulk of world savings, the opposite of what is optimal from a poverty-alleviation perspective. Given the globalization of finance—particularly in the market for government bonds—addressing fiscal imbalances in the United States is of key importance to freeing up capital flows that alleviate poverty.

Notes

1. Dobson and Hufbauer (2001) estimate that developing country GDPs are, on average, approximately 5 percent higher due to both trade and capital flows. They themselves call this a “cautious assertion” to be taken as a first approximation. Prasad and others (2003) are even more cautious, noting that “if financial integration has a positive effect on growth, there is as yet no clear and robust empirical proof that the effect is quantitatively significant” (p. 5). For a general review, see chapter 4 of Hossain and Chowdhury (1998).

2. The latter point regarding cost and poor people is emphasized by Stiglitz and Bhattacharya (2000).

3. In the case of firms, it is not only financial portfolios that are involved, but also productive portfolios such as plant and equipment in the form of foreign direct investment.

4. The precise definition in the World Bank’s World Development Indicators, for example, is “lasting interest in or effective managerial control over an enterprise in another country” (World Bank, various years).

5. Szirmai (2005), for example, notes the “overwhelming impact of Western culture, which is frequently transferred only in the rudimentary form of consumption-oriented behavior and technology” (p. 509).

6. We discuss noncommercial lending, such as that of the World Bank, in chapter 5 on foreign aid.

7. Chuhan, Claessens, and Mamingi (1998) examine some factors behind bond and equity flows to Latin America and Asia during the late 1980s and early 1990s. They suggest that equity flows tend to respond to price-earnings ratios and relative rates of return, while bond flows tend to respond to credit ratings and debt prices in secondary markets.

8. One potential point of confusion can arise in interpreting figures 4.1 and 4.2. As we will see in the next section on financial development, beginning in 2000, the developing world became an exporter of capital to the developed world rather than an importer. How then can the positive inflows of capital into the low- and middle-income countries take place? The answer is that the private capital inflows of figures 4.1 and 4.2 are offset by official capital outflows that take place through the actions of central banks, especially in the case of the middle-income countries.

9. de la Torre and Schmukler (2004) note that privatization proceeds in developing countries increased from US$ 2.6 billion in 1988 to US$ 25.4 billion in 1996.

10. On this last point, see also Osei, Morrissey, and Lensink (2002).

11. See, for example, Taylor and Williamson (1994).

12. See, for example, Lucas (1990).

13. This insight is reviewed at some length in chapters 1 and 2 of Caves (1996).

14. See World Bank (2004b, 2005a). The World Bank (2005a) notes that “The buildup of foreign exchange reserves in the hands of developing countries’ central banks and monetary authorities—and its use in financing global payment imbalances—marks a new phase in the postwar system for financing international payments” (p. 59).

15. This possibility is discussed in Pagano (1993). Evidence is provided by King and Levine (1993).

16. On productivity effects, see Beck, Levine, and Loayza (2000). This research supports what is known as the “Schumpetarian view” of the role of finance in development (after Schumpeter, 1934) in which the primary effects of financial development are on the productivity of firms. On external finance effects, see Rajan and Zingales (1998).

17. See Levine and Zervos (1998).

18. See Reisen and Soto (2001). The overall growth effects of FDI had been called into question by Caves (1996), who wrote: “Some researchers have tried to identify the overall effects of MNEs’ presence in developing countries on … subsequent rates of economic growth. The possible causal connections are numerous but speculative and ill-defined in terms of economic models. Empirical investigations, whether by those disposed to think good or ill of the MNEs, have employed inadequate research procedures and have yielded no trustworthy conclusions” (p. 242). This ambiguity is reiterated in the case of Africa by Bhinda and others (1999).

19. For example, the World Bank (2001) notes that “Globalization … challenges the whole design of the financial sector, potentially replacing domestic with international providers of some of these services, and limiting the role that government can play—while making their remaining tasks that much more difficult” (p. 1). See also Calvo, Leiderman, and Reinhart (1996).

20. These estimates are from Collier, Hoeffler, and Pattillo (2001).

21. Reviews of the economic literature on crises can be found in more- and less-technical terms, respectively, in Appendix B of Eichengreen (1999) and chapter 17 of Reinert (2005). For a specific focus on the role of exchange rate regimes in crises, see Cordon (2002).

22. See, for example, Stiglitz and Weiss (1981) and Williamson (1987). For the role of asymmetric information in influencing institutional investors, see Frenkel and Menkhoff (2004).

23. International Monetary Fund statistics record 64 banking crises between 1970 and 1999 (Crook 2003). Crook writes that “breakdowns in banking lie at the center of most financial crises. And banks are unusually effective at spreading financial distress, once it starts, from one place to another” (p. 11). The World Bank (2001) notes that “If finance is fragile, banking is the most fragile part” (p. 11).

24. For a critique of premature capital account liberalization, see Stiglitz (2000). As the World Bank (2001) notes, “Poor sequencing of financial liberalization in a poor country environment has undoubtedly contributed to bank insolvency” (p. 89). Hanson, Honohan, and Majnoni (2003) also note that “the riskiness of capital account liberalization without fiscal adjustment…, and without reasonably strong financial regulation and supervision and a sound domestic financial system, is well recognized” (p. 10). See also Bhattacharya and Miller (1999).

25. See Wolfensohn (1998).

26. See Fiszbein, Giovagnoli, and Thurston (2003).

27. Even short of actual crises, there is some reason for concern about the potentially negative impact of capital flows on poor people. For people at or near poverty lines, any volatility in consumption can be quite detrimental or even disastrous. From a theoretical point of view, capital flows can reduce the volatility of consumption by de-linking it from national output volatility. Unfortunately, there is empirical evidence that increased financial integration through expanding capital flows can increase rather than decrease consumption volatility in developing countries. See Prasad and others (2003) and Kose, Prasad, and Terrones (2003).

28. These risks are all the more significant when countries are characterized by “currency mismatches” in which assets are denominated in the local currency and liabilities in foreign currencies. Consequently, net worth is directly tied to the value of the local currency. On this issue, see Jeanne (2000), Eichengreen, Hausmann, and Panizza (2003), and Goldstein and Turner (2004). As noted by Eichengreen (2004), “Currency mismatches are widely implicated in financial crises in developing countries” (p. 27).

29. Stiglitz and Bhattacharya (2000) discuss the role of food subsidies, education subsidies, rural infrastructure, and microfinance in this regard.

30. There is a long-standing inquiry into this issue, the results of which are summarized by Caves (1996): “Survey evidence indicates that MNEs do some adapting (of technologies to labor-abundant conditions), but not a great deal, and it appears that the costs of adaptation commonly are high relative to the benefits expected by individual companies” (p. 241).

31. “With the exception of some European-based companies, the proportion of R&D activity by MNEs undertaken outside their home countries is generally quite small and, in the case of Japanese firms, negligible” (Dunning, 1993, p. 301).

32. See also Moran (1998).

33. For an introduction to the role of clusters in technological upgrading, see chapter 11 of Reinert (2005) and the references therein. For the role of clusters in natural resource–based development, see Ramos (1998).

34. Evidence presented by Hejazi and Safarian (1999) indicate that spillovers for research and development are more important for FDI than they are for international trade.

35. See te Velde (2001) and references therein.

36. See Tan and Batra (1995), for example.

37. For readers familiar with “growth and poverty” research, we note that Borensztein, De Gregorio, and Lee (1998) find that it is the combination of FDI and education that has a statistically significant impact on growth.

38. For a review of both benefits and costs of hosting FDI, see chapter 21 of Reinert (2005).

39. See Plender (2004).

40. Unilateral policy options exist, but “because there is competition for MNE activity between home and host countries, and between different host countries, the opportunities for MNEs to play one nation against another are enhanced without the establishment of supra-national institutions and harmonized inter-governmental action towards (transfer pricing)” (Dunning, 1993, p. 523).

41. This evidence is reviewed in chapter 1 of Dobson and Hufbauer (2001).

42. For the case of Korea, see Choe, Kho, and Stulz (1999) and Kim and Wei (2002).

43. See also Hanson, Honohan, and Majnoni (2003).

44. As of January 2005, the Argentine government had made a “final offer” to its bondholders. By the end of February 2005, the country had negotiated a debt swap, which replaced a portion of this debt with lower-value bonds. The Argentine government hopes that this restructuring will begin to bring home the approximately US$150 billion its citizens hold abroad in the form of flight capital. See The Economist (2005a, b).

45. Mistakes made in these areas have proved to be too costly to poor people for countries to relax their vigilance. Prasad and others (2003) conclude that “The relative importance of different sources of financing for domestic investment, as proxied by the following three variables, has been shown to be positively associated with the incidence and the severity of currency and financial crises: the ratio of bank borrowing or other debt relative to foreign direct investment; the shortness of the term structure of external debt; and the share of external debt denominated in foreign currencies” (p. 49).

46. Eichengreen (1999) notes that “the sad truth in all too many countries is that banks have a limited capacity to manage risk and that regulators have limited capacity to supervise their actions” (pp. 11–12).

47. This overborrowing is described by McKinnon and Pill (1997).

48. We take up the accumulated debt burdens of developing countries, as well as the issue of debt relief, in chapter 5.

49. China, for example, has focused on equity rather than debt, and inroads into poverty reduction have been significant.

50. This group of countries, as well as those at point “2,” would no doubt also be relying on foreign aid, but we leave this discussion to chapter 5.

51. According to the results of Wei (2000), corruption has a tendency to bias capital flows away from FDI and toward commercial bank lending.