email icon Email this citation


Mixed Blessing: Foreign Capital Flows and Democracy In Emerging Markets

Leslie Elliott Armijo

Northeastern University

International Studies Association

March 1998

Foreign Capital Flows and Democracy In Emerging Markets 1

Ever tighter and more rapid cross-border financial links gird the globe. High-powered international finance reaches deeply into the same developing countries in which basic local phone service works only sporadically. Many of these countries, moreover, are newly democratic or democratizing, with the attendant explosion in citizen demands of governments for improved jobs, educational opportunities, and better lives. Is the result- large and often volatile foreign capital inflows into democratizing "emerging market" countries- a fortuitous coincidence of need and supply? Or is it, instead, a perversion of justified popular hopes for accountable government, as public policies become skewed toward the orthodox macroeconomic policies global investors are well known to favor? Furthermore, do large foreign capital inflows, other things being equal, tend to promote or inhibit democratic transitions and the consolidation of electoral norms?

The forms of cross-border capital flows have altered in recent decades. In the 1950s through the mid-1960s, foreign aid provided more than half of all capital flows between advanced industrial and developing countries. In 1965, for example, foreign aid constituted 64 percent of net resource flows to developing countries (McCulloch and Petri 1994). In the 1970s the share of medium and long-term bank loans increased dramatically, supplying about half of net resource flows by the end of the decade. Meanwhile, between 1970 to 1980, total net inflows to developing countries almost doubled, to just under 4 percent of their combined economies (see Table 1). 2 Long-term bank lending disappeared abruptly in 1982, in response to Mexico's near default on its external debt. By 1988, foreign aid was again the largest single category of net resource flows (43 percent), followed by direct investment (25 percent), and medium and long-term bank loans and trade credit (14 percent). 3 Total flows plummeted to a little over half of their pre-debt crisis high. Moreover, net transfers (net resource flows minus interest payments on past debt) from advanced industrial to developing countries as a group in the 1980s were approximately nil. 4

Table 1 1 Net Resource Flows as Percent of GNP
Country Year Foreign Aid FDI Med/Lng
Debt to Govt
Med/Lng
Debt to Priv.
Portfolio Flows
to Govt.
Portfolio Flows
to Priv.
Total
Brazil 1970 0.39 1.20 1.27 2.00 0.00 0.00 4.86

1980 0.37 0.83 1.39 0.10 0.15 0.00 2.83

1988 -0.10 0.94 1.08 -0.15 0.03 0.08 1.87

1990 -0.29 0.21 -0.10 -0.03 0.03 0.02 -0.16

1992 -0.32 0.56 -0.05 0.87 0.14 2.21 3.41

1994 -0.40 0.57 0.02 0.34 -0.04 2.53 3.01
Mexico 1970 0.39 0.85 0.45 0.16 -0.04 0.00 1.82

1980 0.36 1.15 2.24 0.90 0.06 0.00 4.71

1988 0.59 1.56 0.82 -1.64 -0.68 1.25 1.91

1990 2.19 1.07 1.12 0.72 0.21 3.41 8.72

1992 0.02 1.35 -0.06 0.14 -1.06 3.28 3.67

1994 -0.49 2.19 -0.13 -0.38 0.86 3.41 5.46
China 1970 0.00 0.00 0.00 0.00 0.00 0.00 0.00

1980 0.10 0.00 0.83 0.00 0.02 0.00 0.96

1988 0.51 1.04 1.52 0.00 0.25 0.19 3.51

1990 0.42 0.98 1.32 0.00 -0.01 -0.04 2.67

1992 0.64 2.67 2.14 0.00 -0.05 0.46 5.86

1994 0.60 6.48 1.15 0.00 0.55 1.11 9.89
India 1970 0.96 0.08 -0.02 0.00 0.00 0.00 1.02

1980 1.49 0.05 0.34 0.11 0.00 0.00 1.99

1988 0.76 0.03 1.05 -0.04 0.24 0.26 2.31

1990 1.29 0.05 0.54 -0.03 0.10 0.38 2.33

1992 1.44 0.06 0.72 -0.02 -0.08 -0.18 1.93

1994 0.14 0.21 0.20 -0.15 -0.13 2.27 2.55
Indonesia 1970 4.91 0.86 0.30 1.38 0.00 0.00 7.44

1980 1.22 0.24 1.02 0.00 0.05 0.00 2.54

1988 3.59 0.68 -0.89 1.15 -0.19 0.43 4.77

1990 2.30 1.00 -1.62 3.36 -0.09 3.29 8.24

1992 2.37 1.34 -0.15 2.15 -0.07 3.09 8.74

1994 1.00 1.26 0.07 0.61 0.00 1.96 4.89
Korea 1970 2.30 0.73 1.09 0.28 0.00 0.00 4.39

1980 2.03 0.01 2.02 0.79 0.07 0.00 4.93

1988 -0.86 0.49 -0.70 -0.13 -0.29 0.27 -1.23

1990 0.14 0.36 0.25 -0.44 -0.11 0.88 1.07

1992 0.01 0.24 0.15 0.23 0.44 1.63 2.70

1994 -0.21 0.25 0.09 0.11 0.46 2.09 2.78
Thailand 1970 0.39 0.61 0.10 0.87 0.00 0.00 1.97

1980 2.01 0.59 1.72 2.11 0.14 0.00 6.58

1988 -0.88 1.82 0.39 0.29 0.07 4.32 6.01

1990 -0.10 2.89 -1.01 3.19 -0.10 3.14 8.02

1992 -0.12 1.93 0.20 1.06 0.26 2.28 5.62

1994 0.33 0.45 -0.10 0.29 0.43 8.53 9.93
Vietnam 1970 - - - - - - -

1980 - - - - - - -

1988 - - - - - - -

1991 -0.22 0.33 0.52 0.00 0.00 -0.67 -0.03

1992 2.71 0.24 1.42 0.00 0.00 4.50 8.87

1994 5.87 0.64 -0.71 0.00 0.00 0.85 6.66
Russia 1970 - - - - - - -

1980 - - - - - - -

1988 - - - - - - -

1990 0.87 0.00 0.66 0.00 0.05 -1.71 -0.13

1992 1.21 0.14 2.03 0.00 0.00 0.19 3.57

1994 1.00 0.27 -0.16 0.00 -0.01 -0.06 1.04
All
LDCs
1970 0.95 0.42 0.33 0.31 0.00 0.00 2.02

1980 1.60 0.21 1.47 0.40 0.11 0.00 3.78

1988 0.92 0.53 0.39 -0.08 0.05 0.31 2.13

1990 1.27 0.55 0.08 0.19 0.05 0.47 2.61

1992 1.20 1.00 0.31 0.25 0.10 1.32 4.19

1994 0.98 1 57 0.08 0.14 0.29 1.50 4.56

Source: Computed from World Bank, World Debt Tables, 1994-95 & 1996

Notes:

1. Foreign Aid equals sum of grants, multilateral loans, bilateral loans, and IMF loans
2. Debt to Govt. equals sum of medium/long term guar. debt from foreign commercial banks and other sources
3. Debt to Private Sector equals medium/long term non-guaranteed debt from foreign commercial banks
4. Portfolio flows to Govt' equals guaranteed bonds
5. Portfolio Flows to Private Sector equals sum of non-guaranteed bonds, portfolio equity, and short term debt
6. Note that Total Net Resource Flows, unlike World Debt Tables concept, includes IMF credit and short-term debt.
7. For 1970, 1980 and 1988, where non-guaranteed debt is not broken out into Bonds and Commercial Banks, the latter are assumed.

Meanwhile, international financial markets evolved dramatically in the 1980s. Jointly known as "financial globalization" these changes meant that more money, both absolutely and as a share of world economic output, was more mobile across borders than ever before. The total stock of financial assets traded in global capital markets, the so-called "eurocurrency assets" rose from about $5 trillion in 1980 to $35 trillion in 1992. In 1994 the McKinsey Global Institute projected they would reach almost $83 trillion by 2000, about three times the combined gross domestic product of the advanced industrial countries (Woodall 1995, p. 10). Some commentators argued that, if measured properly, the total stock of outstanding global financial assets was no more overwhelming, as compared to either global output or world trade, than during the turn of the century epoch of British financial hegemony and the gold standard (Bradsher 1995). However, the volatility of world capital markets in the last decade of the twentieth century unquestionably has had no precedent. 5 In 1973 daily foreign exchange trading was about $10 to $20 billion; by 1992, it was $900 billion; and by mid-1995 it had reached around $1.3 trillion of "hot money" backstopped only by the combined foreign currency reserves of government's of the advanced industrial countries, an apparently inadequate $640 billion. Similarly, the ratio of global trade to foreign exchange transactions was about 1:10 in 1982, but as much as 1:60 a decade later (Woodall 1995, p. 10).

What did these changes mean for developing countries? The largest share of globally mobile capital, of course, flowed among advanced industrial countries, and the largest single destination was the United States, which by mid-1996 owed around $800 billion abroad (Prestowitz 1996). Nonetheless, after the hiatus of most of the 1980s, large flows again began to go to developing countries. Beginning in the late 1980s, Mexico and other erstwhile pariahs suddenly found themselves again receiving net private voluntary capital inflows again- but this time the forms had altered dramatically. In 1980, portfolio capital flows (defined here to include portfolio equity, bonds, and short-term debt) were only 3 percent of flows to developing countries, but by 1988 they were 17 percent, and by 1994, had become the largest single category of flows, with 39 percent. Foreign direct investment, also enjoying a revival, comprised 34 percent (Table 1). Total net flows, meanwhile, had shot up to 4.6 percent of developing countries' GDP. The particular profiles of the "new" cross-border capital flows continues to be subject to change: although the figures are not available yet, the consequence of the East Asian financial crises of 1997 will almost certainly be reduced capital flows to emerging markets. What is unlikely to go away is the much larger share, as compared to the other post World War Two decades, of the new (or old, dating back to the late nineteenth and early twentieth century) form of cross-border investments: portfolio flows.

The risks of portfolio capital flows were illustrated by the dramatic events of Mexico's "peso crisis" in December 1994, when close to a billion dollars exited the Mexican economy in a single day. As had happened with the 1982 debt crisis, in early 1995 the home governments of Mexico's major private creditors, in the interests of preventing a global financial crisis, stepped in with a rescue plan. One consequence of the peso crisis was that the rich countries had to begin thinking about which countries they would bail out should they be threatened with a financial crisis, and whom they would let crash. The rapid spread of Thailand's July 1997 crisis throughout East Asia has been a wake-up call about the vulnerability of interlinked financial markets. Although the potential economic costs of these new, highly liquid, capital flows have been widely discussed (Calvo, Leiderman, and Reinhart 1993; Fernandez-Arias 1995; French-Davis and Griffith-Jones 1995; Folkerts-Landau and Ito 1995; Nunnenkampt and Gundlach 1996; Rojas-Suarez and Weisbrod 1995), 6 the political implications for emerging market countries have been less debated.

This essay tries to think through the implications of these recent shifts in the institutional form of international financial flows for another global trend of the 1980s and 1990s: the turn from authoritarian to democratic rule in developing and post-Communist countries around the world. Is the relative decline of official development assistance, for example, on the whole "good" or "bad" for new democracies like South Africa or the Czech Republic? Why? The second issue of "financial globalization" the recent increase in net resource flows as a share of the total economies of developing countries, and the potential for further increases as financial investments become ever more mobile across national borders, I leave for another discussion.

The essay's first major section sets out definitions, first of "democracy" then of six ideal types of international capital flows. 7 The second section deductively considers the likely impacts of each of the six broad types of financial instruments on four intermediate variables: (1) the rate of economic growth, (2) the fortunes of four players in the national game of politics in most emerging market countries (foreign governments, the host country government, foreign business, and local big business), (3) the risk of a balance of payments crisis arising sometime in the future, and (4) pressure for neoliberal economic reforms. Section three then links the four intermediate variables with some possible consequences for democratic development. Section four briefly identifies some illustrative country cases, and section five concludes.

Variables Defined: Democracy And International Financial Flows

The effects of foreign capital inflows on democracy in emerging market countries are what I hope to explain or predict. This section briefly defines the dependent variable, democracy, and distinguishes a hitherto relatively underexamined source of variation in the independent dimension, foreign capital flows.

Numerous definitions of democracy each have their defenders. The most used, and most achievable, is procedural political democracy. Here democracy is defined by wide access (all adult citizens have the vote, with no restrictions on either citizenship or voting imposed by property ownership, race or ethnicity, literacy, or other demographic or economic criteria) and specified, universalistic procedures, including freedom of speech and organization, multiple political parties and candidates, secret ballots, fixed terms of office, and limited authority of elected officials, who themselves are subject to the law of the land. Robert Dahl (1971) aptly suggested defining degrees of democracy along two dimensions: "participation" or breadth of inclusion of the population in the franchise and the group of those with potential to lead, and "contestation" or degree to which elections offer voters genuine alternatives between viable candidates with differing public policy preferences. Another crucial component of procedural political democracy is a guarantee of basic civil and civic rights, including freedom of expression, association, and religion, equal protection under the law, the right to be charged with a specific crime if arrested and to have an impartial trial within a reasonable time period, and the right to retain one's life and property, except under carefully specified circumstances, as in, for example, the military draft or the government's limited right to seize property for a compelling public purpose under the rule of "eminent domain." Finally, it must be true that in a procedural political democracy the elected leaders, assisted often by both appointed advisors and career civil servants, control the major public policy decisions in society; civilian leaders, that is, are not mere figureheads for military authoritarian rulers who exercise the real power. 8

It should be noted that what the above definition lacks is any limits or comment whatever on the economic conditions that must or should obtain in a "democracy." It presumes that neither minimums of economic security nor economic equality are either necessary or sufficient for democratic government. Implicitly, the definition assumes that the one person, one vote rule, combined with the legal imperative of equality before the law, are sufficient to ensure justice in society. Procedural political democracy thus explicitly is a minimalist definition of democracy, requiring only that major conflicts among persons and groups are settled peaceably and through the political process, in theory open to all. All of the advanced capitalist countries, and increasing numbers of developing and postcommunist countries, meet these minimum requirements, at least most of the time. Procedural political democracy thus seems to be an achievable goal. Furthermore, it also serves as an essential, necessary although not sufficient, component of most contemporary definitions of economic democracy. 9 This essay focuses the bulk of its attention on exploring the possible consequences of foreign capital flows for procedural political democracy.

At this point a further dimension must be introduced. Thus far, democracy has been considered only as a static condition. Yet for the vast majority of emerging market countries, the crucial question is not how certain conditions might affect a long-standing and well-established democracy, but rather how they might shape a new or fragile democracy, or influence the possible transition of an existing authoritarian regime in the direction of becoming a future democracy. Reasonably stable and enduring democracies in developing countries, such as those in India, Jamaica, or Costa Rica, have been few. It should be important to review briefly some of what is known about conditions for successful transitions to democracy.

Three observations about transitions to democracy are relevant. First, it is harder for a country to become democratic than to maintain a preexisting democracy. Above all, procedural political democracy requires the consent of all essential political actors- defined as those individuals and groups that can exercise effective veto power over the outcome of democratic decisionmaking- to abide by the outcomes of elections, parliamentary debates and lawmaking, and like democratic procedures. Adam Przeworski (1991) usefully has formalized these conditions by noting that the rational calculations made by all relevant political actors must suggest to them that the payoffs expected from playing the democratic game, over the medium run, will be greater than those to be anticipated from subverting democracy, even though, and by definition, most players will not achieve their most preferred policy outcomes most of the time. Democracy is about institutionalized uncertainty and continuous compromise; each player's rational calculation must lead to the conclusion that mutual compromise is superior to no-holds-barred conflict, which holds out the possibility of total victory, but also of total defeat (see also O'Donnell and Schmitter 1986). Once democracy has functioned for awhile, people typically began to attach a positive normative significance to it, and also to see it as the "normal" state of affairs. After these points have passed, maintaining stable democracy becomes much easier.

Second, heightened economic insecurity usually is not auspicious for a successful democratic transition (Haggard and Kaufman 1995). It is true that conditions of economic crisis can erode support for authoritarian incumbents. However, in and of itself, there is no reason to believe that a crisis will favor democratic successors over another authoritarian government with a change of personnel. Furthermore, feelings of personal vulnerability are likely to make many key political players more defensive, and thus less willing to compromise, that is, less willing to settle for the second best solutions inherent in the democratic process, than they would be if economic conditions were more settled or promising. Political liberalization that occurs in conjunction with significant economic liberalization, as has happened in Latin America, Eastern Europe, and elsewhere in the 1980s and 1990s, also poses particular problems beyond those associated with making a political transition separately (Armijo, Biersteker, and Lowenthal 1994). Since new forms of foreign capital inflows to developing countries often are associated with a larger program of market-oriented economic reforms, these problems of transitional incompatibility between political and economic reform can make democratization significantly more difficult. For example, even scholars who believe that economic liberalization ultimately will improve both overall economic growth and increased economic opportunities for all sectors, including the poorest, recognize that the short-term effects of market reforms are likely to involve increased unemployment, structural dislocation, and, quite often, heightened income inequality, whether among quintiles of income earners or diverse geographic regions. Increased insecurity and/or inequality, even if policymakers plausibly expect it to be temporary, easily can provoke societal responses from groups that perceive themselves as losers, ranging from diffuse anti-system radicalism and even violence, to focused xenophobia, nationalist chauvinism, exaggerated protectionism, and religious fundamentalism. The ideas of a Patrick Buchanan or a Jean-Marie Le Pen pose little to threat to an established democracy as in the U.S. or France. It was harder to be quite so confident about, for example, the election of a Vladimir Zhirinovsky or Gennadi Zyuganov in Russia in mid-1996, both of whom attacked the incumbent, Boris Yeltsin, for being excessively neoliberal and insufficiently Russian nationalist.

Third, the process of democratization may be conceived of as typically including three stages (O'Donnell and Schmitter 1986; Przeworksi 1991). The first, political liberalization, signifies a loosening of overt authoritarian controls on the exercise of basic civil and civic liberties, such as freedom of expression, association, worship, and travel, and perhaps greater tolerance of criticism by the regime, but stops short of formalizing full democratic procedures. The second stage, formal transition to democracy, occurs when a country adopts new laws and procedures marking an official, legal transition to democratic rules of the game. The third stage is that of democratic consolidation, which implies internalization and normative acceptance of the new democratic procedures by all of the major political actors. One of the problems of some democratic transitions, not surprisingly, is that they get "stuck." Brazil, for example, spent at least a decade, more or less from the mid-1970s to the mid-1980s, moving from political liberalization to the formal transition to democratic rules. Russia in mid-1997 clearly was somewhere between stages two and three, but probably closer to the former than the later. Transitional political leaders in the Philippines and South Africa, Corazon Aquino and Nelson Mandela respectively, have tried hard to lodge their polities firmly in stage three before the leaders had to leave office. These leaders recognized the overriding importance of creating a shared normative commitment to the politics of institutionalized compromise.

The above definition of democracy as procedural political democracy is reasonably well-accepted. That is, the analytical concept is clear, although cross-national measurement is more tricky. The other component to be defined is foreign capital flows. Inquiries as to the effects of more or less foreign capital, or its presence or absence in an economy, are fairly standard in the political economy literature. By contrast, the premise of this essay and this book is that the institutional form of cross-border financial flows makes a difference for political and policy outcomes in the capital-recipient emerging market country- that is, that the form or quality of the capital flow may be as important in some cases as its sheer quantity. I define six types of capital flows based on the answers to three questions: Is the source of the capital in the advanced industrial country the public or the private sector? Is the recipient of the foreign capital in the emerging market country the public or the private sector? Third and finally, how intrinsically volatile is the financial instrument employed? Table 2 provides a summary description of these six ideal types of international capital flows.

Foreign aid flows from the foreign public to the recipient country public sector and has comparatively low volatility. The category includes both grants and low-interest loans offered by developed country governments- either bilaterally or through a multilateral financial institution such as the World Bank or Inter-American Development Bank- to less developed countries, almost always directly to their governments. 10 The numbers reported in the book's introduction consider credit from the International Monetary Fund to be multilateral foreign aid. Foreign direct investment (FDI) flows from the foreign private sector to the private sector within the emerging market country, in which the multinational corporation becomes, de facto, a local player. FDI also has low volatility. Through FDI multinational corporations set up new businesses or purchase existing firms located in the recipient country. The defining feature of direct investment is that the foreign owner assumes a long-term managerial commitment to the business in which he or she invests. Direct investment need not always mean majority ownership, which developing countries frequently have prohibited; foreign control can be exercised via a plurality of shares. Purchase by foreigners of a dominant interest in a privatized state-owned enterprise, for example, thus constitutes direct investment. The investor is a private corporation based in an advanced industrial country. The in-country recipient of funds is the subsidiary or affiliate of a multinational corporation (MNC), and thus also a private sector entity. 11

Commercial bank loans to the incumbent government come from the foreign private to the emerging market country public sector, whether directly to the central bank or finance ministry, or to state-owned enterprises, ostensibly operated at some remove from national budgetary accounts, but nonetheless ultimately responsible to the political authorities. 12 Medium and long-term loans should have medium volatility. Commercial bank loans to local big businesses originate with the foreign private sector and are spent by the local private sector.

Table 2. International Investment Instruments: Ideal Types
Financial Instrument Investor In-Country Recipient Of Funds Volatility
Foreign Aid Public sector Private sector Low
Foreign Direct Investment Private sector Public sector Low
Bank Loans To Government Private sector Public sector Medium
Bank Loans To Private Firms Private sector Private sector Medium
Portfolio Investments W/ Government Private sector Public sector High
Portfolio Investments In Private Firms Private sector Private sector. High

In category four, loan recipients are creditworthy large private firms with a high enough international profile to borrow long-term funds directly from multinational banks. Typically, although not invariably, these firms are exporters, thus providing some assurance to the lender that they will have access to the foreign exchange needed to repay the loans.

Portfolio investments in securities of the incumbent government move funds from foreign private investors 13 to the recipient country public sector, as with foreign investment in treasury bonds of the emerging market goverment or minority shareholding in public sector firms. These securities may be sold directly in international financial markets or to foreign investors who buy them in the capital markets of the capital-importing country. Their defining characteristic is that the funds raised become the responsibility of the emerging market country's government. The sixth and final category, portfolio investments in securities of local businesses, refers to the transfer of resources from the foreign private to the local private sector. Securities in this group include both those traded only within the emerging market country and those floated directly in global markets, as with corporate equities or debentures of private firms of developing countries sold on European exchanges in the form of global depository receipts (GDRs) or traded in the U.S. through American depository receipts (ADRs). I include short-term debt with portfolio investments in the local private sector. 14 Portfolio investments with either the public or private sector have the highest potential volatility. 15

The six ideal types of cross-border capital flows to developing countries just specified have been constructed based on three types of distinctions: the identity of foreign investors (public or private sector), of in-country borrowers (public or private sector), and the hypothesized volatility of the financial instrument or modality (low, medium, or high). 16 In recent years there have been substantial shifts in aggregate flows on all three dimensions. From 1950 through the mid-1960s, foreign aid flows, originating in the public sector of the advanced industrial countries, were substantially larger than flows originating with private investors. By 1970, as shown in Figure 1, public and private sources of foreign funds were of approximately equal importance. In 1980, which represents the profile of the 1973-81 peak years of multinational commercial bank lending, the foreign private sector contributed about 25 percent more funds than did foreign aid. Following the 1982 debt crisis, the share of bilateral and multilateral foreign aid again rose, as shown in the data for 1987. However, by the 1992 private flows of all types dwarfed official flows, a trend that continued through this writing in late 1997. Figure 2 shows an equally pronounced shift in the recipient sector of capital flows, as the share of net flows coming to the in-country public sector dropped toward the end of the 1980s, while that of the in-country private sector rose. Figure 3's trends in predicted volatility reflect, first, the sharp decline in medium and long-term commercial bank lending, second, the recent sharp growth in portfolio flows, now approaching the magnitude of official loans and grants, and, third, recent a recovery of foreign direct investment in the early 1990s. As compared to earlier postwar decades, that is, trends in the 1990s revealed dramatic movement in the direction of private sector to private sector flows, at least a third of which was in a highly liquid form.

The argument thus far has specified an outcome, procedural political democracy, which I suggest may be affected by the composition of foreign capital flows from advanced industrial countries to emerging markets. I assume for purposes of argument that the sheer magnitude of capital inflows among different hypothetical cases does not differ greatly. Rather, the institutional form of its transfer is what shifts. It is difficult or impossible to suggest logic that would propose a direct link between, for example, foreign direct investment and the strengthening or weakening of democracy in a developing country. The essay's next section, instead, looks at four intermediate dimensions to which the institutional form of cross-border capital flows may have a direct link.

Direct Consequences Of Foreign Capital Flows: Likely effects on growth, political resources, financial crises, and pressure for neoliberal policies

Deductive reasoning suggests likely differences among the six types of foreign capital flows in their effects on four important intermediate outcomes- economic growth, the balance of power among nationally relevant political actors, the risk of provoking a balance of payments crisis, and externally imposed pressure for the adoption of neoliberal public policies- each of which, in turn, plausibly is related to democracy. That is, this essay constructs a two stage argument: first, from alternative instruments of foreign capital flows to the four intermediate variables just listed, and second, from each of these intermediate variables to supportive or inauspicious conditions for democracy. Table 3 summarizes the first set of hypothesized links.

The expected relationship between various cross-border financial instruments and economic growth turns on the point that the in-country recipients of foreign resources may be either the government or private firms. From a purely economic viewpoint, foreign resources that come directly to governments probably are less efficient in directly producing growth. There is no reason to doubt the conclusions of the extensive literature of the public choice school, which argues that rational public sector bureaucrats and managers, as compared to private owners, have fewer incentives to maximize profits and thus efficient operations, but many more incentives and opportunities to reap monopoly rents at the general population's expense (See Bates 1988; Bates and Krueger 1993; Meier 1991). Examples include awarding public sector procurement contracts not to the lowest bidder but rather for a consideration, charging illicit fees for government services that ostensibly are free, and so on. In addition, politicians and the bureaucrats who work for them may choose to allocate a portion of the investment resources that come under their control to alleviate social tensions such as those caused by interregional inequities within the country. That virtually all governments are at least at the margin influenced by political considerations in the allocation of public investment funds suggests that public sector disbursement of international resources is likely to be on average less efficient than private sector investment.

It is, of course, reasonable to counter that the private sector, though likely to run more microeconomically efficient firms, is much less well equipped than the central government to solve collective action problems (see Olsen 1965). To the extent that society's economic problems are dominated by such typical collective action challenges as environmental degradation, or underinvestment in future generations and in infrastructure, resource allocation by the central government may well be more macroeconomically efficient in the long run. Making this judgment about a particular society is an empirical task, and cannot be done on the general, analytical plane on which this essay operates. Possibly different countries have different likelihoods of encountering collective action problems.

Nonetheless, for our present purposes, the point is that resources that come directly to the central government tend to exhibit certain characteristics: in brief, the incumbents will employ resource allocation in ways that reflect their public policy priorities, which in turn will be both developmental and directly political. Some expenditures will be directed toward furthering economic growth. Other expenditures, however, are likely to go for other goals, from enhancing equity (arguably laudable, but in the short run less growth-enhancing than a single-minded focus on profits) to political patronage. On the other hand, private sector investors will tend to maximize profits, period. Thus, foreign aid, bank loans to the government, and portfolio investment in government securities all might be expected to be comparatively less efficacious with respect to yielding overall economic growth than foreign direct investment or bank loans or portfolio investments coming to the local non-financial private sector. The judgments in column 1 of Table 3 reflect these suppositions.

Table 3. International Investment Instruments: Direct Implications
Financial Instrument Likely Incremental Stimulus To Economic Growth Political Actor Whose Influence Increases (Assuming no BOP crisis.) Increased Risk Of Balance Of Payments (Bop) Crisis Pressure On Govt. For Neoliberal Economic Reforms (Even if no BOB crises
Foreign Aid Low
  • Foreign government.
  • Incumbent government.
Low Low.
Foreign Direct Investment High.
  • Foreign business.
  • Incumbent government.
Low. Moderate.
Bank Loans To Government Low.
  • Incumbent government.
Moderate. Low.
Bank Loans To Private Sector High.
  • Local big business
  • Incumbent government.
Moderate. Low.
Portfolio Investment With Government Low.
  • Incumbent government.
High. High.
Portfolio Investment In Private Firms High
  • Local big business
  • Incumbant Government
High. High.

The second reasonably direct link is that between the institutional form of foreign investments and the balance of resources and influence among nationally relevant political actors. A definition may be helpful. National politics may be conceived of as a political "game" in which various "players" (social groups and occasionally individuals) jockey for position. Each group's goal is to use whatever politically relevant resources it has available (which may be guns, money, the ability to command votes, ideas, the ability to turn out masses into the street to demonstrate, specialized knowledge, inherited high status, and so on) in order to pursue its preferred public policy agenda (see Wynia 1990). Thus, for example, organized industrial workers possess the resources of strikes, potential mass action, and a reasonably large bloc of votes; maintaining high wages is one of this group's most important policy goals. Capitalists possess the resources of control over funds, which may be channeled to candidates in democratic political systems, or invested abroad, if the business community becomes sufficiently disenchanted with local prospects. Their policy preferences run to lower wages and light government regulation. Interest groups, which may be constituted on the economic basis of class or occupation, or around other identities such as ethnicity or geographical origin, will have competing policy preferences in all political and economic systems. There is nothing per se that is insidious or abnormal about the existence of social groups competing to influence public policies, although many political ideologies (from Iberian or Latin American corporatism to Communism) prefer to pretend that society is intrinsically harmonious, and thus intergroup conflict nonexistent or at least illegitimate.

Social groups, or political players, compete over public policy within the framework of political institutions, including both the formal rules of the game (laws and constitutions) and its informal rules (well-understood although unwritten behaviors and practices, ranging from innocuous social conventions to the never-articulated but strictly enforced rule that the dictator-for-life is not to be contradicted on pain of imprisonment or death). Military authoritarianism is one set of rules, which of course exists in many variants, while procedural political democracy, which also has numerous specific incarnations, is another. Both formal and informal rules of the national political game may evolve gradually or instead shift dramatically because of some discrete event, such as a military coup. Circumstances that alter the balance of relative power among political actors constitute one important source of dynamism and change in national political rules, particularly informal ones.

There are our main political actors whose influence, overt or indirect, over local policymaking and political choices may be differentially enhanced by diverse forms of foreign capital inflows: foreign governments and international organizations, the incumbent government of the capital-recipient country, foreign business, and local big businesses sufficiently well-known to attract foreign private investors. These obviously are not the only politically relevant players, nor will their initial degree of influence be the same across all emerging market countries. (For example, as of the early 1990s local big business was a more powerful political player in most Latin American than most Eastern European countries, because capitalists had been deemed criminals under Communist rule.) The six types of cross-border financial instruments differentially apportion additional resources among these four political and policy players. Column 2 of Table 3 reflects my expectations of marginal increments to the resources controlled by locally relevant political actors. In all cases, I am discussing the pattern of political resources in the absence of a balance of payments crisis associated with the cross-border capital flows.

Foreign governments and multilateral financial institutions (whose policies wealthy foreign governments tend to dominate) enhance their direct influence in local politics through foreign aid. All foreign capital inflows to some extent bolster the incumbent political "regime" (that is, type of political system, as in democratic or authoritarian) and the incumbent "government" or "administration" (that is, the set of politicians currently in power). Capital inflows signify the outside world's confidence in the economic management skills of the ruling team. Foreign business enhances its direct influence in an emerging market country only through foreign direct investment (FDI). The influence of MNCs is ostensibly strictly economic, but in practice includes the right to have an indirect input into public policy and regulatory decisions that affect them. Once installed within the country, MNCs become local political actors. Foreign investors with a majority stake in a local business enterprise are bound to have preferences on local issues, whether or not the formal rules of the game permit them to participate, as their profits and other legitimate interests will be strongly affected by national economic regulatory policies. Finally, local big business irms augment their politically relevant resources when they are able to attract foreign capital flows directly, as in commercial bank loans borrowed directly from multinational banks (category four) or portfolio flows (equity, bonds, or short-term debt) raised by private corporations directly. Where local big business firms have a sufficient international reputation to borrow on their own hook, this enhances their bargaining resources vis-à-vis their own governments, putting them in a position analogous to that of private sector entrepreneurs who account for a large share of a country's exports. (Often they are the same people.)

The comments just made apply to cross-border capital flows in the absence of a balance of payments crisis. Following such a crisis, however, the direct influence of all foreign actors in the emerging market country increases, including foreign investors or lenders, international financial institutions, and the home governments of foreign private investors and lenders.

The six types of cross-border capital flows also differ in their degree of potential volatility, that is, in the ease with which the foreign investor can turn around and repatriate his/her/its funds. Consequently these alternative financial instruments imply differing degrees of balance of payments risk, s shown in column 3 of Table 3. Foreign aid and foreign direct investment have low volatility; they are unlikely to be reversed overnight. Foreign aid flows largely respond to non-economic criteria in any case; it usually takes a very dramatic, and rare, political change within the recipient to have aid flows staunched suddenly. In contrast, the reasons that FDI is of low volatility largely are practical. Factory owners cannot sell out immediately without confronting large losses, particularly in the context of an economic or political crisis. They thus tend to remain in the host country, often attempting to influence public policy choices themselves, whether overtly or covertly. However, although FDI is unlikely to be volatile in the sense of participating in overnight capital flight, political or economic crises in the capital-importing country can cause potential investors to delay or cancel long-planned inward FDI. 17

Medium and long-term commercial bank loans (categories three and four) in principle are of intermediate volatility, and thus pose an intermediate risk of external payments crises. Loans for capital projects cannot be called in, barring gross violation by the lender of the terms of the contract, before they mature in six to ten years. Nonetheless, in practice, shorter-term loans, including medium-term trade credit, tend to become linked to the long-term debt. Within six months following Mexico's August 1982 admission of its inability to make its quarterly debt payments, the major commercial bank creditors had generalized the crisis to most other Latin American countries by refusing to renew entirely standard and customary forms of commercial credit, thus provoking liquidity crises and the borrower countries' consequent inability to meet their debt service payments on the longer-term loans.

Portfolio investments (categories five and six) pose the greatest risk of an external payments crisis. Stocks and bonds are designed to be traded regularly, even daily or hourly: this is the reason their prices fluctuate. If they also trade across national borders in sufficient quantities, then the price of a nation's currency can fluctuate with the fortunes of its domestic capital market. Short-term debt is extremely sensitive to interest and exchange rate movements. Thus, a serious balance of payments and currency crisis can accompany a market downturn. A crisis in one country in a region can be rapidly generalized to its neighbors. Finally, if foreign investors (and/or domestic investors intending to move their capital abroad- it makes no difference) also hold large chunks of the government's domestic public debt, then a fiscal crisis ensues as well. Thus Mexico's financial crisis in the months following December 1994 was both a balance of payments and a fiscal crisis, because exiting foreign investors a large share of the public debt.

The final hypothesized direct effect of foreign capital flows, shown in column 4 of Table 3, is pressure for developing country policymakers to manage their economies to suit foreign investors, rather than local needs or preferences. 18 Pressure for neoliberal economic policies can be expected to vary dramatically according to the type of financial instrument. 19 Both foreign aid and commercial bank loans sometimes went to countries whose economic policy capacities were marginal; donor governments often didn't care, since their major goals were strategic, not economic, and multinational banks during the 1970s lending boom, in retrospect, did not vet their sovereign borrowers carefully (Devlin 1989, Cohen 1986). In fact, developing country governments often used aid and loans to delay or avoid painful economic reforms. The pressure for neoliberal economic reforms from these sources of capital inflow typically is low.

Multinational corporations that invest directly in a developing country are more likely than aid donors or bank lenders to care about the host country policy environment. As local actors, however, as well as foreign investors, the policy preferences of multinationals are not likely to reflect a cookie-cutter, one size fits all orthodox mentality. Instead, they tend to prefer a stable regulatory environment. The pressure exerted by foreign direct investors on capital-importing country governments for orthodox economic reform is moderate. 20

The portfolio investors in emerging markets of the late 1980s and the 1990s differ from their predecessors in two respects. First, the possibility of rapid movement of funds in and out of specific securities, and also in and out of countries and currencies, inspires investors to seek information that will enable them to make profits from rapid arbitrage. Meanwhile, each investor, whether as an individual or institution, must be prepared for rapid movement out of a multitude of idiosyncratic markets. It is very difficult, if not impossible, for investors to have good information on all of them. Thus, summary comparative measures- such as the government budget deficit, trade balance, public debt, growth rates, and inflation- tend to be seized on as a means of selecting among alternative investment venues. Furthermore, comparatively ignorant international investors tend to assume that countries that share similarities of geography, history, or culture (as in "Latin America" or "Eastern Europe") follow similar macroeconomic strategies with similar potential for success. Thus, the same herd mentality that caused multinational banks to react to Mexico's de facto default in August 1982 by pulling loans from all other Latin American borrowers was again in effect in Latin America in early 1995 and in East Asia in late 1997, but at an intensified level. From the viewpoint of investors, of course, such analytical shortcuts are entirely rational; it is better to rank possible investment destinations by imperfect data than to invest with no good information at all. From the viewpoint of emerging market countries, on the other hand, such investor behavior ties their hands very rigidly, forcing them into politically (and sometimes economically) risky behavior such as 30 to 50 percent slashes in already low levels of social spending, often just as democratization finally is giving the poor a voice in politics. 21 This is the "tyranny of the bond traders" that moves exchange rates even in the advanced industrial countries. 22

In addition, global capital markets today contain many decentralized investors, even if one notes the important leadership role of large international institutional investors (see Haley forthcoming). In this situation, the lowest common denominator of information tends to prevail. Hypothetically, even if a given portfolio investor was aware that a country's heterodox stabilization program, for example, was appropriate given conditions in that economy, he or she might have to divest of that country's treasury securities if the investor suspected that less well-informed investors might flee and drive the value of the country's government debt down anyway. Even in the absence of direct, overt pressure from foreign institutional investors, that is, countries with large quantities of portfolio inflows find themselves under high implicit pressure to adopt neoliberal economic policies.

Indirect Links: Cross-border financial flows and democracy in emerging market countries

The previous section suggested relatively straightforward links between the six types of capital flows and four important intermediate variables- economic growth, the relative power of four important political players, the likelihood of a balance of payments crisis, and external pressure for neoliberal economic policy choices. We now return to our core question: how might the institutional form of cross-border capital flow affect the prospects for a successful transition to, or consolidation of, procedural political democracy?

Let me first make an obvious point: the nature of the incumbent political regime (that is, the type of existing political system) is the most important influence on whether foreign capital inflows promote or retard democracy. Ceteris paribus, foreign investment in a democracy will strengthen democracy, while foreign financial resources flowing into a country governed by dictator will tend to enhance his or her position. Similarly, balance of payments crises undermine the credibility of governing incumbents- even if the main borrowers have been private businesses or banks. It makes no sense to discuss the import, for democracy, of any of the six types of financial instruments without also considering the political status quo of the recipient country. The following discussion, summarized in Table 4, returns to this point frequently.

I am skeptical about a commonly posited relationship. In my view, forms of cross-border capital flows that improve economic growth don't necessarily aid in the promotion of democracy. There is no direct causal relationship between economic growth, on one hand, and the transition to or consolidation of procedural political democracy, on the other hand (see Table 4, Row 1).

This is an old argument for political scientists. We can begin with the reason that so many observers have believed that economic growth and political democracy were complementary and mutually reinforcing (see Rostow 1960). The majority of contemporary liberal democracies are wealthy, industrialized countries. Therefore, it is easy to conclude, changes that move countries in the direction of becoming wealthier or more powerful also help them democratize. However, as Guillermo O'Donnell (1973) conclusively demonstrated for Latin America in the 1950s through the 1970s, it also is possible that progress toward industrialization could lead away from democracy. 23 Many more recent treatments also argue that the relationship between economic or industrial growth, on the one hand, and political democracy, on the other, is highly contingent at best, and may even be negative under some conditions (Haggard and Kaufman 1992; Haggard and Kaufman 1995; Armijo, Biersteker, and Lowenthal 1994). A comparison of India and South Korea, for example, which had roughly the same per capita income around 1950, suggests that Indian democracy since 1947 did not produce much in the way of rapid economic growth, at least before the 1980s, while Korea's admittedly successful economic strategy did not lead to democracy before the late 1980s (Varshney 1984). In general, high economic growth under authoritarian political auspices has the effect of legitimating the incumbent non-democratic regime, and thus delaying the democratic transition in the short to medium run. Rapid growth in a democracy similarly validates the current political leaders and their economic policy team.

The second dimension explored in this essay's previous major section linked certain institutional forms of international financial investment to shifts in the distribution of locally relevant political resources controlled by key players in the game of national politics. We cannot plausibly argue that an increase in the resources available to player X will either create or destroy procedural political democracy. However, we may be able to assert that, all other things being equal, an increase in the resources available to player X heightens or diminishes the prospects for democracy, particularly under circumstances Y. (See Table 4, rows 2 through 5.)

Table 4. Intermediate Variables: Implications for Democracy
Intermediate Variable: If Incumbents Are Authoritarians If Incumbents Are Democrats
High Economic Growth May delay democratic transition by legitimating incumbents. Good for democracy.
Foreign Governments' Influence Increased Depends on goals of foreign government. Depends on goals of foreign government.
Incumbent Government's Influence Increased Strengthens authoritarian regime. Strengthens democratic regime.
Foreign Business' Influence Increased Promotes some political liberalization, but not full transition to democracy. Pessimist hypothesis:
  • Dangerous for weak democratic govts. with leftist policy goals.
Optimist Hypothesis
  • Protects foolish leftist govts. from policy mistakes that invite military coups.
Local Business' Influence Increased Promotes some political liberalization, but not full transition to democracy. Depends on characteristics of local business.
Balance Of Fayments Crisis Occurs Discredits authoritarian incumbents with citizenry and international community. Discredits democratic incumbents with citizenry and international community.
Neoliberal Economic Reforms Occur May delay democratic transition by weakening pro-democracy actors (organized labor, the poor) Good in long run. In short/medium run may strain or reverse democratic transition.

Increases in the leverage of foreign governments and international organizations, as through a country's acceptance of foreign aid, heighten the impact of the preferences of these foreign actors within the national political game of the emerging market country. Unfortunately, democratization or democratic consolidation is only infrequently a top priority of aid donors. Bilateral aid donors typically rank achievement of their other strategic objectives- such as access to military bases, cementing alliances, commercial reciprocity, or drug interdiction- above democratization. Multilateral financial organizations usually allocate resources according to overt macroeconomic criteria, and sometimes also according to the unacknowledged strategic criteria of their major donor countries, rather than as a reward for or inducement toward democracy. Increased influence in the politics of emerging market countries for foreign governments, even when they these governments are democratic at home, is, at best, a weak recipe for democratization.

Increased influence for the incumbent government promotes democracy only when the incumbents are democratic already. Monies coming directly to the incumbent government augment the resources (often in practice fungible across investment, consumption, and political patronage spending) that ruling politicians can deploy vis-à-vis their political opponents. New or weak democratic governments may have a particularly acute need for resources, since previously marginalized classes almost always expect a rapid improvement in their material circumstances, now that an oppressive government has gone. The 1990s shift toward private sector in-country recipients of funds, shown in Figure 2 above, thus may be particularly unfortunate for new democracies. However, as noted, all types of capital inflows strengthen governing incumbents to some degree- at least so long as they do not generate a balance of payments crisis.

What are the consequences of local political influence for multinational direct investors? This is, of course, a highly polemical topic on which much has been written (see Chase-Dunn and Bornschier 1985). There are two contending expectations. We may call these contrasting predictions the "pessimistic hypothesis" and the "optimistic hypothesis." The pessimistic hypothesis comes in two versions, which are not mutually exclusive. The first version suggests that transnational direct investors will have no patience with democratically elected leftist reformers. Left-leaning governments that promote policies like land reform or preferences for small-scale local production, thus may meet resistance from foreign direct investors in their midst. Foreign direct investment that originally entered the country during previous authoritarian and/or politically conservative times may be especially problematic for a new or weak democratic government in an emerging market country. Democratically arrived at (that is, politically popular) public policies might be excessively economically populist. The foreign businesspersons may feel that the regulatory environment suddenly- and from their viewpoint, unfairly- has been switched under their noses, thus justifying to themselves a decision to retaliate. The political activities of multinationals in host countries have ranged from the relatively innocuous, as in joining or sometimes organizing local chambers of commerce and or other business lobby groups, to the illegal but not unexpected, as in secretly contributing to the campaign funds of pro-business candidates, to the egregiously inappropriate, including involvements in coups and assassination attempts.

A second version of the pessimistic hypothesis reasons from the presence of transnational direct investors to the likely social class structure of a currently authoritarian or semidemocratic country, and concludes that the possibility for a future transition to full democracy has become less likely. Thus, some political sociologists have postulated that full electoral democracy with nearly universal adult participation is most likely to arise in societies that are at certain stages in their economic development, and possess specific sorts of class structures. If this hypothesis is correct, then certain alterations in groups' relative economic power that are non-threatening to an established democracy nonetheless may impede a full democratic transition. Rueschemeyer, Stephens, and Stephens (1992), for example, worry about the domestic political consequences of a developing economy in which multinational direct investors dominate core productive sectors. 24 The three believe a numerically strong and politically assertive industrial working class to be a necessary although insufficient condition for the successful transition to mass democracy. In the absence of effective lower class demands for political participation, they argue, political reform may cease with partial liberalization of overt authoritarian controls on middle class civil liberties, stopping well short of full democratization. For example, they suggest that the particular class characteristics of Latin American industrialization, whose leading actors for well over a century have been foreign direct investors, directly inhibited the formation of strong labor unions, and thus of stable democracies, in most countries there, at least prior to the late 1980s 25 (see also Karl 1990; Moore 1966).

Contrasted to these predictions is the "optimistic hypothesis" 26 which is the mirror image of the first pessimistic expectation above. Quite so, the optimists argue, foreign entrepreneurs might try to protect their property rights by pressing the government to restore a "sane" investment climate. However, the intervention of foreign investors on the side of neoliberal policies could end up protecting, rather than undermining, formally democratic rules of the game. That is, foreign business leaders, often allied with local big business, could use their influence to push policymakers to adopt more macroeconomically sensible and sustainable policies. Society as a whole thus might avoid just the sort of economic breakdown that ends with a military coup and the installation of military-technocratic ("bureaucratic-authoritarian") 27 political regime justifying their withdrawal of political freedoms in the name of restoring economic growth.

The sequences of events imagined in either the first variant of the pessimistic hypothesis, or the sole version of the optimistic hypothesis, of course, only come into play if the elected government wants to enact leftist, populist policies of economic redistribution, social spending, import-substituting industrialization, and the like- particularly if policymakers are blithely unconcerned with how to pay for their plans. If, on the other hand, democratically chosen policymakers have conservative economic leanings, business actors typically will be more satisfied with the government and less likely to interfere. For reasons outside those analyzed in this essay, neoliberal economic ideas have made a worldwide ideological comeback in the 1980s and 1990s (Biersteker 1995). If the elected government follows conservative economic policies, then both foreign and most local big businesses can support democracy without contravening their class interests and policy preferences. 28

Finally, what can we expect if foreign capital inflows augment the resources available to local big capitalists? Emerging market countries differ greatly from one another: blanket predictions would be foolhardy. Nonetheless, deductive logic and a rational choice perspective suggests that big business should, in most developing countries, favor political liberalization but may be ambivalent about a full transition to competitive electoral democracy. A shift away from the arbitrariness of unrepresentative governance to the rule of law reduces the costs of doing business. Protection of individual freedoms protects businesspersons themselves. Press freedom is consistent with good access to timely business data. However, there are good structural reasons why the attitudes of business leaders toward democratic transitions may be ambivalent. On the one hand, entrepreneurs' personal experience of participation in global markets will tend to broaden the knowledge that they have of the larger world: cosmopolitan, educated, high status individuals tend to have been exposed to, and perhaps imbued with, democratic values. 29 On the other hand, successful local business leaders with a presence in global markets will, almost by definition, have been members of the privileged elite in their home countries. Similarly, multinational direct investors, once committed to a country, tend to become de facto members of the incumbent political regime's support coalition. If an emerging market country's current political regime is authoritarian, then the shift to mass electoral democracy almost inevitably will dilute the public policy influence of these same business elites. If we imagine that business leaders as a group are classic "rational actors" then their strictly selfish interests should lead them to favor a politically liberalized, semi-authoritarian regime- allowing, say, freedom of the print media but not of radio and television (because the masses don't read); freedom to travel abroad, but not to criticize the government openly at home; and well-codified corporate law but not the right to strike. 30

This essay's previous section also ranked the six ideal types of cross-border capital flows according to their risk of provoking an external payments crisis. Table 4, Row 6 suggests that the main consequence of a balance of payments crisis is to discredit governing incumbents. Two further observations seem irrefutable.

First, if the political regime itself is weak or fragile, then such crises may also stimulate a political regime change. However, and second, while debt crises certainly are ominous for fragile democracies, it is not safe to conclude that balance of payments crises occurring under the auspices of authoritarian incumbents necessarily are good for democratic transitions.

When an incumbent military regime weakens and topples because political leaders cannot control the recessionary effects of an external financial crisis, then democracy, of course, may result. However, orthodox stabilization or structural adjustment, and the regressive social class and income shifts normally associated with them, may make it difficult for the successor regime to be a full mass democracy. The argument is as follows. Orthodox structural adjustment policies, implying deep recessions and much pain, seem an inevitable consequence of external payments crises in a contemporary emerging market countries. However, social groups whose political resources increase with a transition to political democracy- notably including organized labor and the poor- 31 may be precisely those that tend to lose out economically during standard neoliberal structural adjustments. For example, recessions mean that first private, then eventually public, sector workers lose their jobs- over the protests of their unions, typically, which then are delegitimized and weakened politically as an actor in constructing future political bargains. In addition, the poor, at least those whose livelihood depends on cash employment or casual earnings (as opposed to subsistence farmers), is the group that tends to suffer most during structural adjustment periods. 32 There are purely economic reasons why the poor should suffer more than organized labor, the middle class, or upper class groups: they lack a diversified portfolio of assets going into the crisis, have little or no safety cushion, and cannot access good information about their options during the lean years.

Loss of relative economic power, however, breeds loss of politically relevant bargaining resources- particularly important during a period of political transition. It is during periods of political transition that the de jure and de facto "rules" of the successor political game are being hammered out among all those groups that can, in effect, claim a place at the table. The bargaining weight of a political player is roughly equivalent to the importance and credibility of its threat to withdraw cooperation and sabotage the entire process of political compromise. If unions are discredited, and potentially mobilizable urban slumdwellers are busier than ever in begging and competing against one another for casual employment, then they (usually) become that much less likely to have the status, time, or energy to contest the rules of the successor political game. Therefore organized labor and the poor lose relative influence in the domestic political game.

Meanwhile, groups whose direct public policy influence would be undercut by the shift from policymaking by insider status to majority votes- such as foreign or local big business, as already discussed, or the military, or the urban middle class (particularly where this class is relatively small)- are unlikely to bear the brunt of structural adjustment cuts. That is, they still will be players of some heft during the transition bargaining. The relative balance of bargaining influence among alternative social and political actors that has been constructed during the time of political regime breakdown and renegotiation tends to be perpetuated into the future, at least until the next system-threatening crisis. 33 Of course, many other factors- prominently including the demonstration effect of Latin American and East Asian democratization in the 1980s, and the end of the Cold War and Eastern European democratization in the early 1990s- also play a crucial role in promoting democratic transitions. Nonetheless, the conclusion is that, even if the previous authoritarian regime has been discredited and overthrown, the experience of a balance of payments crisis in an emerging market country is not likely to be particularly conducive to the establishment of democracy.

The last dimension hypothesized to be directly affected by the form of international capital flows is external pressure on the borrowing country for neoliberal economic reforms, as shown in Table 4, row 7. As noted in this essay's previous section, I expect portfolio types of capital flows to generate the most pressure for preemptive neoliberal policy changes. Neoliberal reforms range from budget-cutting to structural/institutional changes such as shrinking the size of the federal bureaucracy, selling state-owned enterprises ("privatization"), and getting rid of trade barriers. The one-off reform of selling state assets, for example, brings money into the national treasury which can be used to balance the budget, thus attracting capital inflows. Furthermore, the language of market-oriented regulatory changes- free" markets, "liberalization" and so on- parallels that of democratization. It is no wonder that many thoughtful observers have assumed that free politics and free markets are entirely complementary- or even that they are acceptable substitutes for one another (see Bhalla 1994).

However, this easy assumption of automatic complementarity appears to be false, except possibly in the long run. Even if we assume, for the sake of argument, that all such "reforms" actually improve economic performance eventually, there is no linear, necessary connection between market-oriented economic liberalization and the transition to or consolidation of procedural political democracy. A reduction of the state's economic presence typically means not only an end to unnecessary red tape, but also cuts in social spending and the redistributive activities of the paternal state. These social structural shifts are unlikely to promote a transition from authoritarianism to democracy. Under preemptive neoliberal economic reforms we might expect labor unions and the poor to be relatively worse off, at least in the short to medium run, just as we argued they would be following a balance of payments crisis. 34 Moreover, authoritarian leaders, fearing political protests from recessionary reforms and cuts in state spending, may be particularly unlikely to risk political opening while implementing neoliberal economic reforms. Rapid neoliberal reform could make democratic transitions more difficult (Armijo, Biersteker, and Lowenthal 1994).

At the same time, the consensus of economists, at least those practicing in advanced industrial countries, supports the belief that market-oriented reforms will improve economic growth prospects in the medium to long run (Williamson 1990). 35 That is, there is also an "optimistic" hypothesis about external pressure for preemptive neoliberal reforms. Market reforms are sorely needed to turn an authoritarian and inefficient economy around, say some observers, so much so that democracy cannot be safe until tough, and probably politically unpopular, measures are taken to dispossess a huge cadre of rent-seekers within the transitional state. Randall Stone's (forthcoming) essay on Russia makes a strong version of this argument, and Jeffrey Winters (forthcoming) at least raises the possibility that it could apply in Indonesia. If weak or venal domestic leaders are unable to undertake painful reforms, then skittish global portfolio managers (or wealthy locals with the know-how to move their money out through the black market) may, paradoxically, be the common people's best friends. In other words, the credible threat of a balance of payments crisis yields better macroeconomic policies, which is good for democratic transitions. In this case, aspects of both the pessimistic and optimistic expectations are both likely to be valid.

Illustrative Country Cases

I look at country cases illustrating each type of capital flow to see whether the expected consequences manifest themselves. Since no country case is a pure example of a single type of capital flow, these remarks should be understood as impressionistic. 36 Furthermore, since all types of capital inflows tend to strengthen incumbents, while all financial crises tend to weaken them, the most important predictor of the political consequences of net foreign capital inflows is the characteristics of the incumbent political regime. The discussion, summarized in Table 5, thus distinguishes between authoritarian and democratic capital importers that have experienced each type of inflow.

Table 5. Types of Capital Flows and Democracy: Country Cases
Prominent Financial Instrument Authoritarian Capital-Importing Country Democratic Capital-Importing Country
Foreign Aid
  • Philippines under Marcos.
  • Zaire under Mobutu.
  • India under Nehru.
  • Russia under Yeltsin.
Foreign Direct Investment
  • Indonesia under Suharto.
  • Vietnam in 1990's
"Pessimistic" outcome:
  • Chile under Frei,
    then Allende (1964-1973).
  • Brazil under Kubitschek,
    Quadros, Goulart (1956-64).
"Optimistic" outcome
  • Argentina, Brazil in late 1980's,
    1990's.
  • South Africa under Mandela.
Bank Loans To Govt Argentina, Brazil, Mexico in 1970's. India in 1980's.
Bank Loans to Private Firms Chile, Argentina, Brazil in 1970's. ??
Portfolio Investments With Govt Mexico under Salinas (1988-1994). Argentina under Menem (1990's).
Portfolio Investments In Private Firms
  • South Africa under apartheid.
  • Singapore in 1990's (port.equity)
  • Indonesia in 1990's (short-term debt)
  • India, Chile in 1990's
    corporate bonds, port. equity)
  • Thailand, So. Korea in 1990's
    (short-term debt).

Our expectations of foreign aid (see Row 1 of Table 5) were that it would make a rather small contribution to economic growth in the recipient country, that the locally relevant political influence of donor governments (or international organizations) would be strengthened, as would governing incumbents, and that this form of foreign capital transfer posed a comparatively low balance of payments risk. External actors would be relatively unlikely to pressure the sitting government for specific economic reforms, neoliberal or otherwise. The consequences of official grants and credits for democracy in emerging market countries thus mainly hinged upon the goals of the main donor government(s) and the pre-existing characteristics of the government in the recipient country.

United States' aid to the Philippines under Ferdinand Marcos through the late 1970s illustrates foreign aid to an authoritarian government. In return for secure leases for military bases the U.S. considered essential to its East Asian strategy, the U.S. gave generous aid commitments and for many years downplayed credible reports of human rights abuses. Moreover, the Marcos administration apparently used its overall control of resources and influence (although not necessarily foreign aid dollars in a traceable fashion) to perpetuate dense networks of patronage and to corrupt many government-business links. The U.S. government avoided pressuring the Philippines for significant economic reform, although the multilateral development banks sporadically tried. 37 However, in the early 1980s the U.S. altered its perception of "U.S. interests" in the area and gave important support to the "people power" movement of democracy campaigner Corazon Aquino. The large sums of bilateral and multilateral aid to the Philippine government, along with the strong historic ties of the islands to the de facto ex-colonial power, gave the United States government substantial influence over the progress of the eventual democratic transition. In the Philippines, then, foreign aid began to contribute to the circumstances promoting democracy only when this became an outcome important to the country's major aid donor.

Zaire provides a second case of aid to an authoritarian regime. American, French, and other foreign aid from the 1970s through the early 1990s to Zaire under President-for-life Mobutu Sese Seko pursued goals of anti-Communism, strengthening post-colonial francophone ties, and assuring access to huge copper and mineral reserves for commercial uses and possible military contingencies. With such a full agenda of strategic goals, human rights and democratic concerns were consistently and thoroughly downplayed in the relations of the major donors with the authoritarian government. 38 Although Zaire admittedly was an extreme case, it provides a textbook example of the contribution of foreign aid to government economic mismanagement and the propping up of an oppressive government and political regime.

An example of a democratic recipient of foreign aid is India in the 1950s through the 1960s, during these years one of the largest aid recipients among all developing countries, albeit one of the smallest in per capita terms. Aid from the Western democracies, both from multilateral sources such as the World Bank and through bilateral programs like the U.S. Food for Peace donations, provided important budgetary support to the Indian government in its first decades, becoming at least one of the factors that enabled Indian democracy to survive. Large quantities of foreign aid to India not only strengthened the ability of elected rulers to deliver some of the material goods their constituents desired. Foreign assistance also reinforced the government's freedom to pursue its preferred economic policies, which it described as "democratic socialism." These policies included construction of a huge state productive sector, and extensive affirmative action and regional redistribution programs, all intended as a way of enhancing citizen support of the incumbent democracy (and the long ruling Congress Party) within a multiethnic state riven with deep caste and class inequalities. They were the antithesis of neoliberal policies. With hindsight, the majority of Indian economists agree that many of India's statist economic policies also were rather inefficient. However, because the population by and large perceived their redistributive element as just and moral, they played an important role in legitimating democratic government in a setting that many observers initially thought would be an unlikely home for it. 39

The experiences of the Philippines, Zaire, and India generally confirm our expectations of foreign aid. Few observers believe that external assistance made great contributions to economic efficiency in any of these countries. Nor did foreign aid, even from democratic donors, necessarily make great contributions to hastening the transition to democracy. Instead, aid resources gave incumbent governments additional room to maneuver, both economically and politically (so long as the foreign patron government could achieve its de facto core goals), without greatly increasing policymakers' worries that a sudden foreign exchange crisis might develop. More recently, the newly and precariously democratic Russian federation has received large quantities of multilateral assistance, especially- and unusually- from the International Monetary Fund. Randall Stone (forthcoming) puts a compelling, and rather darker, twist on the likely consequences of massive quantities of foreign aid coming to a young and weak democracy. The political incumbent, President Boris Yeltsin, indeed has been strengthened, Stone argues, but at the cost of both sound economic policy and institutionalizing democracy (see also Economist 1996).

Foreign direct investment (Row 2) should have a significant positive impact on economic growth, should heighten the political and policy influence of multinational corporations, and should not be prone to sudden capital flight. I assume that multinational corporations (MNCs) always become players in the domestic politics of the host country, however much they may publicly disavow having any such intent. Most often their political involvement is limited to discreet lobbying on behalf of a "good" investment climate. Thus, MNCs may intrigue against what they view as "politically motivated regulatory red tape" or "irrational" restrictions on their right to hire and fire at will, and so on. If the country's current rulers are authoritarian leaders, they should be strengthened by FDI inflows. However, authoritarian regimes also may experience subtle pressure from foreign corporations to liberalize politically, for example, by expanding freedom of the press. Policymakers may also find multinational direct investors urging them to liberalize the economic policy regime.

Indonesia and, very recently, Vietnam are authoritarian countries that have captured large direct investment inflows (see Winters, forthcoming and Haughton, forthcoming). Loudly anticommunist and authoritarian Indonesia under General Suharto received large amounts of FDI from the late 1960s through the 1990s. Communist and politically authoritarian Vietnam garnered truly astonishing pledges of foreign investment since its embarkation on the path of economic liberalization in the early 1990s. There are some interesting similarities between these otherwise diverse cases. The symbolic vote of confidence from outside, and the economic stimulus, have helped each economy. In both cases generous foreign investment apparently bolstered the political fortunes of the authoritarian incumbents, arguably encouraging them to continue suppressing political dissidents. Foreign direct investment thus did little or nothing to promote democracy. Multinationals have insisted upon some economic reforms, generally ones aiming to make the internal economic regulatory environment more predictable and more similar to business conditions prevailing in advanced industrial countries. Nonetheless, the political incumbents were perhaps encouraged to perpetuate certain economic irrationalities, such as large scale corruption in the management of public sector investments, that did not negatively affect foreign investors' profits. That is, direct investors, like local big business actors, in both countries have been more concerned with ensuring their own freedom of action than with increasing the overall efficiency of the economy. In addition, Jonathan Haughton notes that Vietnam's Communist government has required joint ventures between state-owned enterprises and MNCs, thus strengthening government incumbents more than with ordinary FDI, while diluting the presumed "efficiency effect" of having foreign capital inflows spent by private, rather than public, actors. 40 FDI has the advantage of not being highly volatile. The spread of the 1997 East Asian financial crisis to Indonesia, but not (as of mid January 1998) to Vietnam probably can be attributed to the much greater role of liquid portfolio investment in the former. FDI also gives the host government useful friends: Western and Japanese direct investors in Indonesia have used whatever influence they have over their home governments to encourage them to participate in the rescue package.

If the incumbent government instead is democratic, and if it can limit the political and lobbying activities of foreign businesses to actions that are not regime threatening, then the additional private investment and growth provided by FDI should strengthen democracy. However, emerging market country governments are not always able to control the domestic political activities of foreign business. Chile in the early 1970s illustrates the case of foreign direct investors- who originally had entered the country under a succession of politically and economically conservative though mostly democratic previous administrations- confronted with a newly elected, politically weak, democratic government espousing radically redistributive, populist economic policies. Officers of American multinational companies actively intrigued with representatives of the U.S. government and the Chilean military to overthrow President Salvador Allende, who lost his life in the coup that initiated seventeen years of often brutal military rule (Treverton 1990). The Chilean experience illustrates the pessimistic hypothesis about the increased influence of foreign business in host country politics. Similar events transpired with the Joao Goulart administration (1961-64) in Brazil and the second Peronist government (1973-76) in Argentina, each of which the country's military officer corps, with the tacit support of traditional politicians and business leaders, also displaced by force.

There is another possible twist on these stories, however. Most contemporary economists, including many on the left, probably would agree that the economic policies of Allende, Goulart, and Isabel Peron were, in fact, unsound and unsustainable. Each of the three presidents, for example, raised civil service and public sector union wages when the treasury had no funds to back this move. Arguably, had multinational investors successfully pressured these governments to adopt less radical economic policies, then economic conditions would have improved and the excuse for a military coup could have evaporated. This second line of argument, the optimistic hypothesis, also plausibly applies to the more recent experiences of new democracies in emerging markets that have decided to welcome foreign direct (and portfolio) investors. These include Argentina and Brazil from the mid-1980s, or South Africa in the 1990s. In each case, the policymakers in the new government who historically had been associated with economic nationalism gradually adopted many of the programs of neoliberals (on Argentina, see Armijo 1994, and Gibson 1997; for post-apartheid South Africa, see Daniels and Daniels 1995). Also in each case, this shift in the overall tenor of public policies was initiated by technocrats, politicians, and other policy elites from the "top down" had to be sold to skeptical voters whose first preference was for classically populist economic policies, and was at least partly a response to political incumbents' perceptions of the need to attract foreign capital, especially in the form of direct investment.

The third category of international investments is commercial bank loans to the public sector of the emerging market country (Row 3 of Table 5). I argued that bank loans to government should have a lesser impact on economic growth, while expanding the resources available to the political authorities to use for either developmental or directly electoral purposes with, in practice, comparatively little creditor oversight or external pressure to follow "sound" economic policies, however defined. The risk of a balance of payments crisis, however, is higher with loan finance than with official credits or FDI. During the 1970s the authoritarian governments of Argentina and Brazil, and the civilian, semi-authoritarian government of Mexico each received large quantities of foreign bank loans. In all three cases, foreign loans to nondemocratic incumbents shored up unrepresentative political regimes by allowing political leaders to extend favors to politically crucial groups. 41 Interestingly, another consequence of foreign borrowing was to increase the de facto economic policy autonomy of governments from both foreign and domestic pressure to reform. 42 Finally, capital account surpluses due to foreign borrowing in all three Latin American countries served to make large current account deficits surprisingly painless, at least for upper income groups, and thus probably deprived the democratic opposition of many potential allies among salaried workers and the business community. Until the 1982 debt crisis, the incumbent regimes in each of these countries continued to claim, with some plausibility, that they should be maintained in office because of their singular success in national economic management. Foreign loan inflows not only insulated the military regimes in these countries from foreign pressures to alter their regulatory regimes; they also protected authoritarian incumbents from criticism from domestic political opponents.

However, bank loans, even long-term ones, are a more volatile form of capital flow than foreign aid or direct investment. In Argentina, Mexico, and Brazil the effect of the 1982 financial crisis was a loss of credibility for authoritarian incumbents, leading to dramatic changes of regime in Argentina in 1983 and Brazil in 1985 as each country returned to procedural political democracy. However, had the international environment of the 1980s been less favorable to democracy, the successor regimes in Brazil and Argentina might well also have been authoritarian. Mexico's civilian "soft" authoritarian regime experienced no sudden, dramatic crisis, but rather a gradual erosion of public confidence (important to authoritarian as well as democratic regimes, albeit in somewhat different ways) throughout the 1980s. One consequence was to give previously feeble opposition parties a opportunity to press for fairer, more genuinely democratic, rules of the national political game (see Elizondo, forthcoming).

Commercial bank loans to a democratic government also can augment its resources and bolster its credibility. In the 1980s India, a democratic country which had resisted the blandishments of international bankers throughout the 1970s, became a large commercial bank borrower. Tellingly, one of the hopes of Indian politicians and policymakers was that, by borrowing from the foreign private sector, they could escape the economic policy conditionality imposed by public sector lenders, such as the International Monetary Fund, to which Prime Minister Indira Gandhi very reluctantly had had to turn for a large loan in 1981. As had been the case with Latin American sovereign borrowers of the 1970s, the foreign resources gave additional room for fiscal maneuver (or irresponsibility) to the incumbent government- though this time the incumbents were democratic. India experienced another external payments crisis in 1991, which led the country to another reluctant turn to the IMF and provoked policymakers into beginning long needed economic liberalization. 43 Fortunately for the long dominant Congress (I) Party, it was a short-lived coalition government of the non-Congress left parties that made the politically unpopular decision to turn to the IMF in early 1991; the Congress itself, under newly elected Prime Minister P.V. N. Rao, took the onus of devaluing the rupee by 22 percent in July of that year. Overall, since the external crisis was contained, access to foreign loan capital served to support both the political incumbents during most of the 1980s, that is, the Congress (I) Party, and also India's democratic political regime.

The cases of three large Latin American borrowers in the 1970s, and of India in the 1980s, thus generally confirm our suspicions that cross-border private bank lending to the public sector in emerging market countries increases both the economic and the political bargaining power of incumbent governments- unless and until a serious external payments crisis causes the economic management skills of the political leaders suddenly to be called into question.

This essay has suggested that the effects of commercial bank loans to the private sector (Row 4), are likely to include a greater stimulus to economic and industrial growth and an increase in the political weight of large borrowing firms in domestic policy councils. The inflows should increase the bargaining chips of big business vis-à-vis the incumbent government. If local big business firms bring in significant quantities of foreign commercial bank loans, then authoritarian regimes should experience pressure from their entrepreneurs to liberalize, at least in the areas of respect for property, legal and predictable regulatory enforcement, and sometimes human rights. Of course, many of the most closed regimes, from China to Burma, during the 1970s heyday of commercial lending to developing countries lacked a business class that could directly attract foreign bank loans. Other authoritarian countries with private firms that might have borrowed abroad, such as the East Asian tigers like South Korea, operated domestic regulatory and financial environments that strongly encouraged local firms to depend on state banks instead. 44

As it happens, the main examples of developing countries whose private firms directly contracted large quantities of medium and long-term commercial bank loans abroad are precisely those Latin American countries whose governments also were large borrowers, thus confounding our attempts to distinguish the effects of government versus private sector long-term borrowing. One proposition I suggested above seems not to hold true in these cases, however. There does not seem to be evidence that countries that had a particularly large share of direct private sector borrowing abroad, such as Chile or Argentina, spent their borrowed funds, on average, more wisely than those with relatively smaller shares of private in total foreign borrowing. In fact, most observers have considered that Chile, along with Brazil, apparently invested with a reasonable degree of efficiency, while Argentina wasted a large portion of the funds it borrowed (Frieden 1991, 80). One plausible explanation that is consistent with the logic developed earlier, however, is that the private sector in many of these countries, but perhaps in Argentina particularly (see Lewis 1990), was characterized by what has come to be known as "crony capitalism" in which firms' profits depend less on competitive efficiency and more on political connections that give them access to economic rents. Under these political conditions, private borrowers may expect that any exchange rate or other losses they experience are likely to be "socialized" by the central government.

The expectation that big business would increase its bargaining power with the authoritarian central government, and then would tend to push for political liberalization if not full democracy, seems to have described outcomes in Brazil (where business dissatisfaction with military rule was crucial in accelerating democratization, see Kingstone, forthcoming), but not in either Chile (where big business could not even protect itself from waves of bankruptcy when the debt crisis hit in 1982-1983, much less impose a liberalizing political agenda on military incumbents) or Argentina (where crony capitalism prevailed, linking military officers and big business in non-productive rent-seeking). In these three countries taken together, there is not strong evidence that big businesses used their ability to attract foreign loans to enhance their politically relevant domestic resources, or even that big business' political preferences tended toward political liberalization, though not always the transition to full mass democracy. If the expected effect is present, it has been swamped by other, more powerful, influences. 45

I do not know of a good empirical example of a democratic developing country whose private firms directly tapped multinational commercial banks for large quantities of long-term loans. This cell in Table 5 is thus empty.

The fifth category of international financial flows is portfolio flows to the public sector (Row 5). Our expectations of their effects on the borrowing country's political economy were quite similar to those postulated for commercial bank lending. Portfolio inflows to the public sector should provide a lesser stimulus to economic growth than those going to the private sector, but a greater political resource for the sitting government. As noted, portfolio flows differ from bank loans in one important respect: balance of payments crises provoked or exacerbated by large scale overnight capital outflows are much more likely to occur with securities explicitly marketed to their purchasers as being highly liquid.

Mexico during the early 1990s provided a reasonably clear example of several hypothesized consequences of portfolio flows coming into government coffers (see Gruben forthcoming, Elizondo forthcoming, and Molano, forthcoming). 46 The substantial portfolio capital inflows to the government, beginning in the very late 1980s and accelerating in 1993, on the whole strengthened the semi-authoritarian regime, and its domination by the Mexico's longtime ruling political party, the Partido Revolution.rio Institucional (PRI). Foreign purchases of portfolio capital inflows in the Mexican government peso-denominated cetes securities, and later the dollar-denominated tesobonos, enhanced the freedom of maneuver of the PRI political incumbents, just as the earlier commercial bank loans had done in the 1970s. As Mary Ann Haley and Jeffrey Winters (both forthcoming) emphasize, foreign portfolio investors seek political stability. However, as Carlos Elizondo (forthcoming) notes, sometimes a preference for political stability, in practice, can be a preference for a process of gradual, orderly democratization to continue. Thus, in 1994, the enhanced visibility of Mexican politics in the U.S. as a consequence of the North American Free Trade Agreement (NAFTA) had made U.S. investors aware that a possible source of instability in Mexico was popular protests against the continuing legacies of soft authoritarianism in Mexico, such as election fraud. That is, the need to retain the confidence of notoriously fickle foreign investors put pressure on the out-going Salinas administration to run clean elections. Business confidence prior to the election, of course, was helped by the fact that it looked as though the leftist PRD, whose standard bearer, Cuauhtèmoc C.rdenas, had come close to winning the 1988 election, had no chance in 1994. In August PRI candidate Ernesto Zedillo, as expected, won easily.

Unfortunately, the story did not end there. For a variety of reasons, both economically sound and politically opportunistic, outgoing President Carlos Salinas chose not to devalue despite a widening trade deficit. 47 His administration convinced world markets that it would not precipitously devalue by replacing peso-denominated treasury securities, as they came due, with dollar-denominated bonds offering an attractive interest rate. In December 1994, Zedillo's economic team finally devalued the peso by 15 percent, an intrinsically reasonable amount, in terms relative price inflation prevailing in Mexico as compared to the U.S. Nonetheless, the markets panicked. By early February 1995, the peso's value stabilized at 40 percent below its earlier level, only stopping there due to the firm backing of a $50 billion rescue package engineered by the Clinton administration. 48 The Mexican population, which had endured a drop in the real urban minimum wage of more than 50 percent following its 1982 international debt crisis, 49 and only had begun to see economic recovery in the early 1990s, was stunned to discover that further years of austerity were in store.

The consequences for Mexican democracy over the subsequent three years of the peso crisis were fairly consistent with our theoretical expectations. On the one hand, the PRI stopped looking like a miraclemaker, and the stock of opposition parties rose. Greater transparency of political activity, and incremental progress toward fairer formal rules of the game (election procedures and campaign financing, for example) resulted. 50 These trends were democratizing, and contributed to the opposition victories in the mid 1997 elections. On the other hand, the economic consequences of the peso crisis cleanup have, as also theorized, undercut the social and economic positions of several groups whose active participation in writing new "rules of the game" might be expected to ensure full and fair mass participation in democratic processes. As has happened in country after country, middle class unionized civil servants and public sector industrial workers were laid off in large numbers, weakening these groups as political actors. Post-crisis structural adjustment also has made the poor, both urban and rural, more desperate than ever, encouraging the twin responses of apathy and resignation for the majority, along with a turn to violent protest for a minority. 51 The interim bottom line is thus something like this: Mexico's peso crisis has brought greater political freedom and competitiveness, but has heightened economic insecurity and inequality. The fact that government short-term debt was a large part of the problem has put the public sector on a particularly tight rein.

Turning to the case of portfolio capital inflows to the government of an already democratic emerging market country, we could expect many of the same outcomes: the direct positive impact on economic growth probably would be limited, because public sector borrowers have a political as well as an economic investment agenda; while they lasted, portfolio purchases of government bonds could provide a welcome source of fungible resources to the political incumbents—but portfolio inflows would be prone to sudden reversals, possibly bringing political as well as economic crises for the country and/or its rulers. Argentina in the early 1990s, under its President Carlos Menem of the Justicialist (Peronist) Party, was a newly democratic country that allowed itself to sell large quantities of government debt to buyers bringing capital from abroad. 52 As it happened, the Menem team had occasion to regret its reliance upon portfolio inflows for government deficit financing. When Mexico's peso crisis exploded in late December 1994, first world investors, as had happened in 1982, tended to generalize their pullout across other emerging markets, especially in Latin America, without a great deal of regard for the possible differences among countries. In early 1995 Argentina, Ecuador, Brazil, and others experienced the "tequila effect" of a backlash from Mexico's crisis. 53 While Menem, along with Argentina's decade old democracy, survived the dramatic ebb of capital, the government budget (which thereafter incorporated much higher interest rates on the public debt) and the Argentine economy (where an incipient recovery from the recession of the 1990s was dashed) paid the price.

The tequila effect was not sufficiently severe to have unbalanced Argentina's new democracy. However, it is not too difficult to imagine a scenario in which it might have. President Menem himself was quite lucky in the timing of Argentina's narrowly averted financial crash: the movement to alter the Constitution to allow him to run again, and his victory in that election, already had passed by early 1995 when the run on the Mexican peso began to batter the Argentine peso. Had the Argentine electoral calendar been otherwise, he might well have lost his bid for reelection.

The final type of cross-border capital flow is portfolio investments in private sector securities (Row 6). With an authoritarian regime in the capital-importing country, our expectations would be that such flows would make a potentially large contribution to private investment by the country's largest firms (that is, those whose shares constitute the blue chip investments in the local stock exchange, or who are able to issue bonds or commercial paper in global markets), that the heads of these firms would tend to join foreign direct investors overtly or covertly in pushing for greater political liberalization (in the sense of a rule of law and civil liberties protections) without necessarily risking too much to call for full democratic elections, and that the desire to avoid the ever-present potential of a major capital outflow would lead to a fairly cautious and deferential attitude on the part of the authoritarian rulers toward their business supporters. Preemptive neoliberal economic policy reforms also should be likely.

South Africa in the 1970s and 1980s provides an interesting case. Rich in diamonds, gold, and minerals, it long had one of the most active stock and securities exchanges outside the major industrial democracies. It also had a special kind of authoritarian political system, one that provided liberal democracy for a minority, but used apartheid to exclude the black majority from economic, social, and political power. Generalizing very roughly, in recent decades black Africans were the workers and small farmers, Afrikaaners the civil servants and commercial family farmers, while English-speaking whites dominated the liberal professions and upper reaches of business. British capital long was the largest source of foreign investment, both direct and portfolio, in a tradition dating back to the late nineteenth century. Under these circumstances, foreign capital inflows to the local private sector on the one hand tended to strengthen the white minority, the elite who benefited from authoritarian rule over all non-white residents. At the same time, business links to the outside probably strengthened the English-speaking minority within the overall white minority that ruled the country. Under these particular demographic circumstances, the predicted politically liberalizing effect of capital inflows to the business class was muted by the solidarity of the white community as a whole. Nonetheless, the English-speaking white community was somewhat more open to a gradual transition to democracy than the Afrikaaner one, perhaps partly because of its greater familiarity with the larger world through international business links. As was the case in our examination above of commercial bank lending to the private sector in authoritarian Latin America, there does not seem to be compelling evidence that private portfolio inflows to the South African private sector during the 1970s and 1980s was a great source of pressure for economic efficiency, although my judgment on this point is quite impressionistic (see Daniels and Daniels 1995).

Another obvious case is that of Singapore, in the 1980s and early 1990s both an Asian "tiger" and a hot emerging stock market. As of the mid 1990s, Singapore's longtime strongman and patriarch, Lee Kwan Yew, was respected at home and throughout the region because of the country's great economic advances under his tutelage. At the same time, Lee and his proteges, including Prime Minister Goh Chok Tong, could not afford to alienate their business leaders, on whom the country's prosperity and ultimately their own political tenure depended. In the case of Singapore, many business families within the country also had extensive ties with the overseas Chinese community around the world, including those who had located in the advanced industrial democracies. The Singapore business community was quite far from sharing, for example, the political opinions of radical students who wished to move rapidly to full democracy. However, the internationally competitive business community also was sensitive to its image abroad, collectively wincing when, for example, a Japanese popular song mocked Singapore as the place where chewing gum was a banned substance. Financial globalization in Singapore had resulted in a somewhat freer press, a dawning but vigorous debate over the pros and cons of paternalistic government, and signficant moves toward political liberalization- stopping well short of formal representative democracy- over the decade ending in the mid 1990s. Interestingly, Singapore to mid January 1998 had escaped serious damage from the East Asian financial crisis, at least by comparison with its neighbors. This was probably because there are important distinctions among the various types of new cross-border portfolio investments in the emerging market country's private sector, beyond the obvious differences between portfolio flows and the other less volatile flows discussed in this book. Singapore, although an important emerging stock market, did not suddenly in late 1997 reveal large amounts of short-term foreign borrowing by local banks- a major source of trouble in Thailand, Indonesia, and South Korea.

Indonesia is another authoritarian country that in the 1990s received increasing amounts of portfolio capital inflows into its private sector, particularly the banking system. Like Lee in Singapore, Indonesia's Suharto benefited from large portfolio capital inflows, in that they made his regime and government look like excellent economic managers. However, and unlike Singapore, Indonesia has had a serious financial crisis; as January 1998 the currency had dropped 80 percent against the U.S. dollar, compared to its level in early 1997. What are the apparent implications for democracy? Jeffrey A. Winters (forthcoming) is struck by the much more facile and effective social and political response to the financial crisis in democratic Thailand and South Korea, as compared to authoritarian Indonesia. President Suharto's initial response to the crisis seems to have been to promise the IMF whatever it wanted, and then, in his government's budget of early January, as well as apparent plans for his eventual successor to be his high-spending Vice President Habibie, to have attempted to continue with crony capitalism- with large areas of the economy reserved for his own family- as usual. The incumbent authoritarian government was proving extremely resistent to neoliberal economic reform- even after a violent balance of payments crisis, perhaps partly because, through January 1998, Indonesia's poor and middle sectors, but not yet its top elites, had paid the bulk of the costs of the crisis.

I close the case studies with a look at portfolio flows to the local private sector in three democratic or democratizing emerging market countries: India, Thailand, and South Korea in the 1990s. Although all types of foreign capital inflows tend to assist political incumbents- so long as a balance of payments crisis is avoided- I argued above that those flows whose local disposition is in the hands of the private sector could be more problematic for the often weak or fragile democratic governments in emerging market countries. Portfolio flows controlled by private entrepreneurs should stimulate productive investment while empowering local big business as a potential counterweight to the national political leaders.

India had been securely democratic for decades. Its portfolio flows to the private sector in the 1990s, in contrast to most of the East Asian countries that had trouble in late 1997, had been mainly in the form of equity and bond flows, particularly those associated with large Indian firms raising funds directly by securities floated in international markets, Global Depository Receipts (GDRs). Although even limited entry by private investors directly into Indian capital markets has had an important influence on the shape of Indian financial regulations, as detailed by John Echeverri-Gent (forthcoming), the Reserve Bank of India, the country's central bank and main commercial bank regulator, placed strict limits on the amount of short-term foreign loans and deposits that commercial banks could contract, resulting in a regulatory regime with substantial capital controls still intact. This regulatory caution appears to have paid off; through mid January 1998, India had not yet caught the "Asian flu."

India also is a case where our predictions about the political and policy consequences of capital inflows to private big businesses provide a plausible description of reality. On the one hand, the private sector became increasingly important to the balance of payments in the 1980s and early 1990s, both as a source of exports and as an investment destination for foreign funds. Economic policies, meanwhile, became more market-oriented, particularly since 1991, and the profession of businessperson, by all accounts, has higher status and greater attraction for the "best and the brightest" than it did in the first three decades after independence in 1947. There is a chicken and egg problem here, as it is hard to know whether businesspersons have become more politically and socially prominent as a consequence of economic policy changes initiated by government bureaucrats and senior politicians, or whether the generally pro-market preferences of an increasingly confident private sector wield more influence than previously. Both are probably true. The causes of India's recent economic liberalization are in any case overdetermined (Varshney 1996, Echeverri-Gent 1997). Echeverri-Gent (forthcoming) suggests a new twist on the "optimistic hypothesis" about the effects of private foreign capital flows: by forcing greater regulatory transparency and breaking up cosy oligopolies, democratic integrity is preserved and deepened, as small businesses and small investors get a more level playing field and, most importantly, a corrupting influence from the crony capitalists to the government is dampened. At the same time, he concedes that trade and financial globalization in India in general already have widened interregional income disparities- and probably will continue to do so, with potentially ominous implications for democratic and political stability.

The final examples are Thailand and South Korea, newly democratic countries that received large amounts of portfolio capital inflows to their private sectors in the 1990s, with particularly large flows of short-term hard currency debt, often borrowed by local banks who then would relend in local currency, making their profits on the spread.

Danny Unger (forthcoming) summarizes Thai political economy since the mid-1970s. While agreeing that local big business prefers political liberalization to full democratization, Unger sees the strengthening of local private capital in Thailand in the 1980s and 1990s as an important counterweight to a still very powerful military-bureaucratic-civil service complex, which the proliferating clientelistic political parties had utterly failed to rein in. That is, given Unger's assessment of the stage of Thai democratization, he sees redistribution of economic and political power in the direction of local big business prior to the financial crisis as, on balance, having increased political competition and accountability. Moreover, although he is less explicit on this point, Unger tends toward the "optimistic" interpretation of the post-crisis increased indirect influence of foreign institutional investors on Thailand's economic policy choices, suggesting that sounder macroeconomic and regulatory policies may result. It also is possible to imagine that the balance of payments crisis could weaken Thai democracy, at least in the short-run, by reinforcing the impression that civilian politicians are both incompetent and corrupt, and implying that the Thai military is neither. However, as both Winters and Unger note, Thai adjustment to its financial crisis in late 1997 and through January 1998 was reasonably rapid, arguably because of its comparatively democratic political game.

A related dynamic seemed to be at work in Korea. That country's balance of payments crisis hit before the December 1997 presidential election, in contrast to events in Mexico in 1994, and served to discredit both the conservative government of incumbent President Kim Young Sam (a former dissident who owed his election to a rapprochement with the outgoing military rulers) and the chaebol, powerful and long-favored oligopoly business conglomerates (Kim 1996, Amsden 1989). Longtime dissident Kim Dae Jung won a plurality in a three-way race and used the opportunity given him by the crisis to win striking early negotiating victories even before formally assuming office. 54 As in Thailand, the financial crisis may prove to have a silver lining in that it gives reforming democratic governments some leverage to reduce some of the cosy ties between big business and the state. However, as in Mexico, those who suffer most in the short run will be those in the lower, and to a lesser extent the middle, income groups. Political democratization may be accelerated, but the economic pain will be considerable. Furthermore, if the Rueschemeyer, Stephens, and Stephens (1992) analysis is correct, then the apparent weakening of groups such as labor unions during the transition to democracy might hold worries for the future. The alternative hypothesis is that, by being flexible and "reasonable" in this time of crisis, Korean union leaders may permanently earn themselves a place at the table. Moreover, to the extent that the foreign investor community sees cautious progress in democratization as a key indicator of that highly desired quality, "political stability" (see Elizondo, forthcoming), then the financial crises in Thailand and South Korea may work to secure those countries' democratic openings.

Conclusions

What have we learned? This essay has presented an organizing framework, not a rigorous test. Nonetheless, the following conclusions seem consistent with the evidence presented.

1. I suggested that, contrary to the too-facile expectations of many in the business and policy communities, economic growth does not necessarily promote political democracy, at least not for poor and middle income countries. This lesson seems implicit in the stories told of Mexico and Brazil before the 1982 debt crisis and Indonesia and Vietnam in the more recent period. What economic growth does do, for as long as it continues, is to legitimize incumbents. Conversely, however, where authoritarian regimes have wrapped themselves in the flag of successful economic management, they can be denuded rapidly by financial blowups. This happened to authoritarian rulers in Argentina, Uruguay, and Brazil in the early 1980s (though Chile's General Augusto Pinochet held on), and was a factor in the overthrow of Philippine dictator Ferdinand Marcos and gradual liberalization of Mexico's civilian one-party state.

2. It seemed, and seems, reasonable to suppose that capital inflows controlled by the private sector within the emerging market country, or by foreign direct investors, would be allocated more efficiently, resulting in greater increments to growth. This is not always the case, however. As noted, both Chile and Argentina received a greater share of their foreign bank borrowings in the 1970s as loans to the private sector, while Brazilian loans were contracted both by private firms (as non-guraranteed loans) and by the public sector (Frieden 1991). Chile and Brazil seem to have invested the borrowed funds comparatively efficiently, while Argentina has a reputation for just the opposite results. The difference plausibly results from dissimilar regulatory frameworks (in turn, perhaps, a result of political differences). Walter Molano (forthcoming) is implicitly provocative on this point, as he suggests that the long-run economic consequences of the Thai and East Asian crises of late 1997 could be more serious than the Mexican crisis of 1994-1995, precisely because the borrowers in East Asia were in the private sector, whereas Mexico's peso crisis was first and foremost a crisis of the government's ability to meet its short-term debt obligations. This apparent anomaly- that capital flows coming to the local private sector might produce deeper economic crises than public sector portfolio borrowing- is perhaps explained by the fact that the many of the private borrowers in East Asia were banks and financial institutions operating in newly liberated capital markets with inadequate regulatory oversight (see Amsden and Euh 1997). In fact, it is an emerging truism, if I may be forgiven the expression, that greater transparency and predictability in national financial regulation- often missing in emerging markets—would be a good thing, whether for better economic results in the capital-importing country (see Manzocchi, forthcoming), for foreign investors (Molano, forthcoming), for small investors and ordinary citizens within the emerging market country (Echeverri-Gent, forthcoming), or even for the institutionalization of democracy itself (Stone, forthcoming).

3. Different institutional forms of cross-border capital flows would, I thought, promote the fortunes, and local political influence, of four different political actors: foreign governments and international organizations, incumbent governments in the capital importing country, foreign direct investors, and local big business. The evidence of increments of both political and economic influence accruing to foreign lender/donor governments and/or to borrower governments as a consequence of different institutional forms of cross-border capital flows is straightforward and consistent with these predictions. With the relative decline of foreign aid, for example, not only have borrower governments lost resources, but foreign governments (often to their surprise) also have lost influence. It is worth pondering the consequences of the fact that the currency trader George Soros in the early 1990s gave more aid to promote democracy in Central and Eastern Europe than the U.S. government. 55

With respect to foreign business, my initial decision to call the influence of multinational direct investors "political" but to exclude the indirect influence of global private portfolio investors (before any financial crisis) from this framework seems to have underemphasized important facets of the story. Thus, for example, Haley, Porter, Stone, and Echeverri-Gent (all forthcoming) each detail some of the different ways in which the mostly neoliberal preferences of private global investors (in the case of Russia, Russian owners of wealth who had the connections to engage in capital flight) have constrained public policy choice in emerging market countries, even when the foreign investors did not actively lobby incumbents. Moreover, once a financial crisis looms, private portfolio investors may become directly involved, as in the cases of foreign institutional investors in Mexico in 1994-1995, or private multinational banks who made short-term loans to domestic banks in East Asia in 1997. Tellingly, South Korean president elect Kim Dae Jung granted George Soros an audience in the first two weeks after his December 1997 upset victory, making sure the financial endorsement of Korea implied by the meeting was well photographed (Burton 1998).

The findings on the political consequences of foreign inflows coming under the control of local private business firms were intriguing but difficult to generalize. In some authoritarian countries, from Chile and Argentina in the 1970s to Indonesia in the 1990s, the private business community was so dependent upon the political authorities that even an ability to attract foreign capital inflows seemed not to give it much of an independent voice. Nor was it clear that these borrowers favored even limited political liberalization. In fact, strengthening business families close to the regime has often propped up authoritarian rule indirectly- at least unless and until a financial crisis shook the system (see Winters, forthcoming).

In other developing countries, private business has been a vocal and somewhat independent political actor. Where private capitalists have been able to attract foreign capital without the assistance or intermediation of their government, they have often asserted themselves in domestic policy debates. In both authoritarian and democratic Brazil politicians needed business support at least as much as the reverse (see Kingstone, forthcoming). Private business borrowers in newly democratizing countries such as Thailand and South Korea, or newly market-oriented democratic countries like India, also were flying high before the late 1997 crises, partly because of their importance to the balance of payments.

Danny Unger (forthcoming) suggests that increased influence and independence for the domestic business community was a welcome source of pluralism, given Thailand's long tradition of a centralized, closed, state bureaucracy. With the late 1997 crisis, both government and business leaders lost credibility with the Thai public, although the medium run consequences of this remain unclear. Similarly, Korea's large integrated financial-industrial conglomerates, the chaebol, in the mid 1990s had used their access to foreign funds as yet another advantage vis-a-vis small firms. The chaebol have been humbled by the late 1997 financial crisis, quite plausibly strengthening President-elect Kim Dae Jung's hand in enacting needed regulatory reforms.

India's business community in the 1990s both aided the country's balance of payments by being able to tap global capital markets and increased its public voice in the policy arena, although the precise connection between these two trends is unclear. John Echeverri-Gent (forthcoming) argues that the greater transparency demanded by foreign institutional investors as a quid pro quo for bringing their monies into the country has had the salutory effect of curbing corruption in local business practices. This, he concludes, strengthens democracy, because it reduces the incentives for local business leaders to attempt to suborn politicians to cover up corruption. That is, while Indian big business has been strengthened in some ways, it also has been held to account and disciplined in other respects. We probably can conclude that direct loans to and/or investments with the local private sector does increase big business' control over politically-relevant resources. How much, and with what political consequences, cannot be answered outside of the context of individual countries.

4. There can be little doubt that balance of payments crises are bad for political incumbents, both in terms of the credibility of their economic policies, and the length of their political tenure. On the other hand, and provocatively, democratizing polities that are relatively pluralist and open to policy debate may be more able to cope with financial crises flexibly and effectively than authoritarian regimes. 56 Thus, in late 1997 both South Korea and Thailand changed both economic policies and chief executives while continuing their democratic opening. By January 1998 Indonesia had not made either shift. Both China and Vietnam have been protected from the Asian financial crisis thus far by a combination of reliance on FDI rather than portfolio flows, plus continued capital controls (Haughton, forthcoming). It is hard to imagine the nondemocratic, secretive regimes in either country responding successfully to a financial crisis should one hit.

5. I also expressed concerns over the medium term political consequences of the worsening of income distribution that often- or always?- accompanied post-balance of payments crisis structural adjustment programs. Even more ominously, the same effects on income distribution (and the subsequent distribution of political influence), I argued, might be expected from preemptive neoliberal economic reforms of the sort that emerging market country governments often have had to enact in order to head off a financial crisis that threatened. This hypothesis is hard to evaluate because it requires the construction of counter-factual scenarios, but a few concluding words may be relevant.

The post-1982 debt crisis period of structural adjustment in Latin America is the best source of data, since the waves of financial crisis associated with the portfolio flows of the 1990s occurred too recently to evaluate them. Three trends were apparent. First, poverty and inequality worsened dramatically throughout Latin America during the debt crisis decade. Moreover, neither poverty nor inequality was reduced in the early 1990s, even when most countries of the region began to grow again. 57 Second, procedural political democracy took hold and deepened throughout the region. 58 A contributing factor was the discrediting of military regimes throughout the region that had legitimized political repression in the 1970s by rapid economic growth. Third, the room for public policy maneuver by Latin American governments in the 1990s clearly had shrunk by comparison with previous decades, particularly the 1950s through the 1970s. The neoliberal policy preferences of the late twentieth century's global portfolio investors led them to shun countries that adopted leftist policy agendas that, for example, redistributed land or other assets, made the tax system more progressive, increased social spending, and so on.

From the viewpoint of the distribution of political power within Latin American emerging market countries the news was thus both good and bad. On the one hand, political pluralism and the openness and competitiveness of politics increased as a consequence of the external debt crisis of the 1980s. On the other hand, those groups whose politically-relevant resources already were scarce had their economic power diminished even further just as the rules of the successor democratic games were being negotiated. Meanwhile, the debt crisis also had negative consequences for the international distribution of political power between the governments of developing countries and foreigners, both private investors and their home governments (see Porter, forthcoming). 59

The optimistic comeback to these observations might make some of the following points. Latin America is not the world; there is thus no reason to believe that its experience of structural adjustment predicts anything about East Asia, Eastern Europe, or anywhere else. 60 After all, income distribution was infinitely more unequal in Latin America in 1980 than in East Asia in 1995; perhaps initial inequality is the main determinant of structural adjustment that makes the already poor relatively poorer. Moreover, international trade theory predicts that the locally abundant factors of production- typically including unskilled labor in emerging market countries- will see their relative economic returns rise as a consequence of greater integration with the global economy, because these factors will be relatively scarcer on the world stage (Stolper and Samuelson 1941, Frieden and Rogowski 1996). 61 The recessions and cuts in government spending associated with structural adjustment may be particularly burdensome for the poor, yet inflation stabilization, also demanded by global private portfolio investors, disproportionately benefits lower income groups, since the poor and unsophisticated typically lack access to the diversified asset portfolios that upper income groups use to protect themselves from price volatility (see Stone forthcoming, Armijo 1998). Finally, and most persuasively, neoliberal reforms, although they cause transitional pain, plausibly improve the functioning of the economy in the medium and long-run. In particular, they eliminate most opportunities for politically protected, but economically wasteful, rent-seeking by petty government bureaucrats. 62

6. The most important conclusion, finally, is that little can be said about the likely consequences of the changing forms of foreign capital flows for democratic development in emerging market countries in the absence of knowledge about the current political system. In general, net inflows of all types boosted incumbents, while sudden net outflows destabilized them. Portfolio flows- including stocks, bonds, and short-term debt- have in the mid 1990s, unfortunately, more than lived up to their reputation for volatility.

There thus are some reasons to believe that current trends in global financial markets are not entirely favorable for the oft heralded turn to democracy by developing countries in the late twentieth century. Still, given the conditions of global economic competition today, most developing countries would be worse off in the absence of foreign capital inflows, whatever their form. So long as they do not exit immediately, most capital inflows do promote economic growth in the recipient country, particularly when the funds are invested with an eye to market rather than political criteria. In the long-run (if not necessarily during the interim) capitalist economic growth does seem to be associated with liberal democracy. Liberal democracy ("procedural political democracy"), in turn, provides at least limited protections to the economically disenfranchised, and is a political system superior to any variety of authoritarianism (military-technocratic, fascist, Communist, theocratic, or whatever) for all but the dictator and his or her group. This conclusion's implicit policy recommendation, therefore, is not the naive advice that emerging market countries somehow ought to resist the blandishments of foreign portfolio investors. Rather, the lesson may be that developing countries (both governments and citizens) consider opening their markets up to global financial flows at least somewhat cautiously and with an educated awareness of the risks involved. 63

Notes

Note 1: I thank Shahid Alam, Thomas J. Biersteker, Thomas Callaghy, Susan Christopherson, John Echeverri-Gent, Mary Ann Haley, Rebecca Hovey, Atul Kohli, Mukul Majumdar, Luigi Manzetti, Sylvia Maxfield, Geraldo Munck, Dale Murphy, Sanjay Reddy, Ben Ross Schneider, Moises Schwartz, Danny Unger, Birol Yesilada, Fei-Ling Wang, and the participants in the conference on "Financial Globalization, Economic Growth, and Democracy in Emerging Market Countries" at Brown and Northeastern Universities (November 1995) for their helpful comments on various versions. Back.

Note 2: I use data from the World Bank's annual World Debt Tables, but combine the data slightly differently than the World Bank, which does not change the broad trends. Back.

Note 3: A large share of bank loans in the 1980s was not voluntary lending, but instead "exceptional financing" in which transnational money center banks agreed to loan countries like Mexico and Brazil "new money" so they could make interest and principle payments due on past debt- thus saving both debtor country and creditor banks from the pain of a formally declared default. Back.

Note 4: Many Latin American and African debtors had negative net transfers. In 1986, for example, Mexico's net transfers abroad totaled 5.3 percent of its GDP, while Brazil's summed to 3 percent (World Bank, 1995). Back.

Note 5: The big question is whether heightened volatility is inevitable, because of advances in computer and telecommunications technologies. Helleiner (1994) argues that financial globalization has only come about because of political choices (including the deceptively passive choice not to regulate) made by the leading industrial capitalist economies. Financial globalization also has been propelled by structural changes occurring within the domestic economies of the OECD countries. Two institutional and regulatory trends that began in the early 1980s, securitization (the bundling together and resale of in the capital markets of long-term lending commitments formerly held to maturity by banks and other financial institutions) and the large shift of household savings in OECD countries away from bank deposits and toward investments in mutual funds and other institutional investors, have occurred in tandem with the technology-driven inauguration of twenty-four hour global trading. Domestic and international financial market changes thus reinforced one another. Back.

Note 6: See also Manzocchi forthcoming. Back.

Note 7: This essay is not, of course, the first to theorize about the impacts of different institutional forms of foreign capital inflows on recipient countries. Barbara Stallings 1990 suggested that an important influence on the more positive growth experiences of Korea and Taiwan, as compared to Mexico and Brazil, in the 1980s might have been the institutional form of foreign capital inflows. The two East Asia countries relied more on capital from public sector sources and on loans. In contrast, the Latin American countries imported more private capital and direct investment by multinational corporations, both of which compromised host country autonomy. Furthermore, by the 1980s, the two East Asian societies had raised domestic savings rates sufficiently to reduce significantly their overall reliance on foreign inflows. A few years later, Stallings and Stephany Griffith-Jones noted that the institutional forms of capital flows again had shifted. In that paper, they considered both long-term loans and FDI more advantages forms of capital inflows, from the viewpoint of capital importers, than portfolio equity, while noting that, as of 1991-1992, Asian developing countries were more fortunate on these grounds than Latin American and Caribbean ones (Griffith-Jones and Stallings 1995: 158). Back.
Jeffery Winters 1994, esp. pp. 446-450, proposed a "framework for analyzing capital control and end use." He noted that recipient country governments have low discretion over the investment uses of either portfolio flows or foreign direct investment, medium abilities to control interstate loans (that is, official credits), somewhat greater options for investing private commercial loans as they please, and the greatest degree of discretion over "state capital" or revenues raised domestically, through taxes, borrowing, etcetera.
Sylvia Maxfield 1995, esp. pp. 12-15, was interested in the degree of influence different types of foreign investors have over the economic policies of recipient governments, positing the greatest leverage for portfolio equity investors, an intermediate amount for longterm bondholders, and the least leverage for foreign direct investors and commercial bank lenders. The determining factors for her were the costs to foreign investors of monitoring recipient country performance and, most importantly, the ease of capital repatriation.
Each of these analysts was concerned with the links between capital inflows and recipient country government autonomy and/or economic growth. None explicitly attempted to pursue the further possible link between the institutional forms of capital inflows and democracy.

Note 8: For an alternative way to think through the meaning of "democracy" in developing countries, particularly in Latin America, see Karl 1990. Back.

Note 9: Obviously the definition of democracy adopted by Communist countries, or "peoples' democracies" does not include procedural political democracy as a component. Some theorists commited to economic equality as the foundation of democracy also see procedural political democracy as, at best, a smokescreen for oppression, and, at worst, an instrument of inequality and thus of the lack of democracy. See, for example, MacPhearson 1966. Back.

Note 10: I exclude military assistance from this discussion for two reasons. Most military aid is in the form of contributions in kind, rather than financial flows. In addition, the rationale for military aid is entirely political and strategic, whereas economic aid may have some investment justifications as well as its overtly political ones. Back.
For the sake of analytical simplicity, I also exclude that small, albeit growing, portion of concessional foreign lending that flows directly to private sector recipients in the developing country, such as the loans made to local entrepreneurs by the World Bank affiliate the International Finance Corporation (IFC).

Note 11: In some countries MNC investors have been encouraged to enter joint ventures with state-owned enterprises (SOEs). In principle, one could work through the analysis separately, first for MNC investors who formed wholly-owned subsidiaries or joint ventures with a local private partner, and second for MNC-SOE joint undertakings. Back.

Note 12: Due to limitations of the data, the statistics reported in Table 1 assume that all "government-guaranteed loans" are loans directly to the government, which of course overstates the size of my third category, although it is true that governments in capital-importing countries tend to exercise greater oversight over those foreign loans to local private firms that they guarantee. Categories three and four both bundle medium and long-term trade credit ("other guaranteed medium and long-term debt") with commercial bank loans. Back.

Note 13: In fact, some of the largest institutional investors in global markets actually are the pension funds of public sector employees, especially state and local government workers in the U.S. To the extent that pension fund managers attend only to maximizing profits and minimizing risks, their behavior mimics that of the managers of private sector pension or mutual funds. Back.

Note 14: It is difficult to know whether short-term debt flows are being borrowed by the emerging market country government or the local private sector, except on a case by case basis. Thus, the decision to include all such flows with my category six, portfolio inflows to the local private sector, has the result of overstating the size of this category relative to category five, portfolio flows to the government. Back.

Note 15: While many economists, perhaps a majority, would agree that portfolio capital flows are more volatile than, say, direct investment, opinions do differ. See endnote #4 in Gruben forthcoming.
Sylvia Maxfield 1997 suggests that portfolio investors may differ among themselves in volatility, with those seeking high yields (typically hedge and mutual fund managers) more likely to bolt at the slightest hint of crisis than investors whose principle goals are value and diversification (such as pension fund managers). Back.

Note 16: There is an important caveat. Good, comparable data on all of these categories is difficult if not impossible to unearth: the World Bank's annual World Debt Tables, which only started reporting non-debt capital flows such as FDI and portfolio equity in the 1990s, probably is the best source. Particularly difficult to measure with existing data is the distinction between public versus private recipients of commercial bank loans or portfolio flows: that is, easily available data sets do not clearly differentiate between my categories three and four, or five and six. Back.

Note 17: I thank Stefano Manzocchi for this point. Back.

Note 18: In what John Gerard Ruggie in 1982 termed the "compromise of embedded liberalism" advanced industrial countries during the postwar decades extended to one another the privilege of deviating in their domestic economic regulatory regimes from the market liberalism they all espoused for the international economy. Thomas Callaghy (1989, 1993) points out that this privilege- which might be summarized as the right to selective domestic protectionism along with mostly free access to other rich countries' domestic markets- has never been extended to most developing countries. The exceptions have been those countries, such as South Korea or Taiwan, or Francophone Africa, that had strategic value for the most powerful advanced industrial countries. See also Stallings 1995. Back.

Note 19: On neoliberal (also known as "orthodox" or "neoconservative") economic policies in Latin America, see Pastor 1992, Foxley 1983. Back.

Note 20: Chase-Dunn and Bornschier 1985 detail some of the ways in which the presence of multinational direct investors skews the host country regulatory environment in a more conservative, neoliberal direction. Back.

Note 21: There also is pressure from global investors for capital-importing countries to construct regulatory frameworks consistent with norms in the advanced industrial democracies. See Porter, forthcoming and Echeverri-Gent, forthcoming. Back.

Note 22: See, for example, the concerns expressed in Cerny 1993; Strange 1986. Back.

Note 23: O'Donnell 1973 hypothesized that the breakdown of previously existing democracy in Brazil in 1964, Argentina in 1966, Chile in 1973, and Argentina again in 1976 (after a three year democratic interlude) came about because these societies already had passed through the "easy" stage of import-substituting industrialization (ISI), during which period the usually combatitive forces of industrial capital and labor both could become wealthy together. The harder stage of ISI, in contrast, would require diversion of all the surplus created by industrial production into profits and new investment, leaving nothing available for added increments to workers' wages. Consequently, went the argument, authoritarianism was the only economically viable political system, because under electoral democracy, workers would demand concessions that inevitably would produce economic stagnation. Whether or not one accepts O'Donnell's explanation of the phenomenon, Latin America's experience during these decades definitely calls into question the assumption of mutually reinforcing and linear political and economic progress. See also Collier, ed. 1979. Back.

Note 24: Obviously, they are far from the first to raise this concern. However, their recent articulation of these somewhat familiar arguments stands out for its modulated tone, careful historical scholarship, and cross-regional empirical and theoretical investigation. Back.

Note 25: Their argument, while different in many particulars from that of O'Donnell referred to earlier, is not inconsistent with either his empirical observations or his theoretical explanation. Back.

Note 26: I thank Ben Ross Schneider for first bringing the "optimistic hypothesis" to my attention. Back.

Note 27: The term was coined by O'Donnell 1973. Back.

Note 28: Some businesses, of course, have profits that mainly depend upon monopoly or oligopoly rents deriving from excessive government regulation with particular characteristics that favors their interests. Oligopolists dislike market liberalization at least as much as those who favor government intervention to redistribute income, protect the weak, or solve collective action problems. That is, my assumption of uniform policy preferences across the business community is a simplication of reality. On balance, it is usually true that the business community as a whole will prefer more of the neoliberal policy agenda than many other domestic political actors in developing countries. However, the greater the degree to which the activies of local big business community are fairly characterized by the term "crony capitalism" the fainter its likely support for truly free markets. (Of course, local oligopolists probably will dislike the policy agendas of leftist reformers even more.) Back.

Note 29: Individual big businesspersons may well prefer mass democracy for altruistic and normative reasons; they also may associate democracy with modernity. There are plenty of empirical examples of such behavior, as in the business tycoons who secretly financed Mohandas Gandhi or Nelson Mandela. Back.

Note 30: The structural position, and consequent political preferences, of ambitious entrepreneurs in post-Communist regimes is more complex. They tend to be strong supporters of political democracy, which they, like many of their fellow citizens throughout the national class structure, rather unanalytically associate not only with political freedoms but also with the capitalist economic prosperity of the Western industrial democracies and Japan. The political attitude of business in many countries in Eastern Europe and the former Soviet Union is further complicated by the connections many new businesspersons have with the old Communist apparatchik class (who often have been the only group with sufficient capital to buy privatized state firms) and/or with criminal organizations (who ran the thriving underground "capitalist" economy during the Soviet era). Back.

Note 31: Any group whose main political resource is sheer numbers (of potential voters) or good organizational skills (useful for building a strong, grass-roots political party base) benefits from a shift away from authoritarian political rules (which benefit an elite of some kind, whether membership in that elite is defined by heredity, loyal membership in the ruling political party, or control of the means of production) to the political rules of mass electoral democracy. Once again, my argument regarding the importance of "labor" as a political actor is somewhat different- and rather more simplistic- than that of Rueschemeyer et al. 1992. It is not, I think, inconsistent with their analysis. Back.

Note 32: For a game theoretic treatment of why this should be so, see Sturzenegger 1995. For evidence in the Latin American case, see Oxhorn and Ducatenzeiler 1998. Back.

Note 33: Social science researchers generally assume that human social systems, whether families or national political systems, assume patterns of behavior that, once regularized, tend to persist until upset by some crisis. See Stinchcombe 1968. Back.

Note 34: An important caveat is that lower-income groups may be disproportionately benefited by an early end to inflation. See Stone, forthcoming, and also Armijo 1998. Back.

Note 35: For a more skeptical view of the growth prospects of neoliberal economic reforms, see Przeworski et al. 1995. Back.

Note 36: The countries assigned to particular cells had larger than modal capital inflows of the type indicated; however, the type of capital flow indicated, as for example "bank loans to private firms" was not necessarily the single most important source of foreign capital inflows for that country in that period. Back.

Note 37: Some observers argue that the World Bank, for example, was a significant influence moving Philippine economic policy in a generally neoliberal direction. See Broad 1988. In my view, the portfolio investment received by the country in the 1990s has been infinitely more effective in producing macroeconomic policy changes in an orthodox direction. Back.

Note 38: On U.S. policies toward sub-Saharan Africa, see Clough 1992. Back.

Note 39: A plurality of bilateral donors also came in handy for India and some other aid-recipients. When Western foreign aid donors attempted to impose economic and political conditions Indian Prime Minister Indira Gandhi found onerous (e.g., U.S. pressure to devalue in 1966, U.S. opposition to Indian involvement in the war over Bangladesh in 1971), then the Indian goverment sought alternative sources of foreign assistance from the Soviet Union. Back.

Note 40: In general, the notion that MNC investors help promote "sound" economic policy should not be taken too far. MNCs also can tolerate large deviations from impersonal, market-oriented economic policies when their own profits are large. For example, many MNCs have been happy to engage in high-cost, tariff-protected production for local markets. Back.

Note 41: Foreign funds permitted, for example, economically questionable but politically useful policies such as the purchase of new weapons for the military (in Argentina), expansion of export and farm subsidies (in Brazil), and extension of generous benefits to members of public sector unions (in Mexico, Argentina, and Brazil). Back.

Note 42: In Brazil, for example, accelerating levels of inflation and public sector debt seemed an acceptable tradeoff to governing elites as long as rapid economic growth continued and foreign capital inflows kept domestic interest rates from rising significantly (See Fishlow, 1989). Back.

Note 43: The payments crisis' proximate causes were the rupee's overvaluation and the sudden strain caused by the several effects of the Persian Gulf War, from higher oil import prices to the loss of substantial workers' remittances from Indians employed in Kuwait. When the financial scare hit, the main capital outflows were private portfolio investments in so-called NRI (non-resident Indian) deposit accounts; nonetheless, the fact that India's total foreign debt had risen to close to $90 billion constituted an additional worry Back.

Note 44: On the role of state control of financing in South Korea, see Kim 1996, Woo 1991. Back.

Note 45: Jeffry Frieden (1991) suggests what the missing variable might be. He explains variations in local business' policy preferences between Brazil, the members of whose business community competed with one another for sector-specific subsidies from government, and Chile or Argentina, where business accepted radically pared down government and neoliberal policies, by reference to Chile and Argentina's higher historical levels of class conflict, which made private business more dependent upon the state's repressive apparatus. Thus the Brazilian business community, less dependent upon the state for internal security, felt free to demand both more economic benefits from government- and more rapid and substantial political liberalization. Back.

Note 46: By early 1994 foreign participants accounted for as much as 70 to 80 percent of daily trading in Mexico's stock market (Fidler and Frasier, 1994). They also held around 40 percent of outstanding federal government debt, according to Banco de Mèxico figures showing breakdown of holders of all outstanding government debt supplied to the author by the International Institute of Finance in Washington, D.C. As of February 1994, 1.41 percent of federal government debt was held by Mexican banks, 25.10 percent by the Banco de Mèxico itself, 40.70 percent by foreigners, and 32.79 percent by the non-financial sector resident in Mexico. I note again that the form in which most international financial statistics are collected makes it difficult to distinguish between portfolio flows destined for the capital-importing country's public and private sectors. My choice to assume that most short-term debt flows were destined for the private sector, as reported in the summary statistics in Table 1, will have understated the Mexican government's borrowing abroad. Back.

Note 47: Many observers have argued that Mexico's December 1994 crisis was more of a liquidity than a solvency crisis. If private investors had not reacted so precipitously, neither the subsequent balance of payments nor fiscal crises need have reached anything like the magnitude that they did. On the economic issues, see Gruben forthcoming; Molano forthcoming; Sachs, Tornell, and Velasco 1995; Roett 1996; Passell 1995. On the politics of Mexican international financial policies in 1994, see Elizondo forthcoming; Starr 1997. Back.

Note 48: The package included $20 billion in loans and guarantees from the U.S., a then unprecedented $17.8 billion from the International Monetary Fund (IMF), and a similarly noteworthy $10 billion from the Bank for International Settlements (BIS), the bankers' central bank. Back.

Note 49: In 1990, the real urban minimum wage in Mexico City was 45.5, where 1980 was 100. Real average wages in manufacturing were 77.9, with the same base year. ECLAC 1993, pp. 34-35. Back.

Note 50: My reading of Mexican politics is that Zedillo himself is sincere about democratizing reform, although progress undeniably has been two steps forward, and one and a half back. See Dillon 1996; Crawford 1997. Back.

Note 51: Even the Zapatista movement in Chiapas state, which predates the 1994 financial crisis, largely is a reaction against neoliberal economic reforms, in this case mainly the land tenure changes that accompanied Mexico's accession to NAFTA. Back.

Note 52: In the case of Argentina, much of the "foreign" capital undoubtedly was returning flight capital, that is, money that Argentine citizens, often illegally, had spirited out of the country from the late 1970s through the 1980s to place in safe havens abroad. From the viewpoint of the present analysis, however, this interesting fact is irrelevant, since the behavior of the owners of the portfolio investments is equivalent. Back.

Note 53: On Brazil's experience, see Kingstone, forthcoming. On the tequila effect, see Roett 1996, Molano 1997. Back.

Note 54: The President-elect prevailed upon the outgoing incumbent to pardon two former presidents, Chun Do Hwan and Roh Te Woo, facing life sentences for corruption and abuse of power. By this act, Kim Dae Jung looked magnanimous (President Chun had planned to execute the dissident, only desisting because of vocal and high-level international pressure) and won friends among the military and traditional economic elites. In early January 1998, Kim Dae Jung got at least initial agreement from normally militant trade union leaders for their acquiesance to some layoffs, in exchange for an extension of Korea's thin social safety net. Back.

Note 55: The Soros Foundation spent more than $123 million in Central Europe between 1989 and 1994, about five times the U.S. government's National Endowment for Democracy (Miller 1997). Back.

Note 56: New York Times columnist Thomas L. Friedman (1998) distinguished among countries with modern, "transparent" domestic financial regulation, that have been hurt least (Taiwan, Hong Kong, Singapore), countries with "democratic, but corrupt, systems" that "were hurt second worst" but already have begun to adjust (Thailand, South Korea), and the "corrupt, authoritarian regime that can't adapt" and "is going to melt down" (Indonesia). Back.

Note 57: The Inter-American Development Bank (1997, pp. 17-18), which favors market liberalization and generally tries to put a positive face on neoliberal structural adjustment, put it this way: "After falling continually throughout the 1970s, poverty increased dramatically in Latin America during the 1980s. Ö [D]uring the 1990s, the distribution of Latin American income did not improve, though the persistent deterioration that characterized the late 1980s was arrested. Ö [P]oorer income groups typically benefit disproportionately from economic recovery, just as they are disproportionately hurt by bad timesÖ.But the relatively well-off groups of Latin American society appear to have benefited from the recovery of the 1990s more than the poorest classes." For more critical views of structural adjustment , see Veltmeyer, Petras, and Vieux 1997, Oxhorn and Ducatenzeiler 1998, or Boron 1995. Baer and Maloney 1997, on the other hand, conclude that neoliberal reforms will not worsen Latin American income distribution in the long run. Back.

Note 58: The literature is vast. See Diamond, Linz, and Lipset, eds. 1989. On democratic consolidation in the 1990s, see Dominquez and Lowenthal, eds. 1996. Back.

Note 59: Interestingly, the perception among policymakers and intellectuals in advanced industrial countries is also that financial globalization has limited public policy autonomy in their countries, shifting "power" abroad (see Underhill ed. 1997, Schrecker ed. 1997, Strange 1986, Cerny ed. 1993). The relative gainers were not emerging market countries, of course, but rather private owners of capital. Dani Rodrik (1997) provides a resolutely moderate synthesis of the overall globalization debate, that nonetheless acknowledges the increased bargaining power of capital. Back.

Note 60: Sub-Saharan Africa's experience of structural adjustment in the 1980s was even more depressing than Latin America's, though the causes of this failure are wildly overdetermined (Poku and Pettiford, eds. 1998). Back.

Note 61: For a contrary argument that is empirically rather than theoretically based, see Cohen 1998. He noted that globalization in the 1990s in Mexico, Brazil, and Argentina, Latin America's three largest emerging markets, seems to have worsened income differentials and decreased the returns to unskilled labor, as governments undermined collective bargaining traditions in order to attract foreign capital, while local capitalists shed workers and adopt the latest (capital-intensive) technology in order to compete globally. Back.

Note 62: On the theory of economic "rents" see Krueger 1974. For arguments in favor of market reforms to eliminate corruption and rent-seeking that were written about India, but apply more generally, see Jha 1980, Bardhan 1984. Back.

Note 63: The crises in portfolio flows of the mid 1990s have led to renewed interest in those countries that have retained substantial capital controls, including Chile (see Fidler 1998) and China. Back.

Bibliography