International financial integration should, in principle, help countries to reduce macroeconomic volatility. The survey presented in this section, including some new evidence, suggests that developing countries, in particular, have not attained this potential benefit. The process of capital account liberalization has often been accompanied by increased vulnerability to crises. Globalization has heightened these risks, since financial linkages have the potential to amplify the effects of both real and financial shocks.
Macroeconomic Volatility22
One of the potential benefits of globalization is provision of better opportunities for reducing volatility by diversifying risk. Indeed, these benefits are presumably even greater for developing countries that are intrinsically subject to higher volatility on account of their having less diversified production structures than industrial economies. Recent crises in some MFIs suggest, however, that financial integration may, in fact, have increased volatility.
What is the overall evidence of the effect of globalization on macroeconomic volatility? In addressing this question, it is important to draw a distinction between output and consumption volatility. In theoretical models, the direct effects of global integration on output volatility are ambiguous. On the one hand, financial integration provides access to capital that can help capital-poor developing countries diversify their production bases. On the other hand, rising financial integration could also lead to increasing specialization of production based on comparative-advantage considerations, thereby making economies more vulnerable to industry-specific shocks (Razin and Rose, 1994).
Irrespective of the effects on output volatility, theory suggests that financial integration should reduce consumption volatility. The ability to reduce fluctuations in consumption is regarded as an important determinant of economic welfare. Access to international financial markets provides better opportunities for countries to share macroeconomic risk and, thereby, smooth consumption. The basic idea here is that since output fluctuations are not perfectly correlated across countries, trade in financial assets can be used to delink national consumption levels from the country-specific components of these output fluctuations (see Obstfeld and Rogoff (1998), Chapter 5). Appendix II provides a detailed analytical examination of this issue and shows that the gains from consumption smoothing are potentially very large for developing economies.23
Notwithstanding the importance of this issue, the empirical evidence on the effects of globalization on macroeconomic volatility is rather sparse and, in particular, the evidence concerning the effects of financial integration on volatility is limited and inconclusive (see Box 4.1). In addition, the existing literature has been largely devoted to analyzing the effects of financial integration on output volatility, with little attention paid to consumption volatility. Hence, this paper provides some new evidence on this topic.
Table 4.1 examines changes in volatility for different macroeconomic aggregates over the last four decades. Consistent with evidence presented in the September 2002 issue of the IMF’s World Economic Outlook, MFI economies have, on average, lower output volatility than LFI economies. Interestingly, there is a significant decline in average output volatility in the 1990s for both industrial and LFI economies but a far more modest decline for MFI economies. The picture is similar for a broader measure of income that includes factor income flows and terms of trade effects, which are particularly important for developing countries. Figure 4.1’s top panel, which shows the evolution of the average volatility of income growth for different groups of countries, confirms these results and shows that they are not sensitive to the decade-wise breakdown of the data, although there is a pickup in volatility for MFIs toward the end of the sample.24
Volatility of Annual Growth Rates of Selected Variables
(Percentage standard deviations, medians for each group of countries)
Volatility of Annual Growth Rates of Selected Variables
(Percentage standard deviations, medians for each group of countries)
| Full Sample | Decade | |||||
|---|---|---|---|---|---|---|
| 1960-99 | 1960s | 1970s | 1980s | 1990s | ||
| Output (Y) | ||||||
| Industrial countries | 2.18 | 1.91 | 2.46 | 2.03 | 1.61 | |
| (0.23) | (0.26) | (0.28) | (0.30) | (0.14) | ||
| MFI economies | 3.84 | 3.31 | 3.22 | 4.05 | 3.59 | |
| (0.20) | (0.42) | (0.37) | (0.44) | (0.62) | ||
| LFI economies | 4.67 | 3.36 | 4.88 | 4.53 | 2.70 | |
| (0.35) | (0.61) | (1.01) | (0.69) | (0.38) | ||
| Income (Q) | ||||||
| Industrial countries | 2.73 | 2.18 | 2.99 | 2.54 | 1.91 | |
| (0.34) | (0.33) | (0.40) | (0.29) | (0.30) | ||
| MFI economies | 5.44 | 3.60 | 5.43 | 5.45 | 4.78 | |
| (0.50) | (0.47) | (0.45) | (0.65) | (0.72) | ||
| LFI economies | 7.25 | 4.42 | 9.64 | 7.56 | 4.59 | |
| (0.84) | (0.53) | (1.24) | (1.23) | (0.54) | ||
| Consumption (C) | ||||||
| Industrial countries | 2.37 | 1.47 | 2.16 | 1.98 | 1.72 | |
| (0.30) | (0.27) | (0.25) | (0.28) | (0.20) | ||
| MFI economies | 5.18 | 4.57 | 4.52 | 4.09 | 4.66 | |
| (0.51) | (0.49) | (1.04) | (0.94) | (0.46) | ||
| LFI economies | 6.61 | 5.36 | 7.07 | 7.25 | 5.72 | |
| (0.78) | (0.58) | (0.11) | (0.81) | (0.78) | ||
| Total consumption (C+G) | ||||||
| Industrial countries | 1.86 | 1.38 | 1.84 | 1.58 | 1.38 | |
| (0.23) | (0.28) | (0.18) | (0.19) | (0.20) | ||
| MFI economies | 4.34 | 3.95 | 4.19 | 3.43 | 4.10 | |
| (0.47) | (0.51) | (0.54) | (0.84) | (0.53) | ||
| LFI economies | 6.40 | 4.85 | 6.50 | 6.34 | 4.79 | |
| (0.56) | (0.55) | (0.93) | (0.91) | (0.82) | ||
| Ratio of total consumption (C+G) volatility to income (Q) volatility | ||||||
| Industrial countries | 0.67 | 0.75 | 0.56 | 0.61 | 0.58 | |
| (0.02) | (0.09) | (0.03) | (0.06) | (0.06) | ||
| MFI economies | 0.81 | 0.92 | 0.74 | 0.76 | 0.92 | |
| (0.07) | (0.13) | (0.12) | (0.11) | (0.04) | ||
| LFI economies | 0.80 | 0.95 | 0.68 | 0.82 | 0.84 | |
| (0.08) | (0.06) | (0.10) | (0.51) | (0.14) | ||
Volatility of Annual Growth Rates of Selected Variables
(Percentage standard deviations, medians for each group of countries)
| Full Sample | Decade | |||||
|---|---|---|---|---|---|---|
| 1960-99 | 1960s | 1970s | 1980s | 1990s | ||
| Output (Y) | ||||||
| Industrial countries | 2.18 | 1.91 | 2.46 | 2.03 | 1.61 | |
| (0.23) | (0.26) | (0.28) | (0.30) | (0.14) | ||
| MFI economies | 3.84 | 3.31 | 3.22 | 4.05 | 3.59 | |
| (0.20) | (0.42) | (0.37) | (0.44) | (0.62) | ||
| LFI economies | 4.67 | 3.36 | 4.88 | 4.53 | 2.70 | |
| (0.35) | (0.61) | (1.01) | (0.69) | (0.38) | ||
| Income (Q) | ||||||
| Industrial countries | 2.73 | 2.18 | 2.99 | 2.54 | 1.91 | |
| (0.34) | (0.33) | (0.40) | (0.29) | (0.30) | ||
| MFI economies | 5.44 | 3.60 | 5.43 | 5.45 | 4.78 | |
| (0.50) | (0.47) | (0.45) | (0.65) | (0.72) | ||
| LFI economies | 7.25 | 4.42 | 9.64 | 7.56 | 4.59 | |
| (0.84) | (0.53) | (1.24) | (1.23) | (0.54) | ||
| Consumption (C) | ||||||
| Industrial countries | 2.37 | 1.47 | 2.16 | 1.98 | 1.72 | |
| (0.30) | (0.27) | (0.25) | (0.28) | (0.20) | ||
| MFI economies | 5.18 | 4.57 | 4.52 | 4.09 | 4.66 | |
| (0.51) | (0.49) | (1.04) | (0.94) | (0.46) | ||
| LFI economies | 6.61 | 5.36 | 7.07 | 7.25 | 5.72 | |
| (0.78) | (0.58) | (0.11) | (0.81) | (0.78) | ||
| Total consumption (C+G) | ||||||
| Industrial countries | 1.86 | 1.38 | 1.84 | 1.58 | 1.38 | |
| (0.23) | (0.28) | (0.18) | (0.19) | (0.20) | ||
| MFI economies | 4.34 | 3.95 | 4.19 | 3.43 | 4.10 | |
| (0.47) | (0.51) | (0.54) | (0.84) | (0.53) | ||
| LFI economies | 6.40 | 4.85 | 6.50 | 6.34 | 4.79 | |
| (0.56) | (0.55) | (0.93) | (0.91) | (0.82) | ||
| Ratio of total consumption (C+G) volatility to income (Q) volatility | ||||||
| Industrial countries | 0.67 | 0.75 | 0.56 | 0.61 | 0.58 | |
| (0.02) | (0.09) | (0.03) | (0.06) | (0.06) | ||
| MFI economies | 0.81 | 0.92 | 0.74 | 0.76 | 0.92 | |
| (0.07) | (0.13) | (0.12) | (0.11) | (0.04) | ||
| LFI economies | 0.80 | 0.95 | 0.68 | 0.82 | 0.84 | |
| (0.08) | (0.06) | (0.10) | (0.51) | (0.14) | ||
Effects of Globalization on Volatility: A Review of Empirical Evidence
Unlike the rich empirical literature focusing on the impact of financial openness on economic growth, there are only a limited number of studies analyzing the links between openness and macroeconomic volatility. Moreover, existing studies have generally been unable to document a clear empirical link between openness and macroeconomic volatility. Razin and Rose (1994) study the impact of trade and financial openness on the volatility of output, consumption, and investment for a sample of 138 countries over the period 1950-88. They find no significant empirical link between openness and the volatility of these variables.
Easterly, Islam, and Stiglitz (2001) explore the sources of output volatility using data for a sample of 74 countries over the period 1960-97. On the one hand, they find that a higher level of development of the domestic financial sector is associated with lower volatility. On the other hand, an increase in the degree of trade openness leads to an increase in the volatility of output, especially in developing countries. Their results indicate that neither financial openness nor the volatility of capital flows has a significant impact on output volatility.
Buch, Dopke, and Pierdzioch (2002) use data for 25 OECD countries to examine the link between financial openness and output volatility. They report that there is no consistent empirical relationship between financial openness and the volatility of output. Gavin and Hausmann (1996) study the sources of output volatility in developing countries over the period 1970-92. They find that there is a significant positive association between the volatility of capital flows and output volatility. O’Donnell (2001) examines the effect of financial integration on the volatility of output growth over the period 1971-94 using data for 93 countries. He finds that a higher degree of financial integration is associated with lower (higher) output volatility in OECD (non-OECD) countries. His results also suggest that countries with more developed financial sectors are able to reduce output volatility through financial integration.
Bekaert, Harvey, and Lundblad (2002a) examine the impact of equity market liberalization on the volatility of output and consumption during 1980-2000. They find that following equity market liberalizations, there is a significant decline in both output and consumption volatility. Capital account openness reduces the volatility of output and consumption, but its impact is smaller than that of equity market liberalization. They also report, however, that capital account openness increases the volatility of output and consumption in emerging market countries. The IMF’s World Economic Outlook, September 2002 provides some evidence indicating that financial openness is associated with lower output volatility in developing countries.
The consumption rows of Table 4.1 show that average consumption volatility in the 1990s has declined in line with output volatility for both industrial economies and LFI economies. By contrast, for MFI economies, the volatility of private consumption has risen in the 1990s, relative to the 1980s. It is possible that looking at the volatility of private consumption is misleading, since public consumption could be playing an important smoothing role, especially in developing economies. It is true, as is shown in the total consumption rows of Table 4.1, that total consumption is generally less volatile than private consumption. These results, however, confirm the pattern that, on average, consumption volatility for industrial and LFI economies declined in the 1990s. By contrast, it increased for MFI economies over the same period. Figure 4.1’s lower panel, which shows the evolution of the average volatility of total consumption growth over a 10-year rolling period, presents a similar picture. Could this simply be a consequence of higher income volatility for MFI economies?
Volatility of Income and Consumption Growth
(10-year rolling standard deviations; medians for each group of countries)
Source: Kose, Prasad, and Terrones (2003a).Note: MFI denotes more financially integrated, and LFI less financially integrated, economies.Strikingly, for MFI countries, the volatility of total consumption relative to that of income actually increased in the 1990s relative to earlier periods. The bottom rows of Table 4.1 show the median ratio of the volatility of total consumption growth to that of income growth for each group of countries. For MFI economies, this ratio increases from 0.76 in the 1980s to 0.92 in the 1990s, while it remains essentially unchanged for the other two groups of countries. Thus, the increase in the 1990s of the ratio of the volatility of consumption to that of income for the MFI economies suggests that financial integration has not provided better consumption-smoothing opportunities for these economies.25
More formal econometric evidence is presented by Kose, Prasad, and Terrones (2003a), who use measures of capital account restrictions as well as gross financial flows to capture different aspects of financial integration, as well as differences in the degree of integration across countries and over time. This analysis confirms the increase in the relative volatility of consumption for countries that have larger financial flows, even after controlling for macroeconomic variables as well as country characteristics such as trade openness and industrial structure. These authors, however, also identify an important threshold effect—beyond a particular level, financial integration significantly reduces volatility. Most developing economies, including MFI economies, are unfortunately well below this threshold.26
Why has the relative volatility of consumption increased precisely in those developing countries that are more open to financial flows? One explanation is that positive productivity and output growth shocks in these countries during the late 1980s and early 1990s led to consumption booms that were willingly financed by international investors. These consumption booms were accentuated by the fact that many of these countries undertook domestic financial liberalization at the same time they opened up to international financial flows, thereby loosening liquidity constraints at both the individual and national levels. When negative shocks hit these economies, however, they rapidly lost access to international capital markets.
Consistent with this explanation, a growing literature suggests that the procyclical nature of capital flows appears to have had an adverse impact on consumption volatility in developing economies.27 One manifestation of this procyclicality is the phenomenon of “sudden stops” of capital inflows (see Calvo and Reinhart, 1999). More generally, access to international capital markets has a procyclical element, which tends to generate higher output volatility as well as excess consumption volatility (relative to that of income). Reinhart (2002), for instance, finds that sovereign bond ratings are procyclical. Since the spreads on bonds of developing economies are strongly influenced by these ratings, this implies that costs of borrowing on international markets are procyclical as well. Kaminsky and Reinhart (2001) present more direct evidence on the procyclical behavior of capital inflows.28
Crises as Special Cases of Volatility
Crises can be regarded as particularly dramatic episodes of volatility. In fact, the proliferation of financial crises is often viewed as one of the defining aspects of the intensification of financial globalization over the last two decades. Furthermore, the fact that recent crises have affected mainly MFI economies has led to these phenomena being regarded as hallmarks of the unequal distribution of globalization’s benefits and risks. This raises a challenging set of questions about whether the nature of crises has changed over time, what factors increase vulnerability to crises, and whether such crises are an inevitable concomitant of globalization.
Some aspects of financial crises have indeed changed over time, while in other respects it is “deja vu all over again.” Calvo (1998) has referred to such episodes in the latter half of the 1980s and 1990s as capital account crises, while earlier ones are referred to as current account crises. Although this suggests differences in the mechanics of crises, it does not necessarily imply differences in some of their fundamental causes. Kaminsky and Reinhart (1999) discuss the phenomenon of “twin crises,” which involve balance of payments and banking crises. These authors also make the important points that in the episodes that they analyze, banking sector problems typically precede a currency crisis and that the currency crisis then deepens the banking crisis, activating a vicious spiral. In this vein, Krueger and Yoo (2002) conclude that imprudent lending by Korean banks in the early and mid-1990s, especially to the chaebols, played a significant role in the 1997 Korean currency crisis. Opening up to capital markets can thus exacerbate such existing domestic distortions and lead to catastrophic consequences (Aizenman, 2002).
One key difference in the evolution of crises is that although the 1970s and 1980s featured crises that affected both industrial and developing economies, these have become almost exclusively the preserve of developing economies since the mid-1990s.29 This suggests either that advanced economies have been able to better protect themselves through improved policies or that the fundamental causes of crises have changed over time, thereby increasing the relative vulnerability of developing economies. In this context, it should be noted that although capital flows from industrial economies to MFI economies have increased sharply, these flows among industrial economies have jumped even more sharply in recent years, as was noted earlier. Thus, at least in terms of volume of capital flows, it is not obvious that changes in financial integration can, by themselves, be blamed for crises in MFI economies.
Is it reasonable to accept crises as a natural feature of globalization, much as business cycles are viewed as a natural occurrence in market economies? One key difference between these phenomena is that the overall macroeconomic costs of financial crises are typically very large and far more persistent. Calvo and Reinhart (2000 and 2002) document that emerging market currency crises, which are typically accompanied by sudden stops or reversals of external capital inflows, are associated with significant negative output effects.30 Such recessions following devaluations (or large depreciations) are also found to be much deeper in emerging markets than in developed economies. In addition, the absence of well-functioning safety nets can greatly exacerbate the social costs of crises, which typically have large distributional consequences (see, for example, Baldacci, de Mello, and Inchauste, 2002).
Has Financial Globalization Intensified the Transmission of Volatility?
What factors have led to the rising vulnerability of developing economies to financial crises? The risk of sudden stops or reversals of global capital flows to developing countries has increased in importance, since many developing countries now rely heavily on borrowing from foreign banks or portfolio investment by foreign investors. These capital flows are sensitive not just to domestic conditions in the recipient countries but also to macroeconomic conditions in industrial countries. For instance, Mody and Taylor (2002), using an explicit disequilibrium econometric framework, detect instances of “international capital crunch”—where capital flows to developing countries are curtailed by supply-side rationing that reflects industrial country conditions.31 These North-South financial linkages, in addition to the real linkages described in earlier sections, represent an additional channel through which business cycles and other shocks that hit industrial countries can affect developing countries.
The effects of industrial country macroeconomic conditions, including the stage of the business cycle and interest rates, have different effects on various types of capital flows to emerging markets. Reinhart and Reinhart (2001) document that net FDI flows to emerging market economies are strongly positively correlated with U.S. business cycles. In contrast, bank lending to these economies is negatively correlated with U.S. cycles. Edison and Warnock (2001) find that portfolio equity flows from the United States to major emerging market countries are negatively correlated with both U.S. interest rates and U.S. output growth. This result is particularly strong for flows to Latin America and less so for flows to Asia. Thus, the sources of capital inflows for a particular MFI economy can greatly affect the nature of its vulnerability to the volatility of capital flows arising from industrial country disturbances.32
The increase in cross-country financial market correlations also indicates a risk of emerging markets being caught up in financial market bubbles. The rise in comovement across emerging and industrial country stock markets, especially during the stock market bubble period of the late 1990s, points to the relevance of this concern. This is a particular risk for the relatively shallow and undiversified stock markets of some emerging economies. For instance, as noted earlier, the strong correlations between emerging and industrial stock markets during the bubble period reflect the preponderance of technology and telecommunication sector stocks in the former set of markets. It is, of course, difficult to say conclusively whether this phenomenon would have occurred even in the absence of financial globalization, since stock market liberalizations in these countries often went hand in hand with their opening up to capital flows.
The increasing depth of stock markets in emerging economies could alleviate some of these risks but, at the same time, could heighten the real effects of such financial shocks. In this vein, Delias and Hess (2002) find that a higher degree of financial development makes emerging stock markets more susceptible to external influences (both financial and macroeconomic) and that this effect remains important after controlling for capital controls and trade linkages.33 Consequently, the effects of external shocks could be transmitted to domestic real activity through the stock market channel.
Even the effects of real shocks are often transmitted faster and amplified through financial channels. There is a large literature showing how productivity, terms of trade, fiscal, and other real shocks are transmitted through trade channels.34 Cross-country investment flows, in particular, have traditionally responded quite strongly to country-specific shocks.35 Financial channels constitute an additional channel through which the effects of such real shocks can be transmitted. Furthermore, since transmission through financial channels is much quicker than through real channels, both the speed and magnitude of international spillovers of real shocks are considerably heightened by financial linkages.36
Increasing financial linkages have also resulted in contagion effects. Potential contagion effects are likely to become more important over time as financial linkages increase and investors in search of higher returns and better diversification opportunities increase their shares of international holdings and, owing to declines in information and transaction costs, have access to a broader array of crosscountry investment opportunities.37
There are two broad types of contagion identified in the literature—fundamentals-based contagion and “pure” contagion. The former refers to the transmission of shocks across national borders through real or financial linkages. In other words, while an economy may have weak fundamentals, it could get pushed into a financial crisis as a consequence of investors reassessing the riskiness of investments in that country or attempting to rebalance their portfolios following a crisis in another country. Similarly, bank lending can lead to such contagion effects when a crisis in one country to which a bank has significant exposure forces it to rebalance its portfolio by readjusting its lending to other countries. This bank transmission channel, documented in van Rijckeghem and Weder (2000) and Kaminsky and Reinhart (2001), can be particularly potent, since a large fraction of bank lending to emerging markets is in the form of short-maturity loans. While fundamentals-based contagion was once prevalent mainly at the regional level, the Russian crisis demonstrated its much broader international reach (Kaminsky and Reinhart, 2003).38
Pure contagion, however, represents a different kind of risk, since it cannot easily be influenced by domestic policies, at least in the short run. There is a good deal of evidence of sharp swings in international capital flows that are not obviously related to changes in fundamentals. Investor behavior during these episodes, which is sometimes categorized as herding or momentum trading, is difficult to explain in the context of optimizing models with full and common information. Informational asymmetries, which are particularly rife in the context of emerging markets, appear to play an important role in this phenomenon (see Box 4.2). A related literature suggests that pure contagion may reflect investors’ shifting appetite for risk, but it is no doubt difficult to disentangle such changes in risk appetite from shifts in underlying risks themselves (Kumar and Persaud, 2001). Thus, in addition to “pure contagion,” financial integration exposes developing economies to the risks associated with destabilizing investor behavior that is not related to fundamentals.39
Herding and Momentum Trading by International Investors
The emerging market crises of the 1990s have raised concerns about excessive international capital flow volatility. In particular, international investors have been accused of acting in a destabilizing way, displaying a tendency to engage in herding behavior and momentum trading.
Herding is usually defined as investors mimicking each others’ actions, sometimes ignoring socially valuable information. Rationalizations of herding include learning from others and incentive structures for fund managers. Herding owing to learning from others can occur when actions are observable but information is partly private. In such situations, it may be optimal to rely exclusively on others’ actions. If the abilities of fund managers are unknown to investors, investors may choose to compensate managers based on relative performance. This, in turn, provides an incentive for managers to mimic the actions of their peers: fund managers do not tend to deviate too strongly from “benchmark” indices.
A related behavior of investors is given by momentum trading—strategies prescribing buying assets whose prices have been rising and selling assets whose prices have been falling. Such behavior can also be destabilizing.
The empirical evidence concerning herding and momentum trading at the international level is still sparse. A number of studies have looked at the case of the Republic of Korea. Choe, Kho, and Stulz (1999), for example, find evidence for return chasing and herding among foreign investors before the crisis period, but not over the entire sample period. Kim and Wei (2002) examine the transactions of different types of portfolio investors in Korea before and during the Asian crisis, finding that nonresident institutional investors engage in more herding and more momentum trading than foreign investors residing in Korea.
These have begun to be complemented by regional and global studies. Kaminsky, Lyons, and Schmukler (1999) find some evidence for momentum trading among equity mutual funds investing in Latin America, which appears to be accentuated during crises. Griffin, Nardari, and Stulz (2002) and Richards (2002) find that foreign investors’ purchases in East Asian emerging markets is strongly influenced by both the stock return in those markets and the return in developed country markets. Borensztein and Gelos (2002) find moderate evidence for herding behavior and momentum trading among emerging market mutual funds. Gelos and Wei (2002) document that herding is less pronounced in countries that have more transparent macroeconomic policies and corporate sectors.
Despite these recent efforts, the picture obtained remains incomplete. For example, although the focus has been mainly on equity markets, little systematic knowledge has been accumulated on the behavior of banks and fixed-income investors.
Note: This box was prepared by Gaston Gelos.Some Factors That Increase Vulnerability to Risks of Globalization
Empirical research indicates that the composition of capital inflows and the maturity structure of external debt appear to be associated with higher vulnerability to the risks of financial globalization. The relative importance of different sources of financing for domestic investment, as proxied by the following three variables, has been shown to be positively associated with the incidence and severity of currency and financial crises: the ratio of bank borrowing or other debt relative to foreign direct investment; the shortness of the term structure of external debt; and the share of external debt denominated in foreign currencies.40 Detragiache and Spilimbergo (2001) find strong evidence that debt crises are more likely to occur in countries where external debt has short maturities.41 The maturity structure may not entirely be a matter of choice, however, since, as argued by these authors, countries with weaker macroeconomic fundamentals are often forced to borrow at shorter maturities.
In addition to basic macroeconomic policies, other systemic policy choices can affect the vulnerability of MFI economies. Recent currency crises have highlighted one of the main risks in this context. Developing countries that attempt to maintain a relatively inflexible exchange rate system often face the risk of attacks on their currencies. While various forms of fully or partially fixed exchange rate regimes can have some advantages, the absence of supportive domestic policies can often result in an abrupt unraveling of these regimes when adverse shocks hit the economy.
Financial integration can also aggravate the risks associated with imprudent fiscal policies. Access to world capital markets could lead to excessive borrowing that is channeled into unproductive government spending. The existence of large amounts of short-term debt denominated in hard currencies then makes countries vulnerable to external shocks or changes in investor sentiment. The experiences of a number of MFI countries that have suffered the consequences of such external debt accumulation point to the heightened risks of undisciplined fiscal policies when the capital account is open.
Premature opening of the capital account also poses serious risks when financial regulation and supervision are inadequate.42 In the presence of weakly regulated banking systems and other distortions in domestic capital markets, inflows of foreign capital could exacerbate the existing inefficiencies in these economies. For example, if domestic financial institutions tend to channel capital to firms taking excessive risks or having weak fundamentals, financial integration could simply lead to an intensification of such flows.43 In turn, the effects of premature capital inflows on the balance sheets of the government and corporate sectors could have negative repercussions on the health of financial institutions in the event of adverse macroeconomic shocks.
