Financial Crises

Chapter 5. Resilience in Latin America: Lessons from Macroeconomic Management and Financial Policies

Stijn Claessens, Ayhan Kose, Luc Laeven, and Fabian Valencia
Published Date:
February 2014
  • ShareShare
Show Summary Details
José De GregorioThe author is very grateful to Stijn Claessens and Kevin Cowan for useful comments and discussions, and to José Tomás De Gregorio and Bastián Gallardo for valuable research assistance.

Imagine a really big global crisis, second only to the Great Depression. What impact would it have on emerging market economies, particularly in Latin America? If this question had been asked 10 years ago, the unanimous answer would have been “disaster.” Actually the crisis happened, but it was not a disaster in emerging markets. Emerging market economies around the world suffered the crisis, and some had sizable contractions, but overall the damage was limited and the recovery was very strong. Therefore, an appropriate answer today to the above question would be “bad, but not a disaster.”

The purpose of this chapter is to examine the resilience of emerging market economies, with particular attention to Latin America. As argued below, this resistance to disaster is the reward for having implemented good macroeconomic and financial policies, which allowed a significant monetary and fiscal expansion to occur within a resilient financial system.

To tackle this issue, panel regressions could be performed on determinants of economic performance to gauge the main factors behind the recent economic success of emerging markets. Some interesting work, discussed below, has pursued this route. However, this kind of work has some limitations. For one, the sample period is still too short. Much better evidence will be gathered once a full business cycle has taken place. Now, the evidence that may be captured is the deepness of the contraction and the speed of a still-incomplete recovery. For another, dependent variables tend to be too blunt. The econometric work should be complemented with a more detailed analysis of particular cases, given that there are institutional nuances and differences among apparently similar policies that statistical work cannot capture. This is what this chapter will do in exploring the resilience of emerging market economies to the global financial crisis.

Although the focus is on Latin America, issues that go beyond the region are similar and relevant for the entire emerging market world. In several parts the chapter will refer more generally to emerging market economies, and present evidence for a wider set of countries. This also allows the Latin American experience to be placed in a broader perspective.1

The chapter begins with a discussion of the main factors that explain the success of emerging market economies. In particular, it focuses on macroeconomic policies, exchange rate flexibility, and good luck. The next section is devoted in greater detail to the resilience of the financial system. Next is a discussion of international reserves, which are another factor that made emerging markets more resilient. The chapter concludes with some final remarks.

The Resilience of Emerging Market Economies

Emerging markets, in particular Latin American economies, were very resilient to the 2007–09 global financial crisis. Most suffered recessions, the others severe economic slowdowns. But the effects of the crisis were much milder than traditional adjustments following global recessionary shocks. The recovery, in turn, has been very strong. Now that economies have fully recovered and GDP has reached levels consistent with full capacity utilization, and even more in some cases, new challenges appear.

A number of factors help explain the performance of the emerging market economies; this section briefly discusses some of them, leaving the others for the next sections. The main factors behind this unprecedented performance were the following:

  • Good initial macroeconomic conditions allowed for strong monetary and fiscal stimulus in reaction to the crisis.

  • A cornerstone of the macroeconomic framework was exchange rate flexibility. As flight to safety took place, currencies of emerging market economies depreciated sharply, eliminating incentives for speculation against them. The fear of floating of many other previous experiences was over. Flexibility was not extreme—several countries used a combination of exchange rate intervention and capital controls to mitigate the appreciation of their currencies.

  • Good luck. Before the crisis, emerging market economies faced very good international conditions for expanding exports. Latin American countries, most of them exporters of primary commodities, enjoyed very good terms of trade as commodity prices skyrocketed during the second half of the first decade of the 2000s. After a sharp decline in commodity prices during the crisis, they bounced back to very high levels, which persist to this day.

  • Strong, well-regulated, and fairly simple financial systems provided suitable conditions. The exchange rate depreciation did not cause a financial collapse and financial systems were able to resume lending as soon as circumstances improved.

  • High levels of international reserves were one of the few explanatory variables found to be relevant in potential explanations of why emerging market economies performed well during the crisis. Regardless of the reason for accumulating foreign assets, high levels of reserves played an important role as a deterrent to attacks on the currency and fear of foreign insolvency. They reduced the probability of a sudden stop. They also provided a cushion, although not used massively, against a potential lack of foreign financing.

This section focuses on the first three points, while the next section will be devoted to the financial sector, and the following to reserves accumulation.

The ability to conduct expansionary macroeconomic policies hinged on the sound initial macroeconomic conditions that emerging market economies enjoyed. Fiscal accounts were healthy as ever. Levels of public debt were relatively low. Countries that had windfall gains from high terms of trade saved before the crisis, accumulating resources to spend during the downturn. Other countries were able to borrow to finance their fiscal expansions.

On the monetary side, inflation control was key to allowing monetary loosening. Despite the sharp rise of commodity prices during the buildup to the crisis, which led to increases in inflation, the subsequent slowdown put enough downward pressure on inflation to leave room to cut interest rates. The response of inflation to exchange rate developments was much more muted than in previous episodes as a result of exchange rate flexibility that reduced the pass-through from exchange rates to inflation.2

Figure 5.1 shows the levels of public debt in Latin American economies; public debt was not only historically low, but declined in the years before the crisis. Argentina and Brazil had debt close to 60 percent of GDP, while the rest had levels below those of other emerging markets and of course, much below the levels of advanced economies.

Figure 5.1Gross Public Debt, 2008

Source: IMF, Fiscal Monitor database.

Note: Advanced = Canada, France, Germany, Italy, Japan, United Kingdom, and United States. Asia = India, Indonesia, Korea, Malaysia, Philippines, and Thailand. Emerging Europe = Czech Republic, Hungary, Latvia, Lithuania, Poland, and Romania. Latin America = Argentina, Brazil, Chile, Colombia, Mexico, Peru, Uruguay, and Venezuela.

The fiscal policy actions during the crisis were in sharp contrast with the traditional policy responses of Latin American countries to recessionary shocks coming from abroad. In the past, authorities relied on monetary and fiscal tightening. Fiscal policy usually had to be contractionary, not because of bad judgment, but because there was no space to expand fiscal policy. Creditworthiness would deteriorate during periods of bad external conditions and the ability to finance the budget would be severely impaired. Therefore, liquidity constraints became binding, forcing a fiscal adjustment. This gave origin to procyclical fiscal policies, which followed a simple rule that can be summarized as “spend as much as you can finance.”3

In previous external crises monetary policy was usually tightened because of fear of depreciation. The potential inflationary and financial repercussions of the weakening of the currency were so pervasive that authorities were reluctant to allow a full exchange rate adjustment, and defended parity with high interest rates. To a large extent, this was the result of rigidities in the exchange rate regime that induced currency mismatches in the corporate sector and a high response of price setters to infrequent changes in the exchange rate.4

In contrast to previous responses, Brazil, Chile, Colombia, Mexico, and Peru cut rates to historical lows during the global financial crisis. Although in most cases they have since been raised, they still have not returned to precrisis levels (Figure 5.2). From a comparative perspective, the monetary and fiscal expansions of Latin American countries, as well as Asian countries, were sizable (Figure 5.3). In Latin America, the monetary stimulus was relatively large. Chile’s fiscal stimulus was among the largest, owing to large fiscal resources available in the sovereign wealth funds built up during the copper price boom that preceded the crisis.

Figure 5.2Monetary Policy Interest Rates

Source: Bloomberg, L.P.

Figure 5.3Monetary and Fiscal Stimulus

Sources: Bloomberg, L.P.; Central Bank of Chile; IMF, Fiscal Monitor, November 2009; and Ministries of Finance.

Note: MPR = monetary policy rate.

Most Latin American countries use flexible inflation targets to conduct monetary policy. Although this strategy has been questioned in many advanced economies for ignoring financial factors, it was essential to allowing for sharp, credible, and effective monetary expansions. Evidence has shown that since August 2008, inflation targeting regimes have had a positive effect on postcrisis economic performance (de Carvalho Filho, 2010). This should come as no surprise—monetary policy played a secondary role during the crisis because the crisis was mainly the result of severe financial dislocations. Therefore, countries with sound financial systems were able to engineer large expansionary policies.

Before the global financial crisis, the problem with monetary policy was precisely its deviation from the pursuit of price stability. In the United States, when the bubble burst, monetary policy was loosened to provide a safety net to the financial system. This strategy of letting the bubble grow and mopping up the mess after it burst was the so-called Greenspan put. This was a key ingredient to bubble formation. The collapse of the housing bubble was so large that the Greenspan put was ineffective. A parallel can be drawn with emerging markets’ tradition of managing exchange rates. Permanent promises of exchange rate stability, which set bounds on asset prices on the way up when the currency appreciated and on the way down when the currency depreciated, induced currency mismatches in the private sector. The fear of floating provided implicit insurance against sharp currency fluctuations and reduced private sector incentives to hedge currency exposure. In contrast, the period of floating exchange rates has been characterized by the development of foreign exchange forward markets (De Gregorio and Tokman, 2007).5

Exchange rate flexibility played a crucial role in dampening the adjustment. At the height of the crisis, currencies in Latin America had depreciated sharply, and the financial systems were able to accommodate the depreciation without untenable stress. In a period of a few months, depreciations were about 60 percent, something never seen before (Figure 5.4). The figure shows that this depreciation happened not only in Latin America, but also in Asia. The depreciation responded to fundamentals, acknowledging the potential that emerging markets could be hit disproportionately by the crisis. When they were not, currencies strengthened. But the search for safe havens also resulted in depreciation of riskier and less-liquid emerging markets’ assets. Exchange rates’ abrupt adjustment reduced incentives to speculate against further weakening. The war chest that reserves provided also helped exchange rates to adjust without major disruptions.

Figure 5.4Exchange Rates during the 2008–09 Global Financial Crisis

Source: Bloomberg, L.P.

Note: Figure in parentheses indicates maximum depreciation. An increase indicates a depreciation of the currency.

Finally, emerging market economies also faced very good external conditions, the good luck component of the resilience. Rapid growth in developing economies, in particular China, generated strong demand for exports from developing countries. In Latin America, on top of greater market access, several countries that were primary commodity exporters enjoyed significant terms-of-trade gains. These gains did not happen in Asia because terms of trade were stable, owing primarily to the broad industrial production bases in Asia economies. Panel a of Figure 5.5 shows the evolution of the terms of trade since 1980. In panel b, the bars corresponds to the range between the minimum and maximum levels for the period 1980–2010, and the figure shows the last data and the average for the period 2000–05. Every country, with the exception of Mexico and Uruguay, has enjoyed significant gains since 2005.

Figure 5.5Terms of Trade

Source: World Bank, Net Barter Terms of Trade Index,

One of the main risks in Latin America is a decline in the terms of trade. IDB (2012) provides some simulations with a decline in terms of trade. They consider a risk scenario that involves a deepening of the euro area crisis and a slowdown in China, which would result in a decline of 30 percent in commodity prices. The simulation exercise, which yields a worldwide recession, with Chinese growth falling 3 percentage points and recessions in the United States and Europe, would cause a decline in output in Latin America of 0.6 percent, somewhat smaller, but more persistent, than that of the 2007–09 global financial crisis. Overall, this simulation confirms the vulnerability of the region to a global slowdown and a decline in the price of commodities. But, as during the 2007–09 crisis, Latin America would not suffer a major collapse, as it typically would have in the past. Although the simulation does not account for differences across countries, the impact should be differentiated as long as the resilience to terms-of-trade shocks relies on the government budget’s dependence on terms of trade. Not all countries are in the same position, particularly since the crisis, because governments used a significant portion of their fiscal space.

The Resilience of Financial Systems

The purpose of regulation of the financial system is to ensure that institutions are solvent and liquid. Since the 2007–09 financial crisis, much discussion has focused on financial regulation to limit the risk of spillovers from the financial system and particular institutions to the whole economy. Thus, in addition to traditional microprudential tools, the use of macroprudential rules that limit systemic financial risks is a necessary complement.

Systemic financial risk may arise from the behavior of the financial system through the cycle—the time series dimension—or from the interaction of particular institutions with the rest of the financial system—the cross-section dimension. The most important concern with regard to the business cycle is the excess procyclicality of financial activities. With regard to contemporaneous spillovers, the concern is mostly with systemically important financial institutions whose risks can contaminate the entire financial system.

Whether a particular policy is macro- or microprudential is not always obvious. For example, limiting currency mismatches has both a micro component to limit foreign exposure of particular financial intermediaries, and a macro component to minimize the risk of a financial crisis caused by significant aggregate mismatches. A rule that reduces the risk of insolvency of single institutions caused by macroeconomic shocks by definition will also be protecting the integrity of the whole system.

Latin American countries have been applying macroprudential rules for a long time, even if they did not call it that. The remainder of this section discusses some features of Latin American banking systems that help explain their resilience to the global financial crisis.6

A first characteristic of Latin American banking systems is their relatively high levels of capital and low levels of leverage (Figure 5.6). All countries had regulatory capital greater than the 8 percent required by Basel II.7 This is the result of higher regulatory capital requirements and limits on leverage, as well as the internal strategies of local banks that are willing to hold larger levels of capital.

Figure 5.6Banks’ Capital Ratios, 2008 (times, percentage)

Source: Central Bank of Chile, Financial Stability Report, 2009:QI.

Note: The figure considers only commercial banks. Investment banks are excluded; they had leverage ratios in 2008 of about 26.

Of course, the crisis revealed that leverage could have been larger through off-balance-sheet investment operations that artificially reduced leverage. This possibility is at the center of the issue of leverage in the trading book. Proposals along the lines of the Volcker rule, which prohibits commercial banks from engaging in proprietary trading, go in the direction of reducing financial risks. The route followed in emerging markets, which avoids complicated regulation, is to set limits on the instruments that can be held by banks. For example, in Chile, banks can only hold corporate and Chilean bonds. Only interest rate and exchange rate derivatives are allowed. All other operations must be conducted through other financial institutions, which are also subject to financial regulation and can be subsidiaries of banks. Brazil and Mexico have also been limiting the use of derivatives in their banking systems.

The region’s strong financial systems did not prevent a sharp contraction of credit during the peak of the crisis. However, credit expansion resumed as the economies recovered. It is difficult to disentangle how much of the contraction was due to a decline in the demand for credit and how much to restrictions from the supply side. Both factors may have played a role because the increase in uncertainty tightened financial conditions and reduced demand, as surveys on financial conditions showed.

The degree of financial depth of the banking system is usually measured by the ratio of private banking credit to GDP. Latin America ranks relatively low, as do many other emerging markets. Therefore, periods of financial deepening may be associated with credit booms, not only because average household debt is increasing, but also because of the entrance of new households into the banking system.

Figure 5.7 shows the depth of the banking system for Latin America and other regions. The depth is indeed quite low, except for Chile, but in most countries of the region it increased during the 2000s. However, this increase was much milder than those in emerging Europe and in the advanced economies. In the latter regions, it is more appropriate to talk about credit booms. The recent evidence is in sharp contrast to the Latin American experience with liberalization in the 1980s, which featured significant credit booms that were followed by the debt crisis and the so-called lost decade. During that period domestic credit increased rapidly, fueled largely by capital inflows in the form of external debt. Countries that had larger expansions also had greater output losses from the crisis (De Gregorio and Guidotti, 1995). Prudential regulation together with macroeconomic policies that did not pursue unsustainable expansions may have been behind the avoidance of a credit boom in Latin America in the first decade of the 2000s, although more empirical research is needed to contrast the two experiences.

Figure 5.7Private Credit as a Share of GDP

Source: IMF, International Financial Statistics.

Note: Advanced = Canada, France, Germany, Italy, Japan, United Kingdom, and United States. Asia = India, Indonesia, Korea, Malaysia, Philippines, and Thailand. Emerging Europe = Czech Republic, Hungary, Latvia, Lithuania, Poland, and Romania. Latin America = Argentina, Brazil, Chile, Colombia, Mexico, Peru, Uruguay, and Venezuela.

Another myth that the Latin American debt crisis of the 1980s put to rest was the idea that when external imbalances originate privately they are not a problem (the Lawson doctrine). The Chilean case during the debt crisis was a prime example of a large private imbalance that caused a very severe crisis. In the aftermath of the 2007–09 global crisis, privately originated imbalances caused crises throughout Europe.

The instrument used most frequently to reduce credit expansion has been reserve requirements, by which banks are required to hold some fraction of their deposits as liquid reserves. Raising reserve requirements increases the costs of borrowing, but to be used as a macroprudential rule the requirements must change in response to the business cycle. Brazil, Colombia, and Peru have used changes in reserve requirements to stem credit booms. According to Tovar, Garcia-Escribano, and Martin (2012), the effects of this policy are moderate and transitory. Although reserve requirements may be effective, they also have some side effects—banks will look for new forms of funding to counteract higher requirements, which may end up generating vulnerabilities. Reserve requirements may also shift credit to unregulated credit providers, providing incentives to the shadow banking system.

The Achilles heel of financial systems in emerging markets has been currency mismatches. The problem has not been in banks’ balance sheets, but in the exposure of corporations that borrowed in foreign currency and have most of their activities in the nontraded goods sector. In emerging Europe even mortgages were denominated in foreign currency. These activities expose the financial system to weaknesses stemming from the currency mismatches of the borrowers. All Latin American countries surveyed in IDB (2005) have regulations on currency mismatches in the banking system. These requirements range from quantitative limits on currency exposure to including exchange rate exposure in quantifications of credit risk, with its consequences for capital requirements. At the corporate level, regulation also requires that the risk of currency exposure of borrowers be internalized. For example, in Chile, provisioning requirements are higher for foreign currency lending when the borrowers have most of their income in domestic currency. Of course, this requires more forward-looking provisioning, which is currently being implemented in several Latin American countries (Cifuentes and others, 2011). In Peru (a dollarized economy) and in Uruguay, additional capital requirements are applied to foreign currency lending to unhedged borrowers.

As shown in the previous section, the balance sheets of banks were resilient to the unprecedented fluctuations in the exchange rate that accompanied the global financial crisis. Only in Brazil and Mexico in Latin America, and in the Republic of Korea in Asia, were some large corporations exposed to exchange rate derivatives. These derivatives were highly complex, which raised financial stability concerns. The lesson is that markets need more disclosure about effective currency exposures in corporations’ financial statements, especially when dealing with complex instruments. Banks should take this into account when making provisions and extending credit.

Another important source of financial risk is the exposure of the domestic banking system to foreign banks. Cross-border flows are highly volatile. Figure 5.8 shows the evolution of cross-border claims of foreign banks in Latin America and Asia. The cycle was more pronounced in Asia. In Korea and Malaysia, the declines in the fourth quarter of 2008 were 6.7 percent and 5.0 percent of GDP, respectively, whereas the maximum decline in Latin America was in Chile, with a fall of 3 percent of GDP in the same quarter.

Figure 5.8Quarterly Change in Cross-Border Claims on Latin America and Asia.

Source: Consolidated Banking Statistics (immediate borrower basis), Bank for International Settlements.

The composition of foreign claims according to source country is presented in Figure 5.9 for emerging Europe, Asia, and Latin America. Emerging Europe is more exposed to European banks, Asia is relatively equally exposed across regions, and Latin America is exposed primarily to Spanish banks, mainly Santander and BBVA. Spanish banks have followed an arm’s-length strategy, letting their subsidiaries operate as independent units. But the main feature that could have made Latin American countries less affected by exposure to foreign banks is that most foreign banks operating as commercial banks have their domestic affiliates constituted as subsidiaries rather than branches. Thus, the subsidiary operates just as a domestic bank would, with its own capital, its own board of directors, and strict rules for deposits of the subsidiary in the parent bank.8

Figure 5.9Foreign Claims of Bank for International Settlements Reporting Banks, 2011:Q2

Source: Consolidated Banking Statistics (immediate borrower basis), Bank for International Settlements.

Note: Asia = India, Indonesia, Korea, Malaysia, Philippines, and Thailand. Emerging Europe = Czech Republic, Hungary, Latvia, Lithuania, Poland, and Romania. Latin America = Argentina, Brazil, Chile, Colombia, Mexico, Peru, Uruguay, and Venezuela.

Subsidiarization does not necessarily produce full ring-fencing, as the cases of several Central and Eastern European countries show, where foreign banks also operated as subsidiaries. But subsidiarization induces more local funding. Figure 5.10 shows that among the three regions, funding comes more from local sources in Latin America.9Kamil and Rai (2010) examine why Latin American financial systems were resilient to the global financial crisis. Their analysis indicates that the resilience stemmed from the fact that global banks’ credit was mostly channeled in domestic currency and foreign banks operated as local subsidiaries, funded mostly with domestic deposits. Therefore, subsidiarization, strong regulation of currency mismatches, a broad base of deposit funding, low reliance on short-term wholesale funding, and a simple trading book may help explain the strength and resilience of the banking system in the region.

Figure 5.10Composition of Foreign Claims, 2011:Q2.

Source: Consolidated Banking Statistics (immediate borrower basis), Bank for International Settlements.

Note: Advanced = Canada, France, Germany, Italy, Japan, United Kingdom, and United States. Asia = India, Indonesia, Korea, Malaysia, Philippines, and Thailand. Emerging Europe = Czech Republic, Hungary, Latvia, Lithuania, Poland, and Romania. Latin America = Argentina, Brazil, Chile, Colombia, Mexico, Peru, Uruguay, and Venezuela.

Finally, another macroprudential tool under discussion for use in emerging markets is capital controls. Similar to reserves accumulation (discussed in the next section), capital controls are instruments with two purposes: to limit exchange rate appreciation and to foster financial stability. However, concerns with exchange rate appreciation and financial stability arise from two different sources within financial accounts. Pressures on the exchange rate depend on net capital flows, whereas financial stability considerations depend on gross flows.

Net inflows are the counterpart of the current account deficit, and the exchange rate depends on the saving-investment balance of the economy. Since the mid-2000s, and contrary to the surge of capital inflows of the early 1990s, current account deficits in emerging markets have been limited, and even many commodity-exporting countries have built current account surpluses. Therefore, they are net exporters of capital. Of course, the data may reveal large net inflows, but these inflows have mostly been associated with the accumulation of reserves primarily to protect competitiveness.

Gross flows have significantly increased as a result of greater financial integration. These flows can have important effects on financial stability, and the use of macroprudential instruments may be called for, as in Korea, where large inflows to the banking system have been considered a source of potential financial risk. Korea has very large foreign funding. The country implemented a capital levy on noncore liabilities in August 2011. Banks pay a levy of 20 basis points for foreign-currency-denominated liabilities of less than 12-month maturity (CIEPR, 2012). Although this levy may reduce foreign borrowing by banks, it does not necessarily reduce net inflows because portfolio shifts may change the source of external funding. In Latin America, Peru has used a similar levy on foreign borrowing and has also applied fees for the sale of Central Bank of Peru bonds. Brazil has experimented with broader capital controls, such as reserve requirements on foreign borrowing and a tax on transactions of fixed-income instruments and equity. Colombia applied an unremunerated reserve requirement, similar to Chile’s in the 1990s, during 2007–08.

The evidence of the effects of capital controls on financial stability and on exchange rates is inconclusive. Financial stability has been preserved in countries that have and that have not applied capital controls, and exchange rates have behaved in similar patterns regardless of the use of capital controls. The jury is still out, but two final comments are in order. First, capital controls may be a disguise for fighting other distortions that encourage capital inflows, such as very high interest rates. This may have been the case in Brazil before the global financial crisis. Second, Chile, the poster child for capital controls during the 1990s, was able to weather the big financial storm of 2008–09 without the use of controls. Perhaps Chile’s experience of the 1990s was predictive of the more recent occurrences in Brazil, where very large interest rate differentials encouraged capital inflows.10

The Accumulation of International Reserves

As discussed before, emerging markets have accumulated large amounts of reserves since 2000 (Figure 5.11).11 The holding of international reserves is one of the most common macroprudential policies in emerging markets for reducing the risk of a balance of payments crisis, but reserves accumulation also serves for exchange rate purposes.12

Figure 5.11International Reserves

Sources: IMF, International Financial Statistics and World Economic Outlook database.

International reserves play a dual role (Aizenman and Lee, 2005). On the one hand, they provide a buffer against sudden stops of capital inflows. This is the self-insurance, or precautionary, motive for accumulating reserves. International reserves reduce the risk of balance of payments crises. On the other hand, the accumulation of reserves entails intervention in the foreign exchange market, ameliorating the pressures for appreciation. This is the mercantilist motive for reserves accumulation. Although the impact of sterilized intervention is not entirely clear, it may transitorily prevent an appreciation of the currency.

The high level of reserves during the global financial crisis played an important role in the resilience displayed by emerging economies. The level of international reserves has been one of the few variables shown to be relevant in mitigating the output costs of the crisis (Frankel and Saravelos, 2010; Gourinchas and Obstfeld, 2012). During the peak of the crisis, countries that had larger volumes of reserves experienced smaller increases in their credit default swap spreads. This reduced the impact of the crisis on financing costs.

International reserves may have rendered credible the provision of liquidity in some cases and protected the exchange rate in others. In particular, Brazil, Korea, and Mexico intervened in complicated moments during the financial crisis because of difficulties in their corporate sectors and may have seen the credibility of these measures enhanced by their massive reserves holdings. Thus, the level of reserves and the significant depreciation of their currencies may have helped mitigate the effects of the global financial crisis. Most countries did not deplete reserves massively, which could be interpreted as reserves having little impact as insurance. However, the financial resilience of emerging market economies strongly suggests that having a high level of reserves, even if unused, can be a strong deterrent to speculation when facing sharp changes in global financial conditions. The majority of models seeking to determine the adequate level of reserves assume that they are used. Still, their deterrent effect is substantial, whether used or unused.

The dual role of reserves accumulation explains why self-insurance is so prevalent in emerging markets, and the current level of reserves could indicate that emerging market economies are over-insured. But over-insurance is related to the other component of demand for reserves: to affect the exchange rate.

Other forms of insurance exist that are less costly than reserves, but with no incidence on the exchange rate. These forms of insurance could also generate pressure to appreciate the currency by signaling less vulnerability to external financial turbulence. Commodity-exporting countries can use commodity hedges instead of hoarding reserves, and hedging could be a better instrument from a financial standpoint. Multilateral contingent credit lines, such as the IMF’s Flexible Credit Line, can be used. Bilateral agreements on currency swap lines can also be signed, which, although common in relatively large economies, are not available for smaller ones.

This dual effect of reserves accumulation could explain why many countries seem to have invested more than necessary in this self insurance. In fact, interventions in foreign exchange markets originate from fear of having a misaligned exchange rate. The dual role of reserves may also explain why so few countries have made use of the IMF’s Flexible Credit Line.

The Flexible Credit Line is a good idea as insurance. However, what would happen if a country decided to take a contingent credit facility instead of hoarding reserves? First, its economy would be safer, encouraging more capital inflows. Second, it would have less reason to intervene in the foreign exchange market because it would already be overinsured and would have other ways to obtain external funding if there were to be a sudden stop of private sources. But the difficulty that countries face when looking for cheaper insurance mechanisms is that reserves decisions also affect the exchange rate.

In sum, reserves held as an insurance mechanism are security against sudden stops of capital inflows and a deterrent to destabilizing speculation against the currency. But reserves accumulation has also been a tool for exchange rate management. Certainly the holding of reserves has a relevant carry cost that must be appropriately weighed when deciding to intervene in the foreign exchange market.

The optimal level of reserves is an issue that has produced significant amounts of research, but is still not clear. However, the evidence from the crisis shows that emerging market economies were well protected with their actual levels of reserves, so at least they were not underinsured. However, keeping these levels of reserves is costly, which limits the space for further accumulation, especially when the costs of holding reserves are appropriately taken into account.

Final Remarks

Big financial crises have been common in emerging market economies. The debt crisis in the early 1980s in Latin America generated many lessons for financial regulation. A first and major lesson was that more was no better than less financial deepening. The development of financial markets is good, but leaving them to grow unfettered is extremely risky and is an almost sure route to crisis. De Gregorio and Guidotti (1995), come to the conclusion that in Latin America, growth during the 1980s was lower in countries with more highly developed financial systems because the collapse of their economies during the debt crisis was larger. Carlos Diaz-Alejandro (1985) eloquently wrote “Good-bye Financial Repression, Hello Financial Crash.” Recent research also points in the same direction—beyond a certain level, the contribution of financial depth to growth is marginal (Arcand, Berkes, and Panizza, 2012; Cecchetti and Kharroubi, 2012).

Today, serious international efforts are being made to establish guidelines to strengthen financial markets. However, “if it ain’t broke, don’t fix it” is a reasonable starting point for reform in emerging markets. Current regulatory proposals are particularly geared to regulating complex financial institutions, which are not the typical banks in emerging markets. Banks in emerging markets are simpler, which is a strong reason for their resilience. Regulation must follow the complexity of institutions, and a first rule in emerging markets is to examine whether to allow financial innovation. Does its potential benefits outweigh its risks? Once that question is answered, appropriate regulation must then be discussed. A wholesale overhaul of existing regulation is not the best starting point.

Many countries have been creating financial stability boards to coordinate all relevant agencies dealing with financial stability. Those boards have to be given clear mandates and assigned responsibilities to avoid overlap and conflicts among agencies. New layers of financial regulation must be consistent with the existing duties of current regulatory agencies. Central banks have to play a critical role in this area, not only because they should lead the design of macroprudential tools and preserve financial stability, but also because they have the independence, or should be granted it where they do not have it, to perform this task effectively. However, independent central banks with a clear mandate for financial stability and the creation of financial stability boards will not prevent future crises, but it should minimize the probability of their happening, and should facilitate resolution. Several relevant steps have been taken in Latin America in this regard.

Efficient and strict regulation of the banking system is essential for promoting financial stability. However, this endeavor also has its risks. As the banking system becomes better capitalized and more regulated, incentives to move financial intermediation to unregulated institutions increase. Thus, the shadow banking system may become larger and riskier. There will always be tension between the extent and the perimeter of regulation, which must be permanently addressed. This has been a persistent problem with the application of capital controls.

Banking systems in Latin America are small and concentrated. Efforts to increase competition are always welcome, but new tensions will appear because competition also encourages search for yield. Indeed, the search for yield in advanced economies was also responsible for excessive risk taking. Coordination between competition authorities and financial regulators is important. Competition cannot be promoted at the cost of increased vulnerability.

As argued by Haldane and Madouros (2012), rules that become too complex are not necessarily robust.13 Preserving simplicity in a complex financial system is not always possible, but avoiding complexity to accommodate the demands of different segments of the market may lead to inefficient regulation. The potential capture by vested interests may endanger financial stability.

One important attribute of regulators in emerging market economies, after having survived many crises, is a reasonable degree of prudence. In general, regulators allow activities that can be handled appropriately by financial intermediaries, but above all that can be understood and monitored adequately by market participants and financial regulators. The same prudent behavior is followed by a significant part of the private sector, who know firsthand the perils of financial innovation. Indeed, once the private sector has paid the costs of its own mistakes it should become more aware of the risks. Bailing out those responsible for causing a crisis creates moral hazard and does not induce prudence and discipline. However, the experience of some nonfinancial corporations dealing with complex derivatives before the crisis demonstrates that financial policies cannot rely on the good judgment of the private sector. Regulation, as well as appropriate risk management within firms, is central to preserving financial stability and minimizing the cost of disruptions.

Good macroeconomic policies and a strong financial system, enhanced by a little bit of luck, allowed emerging market economies to perform reasonably well during the 2007–09 crisis in the world economy. Persevering in fiscal responsibility, inflation control, flexible exchange rates, and robust prudential regulation of the financial system are essential for proceeding from recovery to sustained economic progress.


    AizenmanJ. and J. Lee2005International Reserves: Precautionary vs. Mercantilist Views, Theory and EvidenceIMF Working Paper 05/198 (Washington: International Monetary Fund).

    ArcandJean-LouisEnricoBerkes and UgoPanizza2012Too Much Finance?IMF Working Paper 12/161 (Washington: International Monetary Fund).

    CecchettiStephen and EnisseKharroubi2012Reassessing the Impact of Finance on GrowthBIS Working Paper No. 381 (Basel: Bank for International Settlements).

    Central Bank of Chile2009Financial Stability Report first half (Santiago).

    CéspedesLuis Felipe and AlessandroRebucci2011Macro-prudential Policies in Latin America: A Survey” (unpublished; Washington: Inter-American Development Bank).

    CifuentesRodrigoRodrigoAlfaroEduardoOlaberría and RubénPoblete2011The Chilean Financial System and Macroprudential PoliciesFinancial Stability Report first half (Santiago: Central Bank of Chile).

    Committee on International Economic Policy Reform (CIEPR)2012Banks and Cross-Border Capital Flows: Policy Challenges and Regulatory Responses” (Washington: Brookings Institution).

    CowanKevin and JoséDe Gregorio2007International Borrowing, Capital Controls and the Exchange Rate: Lessons from Chilein Capital Controls and Capital Flows in Emerging Economies: Policies Practices and Consequences ed. by S. Edwards (Chicago: National Bureau of Economic Research and the University of Chicago Press).

    CowanKevinErwinHansen and LuisOscar Herrera2005Currency Mismatches, Balance Sheet Effects and Hedging in Chilean Non-financial Corporationsin External Vulnerability and Preventive Policies ed. by R. CaballeroC. Calderón and L.F. Céspedes (Santiago: Central Bank of Chile).

    de Carvalho FilhoIrineu2010Inflation Targeting and the Crisis: An Empirical AssessmentIMF Working Paper 10/45 (Washington: International Monetary Fund).

    De GregorioJosé2011International Reserves Hoarding in Emerging MarketsEconomic Policy Paper No. 40 (Santiago: Central Bank of Chile).

    De GregorioJosé2012Comments on Haldane and MadourosPresentation at the Federal Reserve Bank of Kansas City’s economic policy symposiumThe Changing Policy LandscapeJackson Hole WyomingAugust30September1.

    De GregorioJosé and PabloGuidotti1995Financial Development and Economic GrowthWorld Development Vol. 23 No. 3 pp. 43348.

    De GregorioJosé and AndreaTokman2007Overcoming Fear of Floating: Exchange Rate Policies in Chilein Monetary Policy in Emerging Markets and Other Developing Countries ed. by N. Batini (New York: Nova Science Publishers Inc.).

    Diaz-AlejandroCarlos1985Good-bye Financial Repression, Hello Financial CrashJournal of Development Economics Vol. 19 No. 12 pp. 124.

    FrankelJeffrey and GeorgeSaravelos2010Are Leading Indicators of Financial Crises Useful for Assessing Country Vulnerability? Evidence from the 2008–09 Global CrisisNBER Working Paper No. 16047 (Cambridge, Massachusetts: National Bureau of Economic Research).

    FrankelJeffreyCarlosVégh and GuillermoVuletin2011On Graduation from ProcyclicalityNBER Working Paper No. 17619 (Cambridge, Massachusetts: National Bureau of Economic Research).

    GourinchasPierre-Olivier and MauriceObstfeld2012Stories of the Twentieth Century for the Twenty-FirstAmerican Economic Journal: Macroeconomics Vol. 4 No. 1 pp. 22665.

    HaldaneAndrew G. and VasileiosMadouros2012The Dog and the FrisbeePresented at the Federal Reserve Bank of Kansas City’s 36th economic policy symposiumThe Changing Policy LandscapeJackson Hole WyomingAugust30September1.

    Inter-American Development Bank (IDB)2005Unlocking Credit: The Quest for Deep and Stable Bank Lending Economic and Social Progress in Latin America: 2005 Report (Washington: Inter-American Development Bank).

    Inter-American Development Bank (IDB)2012The World of Forking Paths: Latin America and the Caribbean Facing Global Economic Risks The Latin American and Caribbean Macro Report 2012 (Washington: Inter-American Development Bank).

    JácomeLuis I. Erlend W. Nier and PatrickImam2012Building Blocks for Effective Macroprudential Policies in Latin America: Institutional ConsiderationsIMF Working Paper 12/183 (Washington: International Monetary Fund).

    KamilHerman and KulwantRai2010The Global Credit Crunch and Foreign Banks’ Lending to Emerging Markets: Why Did Latin America Fare Better?IMF Working Paper 10/102 (Washington: International Monetary Fund).

    MagudN. C. Reinhart and K. Rogoff2011Capital Controls: Myth and Reality—A Portfolio Balance ApproachNBER Working Paper No. 16805 (Cambridge, Massachusetts: National Bureau of Economic Research).

    MillerMarcusPaulWeller and LeiZhang2002Moral Hazard and the U.S. Stock Market: Analyzing the Greenspan PutEconomic Journal Vol. 112 No. 478 pp. C17186.

    TovarCamilo E. MercedesGarcia-Escribano and MercedesVera Martin2012Credit Growth and the Effectiveness of Reserve Requirements and Other Macroprudential Instruments in Latin AmericaIMF Working Paper 12/142 (Washington: International Monetary Fund).

To limit the number of countries, the analysis includes Argentina, Brazil, Chile, Colombia, Mexico, Peru, Uruguay, and Venezuela. In some cases countries with no data available are excluded, which happens occasionally for Venezuela and Uruguay. For Asia, the analysis considers the Republic of Korea, India, Indonesia, Malaysia, the Philippines, and Thailand. For emerging Europe it considers the Czech Republic, Hungary, Latvia, Lithuania, Poland, and Romania.

For further discussion on exchange rate flexibility and declining pass-throughs, see De Gregorio and Tokman (2007).

Frankel, Végh, and Vuletin (2011) provide an empirical investigation of the procyclicality of fiscal policies in EMs. Their analysis takes decades as periods, so it cannot isolate the cyclicality of the budget during the crisis. However, they still find countries graduating from procyclicality in the period 2000–09.

There is some evidence that even in periods before the floating of exchange rates the corporate sector’s assets and liabilities in foreign currency were fairly well matched. Therefore, the fear of a financial crisis after a depreciation may have been unfounded. For Chile, see Cowan, Hansen, and Herrera (2005).

In the context of the Greenspan put, the issue of excessive risk taking and lack of proper risk management was raised more than 10 years ago by Miller, Weller, and Zhang (2002).

For recent examinations of the use of macroprudential tools and the institutional arrangements of financial regulation in Latin America, see Céspedes and Rebucci (2011), Cifuentes and others (2011), Jácome, Nier, and Imam (2012), and Tovar, Garcia-Escribano, and Martin (2012). Given the complexities of a full comparative analysis, the reviews in general are partial and focused on specific country experiences, as is this section: most of the examples and details come from the Chilean experience.

Chile appears to have the largest leverage ratio in the region. However, it is among the countries that had the lowest volatility of output, and the evidence shows that the lower the volatility, the higher the leverage. See Central Bank of Chile (2009) for further evidence.

Many countries regulate branches and subsidiaries identically. The most relevant difference is that branches do not have local boards of directors, whereas subsidiaries do. With branches, the foreign bank is responsible for any problem in its affiliate, thus, a subsidiary limits contagion. In addition, subsidiaries can have local or other partners. These are strong incentives for banks to use the subsidiary model to expand across regions.

For further discussion on cross-border banking, see CIEPR (2012), and for an analysis of vulnerabilities of Latin American economies to exposure to Spanish banks, see Chapter 8 of IDB (2012).

For a review of the Chilean experience with capital controls, see Cowan and De Gregorio (2007), and for a large set of countries, see Magud, Reinhart, and Rogoff (2011).

The data in Figure 5.11 are expressed as percentages of GDP. In some cases this ratio declined simply because reserves have grown less than GDP valued in dollars, as in Chile and Indonesia, although their levels in dollars have increased over time. The fact that in 2011 most countries had more reserves as a fraction of GDP does not imply that the reserves were never used; some countries actually used a small fraction of their reserves during the 2007–09 crisis, although they resumed accumulating reserves thereafter.

The issues discussed in the remainder of this section are developed in greater detail in De Gregorio (2011).

See also comments by De Gregorio (2012).

    Other Resources Citing This Publication