3. Why Has Latin America and the Caribbean Fared Better This Time? The Value of Being Prepared
- International Monetary Fund. Western Hemisphere Dept.
- Published Date:
- October 2009
Although it has faced larger external shocks this time, the Latin America and Caribbean (LAC) region has fared noticeably better than in the earlier three global downturns since the 1980s. It has also fared better than other emerging markets. This better performance can be attributed to stronger and more credible policy frameworks, which led to lower banking, external, and fiscal vulnerabilities and allowed some LAC countries to react with monetary or fiscal policy easing.
Better Output Performance
The impact of the global crisis has no doubt been large. But the region has not experienced the large-scale banking or balance of payments crises that besieged it in the past. Why has Latin America and the Caribbean fared so differently, and what do we learn from this?
Better than in Past Crises
In the past 30 years, there have been three global recessions (1982, 1998, and 2001), which severely affected LAC countries. These earlier recessions were less severe than the current one but were large enough to trigger major output losses in the region (Figure 3.1). Moreover, on each occasion, LAC output fell more than world GDP. It took the region many years to fully recover from the fall of 1982; output performance was notably weak in both 1998 and 2002, partly owing to homegrown “amplifiers.” A year after the onset of those global crises, the LAC region’s output was on average about 4 percentage points lower than would have been the case if the region had grown at the world rate. This gap persisted and indeed widened in consecutive years.
Figure 3.1.The LAC region is faring relatively better this time, with regional output expected to recover in tandem with the world.
Source: IMF staff calculations.
1/ GDP-PPP weighted average.
2/ The data show the average evolution 3 years before and after the previous crisis episodes (t=0 corresponds to the episodes of years 1982, 1998, and 2001).
3/ The data show the evolution 3 years before and after the 2009 crisis episode (t=0). Dotted lines correspond to projections.
In the current global recession, Latin America and the Caribbean is not expected to repeat this poor performance. GDP data, available through the second quarter, show the LAC region moving broadly in line with the world economy. And considering what we expect for 2009, the region’s output will have moved much closer to the global economy than in the past. Private sector Consensus Forecasts suggest a similar relative picture, which is somewhat more positive for both Latin America and the Caribbean and global growth than is the IMF staff forecast.
This better performance is expected despite external shocks during the current crisis being larger than in earlier global recessions (Figure 3.2):
World output contraction was much larger, dampening external demand. Reduced external demand for LAC exports worldwide arguably hurts the region more this time than in earlier recessions because the region has become more open and integrated with the world economy. Moreover, trade fell more significantly this time, albeit from historically high levels.
On the financial side, the external shock was strong. While U.S. interest rates were at an all-time low, LAC spreads widened substantially following the Lehman collapse, sharply increasing total borrowing cost.
Figure 3.2.Latin America and Caribbean’s better performance is expected despite more severe global shocks.
Sources: IMF, International Financial Statistics; Merrill Lynch; and IMF staff calculations.
1/ The data show the average evolution 3 years before and after the previous crisis episodes (t=0 corresponds to the episodes of years 1982, 1998 and 2001).
2/ The data show the evolution 3 years before and after the 2009 crisis episode (t=0). Dotted lines indicate projections.
3/ In this case, the previous crisis episodes considered are 1998 and 2001.
4/ Data for 2009 is the average of the latest data available for 2009.
Better than Other Regions
Latin America and the Caribbean’s performance during this crisis is also considerably stronger than that of other emerging markets. A comparison of the (downward) revisions to forecasts of 2009 growth among a set of almost 40 emerging market economies from around the world illustrates this point.
The 12 Latin American countries in the sample were not among those experiencing the largest output shocks in 2009. In fact, their median growth revision was about 3 percentage points smaller than that of the other emerging countries (Figure 3.3).
Figure 3.3.Revisions in growth forecasts have been smaller in Latin America than in other emerging markets.
Sources: Consensus Economics; and IMF staff calculations.
1/ 2009 growth rate expected in August 2009 minus 2009 growth rate expected in August 2008.
Given that it was facing larger shocks, does Latin America and the Caribbean’s better relative performance this time reflect higher vulnerabilities in other regions or improved resilience in LAC?
Preliminary evidence suggests that some LAC countries really have changed, that they are now better able to cope with external shocks of any given size. In particular, if the LA5 (Brazil, Chile, Colombia, Mexico, and Peru) had shown the same sensitivity to external shocks as they did in 1994–2002, their output would have been far weaker during the first half of 2009 relative to their actual performance (Figure 3.4). Model-based estimates considering the behavior of such external variables during the current crisis suggest that these countries “saved” about 4 percentage points of GDP (see Box 3.1).
Figure 3.4.Improved resilience in the LA5 helped mitigate the decline in growth.
Sources: IMF staff calculations.
1/ Simple average of 4 quarter percent changes in growth rates among the LA5 countries (Brazil, Chile, Colombia, Mexico, and Peru).
Box 3.1.Did Latin America Show Greater Resilience during the Current Global Crisis? A Counterfactual Exercise
External shocks played a dominant role in slowing and stopping GDP growth during the recent crisis. Model-based estimates suggest that shocks to external trade and financing explain between 80 percent and 90 percent of the sudden drop in average GDP in LA5 countries (Brazil, Chile, Colombia, Mexico, and Peru) seen in the three quarters following the onset of the crisis in September 2008.1 Of these, negative foreign trade shocks (the combined effect of world GDP slowdown and lower commodity export prices) accounted for more than two-thirds of the contribution of external shocks during this period.
Contribution of Shocks to the Reduction in LA5 GDP between 2008:Q3 and 2009:Q2
Source: IMF staff calculations.
Given the large size of the recent external shocks, the resulting fall in LAC output growth was much less than it would have been if the region’s countries had not achieved fundamental improvements in the credibility of their policy frameworks and reduced financial vulnerabilities. To illustrate this point, IMF staff calculated how these countries would have reacted in the 1990s and early 2000s (a period of weaker fundamentals, greater dollarization of liabilities, and banking systems that were often poorly regulated and capitalized) to a global downturn similar to that in recent quarters. For these purposes, IMF staff used the coefficients of the BVAR model estimated over the period 1994:Q2–2002:Q4 and the actual changes in world demand, commodity prices, and financial variables observed during the period 2008:Q4–2009:Q2.2
According to model-based estimates, the payoff for stronger fundamentals has been very large, suggesting that Latin American saved about 4 percentage points of GDP during the current crisis. If Latin America had responded to external shocks in the same way as during the 1990s and early 2000s, the model predicts that the level of GDP in these countries would have fallen by 7.9 percent between end2008:Q3 and end-2009:Q2. The observed drop in output induced by these external factors, however, was only about half as large as this, at 4 percent. This strongly suggests that reduced financial vulnerabilities, war chests, and the ability to use countercyclical policies prevented the magnification of external shocks that prevailed in previous world downturns.Note: This box was prepared by Herman Kamil. The author thanks Pär Österholm for his generous help in estimating the model.1 For a description of the model used, see Österholm and Zettelmeyer (2007). Historically, 55 to 60 percent of the variation in Latin American GDP growth between 1994 and 2007—a quieter time, on average—has been accounted for by external trade and financial shocks.2 A caveat to this exercise is that initial conditions in 2002:Q4 and 2008:Q3 could have been different also, making the process of comparison somewhat more complex than illustrated here.
As noted earlier, the region faced the crisis with much stronger external positions (Figure 3.5). Many countries reduced external current account deficits and external debt in recent years. Some had substantial surpluses, particularly commodity exporters that benefited from the boom in commodity prices. International reserves were also at much more comfortable levels.
Figure 3.5.Vulnerabilities have been lower this time. International reserve growth was faster and public debt burdens and banking sector leverage were lower.
Sources: IMF, International Financial Statistics; and IMF staff calculations.
1/ Annual medians of ARG, BOL, BRA, CHL, ECU, PRY, PER, URY, VEN, MEX, CRI, PAN, DOM, and TTO.
2/ The data show the average evolution 3 years before and after the previous crises episodes (t=0 corresponds to the episodes of years 1982, 1998 and 2001).
3/ The data show the evolution 3 years before and after the 2009 crisis episode (t=0). Dotted lines indicate projections.
Fiscal vulnerabilities also were lower. The current crisis met the region with much lower public sector borrowing requirements than in the past, because public sector balances had been significantly improved. Lower financing needs resulted from lower debt levels and also from less reliance on very short-term debt. In addition, many countries had increased domestic financing opportunities through the development of domestic capital markets.
The composition of debt was safer. In past global recessions, highly dollarized public debt increased on impact owing to the large depreciation of the currency. Market access was interrupted for some time, and entrenched inflation expectation kept borrowing costs high, resulting in an onerous debt service. Public sector borrowing requirements were larger also because more debt was indexed to international interest rates. This time, the impact on the debt stock was smaller, market access was not interrupted for long, and the increase in interest rates for new borrowing generally did not significantly increase public sector borrowing requirements and external interest rate payments.
On the banking side, the rapid credit expansion observed in many LAC countries before the global crisis did not turn out to be a major vulnerability. Credit growth was being financed mostly from domestic deposits and authorities had put in place prudential measures that limit excessive risk taking (see May 2009 Regional Economic Outlook). Moreover, capital adequacy levels were comfortable this time, notwithstanding rising nonperforming loans in response to the downturn (see Chapter 2). As a result, the region did not suffer a banking crisis—unlike earlier global financial crises.6
Altogether, these lower vulnerabilities made a difference. Our analysis suggests that because of these changes, output losses in LAC countries were lower than elsewhere. Countries with more leveraged financial systems and faster credit growth experienced the largest output losses (Figure 3.6). In all these areas, the LAC region also did better than other emerging market countries (Box 3.2).
Figure 3.6.Countries with more-leveraged financial systems, limited exchange rate flexibility, and higher credit growth experienced sharper revisions to growth.
Sources: IMF staff calculations.
1/ Estimated contribution of each independent variable to the predicted change in expected growth arising from increasing each independent variable from its first to its third quartile value, in a sample of 40 countries.
Cross-country analyses such as the one discussed in Box 3.2 have some difficulty explaining output dynamics in Mexico during this crisis. One reason may be that cross-country econometric models cannot pick up factors that seem to be special to Mexico, including the tight integration of its economy with the U.S. industrial sector, which plunged in 2009. Box 3.3 discusses this and other factors that also played a role in Mexico’s weaker performance in late 2008 and the first half of 2009. Available indicators suggest that Mexico’s economy returned to growth in 2009Q3.
Stronger Policy Frameworks
Policy frameworks in many LAC countries have improved substantially during the last decade, particularly among the largest economies. Countries in the financially integrated commodity exporting group, for example, adopted inflation targeting and more flexible exchange-rate regimes. Several countries also have adopted fiscal frameworks that establish fiscal and debt sustainability rules.
Stronger and more credible policy frameworks, which led to lower external and public sector borrowing requirements, allowed those countries to implement active countercyclical policies to smooth the impact of external shocks.
Box 3.2.The Global Financial Crisis: Why Were Some Countries Hit Harder?
In the aftermath of the collapse of Lehman Brothers in September 2008, some countries fared better than others, suggesting that specific country features may have influenced the transmission of the crisis. A recent IMF staff study focuses on “growth forecast revisions”—the change in forecasts for 2009 between 2008 and 2009—and looks at whether trade and financial openness, vulnerabilities, and policies can explain why some countries weathered the global crisis better than others. The focus on growth revisions carries several advantages: they are not affected by cyclical corrections or other anticipated adjustments in growth, allow for a flexible lag in the transmission of the global shock to each country, and take into account expectations of the likely success of policy responses. According to this measure, the countries most affected by the global crisis would have larger downward revisions to their growth forecasts for 2009.
The study finds—for a sample of 40 emerging market countries—that financial vulnerabilities contributed to larger negative growth revisions, while exchange rate flexibility served as a shock absorber. Countries with greater financial vulnerabilities—more-leveraged domestic financial systems and more rapid growth in lending to the private sector—tended to suffer larger downward revisions to their growth outlooks. Also, countries with less flexible exchange rate regimes experienced a larger downward revision in their growth outlooks for 2009. In countries with more flexible regimes, currency depreciation seems to have helped cushion the effect on growth by curtailing the incentives for capital outflows and supporting the profitability of exports.
Trade linkages also appear to have an effect on growth performances, although the trade variables are statistically significant only in a larger sample of countries that includes lower-income as well as emerging market countries. In particular, the composition of exports rather than overall trade openness seems to explain different growth patterns across countries. While the share of commodities (both food and overall) in exports is associated with smaller negative growth revisions, the share of manufacturing (both advanced and overall) in exports is associated with larger negative revisions.
Interestingly, lending from advanced economies appears to matter in the larger group of countries, though this is not detected in the smaller sample of emerging market economies. Using the larger sample, the study finds some support for the notion that countries with stronger financial linkages with advanced economies, measured as their borrowing from advanced economies as a share of GDP, experienced larger negative growth revisions.
Changes in expected growth for 2009 by different country groups
Sources: Consensus Economics; and IMF staff calculations.
Current Acount Balance
Source: IMF staff calculations.
Primary Fiscal Balance Gap
Source: IMF staff calculations.
There is also some evidence that several other factors helped cushion the blow. Specifically:
External vulnerabilities. Countries with stronger external current account balances tended to experience smaller growth revisions, most likely by reducing their vulnerability to a sudden stop of capital flows.
Strong fiscal position. Some of the results suggest that a stronger primary balance prior to the crisis, as measured by the primary balance gap (the difference between the actual primary balance and the balance consistent with a constant debt-to-GDP ratio), helped reduce the magnitude of the output decline, whereas other fiscal indicators, such as public debt as a share of GDP, appeared to have little effect. This is consistent with the notion that a strong primary balance gives a country more room to undertake countercyclical fiscal policies without raising concerns about debt sustainability.
Interestingly, other measures of the strength of economic policies (besides exchange rate and fiscal policies) did not appear to dampen the effects of the global crisis on growth in emerging markets. Variables capturing differences in economic policies, such as the adequacy of international reserves, the level and variability of inflation, and credit ratings, did not yield significant results.
On balance, these results suggest that the global financial crisis hit the large emerging market countries at their point of weakness. Larger corrections in growth were experienced by those countries with more-leveraged financial systems, faster credit growth, and limited exchange rate flexibility. There is some support for the view that strong current account and fiscal positions helped cushion the blow. At the same time, there is some evidence that export linkages mattered.Note: This box was prepared by Pelin Berkmen, Gaston Gelos, Robert Rennhack, and James P. Walsh. The results of the study will be described fully in a forthcoming working paper.
Box 3.3.¿Qué Pasó? Behind Mexico’s Cycle, by Way of Comparison to Canada
The economies of both Mexico and Canada are expected to contract this year, reflecting their strong real and financial linkages with the U.S. economy. However, whereas the Canadian economy is projected to shrink less than that of the United States (unlike in past recessions), the Mexican economy is projected to experience its largest contraction since the Tequila crisis and shrink by more than 7 percent in 2009. Stronger spillovers from the manufacturing sector and tighter financing conditions are likely among key contributors to Mexico’s weaker performance.
|Total exports of goods||31||27|
|Exports of goods to the U.S.||23||21|
|Cross-holdings of financial assets||91||49|
Variance Decomposition of Real GDP 1/
Real GDP Growth
Source: Haver Analytics.
Canada and Mexico are both strongly exposed to the U.S. economy. This includes real links—more than three-fourths of their exports are directed to the United States—as well as financial links, with significant funding for both Canadian and Mexican firms sourced in the United States. Indeed, reflecting strong integration of production in North America, the decline in manufacturing activity has been very similar across North American Free Trade Agreement (NAFTA) countries. The crisis in the North American auto industry (in which production and trade flows have fallen by 40 percent or more) has played a role, but the correlation also holds for the rest of manufacturing.
Mexico is projected to contract more than would be expected in 2009, whereas Canada is proving more resilient. The volatility of growth in Mexico has historically been greater than that in Canada, and spillovers from activity in the United States have been larger. For the post-Tequila period, the response of Mexican real GDP to U.S. factors is estimated to have ranged between 0.9 and 1.5 times the size of the shock to U.S. industrial production and real GDP, respectively, reflecting increasing trade and linkages with the U.S. economy post-NAFTA.1 In contrast, the spillovers to growth in Canada have fallen. Some estimates suggest a 1–1 reduction in Canadian growth in line with a U.S. growth shock. Comparing these rules of thumb with the IMF’s current projections for growth across NAFTA countries in 2009 suggests that Mexico could contract somewhat more than explained by past relationships with the United States. Canada, on the other hand, appears to be doing better than in the past based on these metrics.2
In part, this may reflect that Mexico’s links to U.S. fluctuations have increased, whereas they have come down in Canada. The research previously cited shows that the variance of Mexican output explained by that in the United States has increased in the post-Tequila crisis period, whereas it has fallen in Canada. This may reflect the growing resilience of Canada in the last decade because policy frameworks were significantly strengthened. Although macroeconomic frameworks in Mexico have also been substantially strengthened, Mexico’s integration into the U.S. manufacturing sector increased sharply following the introduction of NAFTA.
|% of total||y/y change||% of total||y/y change|
Also, the decline in manufacturing appears to have had larger cross-sectoral effects in Mexico than in Canada. Spillovers to services from the sharp decline in manufacturing production have been much more significant in Mexico than in Canada. Activity in transportation and trade,3 which are relatively more important in Mexico, has fallen far more sharply in Mexico than in Canada. Indeed, IMF staff analysis shows that more than one-fourth of the variation in all services output in Mexico can be explained by shocks to U.S. industrial production despite the absence of strong trade links to these sectors, suggesting significant spillovers. However, the mechanism through which they operate remains an area for further research (Sosa, 2008).
The substantial tightening of financing conditions has also weighed on growth in Mexico relative to that in Canada and in part explains Mexico’s weakerthan-expected outlook. The cost of financing (for example, corporate spreads) has risen sharply in both countries but more so in Mexico. Indeed, although financial markets in both countries have escaped substantial disruption, a number of large Mexican firms experienced a sudden stop of financing in the aftermath of large losses on derivative operations in late 2008.4 Moreover, overall credit extension by the domestic banking system has decelerated sharply in Mexico, to 4.2 percent year on year in 2009:Q2 from 18.8 percent year on year in 2008:Q1, though it is unclear to what extent this reflects balance sheet pressures on Mexican subsidiaries of troubled global banks, as opposed to reduced credit demand. In Canada, the deceleration in credit growth has been much less pronounced, to about 6 percent during the first half of 2009 from 8.3 percent in 2008.
Corporate Bond Yields
Sources: Merrill Lynch; and Credit Suisse.
Sharply slowing consumer credit, deteriorating labor market conditions, and effects of the H1N1 flu are additional Mexico-specific sources of drag. Available data suggest that labor market conditions have deteriorated substantially in both countries, with unemployment at an 11-year high in Canada, and substantial declines in employment in Mexico. The impact on consumption, however, is ameliorated in Canada by comprehensive unemployment benefits.5 Meanwhile consumer credit has decelerated sharply in Mexico, across both foreign and domestic banks, whereas it was essentially unchanged from 2008 levels in Canada. Combined with a greater weakening in confidence, these factors have undermined consumption demand. More broadly, part of the weak performance in Mexico in 2009 is explained by the effects of the H1N1 flu, which could lower GDP growth by about 0.5 percent.
Consumer Credit Impulse
Sources: Haver Analytics; and IMF staff calculations.
Differing policy constraints. As has been the case elsewhere, macroeconomic policies have been eased substantially to cushion the impact of the global crisis in both Canada and Mexico. This has been done in the context of strong policy frameworks—inflation targeting and fiscal rules (a gradual debt reduction rule in Canada and a balanced-budget rule in Mexico). Indeed, the policy response has been exceptional by historical standards (policy rates are at historical lows in both countries, and the fiscal impulse is also sizable) and has likely acted to reduce the downturn in both countries relative to their historical experience. However, the extent of market disruption engendered by the global crisis and different starting conditions has restricted room for maneuver, more in Mexico (and other emerging economies) than in Canada. As such, policies have been eased faster and to a greater extent in Canada than in Mexico, and also explain part of the different growth outcomes in 2009, including Canada’s better-than-historical performance.Note: This box was prepared by Kornelia Krajnyak, Evridiki Tsounta, and Ivana Vladkova-Hollar.1Schmitt-Grohe (1998), Sosa (2008), and Swiston and Bayoumi (2008) estimate the rules of thumb discussed here. U.S. industrial production has historically been more volatile than GDP.2 GDP growth in 2009 is projected to contract by −7.3 percent in Mexico, −2.1 percent in Canada, and −2.8 percent in the United States. Mexican GDP fell by −9.1 percent year on year in 2009:H1, during which period U.S. industrial production fell by −12.4 percent.3 A split into retail and wholesale trade activity (the latter more directly related to manufacturing output) is unavailable for Mexico.4 Total external financing flows to the domestic private sector turned strongly negative in 2008:Q4, resulting in a total outflow of about US$7 billion by the end of 2009:Q1.5 Although remittances have fallen sharply in U.S. dollar terms, their impact on household incomes has been muted by their increased Mexican peso value arising from the currency depreciation.
Flexible exchange rates acted as a key shock absorber. Domestic currencies depreciated rapidly in many countries following the collapse of Lehman Brothers. In contrast to past global recessions, the depreciations did not have destabilizing effects on domestic balance sheets, avoiding more serious disruptions in economic activity.
The gains made on the monetary front are one key reason why those effects were not present this time. New monetary regimes encouraged lower levels of dollarization, which led supervisors to adopt and carefully enforce prudential regulations that limit exchange rate risk, and in some cases prompted the development of markets that allow the diversification of such risk (Figure 3.7).
Figure 3.7.Monetary and financial policy frameworks have encouraged lower exchange rate risk exposure.
Source: Bank for International Settlements, IMF, International Financial Statistics; and IMF staff calculations.
1/ The data show the average evolution 3 years before and after the previous crise episodes (t=0 corresponds to the episodes of years 1982, 1998 and 2001).
2/ The data show the evolution 3 years before and after the 2009 crisis episode (t=0). Dotted lines indicate projections.
3/ Simple average for ARG, BRA, CHL, PER, MEX, COL, and VEN. A country has a net foreign currency asset position, as a share of export values. Negative numbers indicate a net debtor position in foreign currency.
4/ Simple average for ARG, BRA, CHL, CRI, DOM, MEX, PER, URY, and VEN. Dollar deposits meassured at constant nominal exchange rates.
The new monetary policy frameworks also allowed for better deployment of international reserves. During this crisis, reductions in international reserves in LAC countries were moderate. Instead of committing reserves to defend an exchange rate peg, these were used prudently to help LAC corporate and financial sectors meet their immediate external obligations.
More credible fiscal frameworks also played a role. In contrast to the situation in earlier years, and thanks to prudent behavior in the last decade, many countries were able to allow the fiscal deficit to rise by letting automatic stabilizers work, mainly on the revenue side. Fewer countries were able to increase discretionary spending, in an active countercyclical fiscal policy response (Chapter 4). This shielded the domestic economy from the additional fiscal contraction observed in earlier worldwide recessions. In addition, some governments used their savings (liquid assets) abroad to finance their larger overall fiscal deficits, allowing fiscal stimulus with little impact on the gross debt burden.
In sum, this time, many LAC countries were able to react to the crisis with monetary or fiscal policy easing. And a number of countries were able to ease both policies (see lower right quadrant below in Figure 3.8).
Figure 3.8.Several countries were able to ease fiscal and monetary policies.
1/ Bubble size is proportional to 2008 GDP at purchasing power parity.
2/ Primary deficit based on domestic tax revenues excluding commodity revenue and grants, in percentage points of GDP.
Source: IMF staff calculations.
Globally and in Latin America and the Caribbean, the recent experience is very rich and will certainly inform future policy agendas. In some respects, the crisis has brought some new lessons and challenges that will require new thinking—starting with financial regulation and supervision. At the same time, this global recession has confirmed older lessons, underscoring policies and conditions that make a country less vulnerable to external shocks. In particular, the global crisis has put to the test policies and frameworks that have been increasingly adopted by LAC countries in the last decade. It has served to show that economies that are open to global trade and finance can withstand severe external shocks without falling into a crisis—and can use the “policy space” gained in normal times to counter effects of external shocks. The region’s progress in reducing vulnerabilities and building capacity for countercyclical policies needs to continue.
Drawing also on the discussions in earlier chapters, we see policy implications in four areas.
1. Financial sector sins and solutions: excess risk taking in the financial sector can bring down any economy; ongoing efforts to contain systemic financial risk are therefore essential.
Recently identified gaps and weaknesses in financial regulation and supervision in advanced economies—the origin of the recent crisis—need to be corrected.
Past efforts at strengthening LAC financial regulation and supervision have paid off, as confirmed by its resilience to recent shocks. But progress has not been uniform, and even countries now in the lead need to keep moving ahead.
Although LAC financial sectors were not undertaking the complex financial operations that were at the core of the U.S. financial crisis, the regulatory and prudential frameworks of the region should incorporate the newly learned lessons on how to better handle these risks.
In particular, countries should consider (1) countercyclical dynamic provisioning, (2) establishing clearer mandates and responsibility for financial stability, (3) clarifying and widening the perimeter of financial regulation, (4) paying closer attention to factors contributing to systemic risk, so that they can be avoided, and (5) adopting other regulations to limit liquidity risks and other mismatches.
A period of significant capital inflows to some LAC countries is likely in the near term, as discussed in Chapter 2, and this could facilitate the formation of bubbles. This possibility also underscores the importance of financial policies that align private incentives correctly to avoid excess risk taking.
2. Well-established, predictable, and transparent frameworks for monetary and fiscal policy can make a difference in mitigating the effect of shocks on activity.
Credible monetary and fiscal policies can play a countercyclical role in dire times. Monetary policy is normally the first line of defense for countries in a position to use it. Monetary policy decisions can be taken quickly and with an immediate effect on expectations (see chapter 4).
The record of previous fiscal policies, as reflected in public debt levels and perceptions of credibility, will determine whether a government can act as a stabilizing force during downturns. The first challenge is to be able to maintain government expenditure when revenues decline temporarily.
In countries with the strongest fiscal frameworks, it will be possible to let automatic stabilizers operate, and often this may be all the fiscal support necessary. In cases of severe shocks, discretionary fiscal policy will be a useful tool. These governments will be able to borrow even in bad times, and may also choose to draw down previously accumulated financial assets.
Better yet is to increase automatic fiscal stabilizers or to have crisis response expenditure plans prepared, ready to be implemented when certain contingencies arise.
Fiscal rules can play a significant role in ensuring sustainability while allowing some degree of fiscal support during downturns. Many fiscal rule designs are possible, but one that works well is to keep public spending in line with potential output growth in normal times, so that boom times generate the resources that can be used during downturns.
Transitions from rule-based fiscal policy reactions and (exceptional) discretionary fiscal policy should be defined in escape clauses established in advance. This will allow fiscal policy, during future extreme situations, to react and help monetary policy support activity without undermining credibility.
3. Preparations for external shocks: the recent experience confirms the interrelated values of strengthening balance sheets, reducing currency mismatches, and allowing significant flexibility in the exchange rate.
Across emerging economies worldwide, those with larger fiscal and external deficits and debt, significant foreign currency mismatches, and less flexible exchange rates tended to have greater difficulties coping with the global crisis.
Countries with intermediate exchange rate regimes should consider moving in the direction of greater exchange rate flexibility. At the same time, for some countries, a hard peg to a foreign currency will continue to make sense. These countries should try to keep relatively larger fiscal policy space to confront shocks successfully.
“Fear of floating” can be overcome. The recent experience confirms that currency depreciation need not be a channel through which a foreign crisis becomes a domestic one. With the right steps and policies, the ability to let the exchange rate move rapidly in response to external shocks can be part of the solution to a capital account shock. Quick overshooting of exchange rates to levels clearly below equilibrium helps contain capital outflows, by establishing expectations of eventually large gains. The key is to establish the right incentives to keep exposure contained—this means good data transparency as well as appropriate regulation of currency mismatches. Central banks must also provide a credible anchor to contain passthrough of depreciation to inflation.
That said, the availability of substantial foreign exchange liquidity was also a stabilizing factor in the recent global recession—even in countries with high exchange rate flexibility. Although such countries made use of their reserves to ease foreign exchange pressures, intervention occurred in controlled amounts, still allowing the market to continuously determine exchange rates.
4. An issue that will need more consideration: determining the optimal level of a country’s international reserves and the rules of access to foreign exchange liquidity.
The recent crisis has led some to conclude that a very large stock of official reserves is a necessary preparation for facing external shocks—but this conclusion is unwarranted and a worrisome development in itself.
Reserves provide a form of self-insurance, but this insurance has a cost. One such cost is the spending that does not take place owing to the reserve accumulation, as well as the implicit loss from earning low rates of return on reserves. Moreover, there is an international coordination problem: what might be seen as good for one country may be counterproductive collectively.
The difficult issues ahead are how much foreign exchange liquidity countries should have, and how best to accumulate this liquidity, without creating distortions (including on the pricing of the exchange rate).
Within the LAC region, a number of countries already have sizable reserves, and there is no presumption that they should strive to reach the still higher levels of reserves of some countries in other regions.
In the end, securing access to foreign exchange liquidity through international cooperation may be the best way for all to insure against future shocks.
Methodological Appendix: Box 3.2
This study draws on data from a range of sources to estimate several regressions. The growth revisions come from the Consensus Forecast database and are defined as the difference between the averages of January-June 2008 and January-June 2009 projections for growth in 2009. These revisions were calculated for the 40 emerging market countries in the Consensus Forecast database. Data on trade, bank credit and deposits, exchange rate regimes, and public finances come from IMF databases. International lending data are from the Bank for International Settlements. Composition of trade data are published by the World Trade Organization. Various indicators on governance and country risk were compiled by The PRS Group, Inc. The regressions were conducted in ordinary least squares. The study uses precrisis values for the explanatory variables to limit any problems of endogeneity. Similar results are obtained for regressions using growth revisions defined as the difference between the August 2009 projections and April 2008 projections (see Table 3.1).
|Specification Dependent variable: Change in consensus forecast||(1)||(2)||(3)||(4)||(5)|
|Leverage||-0.04 *** (0.01)||-0.04 *** (0.01)||-0.02 (0.01)||-0.04 *** (0.01)|
|Leverage * EU accession dummy||-0.04 ** (0.02)||-0.02 *** (0.01)|
|Exchange rate peg dummy||-1.94 ** (0.87)||-1.82 ** (0.81)||-1.28 (0.81)||-2.45 *** (0.78)||-4.27 *** (1.05)|
|Cumulative credit growth||-0.01 * (0.00)||0* (0.00)||-0.01 ** (0.00)|
|EU accession dummy||-2.29 ** (0.88)||3.29 (2.66)|
|Current account balance||-0.08 * (0.04)|
|Primary gap||0.48 *** (0.15)|
|Constant||0.72 (1.30)||0.44 (1.21)||-1.82 (1.54)||0.24 (1.23)||-7.06 *** (0.98)|
Columns (1)–(3) suggest that leverage in the financial system, credit growth, and a fixed exchange rate were associated with the largest negative growth revisions. Leverage is defined as the ratio of bank credit to the private sector to deposits; credit growth is the cumulative growth in bank credit to the private sector between 2004 and 2007; and the dummy variable indicates a currency peg set to 1 for exchange regimes classified as 1–3 on the IMF’s scale and 0 otherwise. Column (2) suggests that ten European accession countries in the sample experienced unusually large growth revisions, possibly because of their stronger trade and financial linkages with advanced economies in Europe, but column (3) suggests that this regional effect captures the stronger impact of leverage created by financial integration.
Columns (4)–(5) test for the effects of external linkages and fiscal policy. Column (4) suggests that a stronger current account balance tended to limit downward revisions to growth. Column (5) suggests that the primary gap (defined as the difference between the actual primary balance and the primary balance required to hold public debt constant) also helped limit growth revisions. In this set of regressions for emerging market economies, foreign trade variables did not enter the specifications in a consistently significant manner. Beyond exchange rate flexibility, the stability or credibility of monetary policy did not appear to have much explanatory power in reducing the impact of the crisis.