As Marcio García observed in a recent blog in the Latin America Economonitor, the change in Latin American economic performance over the last decade is best described with one word: “Wow.” Inflation is down. Growth is up. Fiscal accounts are strong. Debt is declining. Current accounts are in surplus. Many countries are de-dollarizing. Who would have expected this ten or even five years ago? Latin America no longer seems the region of “original sin” and recurrent crises.
But not so fast. Latin America’s macroeconomic improvements may look extraordinary, but so does the external environment that produced it, with high world growth, ample private financing, historically low emerging market risk premiums, and high commodity prices. Events since July have reminded us that this environment cannot be taken for granted. How “fundamental” are Latin America’s economic improvements really if they were mainly the product of good luck? Perhaps very little, suggest some skeptics, led eloquently by Ernesto Talvi of CERES. Perhaps Latin America’s upbeat numbers are just a veneer, concealing frailties that lurk just below the surface.
Better fundamentals are for real
Perhaps—but probably not. Ernesto and his coauthors are right to remind us that all that glitters is not gold. And they have an important point on long-run economic growth, to which I will return later. But as far as Latin America’s macroeconomic and financial vulnerabilities are concerned, first impressions—namely that Latin America has really changed, and is much less likely today to experience a 1990s style crisis—appear to be surprisingly close to the mark, for three reasons.
First, one important element of Latin America’s good luck—the improvement in its terms of trade—is very unlikely to reverse over the medium term, in light of emerging market demand for commodities, and limited supply. Granted, Latin America could well experience downside shocks—say, a 20 percent drop in commodity prices (for example, in the context of an industrial country slowdown). But simulations I have conducted with my colleague Pär Österholm suggest that this would not have a large impact. A 50 percent percent drop in commodity prices would be a different matter, but that the probability of that happening—given the statistical properties of commodity prices—is extremely low.
Second, even if all of Latin America’s improvements were due to good luck, this does not mean that a sharp deterioration of the external environment would necessarily put it back in crisis. Think stocks versus flows. Bad luck—lower commodity prices, lower external growth and higher financing costs—means that current account and fiscal flows would deteriorate. But the region’s improvements in fundamentals are about stocks as much as flows. Number one is a significant improvement in the structure of public balance sheets, with much lower foreign currency debt, and significantly lower public debt rollover requirements. Number two is much higher international reserves. Regardless of their cause—better policies or just good luck—these stock improvements are likely to have a protective effect in the face of external shocks, at least in the short run.
Third, beyond good luck, there have in fact been significant improvements in macroeconomic institutions and policies in Latin America in the last decade—both monetary and fiscal.
The most significant area is monetary policy. Of the seven large Latin American economies today, five—Brazil, Chile, Colombia, Mexico, and Peru—are by now full-fledged inflation targeters. Monetary policy credibility has greatly improved in these countries, as it has in some of the smaller economies. Inflation expectations are much better anchored. As a result, monetary policy makers react less nervously to exchange rate pressures than they used to in the past—as witnessed in the summer, when exchange rates were allowed to depreciate by 10 percent in Brazil and 13 percent in Colombia, and none of the inflation targeting central banks felt compelled to sell reserves or raised interest rates in defense of the exchange rate. Hence, these central banks are in a much better position to insulate their economies from external shocks than they were in the 1990s.
Real improvements have taken place on the fiscal side too, particularly in revenue collection. In background work for the IMF’s most recent Regional Economic Outlook: Western Hemisphere my colleague Ivanna Vladkova-Hollar and I analyzed fiscal developments in the main Latin American countries, and concluded that the improvements in Latin American revenue-to-GDP ratios are mostly driven either by commodity price increases or higher tax collections that are not related to the economic cycle. Because commodity price increases are projected to reverse only in part, recent revenue increases are hence to some extent permanent. The skeptics are right that structural primary balances do not look as glorious as reported actual balances; but they have improved, and for the most part are currently in surplus.
Latin American economies do of course remain vulnerable (some more so than others). A combined U.S. recession and credit crunch, in particular, could do much damage, as explored in the IMF’s most recent Regional Economic Outlook. Public debt remains fairly high. Not all countries are running account surpluses; and where they do exist, they may mask deficits in non-commodity sectors, as Guillermo Calvo and Ernesto Talvi have argued. Furthermore, the gains of the last few years could be frittered away quickly. Public expenditure growth is fast and worrisome, current account surpluses are narrowing, and primary balances (both actual and structural) are estimated to have declined this year after peaking in 2006. But for now, the gains are real, and there can be no question that the region is in a much stronger position to withstand external shocks than it was five years ago.
The markets seem to agree. Latin America has so far managed a decoupling of sorts from financial turbulence in U.S. and European financial markets (left figure below). Though they have risen, Latin EMBI spreads remain low. As Calvo and Talvi pointed out at the October 2007 LACEA meetings, quick action at the center of the world financial systems—namely, liquidity injections and declines in policy interest rates by central banks in the U.S. and Europe—no doubt deserve some credit: these have benefited not just the U.S. and European economies, but also the emerging markets.
However, the reaction of the Latin EMBI has been subdued even relative to measures of U.S. financial turbulence that should reflect the Fed actions. During the 1998-99 LTCM/Russian Crisis period and the 2000-03 U.S. recession and financial volatility periods, the Latin EMBI went up by about 1.6 basis points for each basis point increase of the U.S. high yield bond spread faced by U.S. below-investment grade corporate borrowers. In contrast, since July, the average reaction has been only about 0.4 basis points.
The reaction to VIX, an index of expected volatility of U.S. stocks, has also been much lower compared to previous financial volatility episodes in the U.S., as shown in the right figure below.
And now the bad news …
While this may all be comforting, there is one important point on which the skeptics are dead right. It concerns long-run growth. What portion of the improvement in growth between the poor record of the late 1990 and early years of this decade and the more recent 5-6 percent should be attributed to the external environment? An excellent recent paper by Alejandro Izquierdo, Randall Romero and Ernesto Talvi suggests a simple answer: “all of it.” This is confirmed by calculations that I have done based on my paper with Pär Österholm. If one estimates the relationship between external conditions and growth in the 1990s up to, say, 2003, and then uses this to extrapolate growth between 2004 and 2007 based on the actual realized paths of the external variables during this period, one obtains predicted values for growth that are at least as high as the growth rates that actually materialized.
The fact that the good growth of the last years is mainly a product of favorable external conditions should not be surprising. Better balance sheets and improvements in macroeconomic policies will hopefully prevent crises going forward and make growth more stable. This is in itself a significant achievement, since crises tend to have permanent effects on the level of output, as my colleague Valerie Cerra and other authors have pointed out, and growth and volatility are negatively linked. But macro stability alone is unlikely to give a significant boost to long run growth in Latin America. And the factors that are likely to really matter in this regard—competition and openness; the quality of education, government services and broad institutions; and tax and labor market distortions that can drive businesses and workers into the informal sector—have changed little in recent years. Indeed, in some areas and countries, there has been backsliding.
This is not to say that the dismal growth of the late 1990s is what we should expect for the long run: those were years of bad luck, just like the recent years are atypical for their good luck. My best guess for the long-run rate of growth in Latin America, when fortunes run to neither extreme, is the same that was suggested by Norman Loayza, Pablo Fajnzylber and Cesar Calderon in their extensive study of Latin American growth a few years ago: about 4 – 4½ percent. Respectable perhaps by the standards of the last 30 years—in part as a result of reforms undertaken in the 1980s and 90s—but too slow, and not just in comparison to East Asia. At this rate, the gap in average incomes between Latin America and the industrial countries will not close until the mid-22nd century—even assuming a stable growth process, which we now have more reason to expect than a decade ago. Accelerating growth in Latin America beyond this rate will require more than reducing macro vulnerabilities, and it will be much harder to achieve.
Disclaimer: While the author of this blog is a staff member of the International Monetary Fund, the views expressed are entirely the author’s own. They do not necessarily represent either IMF views or policy, and should not be reported as such.