From the Director General

One of the most important recent economic phenomena in Latin America has been low inflation, often resulting from disciplined central bank policy and more open markets. Because inflation in Latin America was not high in 2002, even in crisis countries where it once exploded regularly, other events have received more attention. These events, however, are part of what makes the inflation story noteworthy. Freer capital movements and trade play much more important roles than is widely recognized.

For Latin America, 2002 began with one crisis, ended with another, and endured still other volatile episodes in the interim. Between mid-December 2001 and mid-January 2002, Argentina defaulted on its $155 billion in outstanding debt, announced three exchange rate regimes, devalued its currency by 100 percent, and had five presidents. In February 2002, the bolivar was devalued about 40 percent. The year ended with Venezuela, stunned by a political crisis and national strike, in economic free fall.

In the interim, Brazil had its own stresses, including a floating exchange rate that depreciated 75 percent against the dollar between February and October and then appreciated 12 percent between October and the end of December. Due largely to presidential election jitters, the spread of Brazil's Emerging Market Bond Index over U.S. Treasury rates moved from about 800 basis points in January to almost 2,300 in October, and fell to about 1,400 by year's end. Meanwhile, Mexico's economy, increasingly tied to the U.S. manufacturing sector, trailed the United States' slow growth rate but did grow in 2002.

The largest inflation targeting countries, Brazil and Mexico, failed to achieve their targets. In trying to do so, however, Brazil engineered a 700-basis-point increase in its benchmark Selic interest rate between September and December. Mexico hewed hard to a target that was not achieved only because of increases in government-administered utility prices.

December-over-December inflation rates in the inflation-targeting countries—which also include Chile, Colombia, and Peru—were relatively low by the standards of the past 20 years, even though the pass-through from Brazil's exchange rate depreciation contributed to a 14 percent increase in that country.

Inflation and monetary expansionism were most strikingly low in the crisis countries, even though inflation was not really very low. Argentina seemed to have the greatest potential for a burst of high inflation. Suddenly, the nation was papered with low-denomination debt instruments that circulated as currency. Argentina's central government began issuing the patacon, which looked and felt just like money. Argentina's provinces did the same. Santa Cruz province, home of former president Carlos Menem, issued debt instruments with a portrait of Eva Perón. Something, however, restrained inflation to 41 percent between December 2001 and December 2002, compared with price increases substantially higher than 1,000 percent between December 1989 and December 1990.

Similarly, despite Venezuela's persistent efforts to reduce its debt without increasing taxes, the extremes of monetary expansionism that might have occurred a decade earlier never materialized. Price controls in Venezuela make comparisons difficult, but data on monetary expansions give us some idea of what was happening. Over December 1990 to December 1991, Venezuela raised its M1 monetary aggregate by 54 percent. In circumstances at least as stressful between December 2001 and December 2002, monetization proceeded rapidly by current Brazilian, Chilean, and Mexican standards. But at 18 percent, Venezuela's rate of monetary expansion was still only one-third what it was at the start of the 1990s.

Openness and Inflation

Different factors explain inflation—or its absence—in different countries. Much econometric evidence suggests that disinflation in the United States is largely due to technological advances. By contrast, in Europe the main cause is a surplus workforce that holds down wages.

However, among the less well-understood relationships observed between inflation and other variables, some of the most robust involve not only openness to capital flows and trade but also the liberalization of capital and current accounts. CLAE staff economists offer evidence that these factors are not only linked but that the liberalization leads and the reduction in inflation follows. This flies in the face of some analysts' thinking, for reasons that are understandable.[1] For if a sudden, unsterilized rush of capital into a country signals out-of-control monetary expansion, how could capital account liberalization be anti-inflationary? However, the more open capital markets are, the easier it is for assets to go abroad when domestic policies would erode their value at home. Currency substitution becomes easier. As a result, the seigniorage-maximizing inflation rate falls. Reactions to this new calculus could include tighter monetary policy—or at least monetary policy that is tighter than it would otherwise be—or making the central bank autonomous.

Latin America has made big strides toward more openness to international capital flows in the past 15 years. The persistent connection between capital account liberalization and lower inflation suggests this relationship contributes significantly to the relatively low inflation rates being seen in the region's crisis countries. Moreover, even where official capital market liberalization is slow, technological advances may still be making capital more footloose. As a famous paper by Stijn Claessens, Michael Dooley, and Andrew Warner concludes, all capital is now "hot" capital.[2] Whether legal or illegal, looser capital movement intensifies currency competition; the ability of capital to relocate can impose greater consequences when one government seeks to inflate its currency and another does not.

The Current Account and Inflation 

Inflation also falls when the current account opens. In the late 1990s, some economic literature claimed the statistical regularity connecting current account openness and inflation was not really very regular. According to some, the connection only reflected factors peculiar to heavily indebted countries during the great debt crises of the 1980s. This detail was purportedly hidden in previous estimations that had sampled large numbers of many different types of countries without correctly disaggregating them. Countries that were not heavily indebted, according to these economists, did not show this relationship.

This interpretation was strongly debated even then, and it continues to be questioned. Research by CLAE staff economists shows that in the 1990s the inverse relation between current account or trade openness and inflation—and even between trade liberalization and changes in inflation—applied to rich countries, poor countries, heavily indebted countries, and not heavily indebted countries. This research also demonstrates that the negative relation between current account openness and inflation has strengthened overall since the 1980s.

Agreement on a paradigm to explain the negative link between current account openness and inflation is less widespread now than a decade ago. However, more trade obviously means more opportunities for businesses to understate the value of their exports and overstate the value of their imports. This offers more avenues for parking assets abroad and creates opportunities for currency substitution and currency competition, just as capital account openness does in other ways.

Some of Latin America's largest countries still do not trade very much. But both imports and total trade as shares of gross domestic product adjusted for purchasing power—a common measure of trade openness—have risen markedly over the past 20 years. During 1986–90 and 1996–2000, trade as a share of this adjusted GDP rose by about one-fourth in Argentina, about one-third in Brazil, and about one-half in Mexico. Chile's and Venezuela's ratios did not increase but have always been high.

While capital and current account liberalization may not have been the only factors lowering inflation in Latin America over the past decade or so, the robustness of the negative relations between them suggests they have been important in making the region's inflation less newsworthy. Inflation targeting, freely fluctuating exchange rates, technological advances in international finance, and industrial country disinflation have had an effect, and the policies resulting in the capital and current account liberalizations may have been determined in conjunction with these factors. The link between liberalization and inflation, however, seems clear.

Notes See Rodrik, Dani (1998), "Who Needs Capital-Account Convertibility?" Harvard University (Cambridge, Mass., February). Claessens, Stijn, Michael P. Dooley, and Andrew Warner (1995), "Portfolio Capital Flows: Hot or Cold?" World Bank Economic Review 9 (January): 153–74.

  |