A quarterly magazine of the IMF

December 2005, Volume 42, Number 4

A Fork in the Road
Arminio Fraga

Latin America faces a choice between populism and deeper reform

On the surface, the IMF's recent growth numbers for Latin America are quite encouraging: almost 4.8 percent average growth for 2004–05, and a forecast of 3.8 percent for 2006. This is an impressive turnaround when measured against the 1.4 percent average for 1999–2003. But the news is not quite as good when measured against the more exuberant backdrop of world growth—estimated at 4.7 percent in 2004–05 and expected to remain at 4.3 percent in 2006. Moreover, if we look at 1997 through what is projected for 2006, Latin America will have grown at 2.8 percent while world growth will have averaged 3.9 percent, split between 2.7 percent for the advanced economies and 5.3 percent for the emerging and developing countries. This lack of convergence in incomes relative to the advanced economies is quite disturbing, as is the underperformance relative to other developing countries, especially in Asia. What is the matter with Latin America? Why has growth lagged other emerging regions by so much? Have we seen progress in recent years? And are we heading in the right direction?

The lost decade(s)

To gain insight into these issues, it pays to begin with a review of the 1980s and 1990s. The 1980s became known in Latin America as the "lost decade" because per capita output growth averaged a shocking negative 0.6 percent a year, after the outstanding 3.8 percent performance of the 1970s (as measured by the GDP-weighted average of the seven largest economies in the region). This decade was marked by economic chaos in most countries in the region, and included multiple episodes of hyperinflation, international debt default, and deep recessions.

All over the region, nonorthodox approaches to macroeconomic stabilization were tried with tragic results. Brazil, for instance, struggled for years with a series of unconventional plans to curb inflation, each one failing as they did not address basic matters such as loose fiscal and monetary policies. Price freezes, asset freezes, and exchange rate pegs of all kinds were attempted in quick succession.

As the 1980s came to an end, it became clear throughout Latin America that certain basic elements of sound economic management had to be present to create an economic background conducive to growth and development. These elements, which became part of the so-called Washington Consensus, included fiscal discipline, a competitive exchange rate, and trade and foreign investment liberalization.

Over the following decade, most nations in the region managed to get their inflation rate down and make progress with avoiding banking and balance of payments crises. According to an index developed by Goldstein, Kaminsky, and Reinhart (2000), the number of crises in the seven largest countries in the region declined from 26 in the 1980s to 9 in the 1990s. Not surprisingly, growth in output per capita in these countries climbed to an annual average of 1.7 percent—a major gain relative to the 1980s—but below the 2.0 percent achieved by the United States and way below developing countries in other regions.

Progress on the macroeconomic front was matched by significant improvement on the social front. With the exception of school enrollment, indicators ranging from illiteracy to infant mortality displayed significant improvement, even during the years of the lost decade. Illiteracy dropped from 15.6 percent in 1980 to 7.9 percent in 2000, with primary school enrollment increasing from 85.2 percent to 95.4 percent. Life expectancy at birth improved from 66.1 percent to 72.1 percent, while the infant mortality rate dropped from 50.1 percent to 22.6 percent.

Similar gains were made on structural issues. A structural reforms index developed by Eduardo Lora of the Inter-American Development Bank shows that reforms increased from an average regional level of 318 in 1985 to 583 in 1999 (out of a maximum possible score of 1,000). More important, gains were recorded in total factor productivity (TFP). Using a measure that captures investments in education (and thus is lower than the usual labor productivity numbers, but offers a better measure of the true gains), TFP advanced from a shocking negative 2.28 percent a year in the 1980s to a modest but positive 0.33 percent a year in the 1990s (Fraga, 2004). Notice that here the regional average is misleading, as Colombia and Venezuela experienced declines of around 2 percent a year while Argentina and Chile showed gains of a similar magnitude.

Still, despite the major macroeconomic and social gains obtained in the 1990s, and with the notable exception of Chile, many in Latin America still felt like another decade had been lost, probably because the old ways of fast growth remained elusive. Oddly, some observers in the region "blamed" the Washington Consensus for the modest performance.

Don't blame the Washington Consensus

My conclusion is different. While the number of crises did decline in the 1990s, most countries in the region failed to implement the main aspects of the Consensus and, as a result, failed to credibly consolidate macroeconomic stability. Take, for example, the high sovereign spreads and low credit ratings that prevailed in most countries in Latin America during the 1990s, especially when compared to Asia.

I would argue that significant progress was achieved in Latin America in the 1990s, especially when compared to the mediocre 1980s. Moreover, the countries that followed the general lines of the Washington Consensus have done better than those that did not. Chile, the star performer, had both the best macroeconomic performance and the best and earliest record in implementing structural reforms. Mexico did well after 1995 on both counts and has achieved reasonable growth since then. Brazil has done well in many ways since 1994, but was unable to avoid two deep confidence crises (in 1999 and 2003), and is now involved in another political crisis. Argentina did shine on the structural reform front in the 1990s, but failed to secure macroeconomic stability and plunged into a deep recession in 2001.

Elections could be decisive

There is, however, a deeper issue underlying Latin America's modest performance since the 1980s: Why have most of these nations failed to get their economic act together in a steady and convincing way? Why have they been unable to avoid frequent crises? And why have they been unable to save and invest more and better? These questions go way beyond economics, and I can only make some conjectures.

Currently, the macroeconomic fundamentals in the region look quite good. Sovereign spreads are very low, and inflation is low in most countries. But, as we know all too well, these reasonably sound fundamentals cannot be taken for granted. Over the next year, elections will take place all over Latin America. And there is a smell of populism in the air in many parts of the region that should not be ignored. A big question is whether Brazil and Mexico, the two largest economies, will follow the successful path of Chile or succumb to populist temptations, be they of the mild Argentine variety (price controls, regulatory uncertainty), or of the more extreme Venezuelan kind that puts democracy at risk as it aspires to the Bolivarian dream of uniting Latin America.

My view here is cautiously optimistic. Argentina and Venezuela are growing fast because they are recovering from deep recessions and are benefiting from fast global growth and vastly improved terms of trade (especially Venezuela). But despite this impressive economic rebound, these countries have yet to inspire confidence in the sustainability of their long-term growth paths.

In contrast, Brazil and Mexico appear to have managed to break away from the vicious cycle of economic crises and populism, each in their own way. In Mexico, the Fox government has refrained from revisiting the electoral tactics that led to a number of crises in the past. Moreover, initial moves by the main potential candidates seem to indicate no break in Mexico's commitment to sound macroeconomic management. In Brazil, while the Lula government has not made progress in several key areas, macroeconomic policies have remained under strict control and are likely to stay that way throughout what is likely to be a hot political campaign in 2006. These are signs of maturity.

However, renewed worries arise when one thinks about the visible difficulties both countries are having in gathering political support for the structural reforms that are still needed. Here we move to the realm of the political economy, where the challenges are related to entrenched special interests, corruption, and poor electoral rules and incentives. These countries have to deal with the vicissitudes of their own political system and history, and in both cases, a consensus for reform has proven hard to reach—perhaps indicating the need for political reform ahead of other reforms.

Herein lies a challenge. Without political reform, structural reforms are unlikely to be enacted. Without reforms, growth is less likely to take off. Without growth, political support for macroeconomic stability will weaken, and the danger of a return to the ups and downs of the past cannot be ruled out.

What then? My guess is that the best strategy for the new governments in Brazil and Mexico will be to aggressively push for structural reforms early in their terms, when the momentum of a successful election may carry the day. Assuming this jumpstarts some growth, other reforms may become feasible. Not easy, but it can be done.