Editor's Summary

The economies of Latin America and the Caribbean grew 6 percent in 2004, the highest rate since 1980. The United Nations Economic Commission for Latin America (ECLAC) estimates that the region's economies expanded by 4.3 percent in 2005. For 2006, ECLAC is forecasting 4 percent growth. If these predictions come true, the region will have completed four consecutive years of growth, accumulating a 10 percent increase in per capita income between 2003 and 2006.

That Latin America should be growing at a time of record-high commodity prices, record-low international interest rates, and robust global demand is not surprising. What is surprising is that the region is not growing more in this environment of the fastest world growth in thirty years. According to the International Monetary Fund (IMF), the region will post the slowest average growth in 2005-06 of any developing region.¹ Developing Asia will grow by 7.4 percent in 2005 and 7.1 percent in 2006, according to the IMF. Even Africa, at 5.1 and 5.4 percent, will amply outgrow Latin America.

The region's growth problem is anything but new. Since 1980, only Chile has registered sustained increases in income. Most other countries have stagnated or grown slightly. By contrast, India has averaged 6 percent economic growth annually for fifteen years, and China's economy has grown by 10 percent a year for twenty-five years.

Consequently, growth is back at the center of the research agenda in Latin America. Applying the standard prescription of "stabilize, liberalize, and open up" may well be necessary for long-term growth, but it certainly does not seem to have been sufficient. Economists and policymakers are casting about for something else to try. One popular alternative is variously known as competitiveness policy, industrial policy, or microeconomic interventions.

In the first paper in this volume, Andrés Rodríguez-Clare claims that such interventions have been commonplace in Latin America since the 1980s, even in countries that generally advocate a hands-off approach to growth. Examples are policies to promote small and medium-sized enterprises, attract foreign direct investment (FDI), and, more recently, encourage technical innovation.

The problem with these policies, Rodríguez-Clare argues, is twofold. First, they have weak theoretical and empirical foundations. For example, it is hard to find the spillovers that would justify an emphasis on FDI. And why, conceptually, are small firms to be favored over large ones? Second, and more damningly, these policies seem to have had a negligible growth payoff.

According to Rodríguez-Clare, a more effective set of microeconomic interventions should specifically address the market failures that hinder accumulation and growth. One such problem arises when a coordination failure prevents firms from taking the necessary actions to increase sectorwide productivity, so that a cluster fails to develop. Rodríguez-Clare focuses on this type of agglomeration economy, and he proposes a set of microeconomic interventions that could be adopted as solutions.

His paper presents a model of a small economy with sector-specific coordination failures. The model suggests that, rather than trying to reallocate resources toward sectors that are seen as offering high clustering possibilities (as was the case with traditional import substitution), policy should aim at fostering cooperation in sectors in which the economy is already showing comparative advantage. A commonly cited example is from the Scandinavian countries, where clusters developed around the forestry industry. Chile seems to be following suit, with clusters now forming around natural-resource-intensive industries such as copper mining, vegetable and fruit farming, and forestry. Whether Chilean public policies in this area have been effective remains controversial, however, as discussant Ronald Fischer argues.

What are the policy implications? One conclusion stressed by Rodríguez-Clare is that general policies to increase innovation across the board are likely to be inferior to policies that take the more selective approach of trying to induce the development of innovation clusters in areas of comparative advantage. But that requires a policymaking apparatus that is able to identify these areas without political interference or corruption. Rodríguez-Clare acknowledges that not all Latin American countries have...