The impact of export diversification in the global financial crisis can be measured across three diversification dimensions: geographic (whether exports go to many or few trading partners), industry/sectoral (whether exports are scattered across many industries), and product (whether many products are produced within industries). The research summarized here argues that industry and product concentration affected Latin American export resilience during the crisis, but geographic diversification did not.

The financial crisis that began with the 2007 collapse of the U.S. subprime lending market and then spread through 2008–09 is remarkable for its global impact on trade. Typical international spillovers, which could stem from portfolio rebalancing or reduced import demand, for example, were compounded during this crisis by financial turbulence that greatly reduced the liquidity of wholesale funding markets. The resulting increase in financial market uncertainty drove credit reductions to firms and households, and consequently led to the most pronounced synchronized trade decline since the Great Depression.

In the wake of this increasingly synchronized international trade, questions have arisen both about this international interdependence and about whether policies can reduce the impact of trade shocks. For example, the IMF has recently turned its attention to conducting surveillance of international outward spillovers from large and systemic global economies in its appropriately titled Spillover Reports. Academic research has continued with increased efforts toward empirically identifying factors that explain how and to which countries a crisis spreads—this is the well-known contagion literature. This literature usually focuses on a cross-section of countries and attempts to identify trade and financial channels of contagion. An equally daunting challenge, however, is to successfully identify country-specific trade characteristics that public policy can influence, and that are empirically linked to the severity of the crisis in each country, as measured, for example, in terms of the decline of its exports.

This is a question of degrees—it is difficult to imagine a set of policies that an economy can realistically implement to insulate itself from a shock of the magnitude of the recent global financial crisis. Nonetheless, the cross-country evidence from the recent crisis suggests that countries and regions were not equally affected. For example, while both advanced and emerging economies had similar peak-to-trough declines in trade indices in the first quarter of 2009 (approximately between 20 and 25 percent), sharp differences are evident across regions, including declines in Africa and the Middle East of just over 10 percent, as compared to a 41 percent decline in Japan. Hence, identifying which country-specific trade characteristics are empirically linked to the severity of the crisis in each country—for example, in terms of the decline of its exports—would be a step forward in developing policies to help soften the impact of international crises.

One useful starting point could be trade diversification—the diversity of products and services that a country sells or buys from the rest of the world. Research going as far back as the mid-1960s has found a positive and significant relationship between how diversified a country’s export product base is and its long-term economic growth rate. More recent research into export diversification and growth often has focused, importantly, on the need to diversify natural resource exports (e.g., Cohen, Joutz, and Loungani 2011 or Ricci and Trionfetti 2011). While the progress in these areas appears promising, a different and perhaps useful direction could be to also consider what broader role export diversification plays in the short term, and particularly, whether it helps or hurts during global downturns.

Costa Neto and Romeu (2011) study this question by assessing the impact export diversification had in the crisis through three different dimensions of trade specialization. First, concentration of exports by geographic destination is considered; that is, whether the bulk of exports from a country go to many or few trading partners. Second, industry/ sectoral export concentration is considered; that is, whether a country’s exports are scattered across many industries and sectors, or concentrated in just a few. Third, product export concentration is considered; that is, whether countries produce many products within their export sectors or just a few. A country’s silk exports, for example, could vary in concentration across the different products classified within this category of exports, such as silkworm cocoons suitable for reeling, raw silk (not thrown), woven fabrics of silk or silk waste, etc.

To find the role that export diversity potentially plays in softening the impact of the global crisis, one can construct a measure of how diverse a country’s trade is across the three dimensions described above. The measure used in Costa Neto and Romeu (2011) is the Herfindahl index, which usually measures the market concentration of firms across industries. This index goes from zero to unity, and represents the squared sum of the market shares. For comparison purposes, the United States Department of Justice considers an industry to be moderately concentrated when the index goes above 0.1.

Using this index, one can then estimate the impact of the three aforementioned types of trade diversity within a trade model. The most successful of these models exploits the “gravity” trade relationship, which, broadly speaking, posits that the volume of trade between countries depends on the geographical distance separating them, the relative size of each country (with size measured by GDP), and a number of factors influencing trade costs, such as free trade agreements, common languages, a shared border, and other factors (e.g., Romeu 2008, Romeu and Wolfe 2011).

In these estimations, the impact of export product diversity can be best understood as influencing trade costs. Export diversity could increase country productivity for the same reason that it increases long-term growth, and the crisis likely had a stronger impact on less productive firms or sectors. Nonetheless, there is strong empirical evidence of non-linearities from trade agreements or restrictions, fixed shipping costs, scale economies, and other trade barriers that complicate incentives to diversifying across products and trading partners and hence could motivate an agnostic view of the role diversification plays in trade (e.g., Henn and McDonald 2011).

Costa Neto and Romeu (2011) estimate this relationship using highly disaggregated bilateral international trade data based on the Harmonized System (HS) of trade reporting at the four-digit level for fourteen Latin American economies. The high level of disaggregation in these data is useful because it provides sufficient observations so as to plausibly capture the dynamics during the crisis within each industry, even though the bulk of the decline occurred across one or two quarters in 2008–09. The estimation also focuses on Latin American countries because they differ greatly in their level of export concentration but are fairly homogeneous in other aspects, thus reducing the risk that some latent country or intra-regional factor is driving any results found.

The degree of export concentration played a statistically and economically significant role during the recent global financial crisis. Specifically, the level of trade concentration is compared across countries after controlling for other global factors in order to identify whether it intensifies or attenuates the global financial crisis on exports. Both product and industry diversification helped attenuate the impact of the crisis, while destination/geographic diversification did not. In the baseline regression, the impact of product, sector, and destination diversification on the quarterly change in trade flows is estimated for Latin American economies, controlling for macroeconomic and trade factors. All else equal, exports are found to decline by approximately 4.7 percent for each decimal unit increase in the (Herfindahl-based) industry trade concentration index (with a similar empirical result found for product diversification within export industries).

The evidence for geographic diversification is weaker and negative, i.e., more geographic diversification worsens the impact of the crisis on exports. As many of the economies in the sample naturally concentrate their trade with the United States because of geographic proximity and other factors, these results suggest that proximity to the United States during the crisis was at least not detrimental to outcomes for indicators of the incidence and severity of the crisis.