Julian di Giovanni and Andrei A. Levchenko

As there has been exponential growth in world trade over the past few decades, the benefits and costs of increased integration remain a hotly debated topic. In particular, the relationship between trade openness and macroeconomic fluctuations has received a great deal of attention in the theoretical and empirical literature. This article reviews the most recent IMF research on two important macroeconomic outcomes that trade can affect—country volatility and cross-country business cycle synchronization. The article focuses on empirical studies that exploit sector-level data for a large group of countries.

Macroeconomic volatility is considered an important determinant of a wide variety of economic outcomes. Numerous studies identify its effects on long-run growth and welfare, as well as inequality and poverty.¹ The question of what are the main determinants of macroeconomic volatility has thus attracted a great deal of attention in the literature. In particular, it has been argued that trade openness plays a role (Rodrik, 1997; and International Labor Organization, 2004). As there has been exponential growth in world trade in recent decades, understanding the relationship between trade and volatility has become increasingly important.

Several studies using cross-country data have highlighted a positive relationship between trade openness and macro-economic volatility. For instance, one study finds that more open countries experience significantly higher GDP volatility (Kose, Prasad, and Terrones, 2003a). However, a companion paper also finds that globalization weakens the negative impact of volatility on growth, providing suggestive evidence that increased volatility may be less damaging in a globalized economy (Kose, Prasad, and Terrones, 2005).

Though cross-country results are informative to some extent, very little is known about the channels through which trade affects volatility. Some of our recent work examines this question using an industry-level panel dataset of manufacturing production and trade (di Giovanni and Levchenko, forthcoming). The main results are threefold. First, sectors more open to international trade are more volatile. Second, trade is accompanied by increased specialization. These two forces imply increased aggregate volatility. Third, sectors that are more open to trade are less correlated with the rest of the economy, an effect that acts to reduce overall volatility. The point estimates indicate that each of the three effects has an appreciable impact on aggregate volatility. Taken together, they imply that the relationship between trade openness and overall volatility is positive and economically significant. The impact also varies a great deal with country characteristics. We estimate that the same increase in openness is associated with an increase in aggregate volatility that is five times larger in developing countries compared with developed ones. Finally, we find that the marginal impact of openness on volatility roughly doubled over the last 30 years, implying that trade has become more closely related to volatility over time.

One channel through which trade can affect a country’s volatility is through the pattern of specialization: countries that come to specialize in particularly risky sectors after trade opening may experience increased macroeconomic volatility. This kind of mechanism is also related to the finding that terms-of-trade volatility is important in explaining cross-country variation in output volatility (Mendoza, 1995). Indeed, differences in terms-of-trade volatility across countries must be driven largely by patterns of export specialization.

However, there currently is no systematic empirical evidence on how countries differ in the riskiness of their export composition. To fill this gap, we develop a measure of the riskiness of countries’ patterns of export specialization, and illustrate its features across countries and over time (di Giovanni and Levchenko, 2007a). The exercise reveals large cross-country differences in the risk content of exports. This measure is strongly correlated with terms-of-trade and output volatility, but does not exhibit a close relationship with the level of income, overall trade openness, or other country characteristics. We then propose an explanation for what determines the risk content of exports, based on the theoretical literature exemplified by the early contribution by Turnovsky (1974). Countries with a comparative advantage in safe sectors or a strong enough comparative advantage in risky sectors will specialize, whereas countries whose comparative advantage in risky sectors is not too strong will diversify their export structure to ensure against export income risk. We use both nonparametric and semi-parametric techniques to demonstrate that these theoretical predictions are strongly supported by the data.

By almost any measure, the world economy exhibits ever-stronger international linkages. Both trade and capital flows have grown dramatically as a share of world GDP over the past few decades. In addition, trade in goods has become more vertical, as intermediates in production account for an increasing share of world trade (Hummels, Ishii, and Yi, 2001). Recent years have also seen newer forms of cross-border economic integration, such as offshoring and outsourcing of different parts of the production chain (Amiti and Wei, 2005).

Has increased trade and financial integration also led to further synchronization of business cycles across countries? There appears to be some evidence that business cycles have become more synchronized over the past two decades, though this finding is only significant for industrialized countries. By contrast, there appears to be no evidence that consumption correlations increased with trade and financial integration, suggesting that the countries are not reaping the benefits of risk-sharing (Kose, Prasad, and Terrones, 2003b). Other work evaluates the relative importance of global, regional, and country-specific shocks in driving business cycles. It finds that a common factor worldwide can explain about one-third of developed countries’ business cycle variation and about 15 percent in a large sample of countries, while regional effects play a minor role (Kose, Otrok, and Whiteman, 2003).

While these studies consider properties of the world business cycle per se, another influential strand of the literature explicitly examines the links between trade and business-cycle synchronization across countries. The seminal paper by Frankel and Rose (1998) established what has become a well-known empirical regularity: country pairs that trade more with each other experience higher business-cycle correlation. While the finding has been confirmed by a series of subsequent studies, the mechanisms underlying this effect are still not well understood.² In light of the rapidly changing nature of global trade, understanding these mechanisms is becoming increasingly important for macroeconomic policy. For instance, Tesar (forthcoming) analyzes business-cycle synchronization of the European Union accession countries in a model of cross-border production sharing.

In some recent work, we examine the mechanisms through which bilateral trade linkages affect business-cycle comovement using an industry-level panel dataset of manufacturing production and trade (di Giovanni and Levchenko, 2007b). We establish that higher bilateral trade in an individual sector increases both the comovement within the sector between trading countries, as well as the comovement between that sector and the rest of the economy of the trading partner. The estimated magnitudes imply that transmission across sectors is responsible for nearly 90 percent of the total impact of higher bilateral trade on the business-cycle correlation.

We also demonstrate that vertical linkages in production within and across sectors are an important force behind the overall impact of trade on business cycle synchronization. The elasticity of within-sector comovement with respect to bilateral trade is significantly higher in industries that use output of the same sector as an intermediate in production. Furthermore, the elasticity of the cross-sector comovement is higher in sectors that are more heavily used as intermediates by other sectors. The importance of vertical linkages found in this paper provides a fruitful area of theoretical research given the failure of standard international business-cycle models to replicate the features of the data.³

As trade integration continues apace, there will be macroeconomic consequences. Our research has highlighted the importance of delving beneath aggregate-level analysis in order to ascertain a more complete picture of the channels through which trade can affect macroeconomic volatility and business-cycle synchronization. Thus far we have focused on patterns at the sector level, which has helped show the importance of different risk characteristics and linkages in disaggregated data. The next step in the agenda is to begin exploring models and data that put firm-level analysis center stage.