Intellectual Merit: Trade as a share of global income has increased dramatically in recent decades, and countries increasingly rely on international trade to grow their incomes. Many studies have examined the import of trade for growth and productivity over long periods of time - often decades or longer. Frequently, however, the impact of changes in trade flows is much more acute. For example, there were sudden and dramatic reversals in trade flows during the Asian crisis in the late 1990s, the Argentine crisis of 2002, and even during the recent global recession of 2008-2009. Little is known about the mechanics of trade adjustment at the firm and product level, and direct links between macroeconomic performance and the reductions in trade during these episodes remain largely unexplored. The PI's propose to use novel transaction-level micro-data that allows - along several dimensions - the most detailed study of the adjustment of the external sector during a crisis episode. The PI's have obtained customs data for all trade transactions for Argentina during 1996-2008 and propose to expand their study to include other countries. An empirical characterization of trade adjustment will allow them to evaluate how costly short-run trade adjustments are in terms of consumer welfare and output productivity. In existing literature the theoretical channels are clear but the empirical evidence is limited. One of their preliminary findings is that the extensive margin, defined as the entry and exit of importing firms or imported product categories, plays a small role in adjustment during the crisis. However, at the firm level, the extensive margin plays an important role as firms drop products that constituted a significant share of their previous imports. Preliminary calculations suggest that the inability of firms to purchase inputs as they did before the crisis has a meaningful impact on their productivity for at least 2-3 years. This has not been uncovered by previous studies because none have examined such disaggregated levels of data through such an acute crisis. Broader Impact: The PI's propose to improve our understanding of mechanically how countries' imports and exports adjust and the impact of such adjustment on the economy. In particular, which margins of trade adjust the most? Do the quantitatively important adjustments occur within firms and within detailed product categories, or do firm entry and exit and product entry and exit play an important role? Does the quality of products traded change? Is adjustment different between intermediate and final goods? How does the response of multinationals differ from that of local firms? Are these responses different for trade flows with developed or developing countries? A clearer description and deeper un¬derstanding of the impact of shocks to international trade on a country's productivity and welfare, particularly during crises, is important because it will inform a country's optimal choice of trade policies and exchange rate regimes. Further, it may give a better sense for understanding the cross-country and cross-sector propagation of international shocks. Given the recent reminder of the potential for large cross-country volatility, this understanding is important for sound policy decisions, to evaluate the impact of potential macroeconomic shocks, and to understand the drivers of welfare declines in recessions in open economies.
The principle outcome of our NSF grant was the research included in our paper "Trade Adjustment and Productivity in Large Crises", which is forthcoming at the American Economic Review. Our work demonstrated that large scale adjustment in international trade occurs within firms and is not due to the country as a whole ceasing imports of a particular type of good or due to large importers halting their international trades altogether. It is important to have demonstrated this in part because it implies that within-firm product churn must be properly addressed in order to evaluate the impact on an economy of various policies which matter for trade, whether they be tariffs or exchange rate policies. For example, our work was used in a policy note as the basis for a quantitative estimate of the damage Argentina’s more recent import restrictions might imply. Other research has started to build on our study of firm-and product-level heterogeneity in import behavior to further hone our models of how changes in international trade matter for overall productivity and wellbeing. A secondary outcome of our grant is our related work "Trade Prices and the Global Trade Collapse of 2008-2009" (with Oleg Itskhoki), which is now published in the IMF Economic Review. There, we demonstrated that price adjustment played essentially no role in the rapid reduction of U.S. import spending on differentiated goods during 2008-2009. This finding has been important in disciplining subsequent models of the trade collapse, some of which counterfactually imply large changes in traded good prices. Taken together, these papers will help the field better hone quantitative models capable of explaining how productivity and economic performance are impacted during episodes – generally crises episodes – when there are rapid changes in the value of international trade.
