Working Papers

• Trade and Real Wages of the Rich and Poor: Cross-Country Evidence  Abstract

  Trade liberalization affects real-wage inequality through two channels: the distribution of nominal wages across workers and, if the rich and the poor consume different bundles of goods, the distribution of price indices across consumers. I provide a unified framework incorporating both channels by allowing for non-homothetic preferences and worker heterogeneity across jobs. Because skill-intensive goods are also high-income elastic in the data, I find an intuitive, previously unexplored, and strong interaction between the two channels. I parametrize the model for 40 countries/regions using sector-level trade and production data, and find that trade cost reductions decrease the relative nominal wage of the poor and the relative price index for the poor in all countries/regions. On net, real-wage inequality falls everywhere. Close

• Imported Inputs and Sectoral Wage Dispersion (with Feiran Zhang) Abstract

  This paper proposes a new mechanism through which trade liberalization affects income inequality within a country: the use of imported inputs. Intuitively, a firm with higher initial productivity is better at using higher quality foreign inputs. This justifies paying the fixed costs for a larger set of imported inputs when input tariff liberalization decreases their relative price. The firm becomes more import intensive, which enhances its productivity advantage. As a result, the firm hires higher quality workers, produces higher quality products and pays higher wages to its workers, increasing within-sector wage dispersion. We find that both the mean and the dispersion of the distribution of firm productivity, markup and size went up during a period when China reduced its tariffs on imported inputs. More importantly, these results still hold when we consider the subset of firms that survived throughout the sample period, from 1998 to 2007. In addition, we develop a partial-equilibrium, heterogeneous-firm model with endogenous imported inputs and labor quality choice that is consistent with these observations. Finally, we provide empirical evidence that supports the model's prediction that the differential change in the import intensity of firms with different productivity levels explains these patterns.  Close

• The Causal Effect of Firm-Level Market-Specific Imports on Exports (with Mi Dai) Abstract

  This paper asks the following question: does importing from a country have a causal effect on a firm's ability to start exporting there by allowing it to establish trade networks, build distribution channels as well as learn about local consumer preferences, business practices and institutional environment in that market? As micro-level data on firm imports became available, there has been a burgeoning research literature on firm import behavior, which hasn’t been intensively studied before. However, most of these studies focus on the connection between imported inputs and firm productivity, not the causal effect of firm-level imports on exports in individual markets. We use one of the most significant trade liberalization episodes, China's accession to the WTO in December 2001, to separate the impact of imported inputs on a firm's productivity from their impact on a firm's ability to enter individual export markets. First, we find that a reduction in import tariffs encourages firms to start importing from new countries. Second, and more interestingly, we find, based on the IV method, that there is a significant positive causal effect of lagged imports on exports to the same country. That is, having imported from a country in the past increases the probability that a firm starts exporting there. We also find that a firm starts exporting to a country that it has imported from on a larger scale. In addition, the duration of such a trading partnership is longer. These results help to explain why large importers are simultaneously large exporters, and shed some light on what the sunk costs of exporting are and how to mitigate them.  Close