Exports and Within-Plant Wage Distributions: Evidence from Mexico

Frías, Judith A.; Kaplan, David S.; Verhoogen, Eric A.

In many developing countries, increasing international integration has been accompanied by rising wage inequality, and traditional Heckscher-Ohlin models, which rely on between-sector reallocations to link trade and labor-market outcomes, are difficult to reconcile with this pattern (Goldberg and Pavcnik 2007). Recently, researchers have proposed a number of potential within-sector explanations based on the behavior of heterogeneous firms, involving technology choice, quality upgrading, search and bargaining, or fair wages, among other mechanisms. There is evidence at the plant level to support a within-sector link between trade and inequality. For instance, Verhoogen (2008) finds that initially larger, higher-productivity Mexican plants had higher export propensity and wages in cross-section in 1993 and that they were more likely to increase exports and wages in response to the late-1994 devaluation of the peso. The shock to exporting thus arguably increased dispersion in wages between plants within sectors. At the plant level, however, many of the proposed within-sector mechanisms carry similar observable implications. Distinguishing among the various mechanisms will require moving to a lower level of disaggregation, and exploiting information at the level of individual workers within plants. In this short article and the longer article to which it is a companion (Frías, Kaplan, and Verhoogen 2011), we use employer-employee data from Mexico and an identification strategy from Verhoogen (2008) to examine the effects of exporting on wage outcomes that are not available in standard plant-level datasets. In Frías, Kaplan, and Verhoogen (2011), we estimate the effect of exporting on wage premia, defined as wages above what individual workers would expect to earn elsewhere in the labor market. Wage premia are estimated as plant effects, controlling flexibly for individual heterogeneity (and allowing the return to worker ability to vary over time), implicitly assuming that the plant effect is the same for all employed workers. In this short article, by contrast, we do not attempt to control for worker heterogeneity, but instead focus on the effect of exporting on the shape of within-plant wage distributions. As we show in more detail below, we find that exporting has little effect on wages at the low end of the wage spectrum within plants, and that it raises within-plant wage dispersion, but not uniformly between all quantiles. The results are consistent with, but add important qualifications to, the finding of Verhoogen (2008) in plant-level data that exporting raised the ratio of white-collar to blue-collar average wages. This article is related to an active theory literature on trade, matching, and organizations which has proposed a variety of mechanisms linking trade and wage distributions within firms. Recent papers using employer-employee data to investigate the consequences of trade for labor-market outcomes (without focusing on the overall within-plant distributions) include Krishna, Poole, and Senses (2011); Hummels et al. (2011); and Davidson et al. (2011); see Frías, Kaplan, and Verhoogen (2011) for a fuller literature review.

Geographic Areas


  • thumnail for exports-within-plant-wage-distributions.pdf exports-within-plant-wage-distributions.pdf application/pdf 518 KB Download File

Also Published In

American Economic Review

More About This Work

Academic Units
International and Public Affairs
American Economic Association
Published Here
April 11, 2014