Occupations and Import Competition

There is a growing concern that many workers do not share in the gains from trade. In this paper, I argue that occupational reallocation plays a crucial role in determining the winners and losers from trade liberalization: what specific workers do \emph{within} an industry or a firm matters. Adjustment to trade liberalization can be protracted and costly, especially when workers need to switch occupations. To quantify these effects, I build and estimate a dynamic model of the Danish labor market. The model features nearly forty occupations, complicating estimation. To reduce dimensionality I project occupations onto a lower-dimensional task space. This parameter reduction coupled with conditional choice probability techniques yields a tractable nonlinear least squares problem. I find that for the median worker, a 1% decrease in income, holding the income in other occupations fixed, raises the probability of switching occupations by .3%. However, adjustment frictions can be large---on the order of five years of income---so that workers tend to move in a narrow band of similar occupations. To quantify the importance of these forces for understanding import competition, I simulate the economy with and without observed changes in import prices. In the short-run, import competition can cost workers up to one half percent of lifetime earnings. Moreover, the variance in earnings outcomes is twice the size of the total gains from trade.