| This dissertation examines the effect of changes in trade policies on the behavior and performance of manufacturing plants. Chapter 1 describes the effects of a temporary increase in tariffs on the performance and behavior of U.S. manufacturing plants. Using antidumping duties as an example of temporary protection, I compare the responses of protected manufacturers to those predicted by new models of trade with heterogeneous firms. I find that apparent increases in revenue-based productivity associated with temporary protection are primarily due to increases in prices and mark-ups. In fact, antidumping duties lower physical productivity among the set of protected plants reporting output data in units of quantity. Moreover, antidumping duties allow for the continued operation of low-productivity plants that likely would have otherwise ceased production. As a result, temporary protection slows the process of output rationalization, with less productive plants producing a greater share of total output, leading to a reduction in aggregate productivity growth. Importantly, plants that are denied protection by the government are no more likely to exit than protected plants. Rather, they adjust by dropping the unprotected product and producing other, potentially higher-productivity products.;Chapter 2 identifies determinants of productivity growth among Colombian manufacturers during a period of unilateral trade liberalization. I find that lower tariff rates were associated with an expansion in the extensive margin of foreign input usage, as well as an increase in investment in new machinery. The expansion in extensive margin of foreign input use is found to be productivity-augmenting, in line with the predictions of Ethier (1982). Higher investment in new machinery had no effect on current productivity, although there is some evidence that these investments have a positive effect on future productivity. Industry concentration, plant scale and the intensity of foreign input usage---three other potential channels for productivity growth---were unaffected by trade liberalization.;Chapter 3 outlines an algorithm for concording ten-digit Harmonized System export and import codes over time, describes the concordances we construct for 1989 to 2004 and provides Stata code that can be used to construct similar concordances for arbitrary beginning and ending years from 1989 to 2007. |
