Despite these clean theoretical results, there is very little empirical evidence on what this transition from short run to long run looks like in reality. This lack of evidence poses a potentially very serious problem for almost all the empirical work to date on the effect of international trade on wage inequality. Every study surveyed in this paper tries to explain inequality changes with contemporaneous product-price changes. Thus, for example, researchers exploring rising inequality during the 1980s analyze trends in product prices during the 1980s. The implicit assumption in all these studies is that the U.S. economy is “sufficiently HO in nature” that price shocks over some time period affect the economy as predicted by HO theory in that same period.
But what if that assumption is incorrect? What if the U.S. economy has sufficiently important short-run “frictions”–imperfect information, high costs of reallocating capital across sectors, people reluctant to relocate geographically, etc.–that price shocks over some time period do not generate HO factor-price effects in that same period? A lot of research indicates that these frictions can matter for several years. For example, Blanchard and Katz (1992) find that in states hit by aggregate-demand drops people can take between five and ten years before deciding to move away to recover economically. The more important these frictions are, for any given shock the longer the economy adjusts in an SF manner before switching over to an HO manner.