If price shocks do not generate HO wage effects over the same time horizon then the basic methodology of linking prices and wages contemporaneously is incorrect such that the common conclusion that trade has contributed very little to rising income inequality needs further exploration. To see this, again consider the finding in Leamer (1996) that during the 1970s the U.S. experienced a sharp decline in the relative price of unskilled-labor-intensive products. By most measures (e.g., college-high school premium) inequality didn’t rise until the 1980s. If price shocks and wage effects occur contemporaneously, then these two facts don’t support the idea that trade mattered. But if the U.S. economy contains extensive frictions which prevent interindustry factor mobility for several years, then this 1970s price shock might not have generated Stolper-Samuelson wage effects until the 1980s. In this case trade-induced price changes might have mattered a lot.
This discussion indicates that a full understanding of trade’s effect on income inequality must address the issue of timing: how slowly does the HO clock tick? If it ticks decade by decade rather than year by year, the literature’s current thinking might need revising.
Potential Limitations of Price-Index Data: What About Cones?
One final issue to consider is the inability of price-index data to identify the crucial issue of how relative factor prices depend on a country’s product mix.