What's the use of factor contents?

The net imports of labor embodied in US international trade has been a fairly small and stable share of the US labor force. From this fact one would like to draw two conclusions: (1) Trade barriers would not have had much effect on earnings. (2) It is technology, not trade that is the driving force for rising income inequality in the US. The first conclusion is valid if the US economy is log-linear and the number of tradables is equal to the number of factors. The second is valid if the global economy is log-linear and the number of tradables is equal to the number of factors. I am inclined to accept the first argument as an adequate approximation, but to reject the second because it relies too heavily on highly implausible assumptions which are incompatible with the data. (C) 2000 Elsevier Science B.V. All rights reserved.