An investigation into a multi-country, multi-factor test of the factor abundance theorem using the 1982 input-output table for the United States
In this dissertation we attempt to test the validity of recent findings that no evidence to support the Heckscher-Ohlin (HO) theorem in a multi-country, multi-factor test can be found using 1967 data. Following the methodology described by Bowen, Leamer and Sveikaskas (BLS) in their 1987 paper, we use "Leonfief-type" tests on the more rigorous Vanek version of HO. We then allow some of the assumptions to vary and use a regression approach to find an optimal relaxed version of HO. After rigorous data collection and manipulation for 1982 data, we find results similar to those found by BLS. While we find some support a weakened HO using regression techniques, the results are contradictory and not intuitively logical. The conclusion is that while the BLS results for 1967 data can be supported using 1982 data, there is suspicion that the amount of manipulation needed to conform the data may, in fact, be leading to the results.
Mielnicki, Lawrence, "An investigation into a multi-country, multi-factor test of the factor abundance theorem using the 1982 input-output table for the United States" (1993). ETD Collection for Fordham University. AAI9324623.