Regional cost function analysis of the logging industry in the Pacific Northwest and Southeast

Nonhomothetic translog cost functions with two variable inputs, labor and capital, were estimated to compare the underlying production structure of the logging industry in the Pacific Northwest and Southeast regions of the United States from 1963 to 1992. The parent cost function and capital share equation for each region were estimated using Maximum Likelihood Estimation techniques. Both equations were significantly different between regions. The industry in both regions was capital-using over the study period. Labor and capital were substitutes in each region, but substitution possibilities were limited. Input substitution was more elastic in the Southeast, and therefore the industry in the Southeast would be better able to adjust to changes in input prices or output demand. Additional tests determined that homothetic, homogeneous, and time invariant functions would be inappropriate for the logging industry in either region.