Factor Biases and Technical Change in Manufacturing: The American System, 1850–1919

Abstract

This paper examines the proposition that U.S. manufacturing experienced biased technical change during the period 1850–1919. Tests for bias, in Hicksian terms, are conducted using the translog cost dual. Redefined census data permit these tests to be made at the two-digit level of industry classification and with four inputs considered—labor, capital, materials, and a residual factor input. The tests demonstrate that labor-saving and capital-using biases existed, but material-using biases also were present. Furthermore, the patterns of bias varied considerably from industry to industry and often were of such a magnitude as to overpower ordinary substitution effects.