Estimation of the CES Production Function with Biased Technical Change: A Control Function Approach.

In this paper, I extend the Olley-Pakes (1996) estimation method to the CES production function with biased technical change. The new semi-parametric approach allows consistent estimation of the degree of returns to scale, the elasticity of substitution, and the bias in technical change. Identification of these parameters is achieved under the assumption that the data generating process reflects not only technologies but also optimizing behavior of producers. Using data from U.S. manufacturing industries over the period 1958-2005, I find strong evidence that industries are characterized by a production technology with the elasticity of substitution below one and with significant biased technical progress.