Assessing the effects of measurement errors on the estimation of production functions

This article explores to what extent the poor results that are often found when estimating parameters of production functions can be attributed to measurement errors, due to the use of common price deflators across firms. Because of the lack of detailed micro‐economic data, econometricians have to rely on industry‐wide deflators when computing outputs and intermediate inputs. A unique feature of the longitudinal data used in this paper is that it reports firm‐level prices. This allows for a comparative assessment of production function parameters where the outputs and intermediate inputs are computed using both firm‐specific prices and industry‐wide deflators. The empirical results presented in this paper show that the use of common deflators across firms leads to lower scale estimates, mainly because of a large downward bias in the estimated coefficients for labour. Copyright © 2006 John Wiley & Sons, Ltd.     

Positive Assortive Merging

This paper addresses the following two questions: (i) Is there any evidence that firms, like human beings, prefer to partner with alike? (ii) Is there any relationship between the ex ante technological and product relatedness of merging parties and the postmerger performances? Using data of patent holdings and product portfolios of big pharmaceutical companies, I find that (i) merger deals are more likely to be signed between firms with related technologies and drug portfolios, and (ii) product relatedness and technology relatedness are positively and negatively correlated with postmerger performances, respectively. The analysis suggests that the negative effect of technology relatedness might be driven by a large human capital depreciation following consolidations. The results have important implications for competition policy, which are discussed in the concluding section.     

Competition & Innovation: Evidence from U.S. Patent and Productivity Data

Is there any evidence that innovation and technological progress are constrained by competition and fostered by monopoly power? Our results, based on a constructed dataset of U.S. manufacturing industries observed over more than two decades, suggest that this is not the case. On the contrary, using both patent statistics and productivity growth as alternative measures of innovation and technological change, we observe faster technological advances in more competitive markets. These results are robust to changes in the econometric techniques used to model nonlinearity in the competition‐innovation relationship and to alternative methods of computing market power.