Competitive conditions and sectors’ productive efficiency: A conditional non-parametric frontier analysis

Abstract

The scope of this paper is to build an appropriate nonparametric frontier framework by using time-dependent conditional efficiency estimators to investigate how competition and time affect the process of technological change and technological catch-up. Our empirical analysis utilizes a sample of 462 U.S. 6-digit manufacturing sectors over the period 1958–2009. Contrary to the existing studies, we apply the probabilistic approach of efficiency measurement to evaluate the dynamic effects alongside with the effects of competition on the sectors’ technological change and technological catch-up levels. Moreover, we estimate the sectors’ idiosyncratic efficiencies, which are an approximation of ‘Solow's residual’ representing the unexplained part of the time-dependent conditional measures. The empirical results shed new light on the topic and signify that the dynamic effect of competitive conditions has a nonlinear effect on the sectors’ estimated performance. Specifically, we unveil a “U-shaped” curve between market concentration and technological change implying that the sectors’ higher concentration levels enhance their technological change levels (e.g. innovation capacity) having however a downsize effect on their technological catch-up ability. Moreover, the relationship between the sectors’ market concentration and technological catch-up although non-linear exhibits an “N-shaped” curve with two turning points. Lastly, the above findings remain robust when we limit our analysis to the capital intensive sectors.