作者: Patricia Byrnes , Rolf Färe , Shawna Grosskopf , C. A. Knox Lovell
关键词: Scale (social sciences) 、 Productivity model 、 Operations research 、 Coal mining 、 Inefficiency 、 Productivity 、 Econometrics 、 Economics 、 Sample (statistics) 、 Differential (mechanical device) 、 Partial productivity
摘要: The purpose of this paper is to compare the abilities two competing analytical techniques-mathematical programming and statistical regression-to shed light on union/nonunion productivity differential in U.S. surface coal mining. approach has virtues being nonparametric thus extremely flexible able provide a decomposition differentials into three components-differences technical efficiency, differences scale congestion. Identification these components provides an aid management its search for sources of, remedies for, gaps. econometric approach, other hand, neither nor does it such decomposition; chief virtue lies fact that, stochastic, allows presence noise measurement error that plagues most if not all empirical data. Thus, approaches have complementary virtues. The techniques are used investigate samples mines. finds large significant positive differential. This due primarily greater congestion occurring smaller nonunion mines one sample, inefficiency sample. analysis nearly same differential, but sheds no composition cure