Minimum Efficient Size and Seller Concentration: An Empirical Problem
指出以往研究用统计代理变量衡量最小有效规模时，这些变量本身更像集中度指标，导致技术因素对集中度的影响可能被高估。
THIS paper concerns the empirical relationship between the concentration of sellers in any market and the so-called minimum efficient scale of production (MEP). This relationship forms the basis for what is often termed the technological explanation of concentration. We shall argue that the results of most previous empirical studies in this area may be misleading due to the widespread use of statistical proxy measures for MEP which are themselves better interpreted as measures of concentration. Following an introductory section, which briefly outlines the underlying theory and results of previous studies, section II describes the two widely used proxies for MEP which are the main concern of this paper. We suggest in section III (and Appendix I) that both exhibit the characteristics desirable for a measure of plant concentration. As such, their use in regression analysis 'explaining' measures of firm concentration poses potential identification problems. Furthermore, section IV (and Appendix 2) explain why both proxies will typically involve systematic measuremenit errors which are, themselves, likely to be correlated with concentration. Section V contains a short comparison of the proxies with two alternative MEP estimators, using data from the I 968 UK Census: the results can be interpreted as offering empirical support for the theoretical arguments of the previous sections. In the light of our arguments, we conclude that the influence of technological factors on concentration may well have been overstated by previous empirical research.