Abstract
T HE measurement of firm size plays a crucial role in applied microeconomics and industrial organization. Firm size has figured prominently in numerous studies of economies of scale in production, advertising, capital market, and cash balances, and in studies of concentration, diversification, profitability, regulation, technological change, and research and development. Even when firm size was not their main concern, many studies often found that size emerged as a robust empirical variable.' All these studies have based their findings on different alternative measures of firm size, often implying that great care in choosing between them is unnecessary since the measures are highly intercorrelated. In a note in this REVIEW, Smyth et al. (hereafter SBP) were the first to recognize that alternative measures of firm size are not interchangeable unless stricter conditions than correlation are met. They have further shown that empirical findings regarding economies of scale are not invariant with the size measure chosen, and that often different conclusions can be reached depending on the particular size measure used. The purpose of this paper is threefold: (I) to offer a general stochastic model that rigorously spells out the conditions for interchangeability among alternative measures of firm size, and of which SBP's deterministic model is a special case; (2) to conduct a statistical test of the interchangeability conditionis using a larger number of size measures, and a far larger sample than the one employed in SBP's empirical test; and (3) to empirically analyze the statistical properties of the most commonly used measures in order to help future investigators in selecting appropriate size measures suitable for their purposes. Section I reviews SBP's work, section II discusses the measurement problem, section III presents our theoretical model, and section IV concludes with some empirical evidence.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have