Modern Evidence on the Firm
反思理论与实证的关系，指出理论若无证据只是猜测，并评述企业实证研究的现状，特别是交易成本与代理理论的有用性。
In the main, empirical research is regarded as subordinate to theory. Theorists perform the difficult and innovative work of conceiving new and sometimes ingenious explanations for the world around us, leaving empiricists the relatively mundane task of gathering data and applying tools (supplied by theoretical econometricians) to support or reject hypotheses that emanate from the theory. To be sure, facts by themselves are worthless, “a mass of descriptive material waiting for a theory, or a fire,” as Ronald H. Coase (1984 p. 230), in characteristic form, dismissed the contribution of the oldschool institutionalists. But without diminishing in any way the creativity inherent in good theoretical work, it is worth remembering that theory without evidence is, in the end, just speculation. Two questions that theory alone can never answer are: First, which of the logically possible explanations for observed phenomena is the most probable? And second, are the phenomena that constitute the object of our speculations important? Modern empirical research on the firm, which dates to Kirk Monteverde and David J. Teece’s articles examining automotive procurement (1982a, b), is now substantial. Rather than try to summarize or discuss particular results of that literature, for which there are already several excellent surveys (the most recent of which is that of Christopher Boerner and Jeffrey Macher [2001]), my aim in the following is to comment more generally on the overall state of knowledge, particularly, the lessons that can be gleaned from the empirical literature about the relative usefulness of transaction-cost and agency theories of the firm and about the value of continued research on economic organization.