Regulation and Deregulation After 25 Years: Lessons Learned for Research in Industrial Organization PAUL L. JOSKOW Elizabeth and James Killian Professor of Economics, MIT, Cambridge, MA 02139 USA. E-mail: [email protected] Abstract. This paper is based on a keynote address given at the 2004 International Industrial Organization Conference in Chicago, April 2004. I draw selectively on the literature from the past 25 years on regulation/deregulation to provide important lessons about the attributes of good research in empirical industrial organization. I. Introduction When I first began doing research on issues related to government regulation of industry in the early 1970s, a significant slice of the economy was subject to some form of government price and/or entry regulation. The affected industries included (a) airlines, (b) trucking, (c) railroads, (d) electric power, (e) natural gas production, (f) pipeline transportation and distribution of natural gas, (g) crude oil and refined petroleum products, (h) cable television, (i) automobile insurance, (j) hospital services, and others. The share of these industries in GDP, while substantial, understates their potential impact on overall economic performance of the economy since they generally supplied important intermediate goods and services to downstream commercial and industrial users. Beginning about 25 years ago legislative and regulatory actions began a process of deregulation, vertical and horizontal industry restructuring and regulatory reform that has now affected, to varying degrees, all of these industries. Many of these industries have been completely transformed during this period of time (e.g. airlines, trucking, and telecommunications). Others are in the midst of more lengthy transitions that involve2restructuring and deregulation of major industry segments and the application of new regulatory mechanisms to those segments that continue to be regulated (e.g. electricity). Technological change, changes in supply and demand conditions and interest group politics have certainly played a central role in fostering these dramatic changes. However, economic research that has examined the effects of government regulation, deregulation and regulatory reform on prices, costs, innovation and consumer welfare has had significant (generally constructive) effects on these policy initiatives. The research focused on detailed analysis of specific industries rather than the superficial analyses of cross-sections of hundred of census “industries” that characterized much of the research in industrial organization 25 years ago. It integrated theoretical modeling with empirical analysis. It provided a stimulus to the development of new econometric techniques for estimating demand and cost functions. It pioneered the use of structural econometric models to measure the effects of changes in governance arrangements on costs, prices, income distribution and welfare. It made important contributions to and supported developments in corporate finance, political economy, mechanism design and incomplete contracts. Finally, it provided a good platform for studying more traditional issues in industrial organization as the structure, behavior and performance of the “deregulated” industries responded to the relaxation of regulatory constraints on prices and entry. Economic research on regulation, deregulation, and regulatory reform has played an important role in the transformation of research in industrial organization and related fields in economics. The purpose of this paper is not to provide a comprehensive overview of the relevant literature about the causes and consequences of regulation, regulatory reform and3deregulation, something that can be found elsewhere (e.g. Joskow and Noll 1981, Joskow and Rose 1989, Peltzman and Winston 2000). Instead, I will draw very selectively on this literature to make a number of general observations about empirical research in industrial organization more generally. There are important lessons to learn from these developments about the attributes of good research in industrial organization generally and the role it may play in the public policy arena. These lessons seem to me to be especially important today as the field seems to be dividing between scholars committed to so-called “structural approaches” and those embracing a wider range of research methods, a division that is likely to hinder the advancement of knowledge of the core issues in industrial organization. Accordingly, I will take the opportunity provided to me by the Industrial Organization Society to reflect on the last 25+ years of research and policy related to government regulation and deregulation of industry and its potential implications for research in industrial organization more broadly today. I will be so bold as to identify what I think are the most important “lessons learned” from the research accomplishments during the last 25 years. I will draw primarily on research related to government regulatory policies that directly control prices for goods and services and restrict entry of new competitors, as well as reforms that have relaxed or removed these regulations. I will not discuss health, safety and environmental regulation, nor the growing body of work on privatization, though much of the research on deregulation outside of the United States is closely linked with supporting privatization initiatives. By way of introduction to the field let me begin with what in some ways is the most important lesson that I have learned after over 30 years of research on regulated industries and other topics in industrial organization:4Lesson #1: Good empirical research in industrial organization has been driven by a set of basic questions about the structure, behavior and performance of markets for goods and services and the institutional environment in which the producers and consumers in these industries interact. A good research program starts with one or more of basic questions about firms, consumers, market and non-market institutions, and the interactions between them. In the research on regulation, regulatory reform and deregulation produced over the last 25 years the primary questions have been reasonably constant and involve both positive and normative lines of inquiry or questions: a. How does regulation affect costs, prices, the rate and direction of innovation, the distribution