The authors are grateful to Avner Ben-Ner, Jacques Defourny, Saul Estrin, Felix FitzRoy, Barbara Lee, Bentley MacLeod, Egon Neuberger, Jeffrey Pliskin, Stephen C. Smith, and three anonymous referees for comments on various drafts of this paper. Jones acknowledges with gratitude thefinancial assistance of the National Science Foundation (Grant # 9010591). Of course, all remaining errors and any omissions are entirely the responsibility of the authors.

We survey the uses of combinatorial auctions that have been deployed in practice, giving emphasis to their key representational and economic aspects. In addition, we discuss behavioral economics considerations on both the bidder and auctioneer sides of the market, and the interrelated topics of simplicity and trust, highlighting key opportunities for future work.

Entrepreneurship is thought to be a key driver of economic growth. While there are myriad forms of entrepreneurship, ranging from self-employment to small and medium size enterprises to technology- and innovation-driven startups, recent research provides evidence that the relationship between entrepreneurship and economic growth is driven not by overall quantity of new firm entry, but rather by a small subset of high-growth startups that are primarily categorized as innovation-driven. This paper provides a survey of the growing literature on the economics of such innovation-driven entrepreneurship. We begin by distinguishing between the various forms of entrepreneurship, which are often confounded in both theory and empirical work. We lay out the current state of knowledge, and describe the challenges faced by researchers in the field, particularly around measurement, data and identification. We conclude with an overview of the major open questions and directions for future research in the area.Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.

I have benefited from comments and criticisms of an earlier draft by Irma Adelman, Peter Balacs, Ronald Dore, Edgar Edwards, Unni Eradi, Michael Faber, Anne Gordon, Keith Griffin, Jill Rubery, Seev Hirsch, Ernest Stern, Frances Stewart, Hugh Stretton, B. R. Virmani, Gordon Winston and Howard Wriggins. To these, and to a research seminar at Queen Elizabeth House, I am very grateful. I am also grateful to the Economic Development Institute of the World Bank and its Director, Mr. Andrew Kamarck, for having provided the facilities and stimulating atmosphere for the early stages of a considerably larger paper, commissioned by Mr. Ernest Stern, of which this paper forms a part. I am grateful to Mr. Stern and the World Bank for permitting me to use the material here.

Difference-in-differences (DiD) is arguably the most popular quasi-experimental research design. Its canonical form, with two groups and two periods, is well understood. However, empirical practices can be ad hoc when researchers go beyond that simple case. This article provides an organizing framework for discussing different types of DiD designs and their associated DiD estimators. It discusses covariates, weights, handling multiple periods, and staggered treatments. The organizational framework, however, applies to other extensions of DiD methods as well.

We survey the literature on estimating risk preferences using field data. We concentrate our attention on studies in which risk preferences are the focal object and estimating their structure is the core enterprise. We review a number of models of risk preferences—including both expected utility (EU) theory and non-EU models—that have been estimated using field data, and we highlight issues related to identification and estimation of such models using field data. We then survey the literature, giving separate treatment to research that uses individual-level data (e.g., property-insurance data) and research that uses aggregate data (e.g., betting-market data). We conclude by discussing directions for future research.

JHE CLIOMETRIC revolution in ecoknomic history wedded neoclassical economics and quantitative methods in order to describe and explain the performance of economies in the past.' Economic history gained in rigor and scientific pretension, but at the expense of exploring a much more fundamental set of questions about the evolving structure of economies that underlies performance.2 Cliometricians have turned their backs on a long tradition stretching back from Joseph Schumpeter to Karl Marx to Adam Smith. These scholars regarded economic history as essential because it added a dimension to economics. Its purpose was to analyze the parameters held constant by the economist. If economics is a theory of choice subject to specified constraints, a task of economic history was to theorize about those evolving constraints. The failure of economic historians to provide their colleagues with a historical dimension to their perspective has reduced the effectiveness of economists in dealing with contemporary problems. Failure of economists to appreciate the transitory character of the assumed constraints and to understand the source and direction of these changing constraints is a fundamental handicap to further development of economic theory. The challenge to the economic historian which has equally compelling implications for the economic theorist is to explain the transformation of the structure of the American Economy in the past century.3 In the rest of this essay I shall explore this issue in order to specify some of the dimensions of the economic historian's task.

In early 1991, the Dutch pig-industry was struck by the so-called mystery swine disease. Large-scale laboratory investigations were undertaken to search for the etiological agent. We focused on isolating viruses and mycoplasmas, and we tested paired sera of affected sows for antibodies against ten known pig viruses. The mycoplasmas M. hyosynoviae, M. hyopneumoniae, and Acholeplasma laidlawii, and the viruses encephalomyocarditis virus and porcine enterovirus types 2 and 7 were isolated from individual pigs. An unknown agent, however, was isolated from 16 of 20 piglets and from 41 of 63 sows. This agent was characterised as a virus and designated Lelystad virus. No relationship between this virus and other viruses has yet been established. Of 165 sows reportedly afflicted by the disease, 123 (75 per cent) seroconverted to Lelystad virus, whereas less than 10 per cent seroconverted to any of the other virus isolates or to the known viral pathogens. Antibodies directed against Lelystad virus were also found in pigs with mystery swine disease in England, Germany, and in the United States. We conclude that infection with Lelystad virus is the likely cause of mystery swine disease.