Chapter 7 Forestalling the demise of empirical economics: The role of microdata in labor economics research

The chapter presents a discussion on the role of microdata in labor economics research. The chapter begins with a review of characteristics of papers published in selected journals (1965-83) in the field of labor economics. This review demonstrates the changing character of the field in terms of topic, extent to which papers are exclusively theoretical, share of papers with a theoretical section which is meaningful and connects to the subsequent empirical endeavor, and type of data used for empirical work. Consonant with Leontief's findings a growth of theoretical papers in terms of both number and share is observed. It is also observed that this growth of theoretical papers does not appear to have crowded out good empirical work; if anything there seems to have been a parallel growth of both significant theoretical and empirical papers as the development of theory and data interact through time. The chapter presents a detailed discussion of the interaction among theory, data base development and policy issues. Topics highlighted in this discussion are earnings functions and intertemporal labor supply models, including retirement and unemployment. The chapter also discusses some of the advantages and disadvantages of different kinds of microdata, including data from social experiments that were a popular and costly type of data gathered in the United States during the 1970s. The chapter discusses some data based pitfalls and the weakness arising from many of the statistical tests in what can be termed second generation work on several topics.