The Limits of Inference without Theory

The Limits of Inference without Theory

Kenneth I. Wolpin
Copyright Date: 2013
Published by: MIT Press
Pages: 200
https://www.jstor.org/stable/j.ctt5vjpxr
  • Cite this Item
  • Book Info
    The Limits of Inference without Theory
    Book Description:

    In this rigorous and well-crafted work, Kenneth Wolpin examines the role of theory in inferential empirical work in economics and the social sciences in general -- that is, any research that uses raw data to go beyond the mere statement of fact or the tabulation of statistics. He considers in particular the limits that eschewing the use of theory places on inference. Wolpin finds that the absence of theory in inferential work that addresses microeconomic issues is pervasive. That theory is unnecessary for inference is exemplified by the expression "let the data speak for themselves." This approach is often called "reduced form." A more nuanced view is based on the use of experiments or quasi-experiments to draw inferences. Atheoretical approaches stand in contrast to what is known as the structuralist approach, which requires that a researcher specify an explicit model of economic behavior -- that is, a theory. Wolpin offers a rigorous examination of both structuralist and nonstructuralist approaches. He first considers ex ante policy evaluation, highlighting the role of theory in the implementation of parametric and nonparametric estimation strategies. He illustrates these strategies with two examples, a wage tax and a school attendance subsidy, and summarizes the results from applications. He then presents a number of examples that illustrate the limits of inference without theory: the effect of unemployment benefits on unemployment duration; the effect of public welfare on women's labor market and demographic outcomes; the effect of school attainment on earnings; and a famous field experiment in education dealing with class size. Placing each example within the context of the broader literature, he contrasts them to recent work that relies on theory for inference.

    eISBN: 978-0-262-31367-4
    Subjects: Economics

Table of Contents

  1. Front Matter
    (pp. i-vi)
  2. Table of Contents
    (pp. vii-viii)
  3. Series Foreword
    (pp. ix-x)
    Philip A. Haile

    The Tjalling C. Koopmans Memorial Lectures were initiated by the Cowles Foundation in 1989 using a fund established by Koopmans’ family, friends, and colleagues shortly after his death in 1985. These lectures offer an opportunity for preeminent scholars of economics to provide synthesis and perspective on a body of ongoing research to a broad audience of economists. The Cowles Foundation is pleased to introduce a monograph series based on these lectures, with the first volume written by Ken Wolpin, the Walter H. and Leonore C. Annenberg Professor in the Social Sciences at the University of Pennsylvania.

    Tjalling Koopmans was a...

  4. Acknowledgments
    (pp. xi-xii)
  5. 1 Introduction
    (pp. 1-4)

    The two lectures that comprised my Tjalling C. Koopmans Memorial Lectures were titled “Ex AntePolicy Evaluation” and “The Limits of Inference without Theory.” I have chosen the second as the title of this book because it encompasses a broad theme that is also illustrated by the content of the first lecture. The title obviously borrows from the famous Koopmans (1947) essay “Measurement without Theory” published in 1947. That essay was a direct response to the book by Burns and Mitchell (1946),Measuring Business Cycles, and, more generally, to the extensive data collection effort being conducted at the National Bureau...

  6. 2 Ex Ante Policy Evaluation—The Role of Theory
    (pp. 5-64)

    The goal of policy evaluation falls into two categories:ex postevaluation andex anteevaluation. The goal of ex post policy evaluation is to determine the impact of policies that have been implemented. This type of evaluation is ubiquitous in economics and in the social sciences more generally. The goal ofex antepolicy evaluation is to determine the impact of prospective or “new” policies. New policies can extend existing policies either in the dimension of a particular policy parameter beyond its current domain or introduce new parameters to an existing policy. An example of the former would be...

  7. 3 The Limits of Inference without Theory
    (pp. 65-148)

    In 1977, theIndustrial and Labor Relations Review(ILRR) published a conference volume, “The Economics of Unemployment Insurance: A Symposium.” The purpose of the volume was to gather together papers that would evaluate the extent to which the unemployment insurance (UI) system affects labor market behavior. Four of the papers were original empirical studies attempting to provide quantitative estimates of the effect of UI benefit generosity on the search behavior of the unemployed (Classen, 1977; Ehrenberg and Oaxaca, 1977; Holen, 1977; Burgess and Kingston, 1976), one of the papers surveyed the empirical literature on that question (Welch, 1977), and one...

  8. 4 A Brief Conclusion
    (pp. 149-150)

    The proper role of economic theory in empirical research has been and remains a controversial issue. In these two lectures, I have tried to illustrate the value for inference of connecting data and theory. In the first lecture (chapter 2), I showed how theory can be used to performex antepolicy evaluation in the most challenging circumstance in which there is no direct policy variation. Theory was shown to be critical in providing a structure within which existing variation that provides an analogue to direct policy variation can be exploited. In the case of a wage tax, wage variation...

  9. Notes
    (pp. 151-164)
  10. References
    (pp. 165-176)
  11. Index
    (pp. 177-184)