Labor Economics: Problems in Analyzing Labor Markets Recent Economic Thought Series Editor: Warren G. Samuels Michigan State University East Lansing, Michigan, U.S.A. Other books in the series: Feiwel, G., Samuelson and Neoclassical Economics Wade, L., Political Economy: Modern Views Zimbalist, A., Comparative Economic Systems: Recent Views Darity, W., Labor Economics: Modern Views Jarsulic, M., Money and Macro Policy Samuelson, L., Microeconomic Theory Bromley, D., Natural Resource Economics: Policy Problems and Contemporary Analysis Mirowski, P., The Reconstruction of Economic Theory Field, A., The Future of Economic History Lowry, S., Pre-Classical Economic Thought Officer, L., International Economics Asimakopulos, A., Theories of Income Distribution Earl, P., Psychological Economics; Development, Tensions, Prospects Thweatt, W., Classical Political Economy Peterson, W., Market Power and the Economy DeGregori, T., Development Economics Nowotny, K., Public Utility Regulation Horowitz, I., Decision Theory Mercuro, N., Law and Economics Hennings, K. and Samuels, W., Neoclassical Economic Theory, 1870 to 1930 Samuels, W., Economics as Discourse Lutz, M., Social Economics Weimer, D., Policy Analysis and Economics Bromley, D. and Segerson, K., The Social Response to Environmental Risk Roberts, B. and Feiner, S., Radical Economics Mercuro, N., Taking Property and Just Compensation de Marchi, N., Post-Popperian Methodology of Economics Gapinski, J., The Economics of Saving labor Economics: Problems in Analyzi n9 labor Markets edited by William Darity, Jr. Cary C. Boshamer Professor of Economics University of North Carolina at Chapel Hill .... " Springer Science+Business Media, LLC Library of Congress Cataloging-in-Publication Data Labar economics: problems in analyzing labor markets I edited by William Darity, Jr. p. cm. - (Recent economic thought series) Includes bibliographical references and index. ISBN 978-94-010-5305-1 ISBN 978-94-011-2938-1 (eBook) DOI 10.1007/978-94-011-2938-1 1. Labor economics. 2. Labar market. 1. Darity, William A., 1953- II. Series. HD4901.L133 1992 331.12-dc20 92-18985 CIP Copyright © 1993 by Springer Science+Business Media New York Originally published by Kluwer Academic Publishers in 1993 Softcover reprint ofthe hardcover Ist edition 1993 AII nghts reserved. No part of this publicat ion may be reproduced, stared in a retrieval system ar transmitted in any form or by any means, mechanical, photo-copying, recording, or otherwise, without the prior written permisslon of the publisher, Splinger Science+Business Media, LLC Contents Introduction I. Unemployment 9 Chapter 1: Keynes, Cambridge, and the New Keynesian Economics 11 by Michael Syron Lawlor Chapter 2: Labor Economics and Unemployment: An Historian's Perspective 59 by Alexander Keyssar II. How Labor Markets Really Work 75 Chapter 3: The Impact of Formal On-the-Job Training on Unemployment and the Influence of Gender, Race, and Working Lifecycle Position on Accessibility to On-the-Job Training 77 by Alfred J. Field and Arthur H. Goldsmith Chapter 4: Empirical Tests of Labor Market Equilibrium: An Evaluation 117 by James Heckman and Thomas MaCurdy Chapter 5: Labor Market Segmentation Theory: Reconsidering the Evidence 141 by William T. Dickens and Kevin Lang III. Race, Discrimination, and Competiton 181 Chapter 6: Labor Markets and Racial Inequality 183 by Jeremiah Cotton Chapter 7: Racial Inequality and Racial Conflict: Recent Developments in Radical Theory 209 by Rhonda M. Williams IV. Culture, Ethnicity, and Poverty 237 Chapter 8: Culture and Human Capital: Theory and Evidence or Theory Versus Evidence? 239 by Stephen A. Woodbury Chapter 9: Labor Economics and Public Policy: Dominance of Constraints or Preferences? 269 by E. Michael Foster Contributing Authors 295 Index 299 Labor Economics: Problems in Analyzing Labor Markets INTRODUCTION William Darity, Jr. In 1984 the Kluwer series in Modern Economic Thought, under the editorial direction of Warren Samuels, brought out a book under my editorship entitled Labor Economics: Modern Views. It consisted of a series of essays and commentaries that sought, in a critical fashion, to assess the state of the art in the field of labor economics with respect to several themes. These included methodology versus practice, the analysis of discrimination by gender and race, the phenomenon of persistent racial differences in un employment exposure, occupational safety and health regulation, dual versus segmented labor markets, and the remnants of the Phillips curve trade-off between unemployment and inflation. Nearly a decade later I was approached by Warren Samuels and Kluwer about editing a new book that would again address where things stand in labor economics. In proceeding with the development of this current book I was a struck by the extent to which the research thrust that was apparent in the early 1980s remains intact as we move toward the 21st century. The vast majority of scholarship in the labor subfield is dominated by the methodological orientation of applied neoclassical microeconomics, supplemented by incursions from the themes that occupy the so-called "pure theorists," particularly of the game theoretic variety. But there also 2 LABOR ECONOMICS is a lively body of work lying outside the mainstream that contests ortho doxy at the most fundamental levels-on grounds oflogical consistency and empirical accuracy. I view the current book as a potpourri of perspectives on the status of labor economics both from those who work in the subfield within the terms of applied neoclassical microeconomics and those who work within the terms of alternative modes of thought. Despite common attachment to a singular conception of how one does economic analysis-the starting point always is with the individual and individual-optimizing behavior-there is substantial dissent among labor economists. Indeed, mainstream labor economists remain in perpetual con flict over the issue of how real-world markets, especially those for employment, actually work. Does the labor market in the aggregate or do sectoral markets behave as contract markets, with comparatively inflexible wages, or as auction markets, with a high degree of wage flexibility? From the mainstream perspective, if the former is the case, there is a basis for explaining the occurence and persistence of unemployment; if the latter is the case, then either unemployment does not really occur or it is at worst a temporary and ephemeral phenomenon. The fact that economists would even engage in debate over the theore tical existence of unemployment is bewildering to the historian Alexander Keyssar. Keyssar (1986), author of an important study of the origins of unemployment as a socially recognized phenomenon of concern to policy makers in the state of Massachusetts, sees the incidence of joblessness as so palpable, real, and devastating that he views the debate among econo mists as a descent into the theatre of the absurd. He says as much in his contribution in the first section of the current book. Also in the first section Michael Syron Lawlor provides a remarkable study of the emergence of the theory of aggregate unemployment in Cambridge, England, in the 1930s. Lawlor argues that the fixprice versus flexprice view of labor markets had been a long discussed issue in Cambridge, dating at least from Alfred Marshall's lectures to the genera tion of dons that were to rise to prominence during the interwar years. Moreover, by making careful use of the Keynes Papers at Kings College at Cambridge University, Lawlor demonstrates that the fixprice view of labor markets and the panoply of arguments for wage rigidity utilized by the con temporary "New Keynesian" economists were well developed by Keynes's favorite whipping boy, A.C. Pigou, well before the publication of The General Theory. Thus, Lawlor doubts that the attribution of rigid wage economics to Keynes is an accurate account of his theory of unemployment and suggests that the contract versus auction market debate has no rele vance to the legitimacy of Keynes's theory of less than full employment INTRODUCTION 3 equilibrium. But since the New Keynesians have a broadly dismissive atti tude toward doctrinal history-their major organ, The Quarterly Journal oj Economics, has an active policy of refusing to consider papers in the history of economic thought-I suspect that Lawlor's chapter will not be addressed by any of the principals. Hopefully, others will read his chapter and be inspired to think more deeply about the limitations of how economists have come to conceive of unemployment. The exploration of the question of how labor markets really work is pursued in a trio of essays that occupy the second section of this book. The timbre of the discussions here indicate ways in which the conventional procedures of mainstream economics can be modified or amended to accommodate potential trouble spots, rather than proposing overturning orthodoxy altogether. The first of the three, by Alfred Field and Arthur Goldsmith, asks a seemingly narrow question that has wide ramifications for the usual em pirical work performed by labor economists. Field and Goldsmith are concerned with an omitted variable from traditional studies of earnings and occupational status, formal on-the-job training. But the scope of their inves tigation suggests how far wrong previous empirical research could have gone by leaving out this variable. Utilizing the Bureau of Labor Statistics' January 1984 Job Training Survey as a supplement to the January 1984 Bureau of the Census Current Population Survey, they are able to demonstrate how dramatically employment and lay-off probability regression equations will be altered when formal on-the-job training is taken into account. The Field Goldsmith investigation also suggests how poorly on-the-job training, whether formal or informal, has been incorporated into empirical labor economics, despite the importance ascribed to such training in theoretical work on human capital accumulation. In the second essay in this group, James Heckman and Thomas MaCurdy undertake an examination of the evidence on whether labor markets are best characterized by equilibrium (in the sense of market clearing) or disequi librium (in the sense of nonmarket clearing) modes of thought. This is the same issue that Goldsmith with an earlier coauthor, Thomas Kniesner, sought to evaluate in a major survey article that appeared in the Journal ojE conomic Literature in 1987. Both articles conclude that the issue cannot be resolved unequivocally on the basis of data available on labor market dynamics or on the basis of existing tests of whether labor markets tend to clear or not to clear. In a related vein, William Dickens and Kevin Lang reexamine the evidence on whether labor markets should be characterized as segmented or singular. In the mid-1980s they first introduced a procedure for testing for the 4 LABOR ECONOMICS existence of dualism in the U.S. labor market (Dickens and Lang 1985). Their procedure, which they argued established that the structure of markets for employment in the United States is, indeed, segmented, spawned a body of research both supportive and critical of their findings. Probably the most negative assessment came from Heckman in collaboration with Joseph Hotz in an article focusing on dualism in the labor market in Panama, where Heckman and Hotz (1986) challenged the Dickens and Lang procedure (rather than necessarily disputing the Dickens and Lang conclusion). Dickens and Lang use the current essay as an opport1l;nity to reply to Heckman and Hotz as well as other critics. While Dickens and Lang see inadequacies in the conventional treatment of labor markets as a continuum of opportunities potentially open to all comers with the requisite skills regardless of ascriptive characteristics, they want to explain the workings of segmented markets in a manner consistent with individual optimizing behavior. Labor markets are segmented, hence imperfect, and there is nothing anyone can do about that; for Dickens and Lang it is a virtual state of nature. But given that fact, individuals then behave in an optimal fashion. Therefore, they see a convergence between the empirical observation that labor markets are segmented and the body of theory that invokes informational asymmetries, efficiency wage doctrine, adverse selection and the like-arguments that have individuals behaving optimally in the face of an immutable imperfection (although precisely why the grand imperfection is immutable is rarely, if ever, explained).' Dickens and Lang view segmented labor market theory as an alternative to human capital-based explanations for wage differentials. But if seg mentation exists-and their evidence seems to be reasonably persuasive that it does-the key question is why some persons or groups have a greater likelihood of being present in one segment rather than another. Is it because they are steered there, independent of their capacity to perform, or is it because their individual abilities best suit them for a particular sector? To be more direct: are black workers disproportionately found in the low-wage segment of the labor market because of discrimination or because of lower levels of human capital accumulation? Obviously, the resolution of this question lies at the heart of how mainstream economists think about the determination of income distribution and the "fairness" with which the market system treats participants . . The fairness of a putatively capitalistic marketplace is the issue that provokes Jeremiah Cotton's inquiry in the section of the book entitled "Race, Discrimination, and Inequality." Cotton contributes an essay that reasserts the real world significance of labor market discrimination while simultaneously establishing the deep range of anomalies the persistence of