Labor Economics
Seventh Edition
George J. Borjas
Harvard University
LABOR ECONOMICS, SEVENTH EDITION
Published by McGraw-Hill Education, 2 Penn Plaza, New York, NY 10121. Copyright © 2016 by McGraw-Hill
Education. All rights reserved. Printed in the United States of America. Previous editions © 2013, 2010, and
2008. No part of this publication may be reproduced or distributed in any form or by any means, or stored in
a database or retrieval system, without the prior written consent of McGraw-Hill Education, including, but not
limited to, in any network or other electronic storage or transmission, or broadcast for distance learning.
Some ancillaries, including electronic and print components, may not be available to customers outside the
United States.
This book is printed on acid-free paper.
1 2 3 4 5 6 7 8 9 0 DOC/DOC 1 0 9 8 7 6 5
ISBN 978-0-07-802188-6
MHID 0-07-802188-X
Senior Vice President, Products & Markets: Kurt L. Strand
Vice President, General Manager, Products & Markets: Marty Lange
Vice President, Content Design & Delivery: Kimberly Meriwether David
Director, Product Development: Meghan Campbell
Managing Director: James Heine
Lead Product Developer: Michele Janicek
Senior Product Developer: Christina Kouvelis
Marketing Manager: Katie Hoenicke
Director of Digital Content Development: Doug Ruby
Digital Product Developer: Meg Maloney
Director, Content Design & Delivery: Terri Schiesl
Full-Service Manager: Faye Schilling
Content Project Managers: Jane Mohr, Mark Christianson, and Sandra M. Schnee
Buyer: Jennifer Pickel
Design: Studio Montage, St. Louis, MO
Content Licensing Specialist: Rita Hingtgen
Cover Image: © hh5800/Getty Images
Compositor: Laserwords Private Limited
Typeface: 10/12 Times LT Std
Printer: R. R. Donnelley
All credits appearing on page or at the end of the book are considered to be an extension of the copyright page.
Library of Congress Cataloging-in-Publication Data
Borjas, George J.
Labor economics / George J. Borjas. — Seventh edition.
pages cm
ISBN 978-0-07-802188-6 (alk. paper)
1. Labor economics. 2. Labor market—United States. I. Title.
HD4901.B674 2016
331.0973—dc23
2014031865
The Internet addresses listed in the text were accurate at the time of publication. The inclusion of a website does
not indicate an endorsement by the authors or McGraw-Hill Education, and McGraw-Hill Education does not
guarantee the accuracy of the information presented at these sites.
www.mhhe.com
About the Author
George J. Borjas
George J. Borjas is the Robert W. Scrivner Professor of Economics and Social Policy at the
John F. Kennedy School of Government, Harvard University. He is also a research associate at the National Bureau of Economic Research and a Research Fellow at IZA. Professor
Borjas received his Ph.D. in economics from Columbia University.
Professor Borjas has written extensively on labor market issues. He is the author of
several books, including Wage Policy in the Federal Bureaucracy (American Enterprise
Institute, 1980), Friends or Strangers: The Impact of Immigrants on the U.S. Economy
(Basic Books, 1990), Heaven’s Door: Immigration Policy and the American Economy
(Princeton University Press, 1999), and Immigration Economics (Harvard University Press,
2014). He has published more than 125 articles in books and scholarly journals, including the American Economic Review, the Journal of Political Economy, and the Quarterly
Journal of Economics.
Professor Borjas was elected a Fellow of the Econometric Society in 1998, and a Fellow
of the Society of Labor Economics in 2004. In 2011, Professor Borjas was awarded the
IZA Prize in Labor Economics. He was an editor of the Review of Economics and Statistics
from 1998 to 2006. He also has served as a member of the Advisory Panel in Economics at
the National Science Foundation and has testified frequently before congressional committees and government commissions.
iii
The McGraw-Hill Series
in Economics
Essentials of Economics
Brue, McConnell, and Flynn
Essentials of Economics
Third Edition
Mandel
Economics: The Basics
Second Edition
Schiller
Essentials of Economics
Ninth Edition
Principles of Economics
Colander
Economics, Microeconomics, and
Macroeconomics
Ninth Edition
Frank and Bernanke
Principles of Economics, Principles
of Microeconomics, and Principles
of Macroeconomics
Sixth Edition
Frank and Bernanke
Brief Editions: Principles
of Economics, Principles of
Microeconomics, and Principles
of Macroeconomics
Second Edition
Karlan and Morduch
Economics, Microeconomics, and
Macroeconomics
First Edition
McConnell, Brue, and Flynn
Economics, Microeconomics, and
Macroeconomics
Twentieth Edition
McConnell, Brue, and Flynn
Brief Editions: Economics,
Microeconomics, and
Macroeconomics
Second Edition
Miller
Principles of Microeconomics
First Edition
Samuelson and Nordhaus
Economics, Microeconomics, and
Macroeconomics
Nineteenth Edition
Schiller
The Economy Today, The Micro
Economy Today, and The Macro
Economy Today
Thirteenth Edition
Slavin
Economics, Microeconomics, and
Macroeconomics
Eleventh Edition
Economics of Social
Issues
Guell
Issues in Economics Today
Seventh Edition
Sharp, Register, and Grimes
Economics of Social Issues
Twentieth Edition
Econometrics
Gujarati and Porter
Basic Econometrics
Fifth Edition
Gujarati and Porter
Essentials of Econometrics
Fourth Edition
Hilmer and Hilmer
Practical Econometrics
First Edition
Managerial Economics
Baye and Prince
Managerial Economics and
Business Strategy
Eighth Edition
Brickley, Smith, and Zimmerman
Managerial Economics
and Organizational
Architecture
Sixth Edition
Thomas and Maurice
Managerial Economics
Eleventh Edition
Intermediate
Economics
Frank
Microeconomics and Behavior
Ninth Edition
Advanced Economics
Romer
Advanced Macroeconomics
Fourth Edition
Money and Banking
Cecchetti and Schoenholtz
Money, Banking, and Financial
Markets
Fourth Edition
Urban Economics
O’Sullivan
Urban Economics
Eighth Edition
Labor Economics
Borjas
Labor Economics
Seventh Edition
McConnell, Brue, and Macpherson
Contemporary Labor Economics
Tenth Edition
Public Finance
Rosen and Gayer
Public Finance
Tenth Edition
Seidman
Public Finance
First Edition
Environmental
Economics
Field and Field
Environmental Economics: An
Introduction
Sixth Edition
International Economics
Appleyard and Field
International Economics
Eighth Edition
Bernheim and Whinston
Microeconomics
Second Edition
King and King
International Economics,
Globalization, and Policy:
A Reader
Fifth Edition
Dornbusch, Fischer, and Startz
Macroeconomics
Twelfth Edition
Pugel
International Economics
Sixteenth Edition
To Sarah, Timothy, and Rebecca
v
Preface to the Seventh Edition
The original motivation for writing Labor Economics grew out of my years of teaching
labor economics to undergraduates. After trying out many of the textbooks in the market, it
seemed to me that students were not being exposed to what the essence of labor economics
was about: to try to understand how labor markets work. As a result, I felt that students did
not really grasp why some persons choose to work, while other persons withdraw from the
labor market; why some firms expand their employment at the same time that other firms
are laying off workers; or why earnings are distributed unequally in most societies.
The key difference between Labor Economics and competing textbooks lies in its philosophy.
I believe that knowing the story of how labor markets work is, in the end, more important
than showing off our skills at constructing elegant models of the labor market or remembering hundreds of statistics and institutional details summarizing labor market conditions
at a particular point in time.
I doubt that many students will (or should!) remember the mechanics of deriving a labor
supply curve or the way that the unemployment rate is officially calculated 10 or 20 years
after they leave college. However, if students could remember the story of the way the labor
market works—and, in particular, that workers and firms respond to changing incentives
by altering the amount of labor they supply or demand—the students would be much better
prepared to make informed opinions about the many proposed government policies that
can have a dramatic impact on labor market opportunities, such as a “workfare” program
requiring that welfare recipients work or a payroll tax assessed on employers to fund a
national health care program or a guest worker program that grants tens of thousands of
entry visas to high-skill workers. The exposition in this book, therefore, stresses the ideas
that labor economists use to understand how the labor market works.
The book also makes extensive use of labor market statistics and reports evidence
obtained from hundreds of research studies. These data summarize the stylized facts that a
good theory of the labor market should be able to explain, as well as help shape our thinking about the way the labor market works. The main objective of the book, therefore, is to
survey the field of labor economics with an emphasis on both theory and facts. The book
relies much more heavily on “the economic way of thinking” than competing textbooks.
I believe this approach gives a much better understanding of labor economics than an
approach that minimizes the story-telling aspects of economic theory.
Requirements
vi
The book uses economic analysis throughout. All of the theoretical tools are introduced
and explained in the text. As a result, the only prerequisite is that the student has some
familiarity with the basics of microeconomics, particularly supply and demand curves. The
exposure acquired in the typical introductory economics class more than satisfies this prerequisite. All other concepts (such as indifference curves, budget lines, production functions, and isoquants) are motivated, defined, and explained as they appear in our story. The
book does not make use of any mathematical skills beyond those taught in high school
algebra (particularly the notion of a slope).
Preface to the Seventh Edition vii
Labor economists also make extensive use of econometric analysis in their research.
Although the discussion in this book does not require any prior exposure to econometrics,
the student will get a much better “feel” for the research findings if they know a little about
how labor economists manipulate data to reach their conclusions. The appendix to Chapter 1
provides a simple (and very brief) introduction to econometrics and allows the student to
visualize how labor economists conclude, for instance, that wealth reduces labor supply, or
that schooling increases earnings. Additional econometric concepts widely used in labor
economics—such as the difference-in-differences estimator or instrumental variables—are
introduced in the context of policy-relevant examples throughout the text.
Changes in the Seventh Edition
The Seventh Edition continues and expands traditions established in earlier editions.
In particular, the text has a number of new detailed policy discussions and uses the evidence reported in state-of-the-art research articles to illustrate the many applications
of modern labor economics. As before, the text continues to make frequent use of such
econometric tools as fixed effects, the difference-in-differences estimator, and instrumental variables—tools that play a central role in the toolkit of labor economists. In keeping
with my philosophy that textbooks are not meant to be encyclopedias, some of the material
that had been a staple in earlier editions has been shortened and sometimes even excluded,
so that the Seventh Edition is roughly the same length as previous editions.
Users of the textbook reacted favorably to the substantial rearrangement of material
(mainly of labor supply) that I carried out in previous editions. The Seventh Edition continues this reframing by tightening up and bringing together much of the discussion on
immigration. Specifically, I have moved the derivation of the immigration surplus model
to the general discussion of international migration in the labor mobility chapter. This
rearrangement of the material gave me the opportunity to add a new section that shows
how the gains from immigration can be greatly increased if immigrants generate human
capital externalities in the receiving country’s labor market. The extension of the immigration surplus model allows for an even more policy-relevant (and economically interesting)
coverage of an important topic—a topic that many students find to be a particularly useful
application of the theoretical models of labor economics.
The last edition introduced a Mathematical Appendix that appears at the end of the
textbook. This appendix presents a mathematical version of some of the canonical models
in labor economics, including the neoclassical model of labor-leisure choice, the model of
labor demand, and the schooling model. It is important to emphasize that the Mathematical
Appendix is an “add-on.” None of the material in this appendix is a prerequisite to reading
or understanding any of the discussion in the core chapters of the textbook. Instructors who
like to provide a more technical derivation of the various models can use the appendix as
a takeoff point for their own presentation. Many instructors welcomed the addition of the
mathematical appendix to the textbook. I, in turn, would truly welcome any suggestions
about how the appendix can be expanded in future editions.
Among the specific changes contained in the Seventh Edition are:
1. Several new “Theory at Work” boxes. The sidebars now include a discussion of how
workers take advantage of the institutional features of the Earned Income Tax Credit
viii Preface
to “bunch up” their hours and ensure they receive the maximum subsidy; the interesting relation between increases in the minimum wage and teenage drunk driving;
the important role that “Rosenwald schools” played in narrowing the education gap
between white and African-American workers; and the labor market impact of explicit
gender discrimination in employment ads in China.
2. A careful updating of all the data tables in the text. To the extent possible, the tables
now include information on the rapidly growing demographic group of “Asians” in the
U.S. labor market and the text often discusses the differences between Asians and other
racial/ethnic groups.
3. A careful summary and discussion of unemployment trends in the United States since
the financial crisis of 2008 and the subsequent Great Recession.
4. New sections that discuss the labor market effects of Obamacare; the experimental evidence on the link between various methods of incentive pay for teachers and student
achievement; the labor market impact of the explosive growth in trade with China; the
potentially important role played by the human capital spillovers presumably generated by
high-skill immigration; and the link between compensating differentials and income taxes.
As in previous editions, each chapter contains “Web Links,” guiding students to
websites that provide additional data or policy discussions. There is an updated list of
“Selected Readings” that includes both standard references in a particular area as well
as recent applications. Finally, each chapter in the Seventh Edition continues to offer
15 end-of-chapter problems, but there is at least one brand new problem in each chapter.
Organization of the Book
The instructor will find that this book is much shorter than competing labor economics
textbooks. The book contains an introductory chapter, plus 11 substantive chapters. If the
instructor wished to cover all of the material, each chapter could serve as the basis for about
a week’s worth of lectures in a typical undergraduate semester course. Despite the book’s
brevity, the instructor will find that all of the key topics in labor economics are covered.
The discussion, however, is kept to essentials as I have tried very hard not to deviate into
tangential material, or into 10-page-long ruminations on my pet topics.
Chapter 1 presents a brief introduction that exposes the student to the concepts of labor
supply, labor demand, and equilibrium. The chapter uses the “real-world” example of the
Alaskan labor market during the construction of the oil pipeline to introduce these concepts.
In addition, the chapter shows how labor economists contrast the theory with the evidence,
as well as discusses the limits of the insights provided by both the theory and the data. The
example used to introduce the student to regression analysis is drawn from “real-world”
data—and looks at the link between differences in mean wages across occupations and
differences in educational attainment as well as the “female-ness” of occupations.
The book begins the detailed analysis of the labor market with a detailed study of labor
supply and labor demand. Chapter 2 examines the factors that determine whether a person
chooses to work and, if so, how much, while Chapter 3 examines the factors that determine how many workers a firm wants to hire. Chapter 4 puts together the supply decisions
of workers with the demand decisions of employers and shows how the labor market
“balances out” the conflicting interests of the two parties.
Preface to the Seventh Edition ix
The remainder of the book extends and generalizes the basic supply–demand framework. Chapter 5 stresses that jobs differ in their characteristics, so that jobs with unpleasant
working conditions may have to offer higher wages in order to attract workers. Chapter 6
stresses that workers are different because they differ either in their educational attainment
or in the amount of on-the-job training they acquire. These human capital investments help
determine the economy’s wage distribution. Chapter 7 discusses how changes in the rate of
return to skills in the 1980s and 1990s changed the wage distribution in many industrialized economies, particularly in the United States. Chapter 8 describes a key mechanism
that allows the labor market to balance out the interests of workers and firms, namely labor
turnover and migration.
The final section of the book discusses a number of distortions and imperfections in
labor markets. Chapter 9 analyzes how labor market discrimination affects the earnings
and employment opportunities of minority workers and women. Chapter 10 discusses how
labor unions affect the relationship between the firm and the worker. Chapter 11 notes
that employers often find it difficult to monitor the activities of their workers, so that the
workers will often want to “shirk” on the job. The chapter discusses how different types of
incentive pay systems arise to discourage workers from misbehaving. Finally, Chapter 12
discusses why unemployment can exist and persist in labor markets.
The text uses a number of pedagogical devices designed to deepen the student’s understanding of labor economics. A chapter typically begins by presenting a number of stylized facts about the labor market, such as wage differentials between blacks and whites or
between men and women. The chapter then presents the story that labor economists have
developed to understand why these facts are observed in the labor market. Finally, the
chapter extends and applies the theory to related labor market phenomena. Each chapter
typically contains at least one lengthy application of the material to a major policy issue, as
well as several boxed examples showing the “Theory at Work.”
The end-of-chapter material also contains a number of student-friendly devices. There
is a chapter summary describing briefly the main lessons of the chapter; a “Key Concepts”
section listing the major concepts introduced in the chapter (when a key concept makes
its first appearance, it appears in boldface). Each chapter includes “Review Questions”
that the student can use to review the major theoretical and empirical issues, a set of 15
problems that test the students’ understanding of the material, as well as a list of “Selected
Readings” to guide interested students to many of the standard references in a particular
area of study. Each chapter then ends with “Web Links,” listing websites that can provide
more detailed information about particular issues.
Supplements for the Book
There are several learning and teaching aids that accompany the seventh edition of Labor
Economics. These resources are available to instructors for quick download and convenient
access via the Instructor Resource material available through McGraw-Hill Connect®.
A Solutions Manual and Test Bank have been prepared by Robert Lemke of Lake Forest
College. The Solutions Manual provides detailed answers to all of the end-of-chapter
problems. The comprehensive Test Bank offers over 350 multiple-choice questions in
Word and electronic format. Test questions have now been categorized by AACSB learning categories, Bloom’s Taxonomy, level of difficulty, and the topic to which they relate.
x
Preface
The computerized Test Bank is available through McGraw-Hill’s EZ Test Online, a flexible
and easy-to-use electronic testing program. It accommodates a wide range of question types
and you can add your own questions. Multiple versions of the test can be created and any
test can be exported for use with course management systems such as Blackboard. The
program is available for Windows and Macintosh environments. PowerPoint Presentations
prepared by Michael Welker of Franciscan University of Steubenville, contain a detailed
review of the important concepts presented in each chapter. The slides can be adapted and
edited to fit the needs of your course. A Digital Image Library is also included, which
houses all of the tables and figures featured in this book.
Digital Solutions
McGraw-Hill Connect® Economics
Less Managing. More Teaching. Greater Learning.
McGraw-Hill’s Connect® Economics is an online assessment solution that
connects students with the tools and resources they’ll need to achieve success.
McGraw-Hill’s Connect Economics Features
Connect Economics allows faculty to create and deliver exams easily with selectable test
bank items. Instructors can also build their own questions into the system for homework or
practice. Other features include:
Instructor Library The Connect Economics Instructor Library is your repository
for additional resources to improve student engagement in and out of class. You
can select and use any asset that enhances your lecture. The Connect Economics
Instructor Library includes all of the instructor supplements for this text.
Student Resources Any supplemental resources that align with the text for student
use will be available through Connect.
Student Progress Tracking Connect Economics keeps instructors informed about how
each student, section, and class is performing, allowing for more productive use of lecture and office hours. The progress-tracking function enables you to:
• View scored work immediately and track individual or group performance with
assignment and grade reports.
• Access an instant view of student or class performance relative to learning objectives.
• Collect data and generate reports required by many accreditation organizations, such
as AACSB.
Diagnostic and Adaptive Learning of Concepts: LearnSmart and SmartBook offer the
first and only adaptive reading experience designed to change the way students read
and learn.
Students want to make the best use of their study time. The
LearnSmart adaptive self-study technology within Connect Economics provides students
with a seamless combination of practice, assessment, and remediation for every concept
Preface to the Seventh Edition xi
in the textbook. LearnSmart’s intelligent software adapts to every student response and
automatically delivers concepts that advance students’ understanding while reducing time
devoted to the concepts already mastered. The result for every student is the fastest path to
mastery of the chapter concepts. LearnSmart:
• Applies an intelligent concept engine to identify the relationships between concepts and
to serve new concepts to each student only when he or she is ready.
• Adapts automatically to each student, so students spend less time on the topics they
understand and practice more those they have yet to master.
• Provides continual reinforcement and remediation, but gives only as much guidance as
students need.
• Integrates diagnostics as part of the learning experience.
• Enables you to assess which concepts students have efficiently learned on their own,
thus freeing class time for more applications and discussion.
Smartbook is an extension of LearnSmart—an adaptive eBook
that helps students focus their study time more effectively. As students read, Smartbook
assesses comprehension and dynamically highlights where they need to study more.
For more information about Connect, go to connect.mheducation.com, or contact your
local McGraw-Hill sales representative.
McGraw-Hill’s Customer Experience Group
We understand that getting the most from your new technology can be challenging. That’s why our services
don’t stop after you purchase our products. You can e-mail our Product Specialists 24 hours a day to get
product-training online. Or you can search our knowledge bank of Frequently Asked Questions on our
support website. For Customer Support, call 800-331-5094, or visit www.mhhe.com/support.
McGraw-Hill Create™ is a self-service website that allows you to create
customized course materials using McGraw-Hill’s comprehensive, crossdisciplinary content and digital products. You can even access third-party
content such as readings, articles, cases, and videos. Arrange the content you’ve selected to
match the scope and sequence of your course. Personalize your book with a cover design
and choose the best format for your students–eBook, color print, or black-and-white print.
And, when you are done, you’ll receive a PDF review copy in just minutes!
Go paperless with eTextbooks from CourseSmart and move light years beyond
traditional print textbooks. Read online or offline anytime, anywhere. Access
your eTextbook on multiple devices with or without an internet connection. CourseSmart
eBooks include convenient, built-in tools that let you search topics quickly, add notes and
highlights, copy/paste passages, and print any page.
Acknowledgments
I have benefited from countless e-mail messages sent by users of the textbook—both
students and instructors. These messages often contained very valuable suggestions, most
of which found their way into the Seventh Edition. I strongly encourage users to contact
me (gborjas@harvard.edu) with any comments or changes that they would like to see
included in the next revision. I am grateful to Robert Lemke of Lake Forest College, who
updated the quiz questions for this edition, helped me expand the menu of end-of-chapter
problems, and collaborated in and revised the Solutions Manual and Test Bank; and Michael
Welker, Franciscan University of Steubenville, who created the PowerPoint presentation
for the Seventh Edition. I am particularly grateful to many friends and colleagues who
have generously shared some of their research data so that I could summarize and present
it in a relatively simple way throughout the textbook, including Daniel Aaronson, David
Autor, William Carrington, Chad Cotti, John Friedman, Barry Hirsch, Lawrence Katz,
Alan Krueger, David Lee, Bhashkar Mazumder, and Solomon Polachek. Finally, I have
benefited from the comments and detailed reviews made by many colleagues on the earlier
editions. These colleagues include:
Ulyses Balderas
Sam Houston State University
Laura Boyd
Denison University
Lawrence Boyd
University of Hawaii, West Oahu
Kristine Brown
University of Illinois–Champaign
John Buck
Jacksonville University
Darius Conger
Ithaca College
Jeffrey DeSimone
University of Texas Arlington
Richard Dibble
New York Institute Technology
Andrew Ewing
University of Washington
Julia Frankland
Malone University
Steffen Habermalz
Northwestern University
xii
Mehdi Haririan
Bloomsburg University of Pennsylvania
Masanori Hashimoto
Ohio State University–Columbus
James Hill
Central Michigan University
Robert N. Horn
James Madison University
Jessica Howell
California State University–Sacramento
Sarah Jackson
Indiana University of Pennsylvania–Indiana
Thomas Kniesner
Syracuse University
Cory Koedel
University of Missouri–Columbia
Myra McCrickard
Bellarmine University
Elda Pema
Naval Postgraduate School
Esther Redmount
Colorado College
Acknowledgments xiii
Jeff Sarbaum
University of North Carolina–Greensboro
Martin Shields
Colorado State University
Todd Steen
Hope College
Erdal Tekin
Georgia State University
Alejandro Velez
Saint Mary’s University
Elizabeth Wheaton
Southern Methodist University
Janine Wilson
University of California–Davis
Contents in Brief
1 Introduction to Labor Economics
2 Labor Supply
3 Labor Demand
1
21
144
5 Compensating Wage Differentials 196
229
7 The Wage Structure
8 Labor Mobility
282
312
9 Labor Market Discrimination
xiv
412
11 Incentive Pay
458
12 Unemployment 494
84
4 Labor Market Equilibrium
6 Human Capital
10 Labor Unions
362
MATHEMATICAL APPENDIX:
SOME STANDARD MODELS IN
LABOR ECONOMICS 541
NAME INDEX 552
SUBJECT INDEX 560
Contents
Chapter 1
Introduction to Labor Economics
Summary 79
Key Concepts 79
Review Questions 79
Problems 80
Selected Readings 83
Web Links 83
1
1-1
An Economic Story of the Labor
Market 2
1-2 The Actors in the Labor Market 3
1-3 Why Do We Need a Theory? 7
Summary 11
Review Questions 11
Key Concepts 11
Web Links 11
Appendix:
An Introduction to Regression Analysis 12
Key Concepts 20
Chapter 2
Labor Supply
21
2-1
2-2
2-3
2-4
2-5
2-6
2-7
2-8
2-9
2-10
Measuring the Labor Force 22
Basic Facts about Labor Supply 24
The Worker’s Preferences 27
The Budget Constraint 31
The Hours of Work Decision 33
To Work or Not to Work? 39
The Labor Supply Curve 42
Estimates of the Labor Supply Elasticity 45
Labor Supply of Women 50
Policy Application: Welfare Programs
and Work Incentives 54
2-11 Policy Application: The Earned Income
Tax Credit 59
2-12 Labor Supply over the Life Cycle 64
2-13 Policy Application: The Decline in Work
Attachment among Older Workers 73
Theory at Work: Dollars and Dreams 40
Theory at Work: Winning the Lotto Will
Change Your Life 43
Theory at Work: Work and Leisure in Europe
and the United States 48
Theory at Work: Gaming the EITC 63
Theory at Work: Cabbies in New York City 69
Theory at Work: The Notch Babies 75
Chapter 3
Labor Demand
84
3-1
3-2
The Production Function 85
The Employment Decision in the Short
Run 88
3-3 The Employment Decision in the Long
Run 94
3-4 The Long-Run Demand Curve
for Labor 98
3-5 The Elasticity of Substitution 105
3-6 Policy Application: Affirmative Action and
Production Costs 106
3-7 Marshall’s Rules of Derived Demand 109
3-8 Factor Demand with Many Inputs 112
3-9 Overview of Labor Market
Equilibrium 114
3-10 Policy Application: The Employment Effects
of Minimum Wages 115
3-11 Adjustment Costs and Labor Demand 127
3-12 Rosie the Riveter as an Instrumental
Variable 133
Theory at Work: California’s Overtime
Regulations and Labor Demand 104
Theory at Work: Minimum Wages and
Drunk Driving 125
Theory at Work: Work-Sharing in
Germany 132
Summary 139
Key Concepts 139
Review Questions 140
Problems 140
Selected Readings 143
Web Links 143
xv
xvi
Contents
Chapter 4
Labor Market Equilibrium
4-1
4-2
4-3
4-4
4-5
4-6
4-7
4-8
Selected Readings
Web Links 228
144
Equilibrium in a Single Competitive
Labor Market 145
Competitive Equilibrium across Labor
Markets 147
Policy Application: Payroll Taxes and
Subsidies 152
Policy Application: Payroll Taxes versus
Mandated Benefits 159
Policy Application: The Labor Market
Impact of Immigration 163
Policy Application: Environmental Disasters
and the Labor Market 177
The Cobweb Model 180
Noncompetitive Labor Markets:
Monopsony 183
Theory at Work: The Intifadah and Palestinian
Wages 146
Summary 190
Key Concepts 191
Review Questions 191
Problems 191
Selected Readings 194
Web Links 195
Chapter 5
Compensating Wage Differentials
5-1
5-2
5-3
5-4
5-5
5-6
Chapter 6
Human Capital
6-1
6-2
6-3
6-4
6-5
6-6
6-7
6-8
6-9
6-10
6-11
6-12
6-13
The Market for Risky Jobs 197
The Hedonic Wage Function 203
Policy Application: How Much Is a Life
Worth? 207
Policy Application: Safety and
Health Regulations 210
Compensating Differentials and Job
Amenities 213
Policy Application: Health Insurance
and the Labor Market 219
Summary 222
Key Concepts 223
Review Questions 223
Problems 223
229
Education in the Labor Market:
Some Stylized Facts 230
Present Value 232
The Schooling Model 232
Education and Earnings 239
Estimating the Rate of Return to
Schooling 244
Policy Application: School Construction
in Indonesia 247
Policy Application: School Quality
and Earnings 249
Do Workers Maximize Lifetime
Earnings? 254
Schooling as a Signal 257
Postschool Human Capital Investments 263
On-the-Job Training 264
On-the-Job Training and the Age-Earnings
Profile 268
Policy Application: Evaluating Government
Training Programs 273
Theory at Work: Destiny at Age 6 243
Theory at Work: Booker T. Washington and
Julius Rosenwald 250
Theory at Work: Is the GED Better
Than Nothing? 261
Theory at Work: Earnings and
Substance Abuse 272
196
Theory at Work: Life On the Interstate 210
Theory at Work: Jumpers in Japan 213
227
Summary 275
Key Concepts 276
Review Questions 276
Problems 277
Selected Readings 280
Web Links 281
Chapter 7
The Wage Structure
7-1
7-2
7-3
282
The Earnings Distribution 283
Measuring Inequality 285
The Wage Structure: Basic Facts
288
Contents
7-4
7-5
7-6
Policy Application: Why Did Wage
Inequality Increase? 291
The Earnings of Superstars 300
Inequality across Generations 303
Theory at Work: Computers, Pencils,
and the Wage Structure 297
Theory at Work: Rock Superstars 302
Theory at Work: Nature and Nurture 306
Summary 306
Key Concepts 307
Review Questions 307
Problems 307
Selected Readings 310
Web Links 311
Chapter 8
Labor Mobility
8-1
8-2
8-3
8-4
8-5
8-6
8-7
8-8
8-9
8-10
8-11
8-12
8-13
312
Geographic Migration as a Human
Capital Investment 313
Internal Migration in the United States 314
Family Migration 319
Immigration in the United States 322
Immigrant Performance in the
U.S. Labor Market 324
The Decision to Immigrate 330
The Economic Benefits from
Immigration 335
Policy Application: High-Skill Immigration
and Human Capital Externalities 338
Policy Application: Intergenerational
Mobility of Immigrants 342
Job Turnover: Facts 346
The Job Match 350
Specific Training and Job Turnover 352
Job Turnover and the Age-Earnings
Profile 353
Theory at Work: Power Couples 322
Theory at Work: Hey Dad, My Roommate
Is So Smart, I Got a 4.0 GPA 343
Theory at Work: Health Insurance
and Job-Lock 351
Summary 356
Key Concepts 357
Review Questions 357
Problems 358
Selected Readings
Web Links 361
xvii
361
Chapter 9
Labor Market Discrimination
362
9-1
9-2
9-3
9-4
9-5
9-6
9-7
Race and Gender in the Labor Market 363
The Discrimination Coefficient 365
Employer Discrimination 366
Employee Discrimination 373
Customer Discrimination 374
Statistical Discrimination 376
Experimental Evidence on
Discrimination 381
9-8 Measuring Discrimination 382
9-9 Policy Application: Determinants of the
Black–White Wage Ratio 387
9-10 Discrimination against Other Groups 395
9-11 Policy Application: Determinants of the
Female–Male Wage Ratio 398
Theory at Work: Beauty and the Beast 372
Theory at Work: Discrimination in the NBA 377
Theory at Work: Employment Discrimination
in China 383
Theory at Work: “Disparate Impact” and
Black Employment in Police Departments 390
Theory at Work: Shades of Black 394
Theory at Work: 9/11 and the Earnings of
Arabs and Muslims in the United States 397
Theory at Work: Orchestrating
Impartiality 401
Summary 405
Key Concepts 406
Review Questions 406
Problems 406
Selected Readings 410
Web Links 411
Chapter 10
Labor Unions
10-1
10-2
10-3
10-4
412
Unions: Background and Facts 413
Determinants of Union Membership 417
Monopoly Unions 423
Policy Application: Unions and Resource
Allocation 425
xviii Contents
10-5
10-6
10-7
10-8
10-9
Efficient Bargaining 427
Strikes 433
Union Wage Effects 439
Nonwage Effects of Unions 445
Policy Application: Public-Sector
Unions 448
Theory at Work: The Rise and Fall of
PATCO 422
Theory at Work: The Cost of Labor Disputes 436
Theory at Work: Occupational Licensing 444
Theory at Work: Do Teachers’ Unions Make
Students Better Off? 449
Theory at Work: Lawyers and Arbitration 450
Summary 451
Key Concepts 452
Review Questions 452
Problems 453
Selected Readings 457
Web Links 457
Chapter 11
Incentive Pay
458
11-1 Piece Rates and Time Rates 458
11-2 Tournaments 465
11-3 Policy Application: The Compensation
of Executives 472
11-4 Policy Application: Incentive Pay for
Teachers 474
11-5 Work Incentives and Delayed
Compensation 477
11-6 Efficiency Wages 480
Theory at Work: Windshields by
the Piece 463
Theory at Work: How Much is Your Soul
Worth? 466
Theory at Work: Incentive Pay Gets You
to LAX on Time 468
Theory at Work: Playing Hard for the
Money 471
Theory at Work: Are Men More
Competitive? 474
Theory at Work: Did Henry Ford
Pay Efficiency Wages? 484
Summary 489
Key Concepts 489
Review Questions 490
Problems 490
Selected Readings 493
Web Links 493
Chapter 12
Unemployment
12-1
12-2
12-3
12-4
12-5
12-6
12-7
12-8
12-9
12-10
12-11
494
Unemployment in the United States 495
Types of Unemployment 503
The Steady-State Rate of
Unemployment 504
Job Search 506
Policy Application: Unemployment
Compensation 513
The Intertemporal Substitution
Hypothesis 520
The Sectoral Shifts Hypothesis 521
Efficiency Wages and Unemployment 522
Implicit Contracts 526
Policy Application: The Phillips
Curve 528
Policy Application: The Unemployment
Gap between Europe and the United
States 532
Theory at Work: The Long-Term Effects
of Graduating in a Recession 501
Theory at Work: Jobs and Friends 507
Theory at Work: Cash Bonuses and
Unemployment 515
Theory at Work: The Benefits of UI 519
Summary 535
Key Concepts 536
Review Questions 536
Problems 537
Selected Readings 540
Web Links 540
Mathematical Appendix: Some Standard
Models in Labor Economics 541
Indexes
552
Name Index 552
Subject Index 560
1
Chapter
Introduction to Labor
Economics
Observations always involve theory.
—Edwin Hubble
Most of us will allocate a substantial fraction of our time to the labor market. How we do in
the labor market helps determine our wealth, the types of goods we can afford to consume,
with whom we associate, where we vacation, which schools our children attend, and even
the types of persons who find us attractive. As a result, we are all eager to learn how the
labor market works. Labor economics studies how labor markets work.
Our interest in labor markets arises not only from our personal involvement but also
because many social policy issues concern the labor market experiences of particular
groups of workers or various aspects of the employment relationship between workers and
firms. The policy issues examined by modern labor economics include
1. Why did the labor force participation of women rise steadily throughout the past
century in many industrialized countries?
2. What is the impact of immigration on the wage and employment opportunities of
native-born workers?
3. Do minimum wages increase the unemployment rate of less-skilled workers?
4. What is the impact of occupational safety and health regulations on employment and
earnings?
5. Are government subsidies of investments in human capital an effective way to improve
the economic well-being of disadvantaged workers?
6. Why did wage inequality in the United States rise so rapidly after 1980?
7. What is the impact of affirmative action programs on the earnings of women and
minorities and on the number of women and minorities that firms hire?
8. What is the economic impact of unions, both on their members and on the rest of the
economy?
1
2
Chapter 1
9. Do generous unemployment insurance benefits lengthen the duration of spells of
unemployment?
10. Why did the unemployment rate in the United States begin to approach the typically
higher unemployment rate of European countries after 2008?
This diverse list of questions clearly illustrates why the study of labor markets is intrinsically more important and more interesting than the study of the market for butter (unless
one happens to be in the butter business!). Labor economics helps us understand and
address many of the social and economic problems facing modern societies.
1-1 An Economic Story of the Labor Market
This book tells the “story” of how labor markets work. Telling this story involves much
more than simply recounting the history of labor law in the United States or in other countries and presenting reams of statistics summarizing conditions in the labor market. After
all, good stories have themes, characters that come alive with vivid personalities, conflicts
that have to be resolved, ground rules that limit the set of permissible actions, and events
that result inevitably from the interaction among characters.
The story we will tell about the labor market has all of these features. Labor economists
typically assign motives to the various “actors” in the labor market. We typically view
workers, for instance, as trying to find the best possible job and assume that firms are trying to make money. Workers and firms, therefore, enter the labor market with different
objectives—workers are trying to sell their labor at the highest price and firms are trying to
buy labor at the lowest price.
The types of economic exchanges that can occur between workers and firms are limited
by the set of ground rules that the government has imposed to regulate transactions in the
labor market. Changes in these rules and regulations would obviously lead to different
outcomes. For instance, a minimum wage law prohibits exchanges that pay less than a particular amount per hour worked; occupational safety regulations forbid firms from offering
working conditions that are deemed too risky to the worker’s health. The deals that are
eventually struck between workers and firms determine the types of jobs that are offered,
the skills that workers acquire, the extent of labor turnover, the structure of unemployment,
and the observed earnings distribution. The story thus provides a theory, a framework for
understanding, analyzing, and predicting a wide array of labor market outcomes.
The underlying philosophy of the book is that modern economics provides a useful
story of how the labor market works. The typical assumptions we make about the behavior
of workers and firms, and about the ground rules under which the labor market participants make their transactions, suggest outcomes often corroborated by the facts observed
in real-world labor markets. The study of labor economics, therefore, helps us understand
and predict why some labor market outcomes are more likely to be observed than others.
Our discussion is guided by the belief that learning the story of how labor markets work
is as important as knowing basic facts about the labor market. The study of facts without
theory is just as empty as the study of theory without facts. Without understanding how
labor markets work—that is, without having a theory of why workers and firms pursue
some employment relationships and avoid others—we would be hard-pressed to predict
the impact on the labor market of changes in government policies or changes in the demographic composition of the workforce.
Introduction to Labor Economics
3
A question often asked is which is more important—ideas or facts? The analysis
presented throughout this book stresses that “ideas about facts” are most important. We
do not study labor economics so that we can construct elegant mathematical theories
of the labor market, or so that we can remember how the official unemployment rate
is calculated and that the unemployment rate was 6.9 percent in 1993. Rather, we want
to understand which economic and social factors generate a certain level of unemployment, and why.
The main objective of this book is to survey the field of labor economics with an emphasis on both theory and facts: where the theory helps us understand how the facts are generated and where the facts can help shape our thinking about the way labor markets work.
1-2 The Actors in the Labor Market
Throughout the book, we will see that there are three leading actors in the labor market:
workers, firms, and the government.1
As workers, we receive top casting in the story. Without us, after all, there is no “labor”
in the labor market. We decide whether to work or not, how many hours to work, how
much effort to allocate to the job, which skills to acquire, when to quit a job, which occupations to enter, and whether to join a labor union. Each of these decisions is motivated by
the desire to optimize, to choose the best available option from the various choices. In our
story, therefore, workers will always act in ways that maximize their well-being. Adding
up the decisions of millions of workers generates the economy’s labor supply not only in
terms of the number of persons who enter the labor market but also in terms of the quantity
and quality of skills available to employers. As we will see many times throughout the
book, persons who want to maximize their well-being tend to supply more time and more
effort to those activities that have a higher payoff. The labor supply curve, therefore, is
often upward sloping, as illustrated in Figure 1-1.
The hypothetical labor supply curve drawn in the figure gives the number of engineers
that will be forthcoming at every wage. For example, 20,000 workers are willing to supply
their services to engineering firms if the engineering wage is $40,000 per year. If the engineering wage rises to $50,000, then 30,000 workers will choose to be engineers. In other
words, as the engineering wage rises, more persons will decide that the engineering profession is a worthwhile pursuit. More generally, the labor supply curve relates the number
of person-hours supplied to the economy to the wage that is being offered. The higher the
wage that is being offered, the larger the labor supplied.
Firms co-star in our story. Each firm must decide how many and which types of workers to hire and fire, the length of the workweek, how much capital to employ, and whether
to offer a safe or risky working environment to its workers. Like workers, firms in our
story also have motives. In particular, we will assume that firms want to maximize profits.
From the firm’s point of view, the consumer is king. The firm will maximize its profits by
making the production decisions—and hence the hiring and firing decisions—that best
1
In some countries, a fourth actor can be added to the story: trade unions. Unions may organize
a large fraction of the workforce and represent the interests of workers in their bargaining with
employers as well as influence political outcomes. In the United States, however, the trade union
movement has been in decline for several decades. By 2010, only 6.9 percent of private-sector
workers were union members.
4
Chapter 1
FIGURE 1-1 Supply and Demand in the Engineering Labor Market
The labor supply curve gives the number of persons who are willing to supply their services to engineering firms
at a given wage. The labor demand curve gives the number of engineers that the firms will hire at that wage.
Labor market equilibrium occurs where supply equals demand. In equilibrium, 20,000 engineers are hired at a
wage of $40,000.
Earnings ($)
Labor Supply
Curve
50,000
Equilibrium
40,000
Labor Demand
Curve
30,000
Employment
10,000
20,000
30,000
serve the consumers’ needs. In effect, the firm’s demand for labor is a derived demand,
a demand derived from the desires of consumers.
Adding up the hiring and firing decisions of millions of employers generates the
economy’s labor demand. The assumption that firms want to maximize profits implies
that firms will want to hire many workers when labor is cheap but will refrain from hiring
when labor is expensive. The relation between the price of labor and how many workers
firms are willing to hire is summarized by the downward-sloping labor demand curve
(also illustrated in Figure 1-1). As drawn, the labor demand curve tells us that firms in the
engineering industry want to hire 20,000 engineers when the wage is $40,000 but will hire
only 10,000 engineers if the wage rises to $50,000.
Workers and firms, therefore, enter the labor market with conflicting interests. Many
workers are willing to supply their services when the wage is high, but few firms are
willing to hire them. Conversely, few workers are willing to supply their services when
the wage is low, but many firms are looking for workers. As workers search for jobs and
firms search for workers, these conflicting desires are “balanced out” and the labor market
reaches an equilibrium. In a free-market economy, equilibrium is attained when supply
equals demand.
As drawn in Figure 1-1, the equilibrium wage is $40,000 and 20,000 engineers will be
hired in the labor market. This wage–employment combination is an equilibrium because
it balances out the conflicting desires of workers and firms. Suppose, for example, that
the engineering wage were $50,000—above equilibrium. Firms would then want to hire
only 10,000 engineers, even though 30,000 engineers are looking for work. The excess
number of job applicants would bid down the wage as they compete for the few jobs
available. Suppose, instead, that the wage were $30,000—below equilibrium. Because
Introduction to Labor Economics
5
engineers are cheap, firms want to hire 30,000 engineers, but only 10,000 engineers are
willing to work at that wage. As firms compete for the few available engineers, they bid
up the wage.
There is one last major player in the labor market, the government. The government
can tax the worker’s earnings, subsidize the training of engineers, impose a payroll tax on
firms, demand that engineering firms hire two black engineers for each white one hired,
enact legislation that makes some labor market transactions illegal (such as paying engineers less than $50,000 annually), and increase the supply of engineers by encouraging
their immigration from abroad. All these actions will change the equilibrium that will
eventually be attained in the labor market. Government regulations, therefore, help set the
ground rules that guide exchanges in the labor market.
The Trans-Alaska Oil Pipeline
In January 1968, oil was discovered in Prudhoe Bay in remote northern Alaska. The oil
reserves were estimated to be greater than 10 billion barrels, making it the largest such
discovery in North America.2
There was one problem with the discovery—the oil was located in a remote and frigid
area of Alaska, far from where most consumers lived. To solve the daunting problem of
transporting the oil to those consumers who wanted to buy it, the oil companies proposed
building a 48-inch pipeline across the 789-mile stretch from northern Alaska to the
southern (and ice-free) port of Valdez. At Valdez, the oil would be transferred to oil supertankers. These huge ships would then deliver the oil to consumers in the United States and
elsewhere.
The oil companies joined forces and formed the Alyeska Pipeline Project. The construction project began in the spring of 1974, after the U.S. Congress gave its approval in
the wake of the 1973 oil embargo. Construction work continued for three years and the
pipeline was completed in 1977. Alyeska employed about 25,000 workers during the summers of 1974 through 1977, and its subcontractors employed an additional 25,000 workers.
Once the pipeline was built, Alyeska reduced its pipeline-related employment to a small
maintenance crew.
Many of the workers employed by Alyeska and its subcontractors were engineers who
had built pipelines across the world. Very few of these engineers were resident Alaskans.
The remainder of the Alyeska workforce consisted of low-skill labor such as truck drivers
and excavators. Many of these low-skill workers were resident Alaskans.
The theoretical framework summarized by the supply and demand curves can help us
understand the shifts in the labor market that should have occurred in Alaska as a result
of the Trans-Alaska Pipeline System. As Figure 1-2 shows, the Alaskan labor market
was initially in an equilibrium represented by the intersection of the demand curve D0
and the supply curve S0. The labor demand curve tells us how many workers would
be hired in the Alaskan labor market at a particular wage, and the labor supply curve
tells us how many workers are willing to supply their services to the Alaskan labor
market at a particular wage. A total of E0 Alaskans were employed at a wage of w0 in the
initial equilibrium.
2
This discussion is based on the work of William J. Carrington, “The Alaskan Labor Market during the
Pipeline Era,” Journal of Political Economy 104 (February 1996): 186–218.
6
Chapter 1
FIGURE 1-2 The Alaskan Labor Market and the Construction of the Oil Pipeline
The construction of the oil pipeline shifted the labor demand curve in Alaska from D0 to D1, resulting in higher wages
and employment. Once the pipeline was completed, the demand curve reverted back to its original level and wages and
employment fell.
Earnings ($)
S0
w1
w0
D1
D0
E0
E1
Employment
The construction project clearly led to a sizable increase in the demand for labor. Figure 1-2
illustrates this shift by showing the demand curve moving outward from D0 to D1. The
outward shift in the demand curve implies that—at any given wage—Alaskan employers
were looking for more workers.
This theoretical framework immediately implies that the shift in demand moved the
Alaskan labor market to a new equilibrium, one represented by the intersection of the
new demand curve and the original supply curve. At this new equilibrium, a total of E1
persons were employed at a wage of w1. The theory, therefore, predicts that the pipeline
construction project would increase both employment and wages. As soon as the project
was completed, however, and the temporary need for construction workers disappeared,
the demand curve would have shifted back to its original position at D0. In the end, the
wage would have gone back down to w0 and E0 workers would be employed. In short, the
pipeline construction project should have led to a temporary increase in both wages and
employment during the construction period.
Figure 1-3 illustrates what actually happened to employment and earnings in Alaska
between 1968 and 1983. Because Alaska’s population grew steadily for some decades,
Alaskan employment also rose steadily even before the oil discovery in Prudhoe Bay. The
data clearly show, however, that employment “spiked” in 1975, 1976, and 1977 and then
went back to its long-run growth trend in 1977. The earnings of Alaskan workers also rose
substantially during the relevant period. After adjusting for inflation, the monthly earnings of
Alaskan workers rose from an average of $2,648 in the third quarter of 1973 to $4,140 in the
third quarter of 1976, an increase of 56 percent. By 1979, the real earnings of Alaskan workers were back to the level observed prior to the beginning of the pipeline construction project.
Introduction to Labor Economics
FIGURE 1-3
Wages and
Employment
in the Alaskan
Labor Market,
1968–1984
Source: William J.
Carrington, “The
Alaskan Labor Market
during the Pipeline
Era,” Journal of
Political Economy 104
(February 1996): 199.
Employment
7
Monthly Salary ($)
4,500
250,000
230,000
4,000
210,000
190,000
3,500
170,000
3,000
150,000
130,000
2,500
110,000
Wage
90,000
2,000
Employment
70,000
50,000
1968
1,500
1970
1972
1974
1976
1978
1980
1982
1984
It is worth noting that the temporary increase in earnings and employment occurred
because the supply curve of labor is upward sloping, so that an outward shift in the demand
curve moves the labor market to a point further up on the supply curve. As we noted earlier, an upward-sloping supply curve implies that more workers are willing to work when
the wage is higher. It turns out that the increase in labor supply experienced in the Alaskan
labor market occurred for two distinct reasons. First, a larger fraction of Alaskans were
willing to work when the wage increased. In the summer of 1973, about 39 percent of
Alaskans were working. In the summers of 1975 and 1976, about 50 percent of Alaskans
were working. Second, the rate of population growth in Alaska accelerated between 1974
and 1976—because persons living in the lower 48 states moved to Alaska to take advantage of the improved economic opportunities offered by the Alaskan labor market (despite
the frigid weather conditions there). The increase in the rate of population growth, however, was temporary. Population growth reverted back to its long-run trend soon after the
pipeline construction project was completed.
1-3 Why Do We Need a Theory?
We have just told a simple story of how the Trans-Alaska Pipeline System affected the
labor market outcomes experienced by workers in Alaska—and how each of the actors in
our story played a major role. The government approved the pipeline project despite the
potential environmental hazards involved; firms that saw income opportunities in building the pipeline increased their demand for labor; and workers responded to the change in
demand by increasing the quantity of labor supplied to the Alaskan labor market. We have,
in effect, constructed a simple theory or model of the Alaskan labor market. Our model is
8
Chapter 1
characterized by an upward-sloping labor supply curve, a downward-sloping labor demand
curve, and the assumption that an equilibrium is eventually attained that resolves the conflicts between workers and firms. As we have just seen, this model predicts that the construction of the oil pipeline would temporarily increase wages and employment in the Alaskan
labor market. Moreover, this prediction is testable—that is, the predictions about wages
and employment can be compared with what actually happened to wages and employment.
It turns out that the supply–demand model passes the test; the data are consistent with the
theoretical predictions.
Needless to say, the model of the labor market illustrated in Figure 1-2 does not do
full justice to the complexities of the Alaskan labor market. It is easy to come up with
many factors and variables that our simple model ignored and that could potentially
influence the success of our predictions. For instance, it is possible that workers care
about more than just the wage when they make labor supply decisions. The opportunity
to participate in such a challenging or cutting-edge project as the construction of the
Trans-Alaska Pipeline could have attracted engineers at wages lower than those offered
by firms engaged in more mundane projects—despite the harsh working conditions in
the field. The theoretical prediction that the construction of the pipeline project would
increase wages would then be incorrect because the project could have attracted more
workers at lower wages.
If the factors that we have omitted from our theory play a crucial role in understanding how the Alaskan labor market operates, we might be wrongly predicting that wages
and employment would rise. If these factors are only minor details, our model captures the essence of what goes on in the Alaskan labor market and our prediction would
be valid.
We could try to build a more complex model of the Alaskan labor market, a model
that incorporates every single one of these omitted factors. Now that would be a tough
job! A completely realistic model would have to describe how millions of workers and
firms interact and how these interactions work themselves through the labor market.
Even if we knew how to accomplish such a difficult task, this “everything-but-thekitchen-sink” approach would defeat the whole purpose of having a theory. A theory
that mirrored the real-world labor market in Alaska down to the most minute detail
might indeed be able to explain all the facts, but it would be as complex as reality
itself, cumbersome and incoherent, and thus would not at all help us understand how the
Alaskan labor market works.
There has been a long debate over whether a theory should be judged by the realism
of its assumptions or by the extent to which it finally helps us understand and predict the
labor market phenomena we are interested in. We obviously have a better shot at predicting
labor market outcomes if we use more realistic assumptions. At the same time, however,
a theory that mirrors the world too closely is too clumsy and does not isolate what really
matters. The “art” of labor economics lies in choosing which details are essential to the
story and which details are not. There is a trade-off between realism and simplicity, and
good economics hits the mark just right.
As we will see throughout this book, the supply–demand framework illustrated in Figure 1-1
often isolates the key factors that motivate the various actors in the labor market. The
model provides a useful way of organizing our thoughts about how the labor market works.
The model also gives a solid foundation for building more complex and more realistic
Introduction to Labor Economics
9
models of the labor market. And, most important, the model works. Its predictions are
often consistent with what is observed in the real world.
The supply–demand framework predicts that the construction of the Alaska oil pipeline
would have temporarily increased employment and wages in the Alaskan labor market.
This prediction is an example of positive economics. Positive economics addresses
the relatively narrow “What is?” questions, such as, What is the impact of the discovery of oil in Prudhoe Bay, and the subsequent construction of the oil pipeline, on the
Alaskan labor market? Positive economics, therefore, addresses questions that can, in
principle, be answered with the tools of economics, without interjecting any value judgment as to whether the particular outcome is desirable or harmful. Much of this book
is devoted to the analysis of such positive questions as, What is the impact of the minimum wage on unemployment? What is the impact of immigration on the earnings of
native-born workers? What is the impact of a tuition assistance program on college enrollment rates? What is the impact of unemployment insurance on the duration of a spell of
unemployment?
These positive questions, however, beg a number of important issues. In fact, some
would say that these positive questions beg the most important issues: Should the oil pipeline have been built? Should there be a minimum wage? Should the government subsidize
college tuition? Should the United States accept more immigrants? Should the unemployment insurance system be less generous?
These broader questions fall in the realm of normative economics, which addresses
much broader “What should be?” questions. By their nature, the answers to these normative questions require value judgments. Because each of us probably has different values,
our answers to these normative questions may differ regardless of what the theory or the
facts tell us about the economic impact of the oil pipeline, the disemployment effects of
the minimum wage, or the impact of immigration on the economic well-being of native
workers.
Normative questions force us to make value judgments about the type of society we
wish to live in. Consider, for instance, the impact of immigration on a particular host country. As we will see in subsequent chapters, the supply–demand framework implies that
an increase in the number of immigrants lowers the income of competing workers but
raises the income of the firms that hire the immigrants by even more. On net, therefore,
the receiving country gains. Moreover, because (in most cases) immigration is a voluntary
supply decision, it also makes the immigrants better off.
Suppose, in fact, that the evidence for a particular host country was completely consistent with the model’s predictions. In particular, the immigration of 10 million workers
improved the well-being of the immigrants (relative to their well-being in the source countries); reduced the income of native workers by, say, $25 billion annually; and increased
the incomes of capitalists by $40 billion. Let’s now ask a normative question: Should the
host country admit 10 million more immigrants?
This normative question cannot be answered solely on the basis of the theory or the
facts. Even though total income in the host country has increased by $15 billion, there also
has been a redistribution of wealth. Some persons are worse off and others are better off. To
answer the question of whether the country should continue to admit immigrants, one has
to decide whose economic welfare the country should care most about: that of immigrants,
who are made better off by immigration; that of native workers, who are made worse off;
10
Chapter 1
or that of the capitalists who own the firms, who are made better off. One might even bring
into the discussion the well-being of the people left behind in the source countries, who are
clearly affected by the emigration of their compatriots. It is clear that any policy discussion of
this issue requires clearly stated assumptions about what constitutes the “national interest,”
about who matters more. In the end, therefore, normative judgments about the costs and
benefits of immigration depend on our values and ideology.
Many economists often take a “fall-back” position when these types of problems are
encountered. Because the immigration of 10 million workers increases the total income
in the host country by $15 billion, it is possible to redistribute income in the postimmigration economy so that every person in that country is made better off. A policy that can
potentially improve the well-being of everyone in the economy is said to be “efficient”;
it increases the size of the economic pie available to the country. The problem, however,
is that this type of redistribution seldom occurs in the “real world”; the winners typically
remain winners and the losers remain losers. Our answer to a normative question, therefore,
will force each of us to confront the trade-off that we are willing to make between efficiency and distributional issues. In other words, normative questions force us to compare
the value that we attach to an increase in the size of the economic pie with the value that we
attach to a change in how the pie is split.
As a second example, we will see that the supply–demand framework predicts that
unionization transfers wealth from firms to workers, but that unionization also shrinks the
size of the economic pie. Suppose that the facts unambiguously support these theoretical implications: unions increase the total income of workers by, say, $40 billion, but the
country as a whole is poorer by $20 billion. Let’s now ask a normative question: Should the
government pursue policies that discourage workers from forming labor unions?
Again, our answer to this normative question depends on how we contrast the gains
accruing to the unionized workers with the losses accruing to the employers who
must pay higher wages and to the consumers who must pay higher prices for unionproduced goods.
The lesson from this discussion should be clear. As long as there are winners and losers—
and most government policies inevitably leave winners and losers in their wake—neither
the theoretical implications of economic models nor the facts are sufficient to answer
the normative question of whether a particular policy is desirable. Throughout this book,
therefore, we will find that economic analysis is very useful for framing and answering
positive questions but is much less useful for addressing normative questions.
Despite the fact that economists cannot answer what many would consider to be the
“big questions,” there is an important sense in which framing and answering positive
questions is crucial for any policy discussion. Positive economics tells us how particular government policies affect the well-being of different segments of society. Who
are the winners, and how much do they gain? Who are the losers, and how much do
they lose?
The adoption of a particular policy requires that these gains and losses be compared
and that some choice be made as to who matters more. In the end, any informed policy
discussion requires that we be fully aware of the price that has to be paid when making
particular choices. The normative conclusion that one might reach may well be affected by
the magnitude of the costs and benefits associated with the particular policy. For example,
Introduction to Labor Economics 11
the distributional impact of immigration (that is, redistributing income from workers to
firms) could easily dominate the normative discussion if immigration generated only a small
increase in the size of the economic pie. The distributional impact, however, would be less
relevant if it was clear that the size of the economic pie was greatly enlarged by immigration.
Summary
• Labor economics studies how labor markets work. Important topics addressed by labor
economics include the determination of the income distribution, the economic impact
of unions, the allocation of a worker’s time to the labor market, the hiring and firing
decisions of firms, labor market discrimination, the determinants of unemployment, and
the worker’s decision to invest in human capital.
• Models in labor economics typically contain three actors: workers, firms, and the
government. It is typically assumed that workers maximize their well-being and that firms
maximize profits. Governments influence the decisions of workers and firms by imposing
taxes, granting subsidies, and regulating the “rules of the game” in the labor market.
• A good theory of the labor market should have realistic assumptions, should not be
clumsy or overly complex, and should provide empirical implications that can be tested
with real-world data.
• The tools of economics are helpful for answering positive questions. The information
thus generated may help in making policy decisions. The answer to a normative question, however, typically requires that we impose a value judgment on the desirability of
particular economic outcomes.
1. What is labor economics? Which types of questions do labor economists analyze?
Review
2.
Who are the key actors in the labor market? What motives do economists typically
Questions
assign to workers and firms?
3. Why do we need a theory to understand real-world labor market problems?
4. What is the difference between positive and normative economics? Why are positive
questions easier to answer than normative questions?
Key
Concepts
derived demand, 4
equilibrium, 4
labor demand curve, 4
Web
Links
A number of websites publish data and research articles that are very valuable to labor
economists.
labor economics, 1
labor supply curve, 3
model, 7
normative economics, 9
positive economics, 9
The Bureau of Labor Statistics (BLS) is the government agency responsible for calculating
the monthly unemployment rate as well as the Consumer Price Index. Their website
12
Chapter 1
contains a lot of information on many aspects of the U.S. labor market, as well as
comparable international statistics: http://stats.bls.gov
The Bureau of Census reports detailed demographic and labor market information:
www.census.gov
The Statistical Abstract of the United States is an essential book that is available online.
It is published annually and contains detailed information on many aspects of the U.S.
economy: www.census.gov/prod/www/statistical_abstract.html
The Organization for Economic Cooperation and Development (OECD) reports statistics
on labor market conditions in many advanced economies: www.oecd.org
The National Bureau of Economic Research (NBER) publishes a working paper series
that represents the frontier of empirical research in economics. Their website also
contains a number of widely used data sets. The working papers and data can be accessed
and downloaded by students and faculty at many universities: www.nber.org
IZA is a Bonn-based research institute that conducts labor research. Their discussion
paper series provides up-to-date research on labor issues in many countries: www.iza.org
Appendix
An Introduction to Regression Analysis
Labor economics is an empirical science. It makes extensive use of econometrics, the
application of statistical techniques to study relationships in economic data. For example,
we will be addressing such questions as
1. Do higher levels of unemployment benefits lead to longer spells of unemployment?
2. Do higher levels of welfare benefits reduce work incentives?
3. Does going to school for one more year increase a worker’s earnings?
The answers to these three questions ultimately depend on a correlation between pairs
of variables: the level of unemployment compensation and the duration of unemployment
spells; the level of welfare benefits and the labor supply; and educational attainment and
wages. We also will want to know not only the sign of the correlation, but the size as well.
In other words, by how many weeks does a $50 increase in unemployment compensation
lengthen the duration of unemployment spells? By how many hours does an increase of
$200 per month in welfare benefits reduce the labor supply of workers? And by how much
our earnings increase if we get a college education?
Although this book does not use econometric analysis in much of the discussion, the
student can better appreciate both the usefulness and the limits of empirical research by
knowing how labor economists manipulate the available data to answer the questions we
are interested in. The main statistical technique used by labor economists is regression
analysis.
Introduction to Labor Economics 13
An Example
It is well known that there are sizable differences in wages across occupations. We are
interested in determining why some occupations pay more than others. One obvious factor
that determines the average wage in an occupation is the level of education of workers in
that occupation.
It is common in labor economics to conduct empirical studies of earnings by looking
at the logarithm of earnings, rather than the actual level of earnings. There are sound
theoretical and empirical reasons for this practice, one of which will be described
shortly. Suppose there is a linear equation relating the average log wage in an occupation (log w) to the mean years of schooling of workers in that occupation ( s). We write
this line as
log w = α + βs
(1-1)
The variable on the left-hand side—the average log wage in the occupation—is called the
dependent variable. The variable on the right-hand side—average years of schooling in
the occupation—is called the independent variable. The main objective of regression
analysis is to obtain numerical estimates of the coefficients α and β by using actual data on
the mean log wage and mean schooling in each occupation. It is useful, therefore, to spend
some time interpreting these regression coefficients.
Equation (1-1) traces out a line, with intercept α and slope β; this line is drawn in Figure 1-4.
As drawn, the regression line makes the sensible assumption that the slope β is positive, so wages are higher in occupations where the typical worker has more schooling.
FIGURE 1-4 The Regression Line
The regression line gives the relationship between the average log wage rate and the average years of schooling of workers
across occupations. The slope of the regression line gives the change in the log wage resulting from a one-year change in
years of schooling. The intercept gives the log wage for an occupation where workers have zero years of schooling.
Log Wage
Change in
Log Wage
Slope = β
α
Years of Schooling
Change in
Schooling
14
Chapter 1
The intercept α gives the log wage that would be observed in an occupation where
workers have zero years of schooling. Elementary algebra teaches us that the slope of a
line is given by the change in the vertical axis divided by the corresponding change in the
horizontal axis or
Change in log wage
β = ________________________
Change in years of schooling
(1-2)
Put differently, the slope β gives the change in the log wage associated with a one-year
increase in average schooling. It is a mathematical fact that a small change in the log wage
approximates the percent change in the wage. For example, if the difference in the mean
log wage between two occupations is 0.051, we can interpret this statistic as indicating that
there is approximately a 5.1 percent wage difference between the two occupations. This
property is one of the reasons why labor economists typically conduct studies of salaries
using the logarithm of the wage; they can then interpret changes in this quantity as a percent
change in the wage. This mathematical property of logarithms implies that the coefficient
β can be interpreted as giving the percent change in earnings resulting from a one-year
increase in schooling.
To estimate the parameters α and β, we first need to obtain data on the average log wage
and average years of schooling by occupation. These data can be easily calculated using
the Annual Social and Economic Supplement of the Current Population Surveys. These
data, collected in March of every year by the Bureau of Labor Statistics, contain a lot of
information about employment conditions and salaries for tens of thousands of workers.
One can use the data to compute the average log hourly wage and the average years of
schooling for men working in each of 45 different occupations. The resulting data are
reported in Table 1-1. To give an example, the typical man employed as an engineer had a
log wage of 3.37 and 15.8 years of schooling. In contrast, the typical man employed as a
construction laborer had a log wage of 2.44 and 10.5 years of schooling.
The plot of the data presented in Figure 1-5 is called a scatter diagram and describes
the relation found between the average log wage and the average years of schooling in the
real world. The relation between the two variables does not look anything like the regression line that we hypothesized. Instead, it is a scatter of points. Note, however, that the
points are not randomly scattered on the page, but instead have a noticeable upward-sloping
drift. The raw data, therefore, suggest a positive correlation between the log wage and
years of schooling, but nothing as simple as an upward-sloping line.
We have to recognize, however, that education is not the only factor that determines
the average wage in an occupation. There is probably a great deal of error when workers
report their salary to the Bureau of Labor Statistics. This measurement error disperses the
points on a scatter diagram away from the line that we believe represents the “true” data.
There also might be other factors that affect average earnings in any given occupation,
such as the average age of the workers or perhaps a variable indicating the “female-ness”
of the occupation. After all, it often is argued that jobs that are predominantly done by
men (for example, welders) tend to pay more than jobs that are predominantly done by
women (for example, kindergarten teachers). All of these extraneous factors would again
disperse our data points away from the line.
Introduction to Labor Economics 15
TABLE 1-1 Characteristics of Occupations, 2001
Source: Annual Demographic Files of the Current Population Survey, 2002.
Occupation
Administrators and officials, public administration
Other executives, administrators, and managers
Management-related occupations
Engineers
Mathematical and computer scientists
Natural scientists
Health diagnosing occupations
Health assessment and treating occupations
Teachers, college and university
Teachers, except college and university
Lawyers and judges
Other professional specialty occupations
Health technologists and technicians
Engineering and science technicians
Technicians, except health, engineering, and science
Supervisors and proprietors, sales occupations
Sales representatives, finance and business services
Sales representatives, commodities, except retail
Sales workers, retail and personal services
Sales-related occupations
Supervisors, administrative support
Computer equipment operators
Secretaries, stenographers, and typists
Financial records, processing occupations
Mail and message distributing
Other administrative support occupations, including clerical
Private household service occupations
Protective service occupations
Food service occupations
Health service occupations
Cleaning and building service occupations
Personal service occupations
Mechanics and repairers
Construction trades
Other precision production occupations
Machine operators and tenders, except precision
Fabricators, assemblers, inspectors, and samplers
Motor vehicle operators
Other transportation occupations and material moving
Construction laborer
Freight, stock, and material handlers
Other handlers, equipment cleaners, and laborers
Farm operators and managers
Farm workers and related occupations
Forestry and fishing occupations
Mean Log Hourly
Wage of Male
Workers
Mean Years of
Schooling for
Male Workers
Female
Share
(%)
3.24
3.29
3.16
3.37
3.36
3.22
3.91
3.23
3.17
2.92
3.72
2.90
2.76
2.97
3.30
2.96
3.39
3.14
2.61
2.93
2.94
2.91
2.75
2.67
2.87
2.66
2.46
2.80
2.23
2.38
2.37
2.55
2.81
2.74
2.82
2.62
2.65
2.59
2.68
2.44
2.44
2.42
2.52
2.29
2.70
15.7
14.9
15.4
15.8
15.6
17.4
19.8
16.2
18.8
16.5
19.7
15.9
14.2
13.8
15.4
13.9
15.1
14.4
13.4
14.8
13.8
13.8
13.8
14.2
13.2
13.4
10.6
13.6
11.4
13.2
11.2
13.4
12.6
11.9
12.3
11.8
12.0
12.1
11.8
10.5
12.0
11.3
12.9
9.9
12.0
52.4
42.0
59.4
10.7
32.2
34.2
31.2
86.2
44.7
75.8
29.3
54.0
83.1
26.0
48.5
37.6
44.7
25.4
64.0
72.4
61.2
57.1
98.0
92.9
41.9
79.2
96.0
18.7
60.0
89.1
48.2
80.4
5.2
2.4
22.5
35.2
36.2
12.7
6.3
3.9
30.4
28.0
20.5
18.5
3.7
16
Chapter 1
FIGURE 1-5 The Scatter Diagram Relating Wages and Schooling by Occupation, 2001
4
Log Wage
3.5
3
2.5
2
8
10
12
14
16
18
20
Years of Schooling
The objective of regression analysis is to find the best line that goes through the
scatter diagram. Figure 1-6 redraws our scatter diagram and inserts a few of the many
lines that we could draw through the scatter. Line A does not represent the general trend
very well; after all, the raw data suggest a positive correlation between wages and education, yet line A has a negative slope. Both lines B and C are upward sloping, but they
are both a bit “off”; line B lies above all of the points in the scatter diagram and line C
is too far to the right.
The regression line is the line that best summarizes the data.3 The formula that calculates the regression line is included in every statistics and spreadsheet software program. If
we apply the formula to the data in our example, we obtain the regression line
log w = 0.869 + 0.143s
(1-3)
This estimated regression line is superimposed on the scatter diagram in Figure 1-7.
We interpret the regression line reported in equation (1-3) as follows. The estimated
slope is positive, indicating that the average log wage is indeed higher in occupations
where workers are more educated. The 0.143 slope implies that each one-year increase
in the mean schooling of workers in an occupation raises the wage by approximately
14.3 percent.
3
More precisely, the regression line is the line that minimizes the sum of the square of the vertical
differences between every point in the scatter diagram and the corresponding point on the line.
As a result, this method of estimating the regression line is called least squares.
Introduction to Labor Economics 17
FIGURE 1-6 Choosing among Lines Summarizing the Trend in the Data
There are many lines that can be drawn through the scatter diagram. Lines A, B, and C provide three such examples.
None of these lines “fit” the trend in the scatter diagram very well.
4
B
Log Wage
3.5
A
3
C
2.5
2
8
10
12
14
16
18
20
Years of Schooling
The intercept indicates that the log wage would be 0.869 in an occupation where the
average worker had zero years of schooling. We have to be very careful when we use this
result. After all, as the raw data reported in Table 1-1 show, no occupation has a workforce
with zero years of schooling. In fact, the smallest value of s is 9.9 years. The intercept is
obtained by extrapolating the regression line to the left until it hits the vertical axis. In
other words, we are using the regression line to make an out-of-sample prediction. It is
easy to get absurd results when we do this type of extrapolation: After all, what does it
mean to say that the typical person in an occupation has no schooling whatsoever? An
equally silly extrapolation takes the regression line and extends it to the right until, say, we
wish to predict what would happen if the average worker had 25 years of schooling. Put
simply, it is problematic to predict outcomes that lie outside the range of the data.
“Margin of Error” and Statistical Significance
If we plug the data reported in Table 1-1 into a statistics or spreadsheet program, we
will find that the program reports many more numbers than just the intercept and the
slope of a regression line. The program also reports what are called standard errors,
or a measure of the statistical precision with which the coefficients are estimated. When
poll results are reported in the media, it is said, for instance, that 52 percent of the
18
Chapter 1
FIGURE 1-7 The Scatter Diagram and the Regression Line
4
Log Wage
3.5
3
2.5
2
8
10
12
14
16
18
20
Years of Schooling
population believes that tomatoes should be bigger and redder, with a margin of error of
±3 percent. We use standard errors to calculate the margin of error of our estimated
regression coefficients.
In our data, it turns out that the standard error for the intercept α is 0.172 and that the standard error for the slope β is 0.012. The margin of error that is used commonly in econometric
work is twice the standard error. The regression thus allows us to conclude that a one-year
increase in average schooling increases the log wage by 0.143 ± 0.024 (or twice the standard error of 0.012). In other words, our data suggest that a one-year increase in schooling
increases the average wage in an occupation by as little as 11.9 percent or by as much as 16.7
percent. Statistical theory indicates that the true impact of the one-year increase in schooling
lies within this range with a 95 percent probability. We have to allow for a margin of error
because our data are imperfect. Our data are measured with error, extraneous factors are being
omitted, and our data are typically based on a random sample of the population.
The regression program will also report a t statistic for each regression coefficient. The
t statistic helps us assess the statistical significance of the estimated coefficients.
The t statistic is defined as
Absolute value of regression coefficient
t statistic = _________________________________
Standard error of regression coefficient
(1-4)
If a regression coefficient has a t statistic above the “magic” number of 2, the regression
coefficient is said to be significantly different from zero. In other words, it is very likely
Introduction to Labor Economics 19
that the true value of the coefficient is not zero, so there is some correlation between the
two variables that we are interested in. If a t statistic is below 2, the coefficient is said to
be insignificantly different from zero, so we cannot conclude that there is a correlation
between the two variables of interest.
Note that the t statistic associated with our estimated slope is 11.9 (or 0.143 ÷ 0.012),
which is certainly above 2. Our estimate of the slope is significantly different from zero.
Therefore, it is extremely likely that there is indeed a positive correlation between the average log wage in an occupation and the average schooling of workers.
Finally, the statistical software program will typically report a number called the
R-squared. This statistic gives the fraction of the dispersion in the dependent variable that
is “explained” by the dispersion in the independent variable. The R-squared of the regression reported in equation (1-3) is 0.762. In other words, 76.2 percent of the variation in the
mean log wage across occupations can be attributed to differences in educational attainment across the occupations. Put differently, our very simple regression model seems to do
a very good job at explaining why engineers earn more than construction laborers—it is
largely because one group of workers has a lot more education than the other.
Multiple Regression
Up to this point, we have focused on a regression model that contains only one independent
variable, mean years of schooling. As noted above, the average log wage of men in an occupation will probably depend on many other factors. The simple correlation between wages
and schooling implied by the regression model in equation (1-3) could be confounding the
effect of some of these other variables. To isolate the relationship between the log wage and
schooling (and avoid what is called omitted variable bias), it is important to control for differences in other characteristics that also might generate wage differentials across occupations.
To provide a concrete example, suppose we believe that occupations that are predominantly held by men tend to pay more—for given schooling—than occupations that are
predominantly held by women. We can then write an expanded regression model as
log w = α + βs + γp
(1-5)
where the variable p gives the percent of workers in an occupation that are women. As
before, log w and s give the log wage and mean years of schooling of men working in that
occupation.
We now wish to interpret the coefficients in this multiple regression model—a regression that contains more than one independent variable. Each coefficient in the multiple
regression measures the impact of a particular variable on the log wage, other things being
equal. For instance, the coefficient β gives the change in the log wage resulting from a
one-year increase in mean schooling, holding constant the relative number of women in the
occupation. Similarly, the coefficient γ gives the change in the log wage resulting from a
one-percentage-point increase in the share of female workers, holding constant the average
schooling of the occupation. Finally, the intercept α gives the log wage in a fictional occupation that employs only men and where the typical worker has zero years of schooling.
The last column in Table 1-1 reports the values of the female share p for the various
occupations in our sample. It is evident that the representation of women varies significantly across occupations: 75.8 percent of teachers below the university level are women,
20
Chapter 1
as compared to only 5.2 percent of mechanics and repairers. Because we now have two
independent variables, our scatter diagram is three dimensional. The regression “line,”
however, is now the plane that best fits the data in this three-dimensional space. If we plug
these data into a computer program to estimate the regression model in equation (1-5), the
estimated regression line is given by
log w = 0.924 + 0.150s - 0.003p
(0.154) (0.011) (0.001)
R-squared = 0.816
(1-6)
where the standard error of each of the coefficients is reported in parentheses below the
coefficient.
Note that a one-year increase in the occupation’s mean schooling raises weekly earnings by approximately 15.0 percent. In other words, if we compare two occupations that
have the same female share but differ in years of schooling by one year, workers in the
high-skill occupation earn 15 percent more than workers in the low-skill occupation.
Equally important, we find that the percent female in the occupation has a statistically
significant negative impact on the log wage. In other words, men who work in predominantly
female occupations earn less than men who work in predominantly male occupations—
even if both occupations have the same mean schooling. The regression coefficient, in fact,
implies that a 10-percentage-point increase in the female share lowers the average earnings
of an occupation by 3.0 percent.
Of course, before we make the tempting inference that this empirical finding is proof
of a “crowding effect”—the hypothesis that discriminatory behavior crowds women into
relatively few occupations and lowers wages in those jobs—we need to realize that there are
many other factors that determine occupational earnings. The multiple regression model
can, of course, be expanded to incorporate many more independent variables. As we will
see throughout this book, labor economists put a lot of effort into defining and estimating
regression models that isolate the correlation between the two variables of interest after
controlling for all other relevant factors. Regardless of how many independent variables
are included in the regression, however, all the regression models are estimated in essentially the same way: The regression line best summarizes the trends in the underlying data.
Key
Concepts
dependent variable, 13
econometrics, 12
independent variable, 13
multiple regression, 19
regression analysis, 12
regression coefficients, 13
regression line, 16
R-squared, 19
scatter diagram, 14
statistical significance, 18
standard errors, 17
t statistic, 18
2
Chapter
Labor Supply
It’s true hard work never killed anybody, but I figure, why take the chance?
—Ronald Reagan
Each of us must decide whether to work and, once employed, how many hours to work.
At any point in time, the economywide labor supply is given by adding the work choices
made by each person in the population. Total labor supply also depends on the fertility
decisions made by earlier generations (which determine the size of the current population).
The economic and social consequences of these decisions vary dramatically over time.
In 1948, 84 percent of American men and 31 percent of American women aged 16 or over
worked. By 2012, the proportion of working men had declined to 64 percent, whereas the
proportion of working women had risen to 53 percent. Over the same period, the length
of the average workweek in a private-sector production job fell from 40 to 34 hours.1
These labor supply trends have surely altered the nature of the American family as well as
greatly affected the economy’s productive capacity.
This chapter develops the framework that economists use to study labor supply decisions. In this framework, individuals seek to maximize their well-being by consuming
goods (such as fancy cars and nice homes) and leisure. Goods have to be purchased in the
marketplace. Because most of us are not independently wealthy, we must work in order to
earn the cash required to buy the desired goods. The economic trade-off is clear: If we do
not work, we can consume a lot of leisure, but we have to do without the goods and services that make life more enjoyable. If we do work, we will be able to afford many of these
goods and services, but we must give up some of our valuable leisure time.
The model of labor-leisure choice isolates the person’s wage rate and income as the
key economic variables that guide the allocation of time between the labor market and leisure activities. In this chapter, we first use the framework to analyze “static” labor supply
1
These statistics were obtained from the U.S. Bureau of Labor Statistics website:
www.bls.gov/data/home.htm.
21
22
Chapter 2
decisions, the decisions that affect a person’s labor supply at a point in time. We will also
extend the basic model to explore how the timing of leisure activities changes over the
life cycle.
This economic framework not only helps us understand why women’s work propensities rose and hours of work declined, but also allows us to address a number of questions
with important policy and social consequences. For example, do welfare programs reduce
incentives to work? Does a cut in the income tax rate increase hours of work? And what
factors explain the rapid growth in the number of women who choose to participate in the
labor market?
2-1
Measuring the Labor Force
On the first Friday of every month, the Bureau of Labor Statistics (BLS) releases its estimate of the unemployment rate for the previous month. The unemployment rate statistic is
widely regarded as a measure of the overall health of the U.S. economy. In fact, the media
often interpret the minor month-to-month blips in the unemployment rate as a sign of
either a precipitous decline in economic activity or a surging recovery.
The unemployment rate is tabulated from the responses to a monthly BLS survey called
the Current Population Survey (CPS). In this survey, nearly 60,000 households are questioned about their work activities during a particular week of the month (that week is
called the reference week). Almost everything we know about the trends in the U.S. labor
force comes from tabulations of CPS data. The survey instrument used by the CPS also
has influenced the development of surveys in many other countries. In view of the importance of this survey in the calculation of labor force statistics both in the United States
and abroad, it is useful to review the various definitions of labor force activities that are
routinely used by the BLS to generate its statistics.
The CPS classifies all persons aged 16 or older into one of three categories: the
employed, the unemployed, and the residual group that is said to be out of the labor force.
To be employed, a worker must have been at a job with pay for at least 1 hour or worked
at least 15 hours on a nonpaid job (such as the family farm). To be unemployed, a worker
must either be on a temporary layoff from a job or have no job but be actively looking for
work in the four-week period prior to the reference week.
Let E be the number of persons employed and U the number of persons unemployed.
A person participates in the labor force if he or she is either employed or unemployed.
The size of the labor force (LF) is given by
LF = E + U
(2-1)
Note that the vast majority of employed persons (those who work at a job with pay) are
counted as being in the labor force regardless of how many hours they work. The size of
the labor force, therefore, does not say anything about the “intensity” of work.
The labor force participation rate gives the fraction of the population (P) that is in
the labor force and is defined by
LF
Labor force participation rate = ___
P
(2-2)
Labor Supply 23
The employment rate (also called the “employment–population ratio”) gives the
fraction of the population that is employed, or
E
Employment rate = __
P
(2-3)
Finally, the unemployment rate gives the fraction of labor force participants who
are unemployed:
U
Unemployment rate = ___
(2-4)
LF
The Hidden Unemployed
The BLS calculates an unemployment rate based on a subjective measure of what it means
to be unemployed. To be considered unemployed, a pers...
Purchase answer to see full
attachment