THE MICROECONOMICS
OF INCOME DISTRIBUTION
DYNAMICS
IN EAST ASIA AND LATIN AMERICA
François Bourguignon
Francisco H. G. Ferreira
Nora Lustig
Editors
THE MICROECONOMICS OF
INCOME DISTRIBUTION
DYNAMICS IN EAST ASIA
AND LATIN AMERICA
class="bi x0 ya w1 h5"
THE MICROECONOMICS OF
INCOME DISTRIBUTION
DYNAMICS IN EAST ASIA
AND LATIN AMERICA
François Bourguignon
Francisco H. G. Ferreira
Nora Lustig
Editors
A copublication of the World Bank and Oxford University Press
© 2005 The International Bank for Reconstruction and Development / The World Bank
1818 H Street, NW
Washington, DC 20433
Telephone: 202-473-1000
Internet: www.worldbank.org
E-mail:
All rights reserved.
First printing September 2004
1 2 3 4 08 07 06 05

François Bourguignon, Francisco H. G. Ferreira,
and Nora Lustig
2 Decomposing Changes in the Distribution of
Household Incomes: Methodological Aspects 17
François Bourguignon and Francisco H. G. Ferreira
3 Characterization of Inequality Changes through
Microeconometric Decompositions: The Case of
Greater Buenos Aires 47
Leonardo Gasparini, Mariana Marchionni, and
Walter Sosa Escudero
4 The Slippery Slope: Explaining the Increase in
Extreme Poverty in Urban Brazil, 1976–96 83
Francisco H. G. Ferreira and Ricardo Paes de Barros
5 The Reversal of Inequality Trends in Colombia,
1978–95: A Combination of Persistent and
Fluctuating Forces 125
Carlos Eduardo Vélez, José Leibovich, Adriana
Kugler, César Bouillón, and Jairo Núñez
6 The Evolution of Income Distribution during
Indonesia’s Fast Growth, 1980–96 175
Vivi Alatas and François Bourguignon
7 The Microeconomics of Changing Income
Distribution in Malaysia 219
Gary S. Fields and Sergei Soares
v
8 Can Education Explain Changes in Income
Inequality in Mexico? 275
Arianna Legovini, César Bouillón, and Nora Lustig
9 Distribution, Development, and Education in Taiwan,
China, 1979–94 313

4.11 Shift in the Distribution of Education, 1976–96 111
4.12 Education Endowment and Demographic Effects 112
4.13 A Complete Decomposition 113
5.1 Average Household Size by Income Decile in
Urban Colombia, Selected Years 135
vi
CONTENTS
5.2 Change in Income from Changes of Returns to
Education, Relative to Workers Who Have
Completed Secondary Education: Male and
Female Wage Earners in Urban Colombia,
Selected Periods 140
5.3 Change in Income from Changes of Returns to
Education, Relative to Workers Who Have
Completed Secondary Education: Male and
Female Self-Employed Workers in Urban
Colombia, Selected Periods 141
5.4 Probability of Being Employed or a Wage Earner
in Urban Colombia according to Various
Individual or Household Characteristics, Various
Groups of Household Members, Selected Years 146
5.5 Simulated Occupational-Choice and Participation
Changes in Percentage Points by Percentile of
Earnings for Urban Males and Females, 1978–88 154
5.6 Simulated Occupational-Choice and Participation
Changes in Percentage Points by Percentile of
Earnings for Urban Males and Females, 1988–95 156
5.7 Changes in Employment Rate by Income Percentile,
Females in Urban Colombia, Selected Periods 157
6.1 Summary Decomposition of Changes in the

8.4 Effect of Labor Choices on Earnings by Percentile
in Mexico, 1984–94 298
8.5 Effect of Educational Gains on Earnings by
Percentile in Mexico, 1984–94 300
8.6 Effect of Changes in Returns to Education on
Earnings by Percentile in Mexico, 1984–94 301
8.7 Effect of Urban-Rural Disparities on Earnings
by Percentile in Mexico, 1984–94 303
9.1 Evolution of Income Inequality, 1979–94 318
9.2 Elasticity of Spouses’ Occupational Choice with
Respect to Head of Household’s Earnings 332
9.3 1979–94 Variation in Individual Earnings Caused
by the Price Effect, by Centiles of the 1979 Earnings
Distribution 336
9.4 Simulated Entries into and Exits from the Wage
Labor Force 337
9.5 Simulation of the 1994 Education Structure on
the 1979 Population 340
9.6 1979–94 Variation in Household Income Caused by
the Price Effect, by Centiles of the 1979 Distribution of
Equivalized Household Income Per Capita (EHIP) 344
9.7 Entries into and Exits from the Labor Force: Overall
Participation Effect 345
9.8 Effects of Imposing the 1994 Education Structure
on the 1979 Population 348
9.9 Effects of Imposing the 1994 Children Structure on the
1979 Population: Relative Variation by Centile of the
1979 Distribution of Equivalized Household Income 351
Tables
1.1 Selected Indicators of Long-Run Structural Evolution 3

4.1 General Economic Indicators for Brazil, Selected
Years 86
4.2 Basic Distributional Statistics for Different Degrees
of Household Economies of Scale 91
4.3 Stochastic Dominance Results 93
4.4 Educational and Labor-Force Participation Statistics,
by Gender and Race 94
4.5 Equation 4.2: Wage Earnings Regression for
Wage Employees 99
4.6 Equation 4.3: Total Earnings Regression for the
Self-Employed 101
4.7 Simulated Poverty and Inequality for 1976, Using
1996 Coefficients 104
4A.1 Real GDP and GDP Per Capita in Brazil, 1976–1996 115
4A.2 PNAD Sample Sizes and Missing or Zero Income
Proportions 116
4A.3 A Brazilian Spatial Price Index 117
4A.4 Brazilian Temporal Price Deflators, Selected Years 118
4A.5 Ratios of GDP Per Capita to PNAD Mean
Household Incomes, 1976–96 118
4B.1 Evolution of Mean Income and Inequality:
A Summary of the Literature 119
CONTENTS
ix
5.1 Decomposition of Total Inequality between Rural
and Urban Areas, Selected Years 129
5.2 Labor-Market Indicators in Urban and Rural
Areas, Selected Years 132
5.3 Changes in Sociodemographic Characteristics
in Urban and Rural Areas, Selected Years 134

Household Income Per Capita 198
6.9 Mean and Dispersion of Household Incomes
according to Some Characteristics of Heads of
Households 200
6.10 Occupational-Choice Behavior, 1980–96 202
6.11 Simulated Changes in Occupational Choices,
Whole Population 205
x
CONTENTS
6.12 Simulated Changes in Occupational Choices, Rural
and Urban Population 206
7.1 Location of Actual Distribution of Per Capita
Household Income, 1984 and 1989, 1989 and 1997 223
7.2 Inequality of Actual Distribution of Per Capita
Household Income, Selected Periods 226
7.3 Occupational-Position Equations for Male Heads
of Household 231
7.4 Occupational-Position Equations for Female Heads
of Household 233
7.5 Occupational-Position Equations for Male Family
Members Who Are Not Heads of Household 235
7.6 Occupational-Position Equations for Female Family
Members Who Are Not Heads of Household 237
7.7 Earnings Functions for Male Wage Earners 240
7.8 Earnings Functions for Female Wage Earners 242
7.9 Earnings Functions for Male Self-Employed Workers 244
7.10 Earnings Functions for Female Self-Employed Workers 246
7.11 Distribution of Per Capita Household Income,
Substituting 1989 Values into 1984 Distribution 259
7.12 Distribution of Per Capita Household Income,

1993–94 335
10.1 A Summary of the Decomposition Results 359
10.2 Interpreting the Decompositions: A Schematic
Summary 381
xii
CONTENTS
Preface
The process of economic development is inherently about change.
Change in where people live, in what they produce and in how they
produce it, in how much education they get, in how long and in
how well they live, in how many children they have, and so on. So
much change, and the fact that at times it takes place at such sur-
prising speed, must affect the way incomes and wealth are distrib-
uted, as well as the overall size of the pie. While considerable efforts
have been devoted to the understanding of economic growth, the
economic analysis of the mechanisms through which growth and
development affect the distribution of welfare has been rudimentary
by comparison. Yet understanding development and the process of
poverty reduction requires understanding not only how total income
grows within a country but also how its distribution behaves over
time.
Our knowledge of the dynamics of income distribution is
presently limited, in part because of the informational inefficiency
of the scalar inequality measures generally used to summarize dis-
tributions. Single numbers can often hide as much as they show. But
recent improvements in the availability of household survey data for
developing countries, and in the capacity of computers to process
them, mean that we should be able to do a better job comprehend-
ing the nature of changes in the income distribution that accompany
the process of economic development. We hope that this book is a

are grateful to the many people in both institutions who supported
it throughout its five-year lifespan. We would like to thank particu-
larly Michael Walton, who supported the birth of the project when
he directed the Poverty Reduction Unit at the World Bank, as well
as Carlos Jarque and Carlos Eduardo Vélez of the IDB, who sup-
ported the project’s completion.
We are also very grateful to Martin Ravallion, who commented
on various versions of the work, from research proposal to finished
papers; to James Heckman, who acted as a discussant for three
chapters at a session in the 2000 Meetings of the American Eco-
nomic Association; to Ravi Kanbur, who provided very useful sug-
gestions at an early stage of the research process; and to Tony
Shorrocks, who gave us many insights into the nature of the decom-
positions we undertook. We are similarly indebted to a number of
participants in seminars and workshops that took place at various
meetings of the Econometric Society (in particular in Latin America
and the Far East); of the European Economic Association (in
Venice); of the Network on Inequality and Poverty of the IDB, World
Bank, and LACEA (Latin American and Caribbean Economic
Association); and at the Universities of Brasília, Maryland, and
Michigan, The Catholic University of Rio de Janeiro, the European
University Institute in Florence, and DELTA (Département et
Laboratoire d’Economie Théorique et Appliquée) in Paris.
Our greatest debt, of course, is to the authors of the seven case
studies, who really wrote the book. Their names and affiliations are
listed separately in the coming pages, and we thank them profoundly
for their commitment and endurance during the long process of pro-
ducing this volume. Finally, the book would not have been possible
without the dedication, professionalism, and attention to detail of
Janet Sasser and her team at the World Bank’s Office of the Publisher.

France
Leonardo Gasparini Director of CEDLAS, as well as pro-
fessor of economics of income distrib-
ution and professor of labor econom-
ics at the Universidad Nacional de La
Plata, Argentina
xvii
Marc Gurgand Researcher at the Département et
Laboratoire d’Economie Théorique et
Appliquée (DELTA) at the Centre
National de la Recherche Scientifique
(CNRS), Paris, France
Adriana Kugler Associate professor of economics at
the Universitat Pompeu Fabra,
Barcelona, Spain, and assistant pro-
fessor of economics at the University
of Houston, Texas
Arianna Legovini Senior monitoring and evaluation
specialist in the Africa Region at the
World Bank, Washington, D.C.
José Leibovich Assistant director of the Departamento
Nacional de Planeación (Department
of National Planning), Bogotá,
Colombia
Nora Lustig President of the Universidad de Las
Americas, Puebla, Mexico
Mariana Marchionni Professor of econometrics at the
Universidad Nacional de La Plata,
Argentina, and researcher at CEDLAS
Jairo Núñez Researcher at the Universidad de los

(Brazilian Geographical and Statistical Institute)
ICV-DIEESE Índice do Custo de Vida–Departamento
Intersindical de Estatística e Estudos Sócio-
Econômeios (Cost of Living Index–Inter Trade
Union Department of Statistics and Socioeco-
nomic Studies, Brazil)
IGP-DI Índice Geral de Preços–Disponibilidade Interna
(General Price Index, Brazil)
INEGI Instituto Nacional de Estadística, Geografia y
Informática (National Institute of Statistics,
Geography, and Informatics, Mexico)
INPC-R Índice Nacional de Preços ao Consumidor–Real
(National Consumer Price Index, Brazil)
MIDD Microeconomics of Income Distribution
Dynamics
xix
OLS Ordinary least squares
PNAD Pesquisa Nacional por Amostra de Domicílios
(National Household Survey, Brazil)
Progresa Programa de Educación, Salud y Alimentación
(Program for Education, Heath, and Nutrition,
Mexico)
xx
ABBREVIATIONS AND ACRONYMS
1
Introduction
François Bourguignon,
Francisco H. G. Ferreira, and Nora Lustig
This book is about how the distribution of income changes during
the process of economic development. By its very nature, the process

cial struggle in modern economies was that between the rival forces
of (a) technological progress—ever raising the demand for (and the
pay of) more educated workers—and (b) educational expansion—
ever raising the supply of such workers. More recently, economists
have developed models with multiple equilibria, each characterized
by its own income distribution, with its own mean and its own level
of inequality.
1
These models show that different combinations of
initial conditions—and of the historical processes that might follow
them—could lead to diverse outcomes.
In this book, we do not suggest yet another grand theory of the
dynamics of income distribution during the process of development.
Instead, we propose and apply a methodology to decompose distri-
butional change into its various driving forces, with the aim of
enhancing our ability to understand the nature of income distribu-
tion dynamics.
2
In fact, rather than searching for a unifying expla-
nation, we explore the incredible diversity in the distributional
experiences and outcomes across economies. Why do changes in
inequality differ so markedly across economies that have similar
rates of growth in gross domestic product (GDP) per capita, such as
Colombia and Malaysia (see table 1.1)? Why do we observe rising
inequality both in growing economies (Mexico) and in contracting
ones (Argentina)? Why do educational expansions sometimes lead
to greater equality (as in Brazil and Taiwan, China) and sometimes
to greater inequality (as in Indonesia and Mexico)?
The microeconomic empirics reported in this volume suggest that
this diversity in outcomes results from the various possibilities that

b
(percent) −1.0 0.2 3.8 5.1 5.2 1.1 5.7
Average years of schooling
Initial year 8.7
c
3.2 4.6 3.8 7.9
c
5.6 6.0
Terminal year 9.8
c
5.3 6.9 6 8.3
c
6.9 9.5
Urbanization rate (percent)
Initial year 86 68 57 23 42 63 70
Terminal year 88 77 61 35 55 58 84
Participation of women in the
labor force (percent)
Initial year 45 28 27 32 60 33 46
Terminal year 56 42 41 48 58 41 50
Family size
Initial year 4.4 4.6 5.4 5.0 4.9 5.3 4.9
Terminal year 4.4 3.6 4.3 4.4 4.4 4.9 4.2
Gini coefficient (household
income per capita, size-
weighted households)
Initial year 0.417 0.595
d
0.502
d

change) are likely to induce some response from households in terms
of the desired level of education for their children. Like all of its rel-
atives in the Oaxaca-Blinder class of decompositions, the technique
discussed in this volume is not designed to model those general equi-
librium effects. It simply separates out how much of a given change
would not have been observed under a well-defined statistical coun-
terfactual (for example, if returns to education had not changed),
without making any statement about the economic foundations
of that counterfactual (for example, the conditions under which
no change in the returns to education would be consistent with
the other observed changes, in an economic sense). Nevertheless, as
we hope the case studies in chapters 3 through 9 will show, the
insights gained from the statistical decomposition and some basic
microeconomic intuition allow analysts to improve their under-
standing of the nature of changes in income distribution in a partic-
ular economy.
The microeconometric approach applied in this volume should
be seen as complementary to the more prevalent macroeconometric
(cross-country) studies of the relationship between growth and
inequality (or the reverse). (See, for instance, Alesina and Rodrik
1994; Dollar and Kraay 2002; Forbes 2000.) Cross-country regres-
sions can, if well specified and run on comparable data, tell us much
about average relationships between measures of income dispersion
4
BOURGUIGNON
,
FERREIRA
,
AND LUSTIG