This expanded edition includes four new chapters, covering recent ideas about human capital, fertility and economic growth, the division of labor, economic considerations within the family, and inequality in earnings.
"Critics have charged that Mr. Becker's style of thinking reduces humans to economic entities. Nothing could be further from the truth. Mr. Becker gives people credit for having the power to reason and seek out their own best destiny."—Wall Street Journal
|Publisher:||University of Chicago Press|
|Sold by:||Barnes & Noble|
|File size:||8 MB|
About the Author
Read an Excerpt
A Theoretical and Empirical Analysis, with Special Reference to Education
By Gary S. Becker
The University of Chicago PressCopyright © 1993 The National Bureau of Economic Research
All rights reserved.
Introduction to the Second Edition
In the preface to the first edition, written about a decade ago, I remarked that in the preceding few years "interest in the economics of education has mushroomed throughout the world." The mushrooming has continued unabated; a bibliography on the economics of education prepared in 1957 would have contained less than 50 entries, whereas one issued in 1964 listed almost 450 entries and its second edition in 1970 listed over 1300 entries. Moreover, this bibliography excludes the economic literature on health, migration, and other nonschooling investments in human capital, which has expanded even faster.
This sustained interest in human capital and the continuing attention shown to the first edition of this book has encouraged me to issue a second edition. Nothing in the first edition has been changed; even the errors remain, conspicuous as they are to me now. I have, however, incorporated three additional papers written within the first few years after the publication of the first edition. One of these three additions has not previously been published and another has not been readily available
Chapter II developed an analysis of postschool investment and used it to explain age-earnings profiles and to interpret data on earnings per hour. That chapter also introduced a distinction between specific and general training to explain the relation between job skills and labor turnover, and the "hoarding" of labor during cyclical swings in business. These concepts have spawned a large and important literature that has successfully explained many aspects of the labor market in the United States and elsewhere.
Chapter III introduced an analysis of the accumulation of human capital over the life cycle to explain, among other things, the shape of age-earnings profiles, the concentration of investments at earlier ages, and the personal distribution of earnings. This chapter also helped stimulate a large and empirically relevant literature.
The personal distribution of earnings is partly determined by the distribution of, and the returns from, human capital. Mincer is responsible for the pioneering analysis that relates the distribution of earnings to human capital. Section 3 of Chapter III extended his analysis by relating the distribution of earnings explicitly to rates of return and investment costs.
The additional material added in the second edition includes a portion of a paper, written jointly with Barry R. Chiswick, which provides a convenient formulation for statistical estimation of the relation between the log of earnings, rates of return to human capital, and the time spent investing in human capital. Regression equations derived from this formulation are developed to estimate the contribution of schooling to earnings inequality in the United States, especially its contribution to the difference in earnings inequality between the South and the North. This line of empirical analysis has more recently been extended to include postschool investment in a major study by Mincer, and in other studies as well.
In the first edition, although Chapter III assumed that individuals maximize their well-being as they accumulate human capital over their lifetime, no explicit model of utility or wealth maximization was developed. Therefore, the factors determining the distribution of investments at different ages were not explicitly analyzed. In my Woytinsky Lecture, published in 1967 and reprinted here as an addendum to Chapter III (see p. 94), a model of wealth maximization is developed that explains the distribution of investments, in particular the decline in investments over time, by (a) the decline in benefits from additional capital as fewer years of life remain, and (b) the rise in investment costs because foregone earnings rise as human capital is accumulated.
Here the analysis goes behind the distribution of human capital and rates of return and examines the underlying distribution of opportunities and abilities. Since the observed distribution of earnings results from the interaction of these underlying distributions, the relative importance of opportunities and abilities is not easily "identified," although some tests are suggested. I have added a supplement to this discussion of "identifiability" that is motivated by many recent attempts to assess the independent effect of family background on earnings. It shows why these attempts understate the effect of background, and overstate the effect of human capital, on earnings, perhaps by substantial amounts.
The Woytinsky lecture also analyzes the effects on inequality and skewness in earnings of more equal opportunity, minimum schooling legislation, and "objective" selection of applicants to scarce places in schools. In it I attempt to explain, too, why earnings are more equally distributed and less skewed than incomes from nonhuman capital. Although the formulation has some unsolved analytical difficulties, I believe that this paper opens up a promising line of investigation that has received insufficient attention.
The models of capital accumulation in the lecture—and in Ben-Porath's paper and several subsequent ones—have several limitations. Since the total hours supplied to the market sector are taken as given, these models do not consider the interaction between changes in wage rates over the life cycle resulting from the accumulation of human capital and the optimal allocation of time between the market and nonmarket sectors. Moreover, human capital is assumed to affect only earnings and the production of additional human capital, and to have no direct effect on utility or consumption.
These and some other restrictions are relaxed in the final essay added to this second edition. This paper, which I wrote and circulated in 1967 but never published, builds on the new approach to household behavior. In this approach, households produce the commodities that enter their utility functions by combining market-purchased goods and services, their own time, and human capital and other environmental variables. With this approach I consider the uses of an individual's time at different ages; in particular I focus on the allocation of time to three activities: the production of nonmarket commodities (nonmarket time); the production of human capital (investment time); and the production of earnings (labor market time). I am also able to treat systematically a direct effect of human capital on consumption by permitting it to affect the efficiency of household production.
The empirical analysis from the first edition is left intact, even though a substantial body of additional evidence has been accumulated since then, because the major findings have stood up remarkably well to the additional evidence. These findings include:
1. The average money rate of return on a college education to white males is between 11 and 13 per cent, with higher rates on a high-school education, and still higher rates on an elementary-school education. This range for the rate of return on college education, as well as the decline in the rate with successive stages of schooling, has also been found in many subsequent studies.
2. The higher earnings of, say, college graduates compared to high-school graduates are partly due to the college graduate's greater ability, ambition, health, and better educated and more successful parents. I concluded from an examination of several kinds of evidence that differences in these and related traits explain a relatively small part of the earnings differentials between college and high-school graduates (but a larger part of the differentials at lower education levels). Hence, rates of return to college graduates that are unadjusted for "selectivity" are not bad guides to the true rates. Subsequent studies have adjusted for selectivity with a variety of data sources, and their conclusions usually have been quite similar to mine.
Several papers in recent years have tried to formalize the rather old notion that education is largely a device to screen out abler persons for employers, and that, therefore, only a small part of earnings differentials by education can be attributed to the education per se. Even if schooling also works in this way, the significance of private rates of return to education is not affected at all. Moreover, it should be noted that virtually no effort has been made to determine the empirical importance of screening. Furthermore, several major empirical issues must be resolved if screening is to be the primary explanation of earnings differentials. For example, college would be a horrendously expensive "employment agency": each year of college cost a typical individual in 1970 at least $6000 and cost society at least $1500 more than that. Surely, a year on the job or a systematic and intensive interview and applicant-testing program must be a much cheaper and more effective way to screen. My own opinion is that schooling-as-screening must occur in a world with imperfect information, but is a relatively minor influence in determining earnings differentials by education.
3. The evidence I examined indicated that rates of return on college and high-school education declined from about 1900 to 1940, but not after 1940, even though the relative number of college and high-school graduates also grew rapidly after 1940. I concluded that demand shifted more toward educated persons after 1940, partly due to the rapid growth of expenditures on R. and D., military technology, and services. The absence of any decline in rates of return after 1940 has been confirmed in a few subsequent studies. Perhaps the current (1973) weak market for highly skilled manpower is the beginning of a resumption of the earlier decline. Note, however, that the absence of any decline after 1940 is not unique in American history; skill differentials, and thus presumably rates of return on education, apparently did not decline from 1860 to 1890.
4. Average money rates of return on education are not the same for all groups; they are higher on college education for urban white males than for black or rural males, and higher for black than for white women. The evidence I examined suggested that these differences in rates led to corresponding differences in the fraction of high-school graduates going on to college. This effect of rates of return on the incentive to acquire education has been found in other studies. For example, a growth in the monetary return to blacks from a college education in the 1960s has apparently sizably increased their number going to college, as well as shifted their fields of specialization: out of professions that cater to segregated black markets, such as clergy and medicine, and into more integrated professions, such as business and engineering.
5. In Chapter VII, I calculated age–human-wealth profiles for different education classes that show the relation between age and the present value of future earnings, and used them to understand, among other things, life-cycle variations in savings. Some studies have continued this analysis of the linkage between the accumulations of human and nonhuman wealth. I also drew on evidence for slaves, the one example of an explicit market that trades and prices human capital stocks rather than simply the services yielded by these stocks. A major and insightful study has recently appeared that interprets the market for slaves in the United States in terms of the theory of investment in human capital.
The continuing vigor of the research in human capital is increasing testimony that this area of study is not one of the many fads that pass through the economics profession, but an important and lasting contribution. The major reason, in my judgment, is that the theoretical and empirical analyses have been closely integrated, with the theory often inspired by empirical findings. The intimate relation of theory and observation has built a strong foundation for future work that cannot easily be torn down or ignored.
Therefore, I am confident that the analysis of human capital will continue to be a fruitful field of research. Although important studies of the effects of human capital in the market sector can be expected, I anticipate that the excitement will be generated by studies of its effects in the nonmarket sector. Major insights into the determinants of fertility, the production of health, the benefits from schooling to women who do not participate in the labor force, the productivity of marriage, and other topics will result from an integration of the theory of human capital with the allocation of time, household production functions, and the theory of choice.
In short, the prospects for the analysis of human capital look almost as bright to me today as they did during its salad days.
Introduction to the First Edition
Some activities primarily affect future well-being; the main impact of others is in the present. Some affect money income and others psychic income, that is, consumption. Sailing primarily affects consumption, on-the-job training primarily affects money income, and a college education could affect both. These effects may operate either through physical resources or through human resources. This study is concerned with activities that influence future monetary and psychic income by increasing the resources in people. These activities are called investments in human capital.
The many forms of such investments include schooling, on-the-job training, medical care, migration, and searching for information about prices and incomes. They differ in their effects on earnings and consumption, in the amounts typically invested, in the size of returns, and in the extent to which the connection between investment and return is perceived. But all these investments improve skills, knowledge, or health, and thereby raise money or psychic incomes.
Recent years have witnessed intensive concern with and research on investment in human capital, much of it contributed or stimulated by T. W. Schultz. The main motivating factor has probably been a realization that the growth of physical capital, at least as conventionally measured, explains a relatively small part of the growth of income in most countries. The search for better explanations has led to improved measures of physical capital and to an interest in less tangible entities, such as technological change and human capital. Also behind this concern is the strong dependence of modern military technology on education and skills, the rapid growth in expenditures on education and health, the age-old quest for an understanding of the personal distribution of income, the recent growth in unemployment in the United States, the Leontief scarce-factor paradox, and several other important economic problems.
The result has been the accumulation of a tremendous amount of circumstantial evidence testifying to the economic importance of human capital, especially of education. Probably the most impressive piece of evidence is that more highly educated and skilled persons almost always tend to earn more than others. This is true of developed countries as different as the United States and the Soviet Union, of underdeveloped countries as different as India and Cuba, and of the United States one hundred years ago as well as today. Moreover, few if any countries have achieved a sustained period of economic development without having invested substantial amounts in their labor force, and most studies that have attempted quantitative assessments of contributions to growth have assigned an important role to investment in human capital. Again, inequality in the distribution of earnings and income is generally positively related to inequality in education and other training. To take a final example, unemployment tends to be strongly related, usually inversely, to education.
Passions are easily aroused on this subject and even people who are generally in favor of education, medical care, and the like often dislike the phrase "human capital" and still more any emphasis on its economic effects. They are often the people who launch the most bitter attacks on research on human capital, partly because they fear that emphasis on the "material" effects of human capital detracts from its "cultural" effects, which to them are more important. Those denying the economic importance of education and other investments in human capital have attacked the circumstantial evidence in its favor. They argue that the correlation between earnings and investment in human capital is due to a correlation between ability and investment in human capital, or to the singling out of the most favorable groups, such as white male college graduates, and to the consequent neglect of women, dropouts, nonwhites, or high-school graduates. They consider the true correlation to be very weak, and, therefore, a poor guide and of little help to people investing in human capital. The association between education and economic development or between inequality in education and income is attributed to the effect of income oneducation, considering education as a consumption good, and hence of no greater causal significance than the association between automobile ownership and economic development or between the inequality in ownership and incomes.
Excerpted from Human Capital by Gary S. Becker. Copyright © 1993 The National Bureau of Economic Research. Excerpted by permission of The University of Chicago Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
Table of ContentsList of Tables
List of Charts
Preface to the Third Edition
Preface to the First Edition
I. Introduction to the Second Edition
Introduction to the First Edition
II. Human Capital Revisited
2. Education and Training
3. Human Capital and the Family
4. Human Capital and Economic Development
Part One: Theoretical Analysis
III. Investment in Human Capital: Effects on Earnings
1. On-the-Job Training
3. Other Knowledge
4. Productive Wages Increases
IV. Investment in Human Capital: Rates of Return
1. Relation between Earnings, Costs, and Rates of Return
Addendum: The Allocation of Time and Goods over Time
2. The Incentive to Invest
Number of Periods
Wage Differentials and Secular Changes
Risk and Liquidity
Capital Markets and Knowledge
3. Some Effects of Human Capital
Ability and the Distribution of Earnings
Addendum: Education and the Distribution of Earnings: A Statistical Formulation
Addendum: Human Capital and the Personal
Distribution of Income: An Analytical Approach
Supplement: Estimating the Effect of Family Backgrounds on Earnings
Part Two: Empirical Analysis
V. Rates of Return from College Education
1. Money Rates of Return to White Male College Graduates
Returns in 1939
Costs in 1939
Rates of Return in 1939
Rates of Return in 1949
2. Some Conceptual Difficulties
Correlation between "Ability" and Education
Correlation between Education and Other Human Capital
3. Rates of Return to Other College Persons
4. Variation in Rates of Return
VI. Underinvestment in College Education?
1. Private Money Gains
2. Social Productivity Gains
3. Private Real Rates
VII. Rates of Return from High School Education and Trends over Time
1. The Rate of Return from High School Education
2. Trends in Rates of Return
VIII. Age, Earnings, Wealth, and Human Capital
1. Age-Earnings Profiles
2. Age-Wealth Profiles
IX. Summary and Conclusions
2. Future Research
3. Concluding Comments
Part Three: Economy-Wide Changes
X. Human Capital and the Rise and Fall of Families, by Gary S. Becker and Nigel Tomes
2. Earnings and Human Capital
Perfect Capital Markets
Imperfect Access to Capital
3. Assets and Consumption
4. Fertility and Marriage
5. Empirical Studies
6. Summary and Discussion
XI. The Division of Labor, Coordination Costs, and Knowledge, by Gary S. Becker and Kevin M. Murphy
2. Division of Labor among Tasks
3. Coordination Costs
4. Knowledge and Specialization
5. Extent of the Market
6. The Growth in Specialization and Knowledge
7. The Division of Labor between Sectors: Teachers and Workers
XII. Human Capital, Fertility, and Economic Growth, by Gary S. Becker, Kevin M. Murphy, and Robert Tamura
2. Basic Properties of the Model
3. Fertility and Growth
4. Comparative Advantage in the Production of Human Capital
6. Concluding Remarks
A. Sources and Methods
a. The Basic Data
b. Under- and Overreporting
d. Coverage in 1939
f. Urban-Rural Distribution
g. Hours of Work
a. Earnings of Students
b. Direct Private Costs
c. Direct Social Costs
B. Mathematical Discussion of Relation between Age, Earnings, and Wealth