To most of you, capital means a bank account, one hundred shares of
IBM, assembly lines, or steel plants in the Chicago area (especially
during a Ryerson lecture). These are all forms of capital in the sense
that they yield income and other useful outputs over long periods
But I am going to talk about a different kind of capital. Schooling,
a computer training course, expenditures on medical care, and lectures
on the virtues of punctuality and honesty are capital too in the sense
that they improve health, raise earnings, or add to a person’s
appreciation of literature over much of his or her lifetime. Consequently,
it is fully in keeping with the capital concept as traditionally defined
to say that expenditures on education, training, medical care, etc.,
are investments in capital. However, these produce human, not physical
or financial, capital because you cannot separate a person from his
or her knowledge, skills, health, or values the way it is possible
to move financial and physical assets while the owner stays put. This
embodiment of human capital in people is depressingly illustrated
by the reactions of Hong Kong residents to the takeover of Hong Kong
in 1997 by China. Many local people are busy protecting against China’s
policies by selling off some of their local financial and physical
assets in order to invest in safer foreign securities and property.
At the same time, however, computer experts, top management, and other
skilled personnel are leaving Hong Kong in droves to seek citizenship
elsewhere. They cannot reduce the risk to their human capital from
China by investing only part of the human capital abroad; they must
go where their capital goes.
It may seem odd now, but I hesitated a while before deciding to call
my book Human Capital—and even hedged the risk by using a long
subtitle. In the early days, many people were criticizing this term
and the underlying analysis because they believed it treated people
like slaves or machines. My, how the world has changed! The name and
analysis are now readily accepted by most people not only in all the
social sciences, but even in the media. I was surprised when a few
months ago Business Week magazine had a cover story titled “Human
Capital.” And more amazing still, this has been their most popular
cover story in several decades.
However, I should add that the concept of human capital remains
suspect within academic circles that organize their thinking about
social problems around a belief in the exploitation of labor by
capital. It is easy to appreciate the problems created for this
view by the human capital concept. For if capital exploits labor,
does human capital exploit labor too—in other words, do some
workers exploit other workers? And are skilled workers and unskilled
workers pitted against each other in the alleged class conflict
between labor and capital? If governments are to expropriate all
capital to end such conflict, should they also expropriate human
capital, so that governments would take over ownership of workers
You can see why an idea developed to understand the economic and social
world has been thrust into ideological discussions. Yet the concept
of human capital has been popular in Communist countries. My book
and those by Schultz and others on human capital are extensively used
in the Soviet Union, Eastern Europe, and China. Even before the recent
reforms, economists and planners there had no trouble with the concept
of investing capital in people.
I will try to avoid technical analysis and jargon, and concentrate
on showing how the analysis of investments in human capital helps
in understanding a large and varied class of behavior not only in
the Western world, but also in developing countries and countries
with very different cultures. My discussion follows modern economics
and assumes that these investments usually are rational responses
to a calculus of expected costs and benefits.
2. Education and Training
Education and training are the most important investments in human
capital. My book showed, and so have many other studies since then,
that high school and college education in the United States greatly
raise a person’s income, even after netting out direct and
indirect costs of schooling, and after adjusting for the better
family backgrounds and greater abilities of more educated people.
Similar evidence is now available for many points in time from over
one hundred countries with different cultures and economic systems.
The earnings of more educated people are almost always well above
average, although the gains are generally larger in less-developed
countries. Consider the differences in average earnings between
college and high school graduates in the United States during the
past fifty years. After being reasonably stable at between 40 and
50 percent until the early 1960s, they rose during that decade and
then fell rather sharply. This fall during the 1970s led some economists
and the media to worry about “overeducated Americans”
(see Freeman, 1976). The concept of human capital itself fell into
But as Kevin Murphy and Finis Welch document in a recent study (1989),
the monetary gains from a college education rose sharply during the
1980s to the highest level during these fifty years. The earnings
advantage of high school graduates over high school dropouts also
increased. Talk about overeducated Americans has vanished, and it
has been replaced by concern once more about whether the United States
provides adequate quality and quantity of education and other training.
These concerns are stimulated by tough economic competition from a
renewed Europe, Japan, Korea, and other Asian countries, by sluggish
rates of productivity advance in the United States during the past
fifteen years, by a large drop in SAT scores, and by the dismal performance
of American high school students on international tests in mathematics.
For those who prefer a monetary bottom line, trends in the earnings
of young persons in the United States provide good reason for concern
about the preparation they are receiving. The trend has been disastrous
for the 15 percent of all students and much larger percentage of inner-city
blacks who fail to complete high school. Their real wage rates have
fallen by more than 30 percent since the early 1970s. Whether because
of school problems, family instability, or other forces, young people
without a college education are not being adequately prepared for
work in modern economies.
A Labor Department commission on labor quality, of which I am a member,
is considering what can be done to improve the quality of workers
in the United States. The concerns that led to the creation of this
commission have stimulated renewed academic interest in the analysis
of human capital, which illustrates how research in social sciences
responds, sometimes excessively, to public policy issues.
The fraction of high school graduates who entered college fell during
the middle of the seventies when benefits from a college education
dropped, and it rose again in the eighties when the benefits greatly
increased. This caused an unexpected boom in college enrollments during
the past few years, despite the relatively few people who are reaching
college age. So, alas, the large rise in applications to our College
in recent years is not due solely to more widespread appreciation
of the superb education it provides. Many educators expected enrollments
in the eighties to decline not only for demographic reasons, but also
because college tuition was rising rapidly. They were wrong because
they failed to appreciate that benefits from college rose even faster
than costs, and that high school graduates respond to changes in both
benefits and costs.
One might believe that enrollments in college would be easy to predict
since the number of persons graduating from high school can be predicted
quite closely. But demographic-based college enrollment forecasts
have been wide of the mark during the past twenty years, as Steve
Stigler and I, especially Steve, showed in a subcommittee report a
few years ago to the Baker Commission. Such forecasts ignored the
changing incentives to women, blacks, and older persons to enroll
That human capital investments tend to respond rationally to benefits
and costs is clearly indicated by changes in the education of women.
Prior to the 1960s in the United States, women were more likely than
men to graduate from high school but less likely to continue on to
college. Women shunned math, sciences, economics, and law, and gravitated
toward teaching, home economics, foreign languages, and literature.
Since relatively few married women continued to work for pay, they
rationally chose an education that helped in household production
and no doubt also in the marriage market. All this has changed radically.
The enormous increase in the participation of married women is the
most important labor force change during the past twenty-five years.
Many women now take little time off from their jobs even to have children.
As a result, the value to women of market skills has increased enormously,
and they are shunning traditional “women’s fields”
to enter accounting, law, medicine, engineering, and other subjects
that pay well. Indeed, women now comprise one-third or so of enrollments
in law, business, and medical schools, and many home economics departments
have either shut down or are emphasizing the “new home economics,”
which is a true branch of economics.
The same trends in women’s education are found in Great Britain,
France, Scandinavia, Taiwan, Japan, Mexico, and other countries with
large increases in the labor force participation of women, even when
attitudes toward women differ greatly from those now prevalent in
Europe and the United States. Whenever the labor force participation
of married women has increased sharply, changes in the gains from
work for pay have had a more powerful effect on the behavior of women
than have traditional ideas about the proper role of women.
Job opportunities for women at first improved slowly as they started
to move up in business and the professions during the past several
decades. But the trend accelerated sharply after the late 1970s. The
ratio of the earnings of full-time working women and men has increased
more rapidly since 1979 than during any previous period in our history,
and women are becoming much more prominent in many highly skilled
jobs. Improvements in the economic position of black women have been
especially rapid, and they now earn just about as much as white women.
Although the civil rights movement clearly contributed to greater
job opportunities for women and other minorities, it is far from
the whole story. This can be seen from the fact that women progressed
most rapidly under the Reagan administration, which was opposed
to affirmative action and did not have an active Civil Rights Commission.
In my judgment, women advanced primarily because of their greater
attachment to the labor force. This in turn was stimulated by a
large decline in fertility, a rapid increase in divorce, and the
growing importance of the service sector. Human capital analysis
assumes that schooling raises earnings and productivity mainly by
providing knowledge, skills, and a way of analyzing problems. An
alternative view, however, denies that schooling does much to improve
productivity, and instead it stresses “credentialism”—that
degrees and education convey information about the underlying abilities,
persistence, and other valuable traits of people.
According to extreme versions of this line of analysis, earnings of,
for example, college graduates exceed those of high school graduates
not because college education raises productivity, but because more
productive students go on to college.
Credentialism obviously exists. But many kinds of evidence suggest
that credentialism does not explain most of the positive association
between earnings and schooling.
The main problem with credentialism is that companies do not want
information on success at schoolwork, but on abilities and performance
in the context of working life: the discipline imposed by factories,
the need to please customers and get along with fellow employees,
and so forth. Success in the flexible, individualistic, and rather
undisciplined university atmosphere in most countries and in high
schools in the United States does not convey much relevant information.
I tell my classes that eccentrics and nuts can last much longer as
students than as workers, and they respond that the same is true of
A cheaper and more efficient way to provide information to employers
is for teenagers to enter directly into the labor force, as they
did prior to the industrial revolution. Far more would be learned
about their work-related abilities and other characteristics after
six years of work experience than after six additional years of
schooling. High school and college education has spread extensively
in modern economies because the additional knowledge and information
acquired in school is so important in technologically advanced economies.
I should add that advocates of the credentialism approach have become
rather silent in recent years with the growing concerns about schools
and labor quality in the United States.
Of course, learning and training also occur outside of schools, especially
on jobs. Even college graduates are not well prepared for the labor
market when they leave school, and they are fitted into their jobs
through formal and informal training programs. The amount of on-the-job
training ranges from an hour or so at simple jobs like dishwashing
to several years at complicated tasks like engineering in an auto
plant. The limited information available indicates that on-the-job
training is an important source of the very large increase in earnings
as workers gain greater experience at work. And recent bold estimates
by Jacob Mincer suggest that the total investment in on-the-job training
may be almost as large as the investment in education.
After a few years of frequent job changes, most workers settle down
and remain with the same company for a long time. Workers and their
employers get bonded together in large part because of the on-the-job
learning and training. Therefore, it is not surprising that job changes
are common among unskilled workers and uncommon among skilled workers.
It also appears that job changes are much less frequent in Japan than
in the United States mainly because on-the-job investments in workers
are greater in Japan. My friends in the humanities like Dick Stern
may complain that so far I have only mentioned “money,”
or they might say "mere money." Is there any place in human capital
theory for education to appreciate literature, culture and the good
life? Fortunately, nothing in the concept of human capital implies
that monetary incentives need be more important than cultural and
Obviously, it is much easier to quantify the monetary side, but,
nevertheless, progress has been made on other aspects. Many studies
show that education promotes health, reduces smoking, raises the
propensity to vote, improves birth control knowledge, and stimulates
the appreciation of classical music, literature, and even tennis.
In an ingenious study that relies heavily on economic theory, Bob
Michael (1972) quantifies some non-monetary benefits of education.
His results and those of others indicate that such benefits of schooling
are quite large, although for most people they are apparently smaller
than monetary benefits.
3. Human Capital and the Family
No discussion of human capital can omit the influence of families
on the knowledge, skills, values, and habits of their children. Parents
who severely beat their children cause lasting damage, while at the
other end of the spectrum, sympathetic and firm parents help motivate
Large differences among young children grow over time with age and
schooling because children learn more easily when they are better
prepared. Therefore, even small differences among children in the
preparation provided by their families are frequently multiplied over
time into large differences when they are teenagers. This is why the
labor market cannot do much for school dropouts who can hardly read
and never developed good work habits, and why it is so difficult to
devise policies to help these groups.
Parents have a large influence on the education, marital stability,
and many other dimensions of their children’s lives. The term
“underclass” describes families in which low education,
welfare dependence, early pregnancy, and marital instability pass
from parents to children. In light of this, it is rather surprising
that although earnings of parents and children are positively related,
the relation is not strong. For example, if parents’ earnings
in the United States are 20 percent above the mean of their generation,
the children’s earnings tend to be less than 6 percent above
the mean of their own generation. Earnings of parents and children
appear to be a little more strongly related when parents are poorer.
It is easy to see why children’s and parents’ earnings
may be closer in poorer families. Richer families can pay for the
training of their children, including the earnings foregone when children
spend time in training rather than at work. Many poorer parents would
be willing to lend their children money to help them obtain further
training if the parents could expect to get paid back later when they
are old. But children may not carry out their part of the bargain,
especially in highly mobile societies where children often live far
from their parents.
One solution is for governments to lend money to students when their
parents are unable or unwilling to finance the training. The federal
government has developed an extensive loan program to help students
finance college education. Unfortunately the program has serious flaws,
including low caps on the maximum amounts that can be borrowed, misplaced
and excessive subsidies, and shockingly high default rates. In addition
to explicit loans, some direct subsidies to schools may, in effect,
also be “loans” to students which they repay later with
taxes that help finance support for the elderly. By combining publicly
subsidized schooling with a social security system, countries may
have found a very crude and indirect, but perhaps reasonably effective,
way to provide loans to children that get repaid when the parents
are old and collect retirement benefits (see Becker and Murphy, 1988).
Families divide their total spending on children between number of
children and the amount spent per child. The number of children and
spending per child tend to be negatively related. The reason is simple.
An increased number of children raises the effective cost of adding
to the spending on each child, because an additional dollar or hour
of time spent on each child then means a larger total addition to
spending. Similarly, an increase in the dollars or time spent on each
child raises the cost of having an additional child. Consequently,
even a modest tax on births can have a large negative effect on the
number of children and a large positive effect on the amount spent
on each child.
China imposed heavy, not modest, taxes and other penalties on large
families during the past decade, especially in urban areas. It is
revealing about the cross-cultural relevance of this analysis that
sharp declines in urban fertility have been accompanied by discussions
in the Chinese press of the “emperor child.” This refers
to only children who receive lavish toys and presents from their parents,
and are pushed toward outstanding educational achievement.
This negative relation at the family level between number of children
and spending per child implies a close and also usually negative relation
at the aggregate level between population growth and investments in
human capital. Differences among ethnic groups in the United States
are fascinating. Groups with small families generally spend a lot
on each child’s education and training, while those with big
families spend much less. The Japanese, Chinese, Jews, and Cubans
have small families and the children become well educated, while Mexicans,
Puerto Ricans, and blacks have big families and the education of children
suffers. (I should add that the Mormons are an interesting exception,
for they have both very large families and high levels of achievement).
It should come as no surprise that children from the ethnic groups
with small families and large investments in human capital typically
rise faster and further in the United States’ income-occupation
hierarchy than do children from other groups.
Malthus’ famous prediction that people marry earlier and birth
rates rise when incomes increase was decisively contradicted by the
industrial revolution, whose effects became evident only shortly after
publication of the second edition of his book on population. This
is a common paradox: a great book gets contradicted by events not
long after publication. The contradiction to Malthus’ theory
is that fertility fell sharply, rather than rose, as per capita incomes
grew in Great Britain, the United States, France, Germany, Sweden,
and other Western countries. Rapid advances in education and other
training accompanied the sharp declines in fertility. Parents did
spend more on children when their incomes rose—as Malthus predicted—but
they spent a lot more on each child and had fewer children, as human
capital theory predicts.
Similar changes occur in other cultures when they experience rapid
economic growth. Taiwan’s birth rate was cut in half from 1960
to 1975, while the fraction of high school graduates doubled after
Taiwan took off in the 1960s toward its remarkable economic growth.
Mexico’s birth rate did not fall much during its rapid economic
growth in the 1950s and 1960s. But since 1975 birth rates have fallen
by more than one-third, and school enrollments have expanded rapidly.
4. Human Capital and Economic Development
Economic analysis has no trouble explaining why, throughout history,
few countries have experienced very long periods of persistent growth
in income per person. For if per capita income growth is caused by
the growth of land and physical capital per worker, diminishing returns
from additional capital and land eventually eliminate further growth.
The puzzle, therefore, is not the lack of growth, but the fact
that the United States, Japan, and many European countries have
had continuing growth in per capita income during the past one hundred
years and longer.
Presumably, the answer lies in the expansion of scientific and technical
knowledge that raises the productivity of labor and other inputs in
production. The systematic application of scientific knowledge to
production of goods has greatly increased the value of education,
technical schooling, and on-the-job training as the growth of knowledge
has become embodied in people—in scientists, scholars, technicians,
managers, and other contributors to output.
It is clear that all countries which have managed persistent growth
in income have also had large increases in the education and training
of their labor forces. First, elementary school education becomes
universal, then high school education spreads rapidly, and finally
children from middle income and poorer families begin going to college.
A skeptic might respond that the expansion in education as countries
get richer no more implies that education causes growth than does
a larger number of dishwashers in richer countries imply that dishwashers
are an engine of growth.
However, even economists know the difference between correlation and
causation, and have developed rather straightforward methods for determining
how much of income growth is caused by a growth in human capital.
In an excellent study for the United States, Edward Denison (1985)
finds that the increase in schooling of the average worker between
1929 and 1982 explains about one-fourth of the rise in per capita
income during this period. He is unable to explain much of the remaining
growth. I like to believe that this is mainly because he cannot measure
the effects on earnings of improvements over time in health, on-the-job
training, and other kinds of human capital.
The outstanding economic records of Japan, Taiwan, and other Asian
economies in recent decades dramatically illustrate the importance
of human capital to growth. Lacking natural resources—e.g.,
they import practically all their sources of energy—and facing
discrimination from the West, these so-called Asian tigers grew rapidly
by relying on a well-trained, educated, hard-working, and conscientious
labor force. It surely is no accident, for example, that Japan’s
system of lifetime employment at large companies originated after
World War II when they began to upgrade their technology rapidly partly
by investing heavily in the training of employees. The lifetime system
is not explained just by the traditional Japanese culture that emphasizes
loyalty toward groups, for job changes in Japan were frequent during
the first half of this century (see Hashimoto and Raisian, 1985).
Compelling evidence of the link between human capital and technology
comes from agriculture. Education is of little use in traditional
agriculture because farming methods and knowledge are then readily
passed on from parents to children. Farmers in countries with traditional
economies are among the least educated members of the labor force.
By contrast, modern farmers must deal with hybrids, breeding methods,
fertilizers, complicated equipment, and intricate futures markets
for commodities. Education is of great value since it helps farmers
adapt more quickly to new hybrids and other new technologies (see
Welch, 1970). Therefore, it is no surprise that farmers are about
as well educated as industrial workers in modern economies.
We have reached the end of my visit. Perhaps I have succeeded in conveying
the enormous energy devoted to the analysis of human capital during
the past quarter-century and the impressive advances of analytical
techniques and the accumulation of empirical regularities. Much is
now known for many countries about the effects of education on earnings,
occupation, employment, and unemployment of both men and women and
various races and ethnic groups. Much too is known about the link
between birth rates and investments in education and training, how
families influence the human capital of their children, and the relation
between investments in human capital and economic progress.
1. I appreciate the helpful comments of Guity Nashat, Sherwin Rosen,
and George Stigler and the assistance of David Meltzer.This text may
be used and shared in accordance with the fair use provisions of US
and international copyright law agreement, and it may be archived
and redistributed in electronic form, provided that this entire notice,
including copyright information, is carried and provided that the
University of Chicago Press is notified and no fee is charged for
access. Archiving, redistribution or republication of this text on
other terms, in any media, requires both the consent of the authors
and the University of Chicago Press.