The idea of capital has long had a strong materialistic bent that is evident in the dominance of material capital in economic thinking. The logical basis of an all-inclusive concept of capital, which includes human capital, was established by Irving Fisher (1906). This concept treats all sources of income streams as forms of capital. These sources include not only such material forms as natural resources and reproducible producer and consumer goods and commodities but also such human forms as the inherited and acquired abilities of producers and consumers. Yet the core of economics with respect to this matter concentrates on producer goods, particularly on structures, equipment, and inventories, with little or no attention to the abilities of human beings, even though human resources are much the larger source of income streams.
An approach to capital that includes human capital has two major advantages. The first arises out of the fact that by taking both human capital and material capital into account, a number of biases in economics would be corrected. The overemphasis of material sources of income streams is one of them. Closely related are the imbalances in investment programs of countries where investment in human capital is not an integral part of such programs. Another is the mistaken inference that the real capital–income ratio is necessarily declining over time when the observed ratio of material capital to income falls. Still another is the belief that the productivity of the economy as a whole increases as rapidly as total output rises, relative to measured inputs, although the estimates of inputs fail to include many improvements in the quality of both material factors and human agents. These improvements in quality are the product of investment and thus are forms of capital. There are strong reasons, both theoretical and empirical, to support the inference that in value terms the productivity of the U.S. economy, for example, has remained approximately constant for many decades. [SeeProductivity.] An all-inclusive concept of capital also provides a framework for determining how closely the private and public sectors of an economy come to an optimum in investing in each of the sources of income streams.
The other major advantage of the concept of human capital is in analyzing the various organized activities that augment those human abilities which raise real income prospects. People acquire both producer and consumer abilities. Many of these abilities are clearly the product of investment. There are important unsettled questions about economic growth, changes in the pattern of wages and salaries and the personal distribution of income, that can be resolved once investment in human capital is taken into account. There are also biases in the way the labor force is measured, and in the treatment of public expenditure for education and medical care that can be corrected by using the concept of human capital.
Inherited and acquired abilities. The philosopher–economist Adam Smith boldly included all useful abilities of the inhabitants of a country, whether inherited or acquired, as part of capital. These two sorts of abilities, however, differ importantly in the formation of human capital.
Migration and population growth aside, inherited abilities of a population are akin to the original properties of land in the sense that they are “given by nature” in any time period that is meaningful for economic analysis. Any genetic drift that affects the distribution and level of these abilities occurs so slowly that it is of no relevance in economic analysis. It seems to be true also that the distribution of inherited abilities within any large population remains, for all practical purposes, constant over time and that the distribution of these abilities is approximately the same whether a country is poor or rich, backward or modern, provided the population is large.
But the picture is quite otherwise in the case of acquired abilities having economic value. The formation and maintenance of these abilities are analogous to the formation and maintenance of reproducible material capital. These abilities are obviously subject to depreciation and obsolescences. The distribution and level of acquired abilities can be altered importantly during a time span that matters in economic analysis. Historically they have been altered vastly in countries that have developed a modern economy. In this respect the difference between poor and rich, backward and modern, countries is indeed great. The level of acquired abilities that have economic value is very high in a few countries while it is still exceedingly low in most countries. The truth is that the amount of human capital per worker, or per million inhabitants, varies greatly among countries.
The acquired abilities that raise income prospects are of many types and differ from country to country, depending upon differences in the demand for these abilities and upon differences in the opportunities to supply them. They also are augmented in different ways, depending in part on the type of abilities and in part on the process of investing in them. Some abilities are acquired through informal and essentially unorganized activities, which is the case with most learning in the home and learning from informal community experiences. Others are acquired through organized activities that are, as a rule, also specialized; these include schooling, most on-the-job training, and many adult programs to improve the skills and knowledge of those participating [seeAdult education; Labor force, article onParticipation]. People also improve their future earning abilities through medical care, by acquiring job and other types of information about the economic system, and by migrating to take better jobs.
The formation of human capital, especially through those activities which have become organized and specialized in a modern economy, is of a magnitude to alter radically the conventional estimates of savings and capital formation. These forms of human capital are the source of many additional income streams contributing to economic growth. They also alter wages and salaries, in both absolute and relative terms, and the share of the national income from earnings relative to that from property over time.
Instead of developing and using a general concept that includes human capital, economists have predominantly used a concept restricted to classes of wealth that are bought and sold. Irving Fisher, in a series of papers published just before the turn of the century and then in his excellent but neglected book, The Nature of Capital and Income (1906), clearly and cogently established the economic basis for an all-inclusive concept of capital. But the prestige of Alfred Marshall was too great; his ideas on this matter prevailed. Marshall dismissed Fisher’s approach in these words: “Regarded from the abstract and mathematical point of view, his position is incontestable. But he seems to take too little account of the necessity for keeping realistic discussions in touch with the language of the market-place” (Marshall , 1916, pp. 787–788). Marshall concluded his appendix “Definitions of Capital” by stating, “… we are seeking a definition that will keep realistic economics in touch with the market-place …” (ibid., p. 790). Marshall’s market place restriction had the effect of excluding all capital that becomes an integral part of a people.
In this respect five of the more serious biases that thwart economic analysis and limit its usefulness require a brief comment.
It has been said that the economists have had a rather unfavorable image in the public mind. Whether true or not, there have been many protests contending that the policy implications of economics are primarily concerned with the value of material things. Unquestionably, economics has been strongly and persistently biased in favor of producer and consumer goods and commodities. For this reason it is justly charged as having a materialistic orientation. This orientation is all too evident in the treatment of capital, where producer goods are treated as if they were the sum and substance of capital and as if economic growth were dependent wholly on investment in such goods. Accordingly, there is much merit in the protest that the rise of human capital in capitalism is not seen or that increases in human capital, which have become a crucial feature of the economic system, are neglected.
Labor inputs inadequately specified
Economists have found it all too convenient to think of labor as a homogeneous input free of any capital components. Much theory rests on a presumed dichotomy between labor and capital. But it is a treacherous dichotomy when analyzing economic growth, for the reason that the acquired abilities of labor that contribute to growth are as much a product of investment in man as growth is a product of investment in material forms of capital. The bias here is also clearly evident in the conventional approach to the measurement of labor as a factor of production.
In this approach it suffices to count the number of workers in the labor force or the number of man-hours worked. Differences in the acquired abilities of a labor force that occur over time are not reckoned. This particular bias has fostered the retention of the classical notion of labor as a capacity to do manual work requiring little skill and knowledge, a capacity with which, according to this notion, laborers are endowed about equally. But this notion of labor is patently wrong. The size of the labor force or the number of man-hours worked is not a satisfactory measure of increases in the productive services rendered by labor over time because of changes in the human capital component.
Misinterpretation of declines in capital–income ratios
The empirical foundation of economics has been much strengthened by studies of wealth and income [seeNational income and product accounts; National wealth]. One of the uses made of these studies has been to show that the capital–income ratio has been declining in countries with a modern economy. The decline in this ratio is then frequently viewed with apprehension because of the inferences that are drawn with respect to savings and investment and with respect to economic growth. Here, too, there is obviously a bias arising out of the restricted concept of capital on which these estimates are based.
There are no compelling reasons why the stock of any particular class of capital should not fall (or rise) relative to national income over time. Producer goods—structures, equipment, and inventories—are such a class. It is this particular class of capital that has been declining relative to income in the case of these estimates. Leaving aside the fact that the estimates of producer goods omit many improvements, they are at best only a part of all capital. The most serious omission in them is human capital, which has been increasing at a much higher rate than that of material reproducible capital. In the United States between 1929 and 1957, for example, while national income was increasing at about 3 per cent per annum, the stock of reproducible tangible capital rose only 2 per cent per annum. But the stock of educational and of training-on-the-job capital in the labor force rose between 4 and 5 per cent per annum. It turns out that the sum of this class of material capital and of the human capital just mentioned rose about 3 per cent per annum, that is, at the same rate as national income. The ratio of this more nearly all-inclusive concept of capital to national income was about the same in 1929 and in 1957; in both of these years it was about six. Thus the apparent substantial declines in capital relative to income, which are based on estimates covering a number of modern countries, are an illusion in the sense that they are not valid indications of what has been happening to the ratio of all capital to income.
Savings and investment–income relation
Another closely related issue has been the concern about the amount of savings and investment relative to income, the concern being that as national income rises, savings and investment decline relatively [seeConsumption function]. Here, too, conventional estimates are very misleading because they omit investment in human capital. They under-state the amount of savings and investment that occurs in any given year, and they show a decline in such savings and investment over time relative to income, when in fact there may have been no decline in all savings and investment in relation to income. Again, an appeal to more all-inclusive estimates for the economy of the United States is instructive. Based on the sum of the investment in reproducible material capital and in educational and on-the-job-training capital in the labor force, the amount of capital thus formed was equal to about 26 per cent of net national product in both 1929 and 1957.
Seeming rise in aggregate productivity
Another bias that has come to thwart economic analysis is the belief that the productivity of capital and labor has been rising very substantially over time, especially in countries that have developed a modern economy. There are estimates to support the belief that output has been rising not only relative to capital and to labor, respectively, but also relative to all inputs of capital and labor treated as an aggregate. The output of a particular industry or sector may, of course, rise in relation to material capital or to the size of the labor force. Nor is it implausible, under the circumstances that characterize economic growth, for an entire industry or sector to lag in its adjustments and thus to operate for a considerable period at a disequilibrium. This would cause the value of its output to decline relative to all inputs valued at equilibrium prices as the disequilibrium becomes established, and then to rise as such an industry or sector reattains an equilibrium.
But there is no strong theoretical or empirical basis for believing that the productivity of all factors of production treated as an aggregate, where the economy grows at an even pace, should either rise or fall. A much more plausible hypothesis is that it remains approximately constant over time. Why are there so many estimates that seemingly show total national output rising relative to total inputs? In a real sense it is because more of the additional capital that is formed over time is concealed than is income. This explanation is undoubtedly too cryptic, and therefore some elaboration is in order.
The analytical game that most economists have been playing in studying economic growth has been to take an index of reproducible material capital that omits changes in quality, in the sense that it abstracts from improvements in material capital. The next move is to take the size of the labor force, or man-hours worked, which also omits changes in quality, in the sense that improvements in the capabilities of labor are not fully reckoned. These two measures of inputs are then aggregated and related to total output. This game always shows the total input of such capital and labor as falling relative to total output over time. The inference is then drawn that the productivity of the economy as a whole rises over time, and this rise in productivity is generously attributed to “technological change,” which according to this game would appear to account for most of the observed economic growth.
The finding that the over-all productivity of the economy increases in this manner is due to two types of illusions. The first is simply a consequence of the fact that many factors of production which are added to the resources of an economy over time are not included among the inputs; they are frequently and conveniently swept under the rug of “technological change.” Here, basically, the analytical problem is one of specifying and identifying the improvements in human and material resources that occur over time. Undoubtedly most of the seeming rise in over-all productivity, which is based on estimates of the conventional measures of material capital and labor, is a result of omission of a large array of these quality components [seeAgriculture, article onProductivity and technology].
The second type of illusion is based on an apparent change in the capital–income ratio of a country. The observed ratio declines for reasons already noted, namely, because only a part of all capital is reckoned and because this part of the stock of capital is not increasing at as high a rate as either all capital or income. To avoid this type of illusion, an all-inclusive concept of capital is necessary. The crucial question is as follows: When all the sources of income streams are treated as capital, is the rate of return on capital, so conceived, rising persistently over time? There is no theoretical basis for an affirmative answer, even though economic instability or forms of disequilibrium are postulated. Nor is there any empirical evidence that would support an affirmative answer to this question. The rate of return to investment that entails a “standard” component of risk and uncertainty is probably no higher presently in the United States than it was, say, during the 1920s. Nor should it come as a surprise that this rate of return has not been rising secularly.
With regard to motives and preferences of people for holding and acquiring sources of income streams, the most plausible assumption is that they have remained essentially constant. With respect to the behavior of suppliers of the sources of income streams, the equally plausible assumption is that these suppliers have been successful in providing enough new sources to increase national income, as it is now measured, at a rate of, say, between 3 and 4 per cent per year; however, they have not succeeded in increasing the supply at so fast a rate as to cause the price of these sources per dollar of income per year to decline, given the growth in demand consistent with the underlying preferences of the demanders. From these two very plausible assumptions its follows that the rate of return to investment would tend to remain approximately constant over time; and in this critical sense, the value of capital in relation to income has not been rising.
Both of these types of productivity illusions are in large part a consequence of the neglect of human capital and of its contribution to production, although improvements in the quality of other forms of capital are also a part.
But there is a sense in which real income can rise in a way that would alter the productivity of an economy that is beyond the two illusions already examined, although it is closely related analytically. There are consumer satisfactions that people derive from better health, from more education, and from more leisure time. These satisfactions also increase in most countries as economic growth occurs. Are they to be treated as a consumer surplus? [SeeConsumer’s surplus.] Or are they concealed income from particular forms of capital that have been augmented over time? Surely a part of education has the attribute of an enduring consumer component that renders a stream of consumer satisfactions. These satisfactions from education, at least in principle, can be treated as a product of investment in schooling, akin to the satisfactions derived from investment in conventional consumer durables. But none of them appear in national income as it is presently measured. If they were included and if the sources were omitted from capital, it would tend once again to reduce the capital–income ratio. Contrariwise, if the stock of additional education capital represented by these enduring consumer abilities were to be identified and measured and thus made a part of the total stock of capital, and if the stream of income from this part of educational capital were omitted from national income, it would tend to increase the apparent capital–income ratio over time. Obviously the reason for the apparent rise of such a ratio would be the use of a partial concept of income. Unquestionably the same reasoning is applicable to consumer satisfactions from better health.
It is not obvious, however, that the additional satisfactions that come from more leisure time, from decreases in hours of work per week or in days worked per week and during a year, can be treated in the same way. But since more time for leisure has a value, the value of it can be included in income. The source of this leisure is an integral part of total production that provides enough income so that people can afford the leisure time. Thus, when all sources of income are treated as capital, the source of leisure is already reckoned. Accordingly, if the income component represented by leisure were to be omitted, and if all capital were reckoned, it would tend to increase the apparent capital–income ratio over time.
Public “welfare” expenditures
A long-standing bias permeates the treatment of public expenditures for health facilities and services and for education, treating them as if they were wholly for consumption, as welfare measures that in no way enhance the abilities of people as producers. Even vocational training and public funds to retrain workers in depressed areas for new jobs are often treated as welfare programs, although they are predominantly an investment in the productive abilities of the recipients. Conversely, most of the notions for attaining an optimum rate of economic growth in poor countries are seriously biased because of their strong emphasis on investment in new steel mills and in other modern industrial structures and equipment, with no comparable emphasis on providing for the complementary investment in human agents to administer and to do the skilled work which these installations require.
Human beings are both consumers and producers. To production they contribute either entrepreneurship or work. In classical economics the producer attribute of human agents is that of a factor of production, referred to simply as labor. In modern analysis it is that of an input, or of a coefficient of production [seeProduction]. In accounting for increases in national income over time, labor is treated as one of the sources of economic growth. As a source of income streams, the acquired abilities of human agents have, as we have argued, the attribute of an investment. All of these attributes of human agents are in one way or another a form of capital. Viewed as capital, they are a stock that renders services of economic value. The services are either consumer or producer services. Human capital is not, of course, bought or sold where men are free and none are slaves, as are the material forms of capital; but its producer services are generally for hire, the price being a wage or a salary. These producer services of human agents can be augmented by means of investment, which increases their income prospects. The additional income that is realized from an investment in human capital implies some rate of return.
The logical approach to determining the economic value of any of these attributes of human agents will differ depending upon the aim of the analysis, the theory and estimating technique that are used, and the limitations of the data. Much, of course, depends upon the aim. To see this, three different aims will be considered briefly.
What are the inputs in an economy, sector, or industry? A basic difficulty arises at once out of the fact that an input has two economic faces; one is the income value of its productive services, and the other is its capital value. In the case of a parcel of land, for example, there is the rent paid for its use and the price at which the land sells.
The aim of many studies is to determine the economic value of the productive services of the inputs. Suppose we begin with a net national product for a given year and ignore all increases or decreases in inputs that occur during the year. Suppose also that the inputs are of two sorts, namely, labor and producer goods. Such a gross dichotomy may show that a fourth of the net national product is functionally contributed by producer goods and three-fourths by human agents. There is then the temptation to transform the productive services of the inputs into stocks of capital by using a naive approach that treats their services as permanent income streams and that capitalizes each of these income streams at the same rate of return. This approach would, of course, imply that the stock of human capital is three times as large as that of producer goods.
But this is obviously a naive way of transforming income streams into capital stocks, since estimates of net national product, as presently determined, are far from net—especially so for human capital —and thus in this important respect these estimates do not represent permanent income streams. It is also true that it would be a rare coincidence in a dynamic, growing economy to find an equality in the real rates of return to investment.
Nevertheless, in pursuing this aim there may be analytical reasons for concentrating on the magnitude and value of the productive services of these inputs while leaving aside the problem of determining their capital value. To do this, what is required is the price and the amount of the respective input services employed. With regard to these requirements, economists have treated human agents more adequately than producer goods. The difference is a consequence partly of inadequacies of the theory used and partly limitations of data. In aggregating producer goods, the differences in the quasi rents, or the relative prices of the services of these goods, are not as a rule reckoned; nor are the improvements in quality of new producer goods generally taken into account [seeRent].
In this sense it is true that most estimates of material capital conceal a part of the additional capital that is formed over time. The problem of aggregation in this connection is not only conceptual but is also confounded by the lack of price data of the productive services of different classes of producer goods [seeAggregation]. Fortunately these analytical inadequacies are not nearly so pronounced in the case of the productive services of human agents. Wages and salaries provide price data; and human agents can be classified into fairly homogeneous groups by occupations, or levels of skills, and by age, sex, and schooling. Nevertheless, this picture of the productive services of inputs will not reveal the differences in investment opportunities among the reproducible inputs.
Sources of economic growth
National income increases at, say, a rate of 3 per cent per year between two dates. What are the sources of that growth? The matter of investment, rates of return to investment, and whether net savings are allocated optimally among investment opportunities can be put aside in determining the part of growth associated with each source. If the income value of the productive services of human agents and of producer goods were known, it would entail only a little simple arithmetic. But such is not the case, mainly for the reasons already considered. For human agents it is fairly straightforward to the extent that there is a linkage between what they contribute to production and what they earn in wages and salaries, although to determine what part of the increase in wages and salaries over time comes from more schooling, on-the-job training, better health, and from still other sources is far from easy. But not all of the additional income from these sources accrues to the individuals who have acquired these abilities. Some of it accrues to their co-workers, employers, and neighbors. In addition, there is an array of consumer satisfactions from these sources that accrues partly to the individuals who have acquired the relevant abilities and partly to others in the community. In general these consumer satisfactions are omitted from national income as presently measured.
For producer goods, it is as yet most difficult to ascertain this linkage because of the manner in which these forms of capital are identified and measured and because of the lack of price information on the services of producer goods. Improvements in the quality of such capital are largely omitted, and they become a major part of a “source” that appears as a residual, an increase in “output per unit of input.” Thus, until this residual is properly allocated, it is obvious that producer goods (material capital) are underrated as a source of growth. Capital embodied in human agents is in this respect on a much stronger footing.
While this knowledge of the sources of economic growth is indeed useful in serving the aim of Edward F. Denison’s comprehensive study (1962), it is not an approach to determining the underlying costs and returns to the investment that produced the additional sources that account for this part of economic growth. The important matter of an optimum allocation of total net savings among investment opportunities is not a part of the aim of this approach.
Acquired abilities that have economic value usually entail identifiable costs. Each process of acquiring abilities that enhance income prospects has the attributes of an investment. Viewed as an investment, what is the rate of return? The aim implied by this question cannot be realized by obtaining a picture of inputs or by ascertaining the sources of economic growth as they are treated above. An investment approach is required to attain this aim, which is important because knowledge about investment and the rate of return in this connection is essential in making the economic decisions necessary to achieve an optimum allocation of savings among investment opportunities. The relevance of this approach for a large array of economic problems is set forth by this writer in “Investment in Human Capital” (Schultz 1961b). The theoretical “relations between earnings, rates of return, and the amount invested” and “how the latter two can be indirectly inferred from earnings” are investigated by Gary S. Becker (1962) in a paper that appears in the supplement referred to below.
The investment approach is central in a number of recent studies, the results of which are presented in a supplement, “Investment in Human Beings,” to the Journal of Political Economy, October 1962. It includes the theoretical analysis by Becker just mentioned and the findings of several major empirical studies. These studies pertain to education, on-the-job training, health, information about the labor market, and migration when migration is treated as investment in human beings. Only two of these forms of investment, on-the-job training and education, will be considered here.
Investment in job training
By treating “training” as an “investment in acquisition of skill or in improvement of worker productivity” and by using a procedure akin to that used in determining investment in education, Jacob Mincer (1962) has identified and measured what appear to be costs of on-the-job training. His study, which also considers returns to this training, is restricted to males in the United States.
The investment in this training during 1939 was $3,000 million and during 1958, $13,500 million. In constant 1954 dollars, it was $5,700 million and $12,500 million, respectively. Mincer’s study reveals two major shifts. One is toward higher skill levels; for example, males who already had a college level of education by 1939 accounted for one-third of all this training acquired that year; during 1958 they received nearly two-thirds of it. The other shift is toward formal schooling relative to on-thejob training; the investment in this training declined from about four-fifths to three-fifths of that in schooling between 1939 and 1958.
Estimates of the rate of return to investment in on-the-job training are very fragmentary. Those reported range from 9.0 to 12.7 per cent per year. The apparent reasons for the two shifts referred to above and the implications of on-the-job training as a factor in income and with respect to employment behavior are also examined by Mincer.
Investment in education
Education is unquestionably the largest source of human capital consisting of acquired abilities. But the road to an analysis of the economic value of education is not paved. The costs of education are surprisingly well concealed. Not all of the benefits accrue to students; they are frequently widely dispersed. The rates of return depend on earning profiles of many different shapes, extending over many years. The responses to new, profitable investment in education are subject to some long lags; they are blunted in the case of public decisions by other matters and in the area of private decisions by incomplete information and by uncertainties that are inherent in a long future. There is also the uncertainty inherent in the fact that no student knows his abilities for schooling prior to putting himself to the test. In addition, the capital market is not well organized when it comes to lending funds for schooling.
Seemingly the task is simplified when it comes to formal education, since it is organized and presumably can be viewed as an industry that produces schooling. The difficulty with this simplification is that the functions of the educational establishment include activities other than schooling. One of the important functions of higher education is research. On-campus research has been increasing rapidly, and much of it is an integral part of graduate instruction. Another function consists of extension activities, notably, in the United States, the farflung state agricultural extension services. There is also activity akin to an advisory service, especially to public agencies. Then, too, universities have been entering upon programs of instruction and research abroad with so-called sister universities in the cooperating country. And not least of these other functions is that of discovering and cultivating talent, which is quite distinct from formal schooling.
Costs. Much has been done recently to clarify the cost components of education. Opportunity costs are large, especially the earnings foregone by mature students, which were concealed in the way costs of schooling were formerly estimated. In the United States, for example, earnings foregone account for fully half to three-fifths of the total costs of high school and higher education. Because of the importance of earnings foregone, education beyond the elementary level is far from free to students. In poor countries and also in some low-income communities in the United States, for instance, in some agricultural areas and in city slums, earnings foregone have been and still are a factor even for children during the latter years in the elementary grades. When earnings foregone are brought into the picture, a part of the educational scene that always appeared blurred becomes clear. The distinction between private costs incurred by the student or his family and total costs to the economy is important analytically in explaining differences in incentives to invest in schooling, the shift in favor of formal schooling relative to on-the-job training over time, and in ascertaining the rates of return that matter in determining optimum investment decisions.
Total costs also provide clues to the amounts invested and changes in stocks. Leaving aside the education of persons who are not in the labor force, in the United States, as already noted, the amount spent on the schooling of persons who are 14 years and older rose at a rate of 4 per cent per annum between 1929 and 1957, measured in constant 1956 dollars. Investment in reproducible tangible wealth rose at 2 per cent per annum. When these rates of growth are applied to the respective stocks of 1957, the net annual investment implied is $21,900 million for this schooling and $25,500 million for this form of material capital.
Benefits. The future benefits from schooling accrue in part to the student and in part to others in society. Burton A. Weisbrod (1962) has substantially clarified the distinction between these two parts, although other investigators are fully aware that there are these two classes of benefits from schooling. As yet there has been little empirical success in determining the value of the benefits from schooling that accrue to co-workers and employers of the students and to the students’ neighbors. There is a strong presumption that universal literacy of a population in a modern economy has large external economies [seeCapital, social overhead; External economies and diseconomies].
Furthermore, not all the benefits from schooling that accrue to the student are revealed in his future earnings, in wages, salaries, and entrepreneurial income from work. The benefits that accrue to the student are of three sorts. One consists of current consumption; the other two are an investment. That which is current consumption consists of satisfactions that the student obtains from schooling while in attendance. This benefit is undoubtedly small, for school days entail much hard work and long hours. There is next a class of enduring consumer abilities acquired through schooling; from these abilities the student derives satisfactions throughout his remaining life, for example, the ability to appreciate and enjoy the fine arts, the masterpieces of literature, science, and logical discourse. The source of these satisfactions is an investment in particular consumer abilities; but the value of the stream of satisfactions from this source, although substantial, is not a part of future wages and salaries. The third set of benefits consists of increases in the student’s productivity, the source being the producer abilities acquired from schooling. While most of these appear in future earnings, there are nevertheless some that are derived from production activities that the student does for himself over the years, like preparing his income tax returns, which do not enter into his earnings or into national income as it is presently measured.
Earnings. Investigations of investment in education have concentrated on earnings while leaving aside the other benefits from schooling. Even so, it is no simple matter to identify and measure these earnings. They are beset by the effects of differences in the inherited abilities of workers, of race, sex, and age; by unemployment; by the content and quality of schooling; and by the effects upon earnings of job training, health, and other forms of investment in human beings. Difficult as it is to isolate and adjust for these effects, some fairly satisfactory estimates have been obtained. These estimates make it possible to determine the rates of return to schooling, which are considered briefly below. They show also that at least one-fifth of the economic growth of the United States between 1929 and 1957 came from additional earnings connected with schooling.
Rates of return. Available estimates on rates of return are limited to money returns, that is, to money earnings from schooling that accrue to the student. Accordingly, all other benefits from schooling are omitted in these estimates; and to this extent the real rates of return to education are underestimated. For a more complete review and appraisal, see The Economic Value of Education (Schultz 1963).
Rates of return to total costs of schooling support several important generalizations. Costs here consist of all direct and indirect costs, including earnings foregone, whether borne privately or publicly; returns are restricted to monetary earnings from schooling. For males in the United States the following generalizations emerge: (1) the rate of return to elementary schooling is higher than to high school education, and in turn the rate of return to high school education is higher than to college education; (2) the rate of return to high school education (completing the twelfth year) rose persistently and very substantially between 1939 and 1958, while that to college education (completing at least the sixteenth year) declined somewhat between 1939 and 1956 and then began to rise; and (3) the lowest of these rates of return has been about 12 per cent per annum.
New knowledge pertaining to investment in human capital is already quite satisfactory with regard to the behavior of the supply and the rates of return to on-the-job training and to education. Little is known, however, about the factors that have been increasing the demand for these acquired abilities—an integral part of economic growth.
Theodore W. Schultz
Anderson, C. Arnold; Brown, James C.; and Bowman, M. J. 1952 Intelligence and Occupational Mobility. Journal of Political Economy 60:218−239.
Ashby, Eric 1960 Investment in Education: The Report of the Commissions on Post-school Certificate and Higher Education in Nigeria. Lagos (Nigeria): Federal Ministry of Education.
Becker, Gary S. 1960 Underinvestment in College Education? American Economic Review 50:346−354.
Becker, Gary S. 1962 Investment in Human Capital: A Theoretical Analysis. Journal of Political Economy 70, no. 5 (Supplement):9−49.
Becker, Gary S. 1964 Human Capital: A Theoretical and Empirical Analysis, With Special Reference to Education. New York: National Bureau of Economic Research.
Benson, Charles S.; and Lohnes, Paul R. 1959 Skill Requirements and Industrial Training in Durable Goods Manufacturing. Industrial and Labor Relations Review 12:540−553.
Blank, David; and Stigler, George J. 1957 The Demand and Supply of Scientific Personnel. National Bureau of Economic Research General Series, No. 62. New York: The Bureau.
Bonner, J.; and Lees, D. S. 1963 Consumption and Investment. Journal of Political Economy 71:64−75.
Bowen, William G. 1964 Economic Aspects of Education: Three Essays. Princeton Univ., Department of Economics, Industrial Relations Section.
Bowman, Mary Jean 1962 Human Capital: Concepts and Measures. U.S. Office of Education, Bulletin  no. 5:69−92.
Colberg, Marshall R. 1965 Human Capital in Southern Development: 1939−1963. Chapel Hill: Univ. of North Carolina Press.
Commission on Human Resources and Advanced Training 1954 America’s Resources of Specialized Talent: A Current Appraisal and a Look Ahead. New York: Harper.
Denison, Edward F. 1962 Education, Economic Growth, and Gaps in Information. Journal of Political Economy 70, no. 5 (Supplement): 124−128.
DeWitt, Nicholas 1955 Soviet Professional Manpower, Its Education, Training, and Supply. Prepared in co-operation with the National Academy of Sciences–National Research Council for the National Science Foundation. Washington: Government Printing Office.
Edding, Friedrich 1958 Internationale Tendenzen in der Entwicklung der Ausgaben für Schulen und Hochschulen; International Trends in Educational Expenditures. Kiel (Germany): Institut fur Weltwirtschaft. → Contains a summary in English.
Education and the Southern Economy. 1965 Southern Economic Journal 32, part 2:1−128.
The Falk Project for Economic Research in Israel 1961 Report: 1959 and 1960. Jerusalem (Israel): The Project. → See especially pages 138−146 on the profitability of investment in education and pages 146−150 on the measurement of educational capital in Israel.
Fisher, Irving (1906) 1927 The Nature of Capital and Income. New York and London: Macmillan.
Friedman, Milton and Kuznets, Simon 1945 Income From Independent Professional Practice. National Bureau of Economic Research General Series, No. 45. New York: The Bureau.
Hansen, W. Lee 1963 Total and Private Rates of Return to Investment in Schooling. Journal of Political Economy 71:128−140.
Harbison, Frederick; and Myers, Charles A. 1964 Education, Manpower, and Economic Growth: Strategies of Human Resource Development. New York: McGraw-Hill.
Investment in Human Beings: Papers Presented at a Conference Called by the Universities–National Bureau Committee for Economic Research. 1962 Journal of Political Economy 70, no. 5: Supplement. → The entire supplement is devoted to the problem of human capital.
Kellogg, Charles E. 1960 Transfer of Basic Skills of Food Production. American Academy of Political and Social Science, Annals 331:32−38.
Kenen, Peter B. 1965 Nature, Capital and Trade. Journal of Political Economy 73:437−460.
Machlup, Fritz 1962 The Production and Distribution of Knowledge in the United States. Princeton Univ. Press.
Marshall, Alfred (1890) 1916 Principles of Economics. 7th ed. New York: Macmillan; London: St. Martins.
Miller, Herman P. 1960 Annual and Lifetime Income in Relation to Education: 1939−1959. American Economic Review 50:962−986.
Mincer, Jacob 1958 Investment in Human Capital and Personal Income Distribution. Journal of Political Economy 66:281−302.
Mincer, Jacob 1962 On-the-job Training: Costs, Returns, and Some Implications. Journal of Political Economy 70, no. 5 (Supplement): 50−79.
Mushkin, Selma J. (editor) 1962 The Economics of Higher Education. U.S. Office of Education, Bulletin , no. 5. → The entire issue is devoted to the topic. See especially pages 69−92 and pages 281−304.
Nicholson, Joseph S. 1891 The Living Capital of the United Kingdom. Economic Journal 1:95−107.
Organization for economic Cooperation and Development 1962 Policy Conference on Economic Growth and Investment in Education. Paris: The Organization. → Also published in Volume 36 of the Bulletin of the International Bureau of Education.
Organization for economic Cooperation and Development 1964 The Residual Factor and Economic Growth. London: H.M. Stationery Office.
Princeton University, Industrial Relations Section 1957 High-talent Manpower for Science and Industry: An Appraisal of Policy at Home and Abroad, by J. Douglas Brown and Frederick Harbison. Research Report Series, No. 95. Princeton, N.J.: The Section.
Schultz, Theodore W. 1960 Capital Formation by Education. Journal of Political Economy 68:571−583.
Schultz, Theodore W. 1961a Education and Economic Growth. Volume 60, pages 46−88 in National Society for the Study of Education, Yearbook. Part 2: Social Forces Influencing American Education. Univ. of Chicago Press.
Schultz, Theodore W. 1961b Investment in Human Capital. American Economic Review 51:1−17.
Schultz, Theodore W. 1963 The Economic Value of Education. New York: Columbia Univ. Press.
Stigler, George J. 1962 Information in the Labor Market. Journal of Political Economy 70, no. 5 (Supplement) : 94−105.
Stigler, George J. 1963 Capital and Rates of Return in Manufacturing Industries. A Study of the National Bureau of Economic Research. Princeton Univ. Press.
Tawney, Richard H. 1938 Some Thoughts on the Economics of Public Education. Oxford Univ. Press.
Vaizey, John 1958 The Costs of Education. London: Allen & Unwin.
Weisbrod, Burton A. 1961 The Valuation of Human Capital. Journal of Political Economy 69:425−436.
Weisbrod, Burton A. 1962 Education and Investment in Human Capital. Journal of Political Economy 70, no. 5 (Supplement): 106−123.
Wiles, P. J. D. 1956 The Nations’s Intellectual Investment. Oxford University, Institute of Statistics, Bulletin 18, no. 3:279−290.
According to marginal analysis, each factor of production is paid according to its contribution to production, and the market constitutes a mechanism that establishes a moral rule of distributive justice in the society. This theory of marginal productivity was put forth around the turn of the twentieth century by the American economist John Bates Clark (1847–1938). In the neoclassical analysis of the labor market, the equilibrium wage is set at the point of intersection of the demand and supply curves of labor. The level of the equilibrium wage set by the intersection of the two curves guides the allocation of workers across firms in such a way that an efficient allocation of resources is achieved. The implication is that wages are equal to the value of the marginal product of labor, which is the same for all workers and every firm. To achieve this condition, firms and workers should operate in a perfectly competitive environment, anonymity for both sides should be present, and the allocation of labor to firms should be random. However, wage differentials are present in the labor market, and several attempts have been made to explain the reasons for these wage differentials.
Adam Smith (1723–1790) was the first economist to introduce the idea of wage differentials and attempt to offer reasons for its existence. According to Smith, one reason for wage differentials is the common idea that wages vary positively, everything else being held constant, with the disutility of labor. This argument was later formalized by William Stanley Jevons (1835–1882) and states that an individual is willing to offer more labor input only if the wage is higher; this is because leisure time gets scarcer and thus becomes more valuable to an individual. Another reason for relative wage differentials put forward by Smith is related to the concept of human capital, which has been introduced only recently into economic analysis. According to Smith, the cost of a person’s education or training can be viewed as an investment in the individual’s future earnings capacity, analogous to an investment in physical capital. To be economically justified, this investment must be recuperated over the lifetime of the student or trainee. Thus, those with education or training will generally earn more than those without. In recent years, this interpretation of wage differentials has given rise to numerous attempts to measure the rate of return on investment in education and training. In general, these efforts have sought to check whether such investments in education and training do, in fact, earn normal profits for an equally valuable capital.
Neoclassical theory has never abandoned the idea that wages tend to equal the net product of labor; in other words, the marginal productivity of labor rules the determination of wages. However, it has also been recognized that wages tend to retain an intricate, although indirect, relationship with the cost of rearing, training, and sustaining the energy of efficient labor. All these factors cause wage differentials, and they make up what is known as human capital in economic literature. Human capital, in mainstream economics, is similar to the physical means of production—for example, factories and machines—and is defined as the knowledge and skills possessed by the workforce that can be accumulated. Human capital, therefore, is a stock of assets that one owns and that allows one to receive a flow of income, similar to interest earned. In modern times, human capital is largely a product of education, training, and experience. Within this approach, investment in human capital is treated similarly to investment in physical capital and is considered, in many cases, as a principal factor for the observed differences in developmental levels between nations. It is argued that investment in human capital can take various forms, including improvements in health services, formal education, and training of employees, and it can help eliminate blockages in productivity enhancements. It is worth clarifying that the acquisition of more human capital is not related to the number in the workforce but to the improvement of labor services and the concomitant increase in labor productivity.
Today, new growth theorists put a lot of emphasis on human capital: “The main engine of growth is the accumulation of human capital—or knowledge—and the main source of differences in living standards among nations is a difference in human capital. Physical capital plays an essential but decidedly subsidiary role. Human capital takes place in schools, in research organizations, and in the course of producing goods and engaging in trade” (Lucas 1993, p. 270). Within this framework, a lot of emphasis is given to the policies promoting human capital, such as investment in education or health. More than that, according to the economists Theodore W. Schultz (1902–1998) and Anthony P. Thirlwall, it is argued that differences in the quality of human capital are the cause of the different levels of development among nations and that the improvement of human capital might reduce the need for physical capital.
In addition, literature about organizational behavior and human resource management has emphasized the role of human capital in promoting the development and implementation of national and corporate strategies. Hence, human capital is a very important asset and may serve as a useful link in the chain between employee and business performance; moreover, human capital, human-resources practices, and firm performance are areas that need to be explored together to design better corporate strategies.
Modern human capital theory was initiated by the work of Jacob Mincer, in which investment in human resources is considered similarly to other types of investments. In Mincer’s first model (1958), individuals are assumed to be identical prior to any training, thereby forcing a differential wage to employment based on the length of the expected training period. The size of the compensating differential is determined by equating the present value of the future net earnings stream (gross earnings minus costs) with the different levels of investment. The formal presentation of Mincer’s first simple model is:
lnw (s ) = lnw (0) + rs
where lnw (s ) represents the log of the annual earnings of an individual with s years of education, lnw (0) represents the log of the annual earnings of an individual with basic years of education, and r is the internal rate of return to schooling years s. According to the above equation, individuals with more training receive higher earnings. The difference between earning levels of individuals with different years of schooling is determined by the second term of the right-hand side of the equation. If one defines the internal rate of return to schooling as the discount rate that equates the lifetime earnings stream for different educational choices, then the internal rate of return to schooling can be estimated by the coefficient on years of schooling. This simple framework offers a number of interesting implications. However, the whole analysis relies on some unrealistic assumptions, which can be summarized as follows:
the individuals are all identical prior to making choices;
the age of retirement does not depend on years of schooling;
the only cost of schooling is foregone earnings;
earnings do not vary over the life cycle; and
there is no post-school on-the-job investment.
Mincer’s second model (1974) allows for the on-the-job investment and yields an earnings specification that is similar to the first. To establish the relationship between potential earnings and years of labor-market experience, assuming that observed earnings are equal to potential earnings less investment cost, the following relationship for observed earnings is produced:
lnw (sx ) = a 0 + ρs s + β 0 x + β 1 x 2
where lnw (sx ) stands for the observed earnings, x is the amount of work experience, ρs is the rate of return on formal schooling, and s represents the years of education. The intercept term is the product of the log skill price and the initial ability of the individual. The coefficients β 0 and β 1 stand for the return to experience.
This second expression is called Mincer’s standard wage equation, which regresses the log earnings on a constant term, on a linear term of the years of schooling, and on a quadratic term of the labor-market experience. Mincer’s standard wage equation transforms a normal distribution of years of schooling into a log-normal distribution of earnings. Under the assumption that post-school investment patterns are identical across individuals and do not depend on the schooling level, Mincer shows that there is an important distinction between age-earnings profiles and experience-earnings profiles, where experience means years since leaving school. More specifically, he shows that the log-earnings-experience profiles are parallel across schooling levels and that log-earnings-age profiles diverge with age across schooling levels.
By the early 1970s, the estimation of the returns on schooling using Mincerian wage regressions had become one of the most widely analyzed topics in applied econometrics. The reason is that the human-capital earning function has several distinct characteristics that make it particularly attractive:
the functional form is not arbitrary, and the identity is based on the optimizing behavior of individuals as captured by the outcome of the labor-market process;
it converts immeasurables (the dollar cost of investment in human capital) into measurables (years of schooling and years of labor-market experience);
it can include instrumental variables to capture a dichotomous variable describing some characteristics such as race or sex; and
the coefficients of the regression equation may be attributed with economic interpretations.
While the human-capital literature has now been generalized to incorporate on-the-job training, it should be noted that the Mincerian wage regression equation is a representation of the statistical relationship between wages and experience (given schooling) for an exogenously determined rate of on-the-job training. The Mincerian wage regression disregards the endogeneity of post-schooling human-capital accumulation and treats schooling and training symmetrically. More precisely, Mincer’s approach ignores the possibility that schooling may change the human-capital accumulation process that takes place on the job.
C. Lester Thurow’s job-competition model serves as an interesting counterpoint to the traditional models in explaining the distribution of earnings. In his book Generating Inequality: Mechanisms of Distribution in the U.S. Economy (1975), Thurow shows that for the period 1950–1970, changes in the educational attainments of white males 25 to 64 years old did not affect their earnings. He finds that educational distribution is equalized through the years while income distribution is not, and he notes that the expected arguments and results from marginal productivity are not fulfilled. To explain the distribution of wage earnings, Thurow rejects market imperfections as a possible explanation of unexpected observations in labor markets. In fact, he argues that individuals compete against one another for job opportunities based on their relative costs of being trained to fill a job position instead of based on wages they are willing to accept. He argues that wages are based on the marginal productivity of the job, not of the worker. Workers who compete for jobs offered at fixed wages determine the labor supply, and workers compete for relative positions based upon their training costs to employers rather than on wages. Hence, job competition prompts people to overinvest in formal education for purely defensive reasons. Moreover, technology, the sociology of wage determination, and the distribution of training costs determine the distribution of job opportunities.
Even though the advantages of education are frequently exaggerated in terms of their strictly economic results, there is no doubt that education is advantageous in earning a higher income and also teaches general skills that improve the quality of labor and, by extension, human capital. However, recent literature on issues related to human capital stresses the need to consider other dimensions of human identity that contribute to the formation of human capital. In fact, modern labor economics has criticized the simple approach that tries to explain all differences in wages and salaries in terms of human capital as a function of knowledge, skills, and education. The reason is that the concept of human capital can be infinitely elastic, including unmeasurable variables such as culture, personal character, or family ties. Many other factors, therefore, may contribute to wage differentials, such as gender or race. The existence of imperfections in labor markets implies mostly the existence of segmentation in labor that causes the return on human capital to differ between different labor-market segments. Similarly, discrimination against minority or female employees implies different rates of return on human capital. Most of these studies in their wage equations use the technique of instrumental (or dummy) variables to introduce specific characteristics of the labor segmentation that takes place. The statistical significance of these instrumental variables indicates their importance in forming human capital and the concomitant wage differentials. In addition, over the past few decades several writers, including Rhonda Williams (1991) and Howard Botwinick (1993), have begun to develop alternative approaches to the analysis of discrimination and wage differentials that are based on a more classical analysis of capitalist competition and accumulation.
The theoretical discussion about the meaning of human capital is vast and extends into the fields of human development and human resource management. For instance, human-development literature often distinguishes between specific and general human capital, where the first refers to specific skills or knowledge that is useful only to a single employer, while the second refers to general skills, such as literacy, that are useful to all employers. Human-development theories also differentiate social trust (social capital), sharable knowledge (instructional capital), and individual leadership and creativity (individual capital) as three distinct forms of human participation in economic activity.
Indisputably, the term human capital is used everywhere in economic and business analysis in which labor has to count as an input in the production process. The term has gradually replaced terms such as laborer, labor force, and labor power in the relevant analysis, giving an impression that these traditional terms are socially degraded. Moreover, in the strict sense of the term, human capital is not really capital at all. The term was originally used as an illustrative analogy between, on the one hand, investing resources to increase the stock of ordinary physical capital (such as tools, machines, or buildings) in order to increase the productivity of labor and, on the other hand, investing in educating or training the labor force as an alternative way of accomplishing the same general economic objective. In both sorts of investment, investors incur costs in the present with the expectation of deriving future benefits over time. However, the analogy between human capital and physical capital breaks down in one important and very crucial respect. Property rights over ordinary physical capital are readily transferable by sale, whereas human capital itself cannot be directly bought and sold on the market. Human capital is inseparably embedded in the nervous system of a specific individual and thus it cannot be separately owned. Hence, at least in regimes that ban slavery and indentured servitude, the analogy between human capital and physical capital breaks down.
Botwinick, Howard. 1993. Persistent Inequalities: Wage Disparity under Capitalist Competition. Princeton, NJ: Princeton University Press.
Gintis, Herbert, Samuel Bowles, and Melissa Osborne. 2001. The Determinants of Individual Earnings: Skills, Preferences, and Schooling. Journal of Economic Literature 39 (4): 1137–1176.
Lucas, Robert E., Jr. 1993. Making a Miracle. Econometrica 61: 251–272.
Mincer, Jacob. 1958. Investment in Human Capital and Personal Income Distribution. Journal of Political Economy 66 (4): 281–302.
Mincer, Jacob. 1974. Schooling, Experience and Earnings. New York: National Bureau of Economic Research.
Schultz, Theodore W. 1962. Reflections on Investment in Man. Journal of Political Economy 70: 1–8.
Thirlwall, Anthony P. 1999. Growth and Development. London: Macmillan.
Thurow, C. Lester. 1975. Generating Inequality: Mechanisms of Distribution in the U.S. Economy. New York: Basic Books.
Williams, Rhonda. 1991. Competition, Discrimination and Differential Wage Rates: On the Continued Relevance of Marxian Theory to the Analysis of Earnings and Employment Inequality. In New Approaches to Economic and Social Analyses of Discrimination, eds. Richard R. Cornwall and Phanindra V. Wunnava. New York: Praeger.
Human capital refers to the knowledge, skills, and capabilities of individuals that generate economic output. Human capital averages about two-thirds of the total value of the capital of most economies, which includes land, machinery, and other physical assets as well as the skills and talents of people. The value of human capital is often apparent after physical destruction, as during World War II—many of the German and Japanese cities that were bombed intensely were able to recover 80 to 90 percent of their previous levels of production within months.
More than two centuries ago, Adam Smith observed in Wealth of Nations (1776) that an educated man must earn more than "common labor" to "replace to him the whole expense of his education." Human capital was first discussed extensively by two Nobel Prize–winning economists, Theodore Schultz (1979) and Gary Becker (1992), to explain how any personal decision to sacrifice today for a return tomorrow can be analyzed in the same way that a business considers an investment decision, such as whether to buy new machinery. A nation's stock of human capital and thus its economic growth potential could be increased, they reasoned, if governments reduced the cost and increased the benefits of schooling, the human capital embodied in individuals.
Human capital is rented in the labor market rather than "sold." Individuals exchange effort for reward, and acquire human capital in the expectation that their incomes will be higher. Human capital takes time to acquire, and the basic model of human capital acquisition compares the income stream from going to school with that of going to work immediately. The costs of going to school include the direct costs, in the form of tuition and books, as well as the indirect costs, the forgone earnings that could have been received from working. The benefits of more schooling are higher earnings in the future.
Costs and Benefits
These costs and benefits must be brought to a single point in time so that the present value of the higher average lifetime earnings with more schooling and the lower average earnings with less schooling can be compared. A rational individual chooses the education and work profile that maximizes the present value of lifetime earnings.
This investment approach to education yields several important predictions about behavior. First, most people will get some education, and most students will be young, because the early years of education have few direct costs, and there are no forgone earnings because most societies prohibit children from working; younger people also have a longer period over which they can recoup their educational investment in the form of higher earnings. The analysis gets more interesting when young people reach the age at which they can work, sixteen or eighteen in most industrial countries, and twelve to fifteen in many developing countries. Youth in industrial countries tend to stay in school because the direct costs are often low and the for-gone earnings may be low (working as a teen is often at the minimum wage), while earnings with a college education can be significantly higher. The interest rate used to compare future and present earnings is also important; if this interest rate is low because of subsidized loans, more people will choose to get more education.
Second, government policies shape individual decisions about how much human capital to acquire by affecting the cost of schooling and the payoff from work. Social attitudes also play a role, encouraging young people to stay in school, or to go to college with friends in industrial countries, but in developing countries often encouraging children to help support the family as soon as possible. More forward-looking people, those most able to sacrifice now for future returns, are likely to get the most education, such as those willing to undergo rigorous and time-consuming medical education.
In the United States a combination of higher lifetime earnings, government policies, and social attitudes has increased the percentage of high school graduates who go to college. In 1960, about 45 percent of all high school graduates enrolled in college in the following twelve months, 54 percent of men and 38 percent of women. By 1980, 49 percent of high school graduates enrolled in college: men dropped to 47 percent, and women rose to 52 percent. In 1999, the most recent data available, 63 percent of high school graduates enrolled in college, 61 percent of men and 64 percent of women. Most studies find that men are more present-oriented to immediate earnings opportunities than women, explaining why more women are in college.
The average earnings of college-educated persons are higher than the earnings of high school graduates, and rose in the late twentieth century. In 1979, male college graduates earned 33 percent more than high school graduates, the so-called college earnings premium, and female college graduates earned 41 percent. By 2000, these college earnings premiums rose to 84 percent for men and 67 percent for women, in part because the real earnings of those with less education fell as a result of globalization, which reduced the wages of many high school graduates employed in manufacturing.
Education is generally a good investment: the private rate of return to a college education ranges from 12 to 40 percent in most countries, more than the return on investments in stocks and bonds. Around the world, the rate of return to primary school education was estimated to be 29 percent in the early 1990s, 18 percent for secondary schooling, and 20 percent for higher education. In developing African countries, these rates were 39, 19, and 20 percent, respectively, while in the industrial countries that are members of the Organisation for Economic Co-Operation and Development (OECD), the rates were 22, 12, and 12 percent, respectively. The high rate of return on primary education suggests that countries should subsidize it most.
It is very hard to compute average rates of return for higher levels of education because it is likely that the most capable individuals go to college, so that colleges transmit knowledge that increase productivity and also serve as screening institutions for employers seeking the best workers; some economists argue that the screening role is more important than the productivity-increasing role. On the other hand, the private rate of return to a college education may be higher if more highly educated individuals have both higher incomes as well as more fringe benefits and better or more prestigious jobs. The social rate of return may be even higher than the private rate if highly educated individuals provide more leadership and commit fewer crimes.
Any personal investment that raises productivity in the future can add to an individual's human capital, including onthe-job training and migration. A job that offers training, for example, may have lower earnings but still be attractive because a person knows that, after completing the military, aviation, or stockbroker course, earnings will rise. On-the-job training that makes the person more useful to the employer who provided it is called job-specific training, while training that transmits general skills that make the trained individual useful to many employers is called general training. Employers are more likely to pay for specific than general training. Employers do a great deal of training; by some estimates, U.S. employers spend almost as much on training as is spent at colleges and universities.
Migration of Human Capital
If people truly are the "wealth of a nation," should developing countries worry about a "brain drain" if their doctors, nurses, and scientists migrate to richer countries? The answer depends on the 3 R's of recruitment, remittances, and returns. Recruitment deals with who migrates abroad: Are the emigrants employed managers whose exit leads to layoffs, or unemployed workers who have jobs and earnings abroad but would have been unemployed at home? Remittances are the monies sent home by migrants abroad: Are they significant, and fuel for investment and job creation, or small and not used to speed development? Returns refers to whether migrants return to their countries of origin: Do they return after education or a period of employment abroad with enhanced skills, or do they return only to visit and retire?
During the 1990s the migration of highly skilled workers from developing to more developed countries increased, reflecting more foreign students as well as aging populations able to pay for more doctors and nurses and the internet-related economic boom. International organizations are exploring ways in which the more developed countries could replenish the human capital they take from the developing world via migration, perhaps by contributing to or backing loans to improve their educational systems. If the human capital that migrates is not replenished, global inequalities may increase.
See also Education ; Globalization ; Social Capital .
Becker, Gary S. Human Capital. A Theoretical and Empirical Analysis, with Special Reference to Education. New York: National Bureau of Economic Research, 1964.
Blaug, Mark. "The Empirical Status of Human Capital Theory: A Slightly Jaundiced Survey." Journal of Economic Literature 14 (September 1976): 827–855.
Organisation for Economic Co-Operation and Development. Human Capital Investment: An International Comparison. Paris: OECD, 1998.
Psacharopoulos, George. "Returns to Investment in Education: A Global Update." World Development 22, no. 9 (1994): 1325–1343.
Smith, Adam. An Inquiry into the Nature and Causes of the Wealth of Nations. Dublin: Whitestone, 1776.
Philip L. Martin
The quality of labor in a country's workforce can directly influence a nation's economic growth. Investment in vocational training and education, which improves the quality of labor, is called investment in human capital. As an individual becomes more skilled and educated, productivity or output of work may increase, along with income. The concept of human capital can provide justification for wage and salary differentials by age and occupation. Education and training in skill development can create human capital just as construction of a building creates physical capital.
Some economists assert that a society should allocate resources to educational and training services similar to the allocation of resources for physical capital. Costs would be incurred in expectation of future benefits. However, unlike physical capital, human capital is not a guarantee and cannot be repossessed in settlement of a debt. The key question has been whether or not benefits exceed expenditures by a sufficient amount.
Until the mid-nineteenth century, education expenditures were primarily generated by the private sector. By the 1850s all states had developed programs for funding public schools. As late as the early twentieth century, most people considered education that was beyond the primary grades to be a luxury—particularly among low-income groups. However literacy rates continued to move upward and since 1940, education levels have consistently climbed. In 1940 24 percent of the U.S. population had high school diplomas and 4.6 percent earned college degrees. By 1996 almost 82 percent had completed four years of high school and almost 24 percent had completed four or more years of college. By attending college or vocational training programs, individuals were able to invest in themselves. Firms invested in human capital with on-thejob training. Government invested in human capital by offering programs to improve health, quality free schooling, including vocational and on-the-job training, and by providing student loans.
See also: Physical Capital