Investment in human capital may be analysed a by more "scientific" approach. By comparing the costs and benefits of an educational investment we are able to arrive at some conclusions as to the profitability of investing in education. Consider a simple model of human capital. Assume that a high school graduate is trying to determine whether to go to college. There are two general types of cost. Direct costs include, tuition, fees, books and supplies; the indirect costs of attending college are the foregone earnings of not entering the labour market after high school and the physical costs of studying and being examined. The costs must be compared to the economic benefits of investment in education, in other words, to the enlarged future flow of earnings. It is important to note that future benefits are worth less to us than the same benefits received today for two reasons. First, people prefer consumption today to consumption tomorrow because uncertainties make future enjoyments problematic. Second, interest can be earned by investing monetary benefits rather than using them for consumption. It is therefore necessary that the net present value (NPV), i.e. the discounted value of a financial sum arising at some future period, of the present and future costs and benefits of a college education be determined as they accrue at different points in time. These costs and benefits can be represented diagramatically (figure 1). We must compare the costs (Areas 1 and 2) and the benefits (Area 3) in deciding if investment in education is profitable [1].
Figure 1.
The discounting formula for costs and benefits over a number of years can be formulated as follows:
Vp = E0 + E1(1+i)-1 + E2(1+i)-2 + ... + En(1+i)-n
The "E"s represent a stream of net incremental earnings; "n" is the duration of the earnings stream, i.e. expected working life; i is the interest rate. This tells us that the more distant the future earnings the greater the discounting. As with any other investment, educational investment should occur if Vp (the NPV) is greater than zero, as the discounted benefits exceed the discounted costs. Another method used in making the investment decision is to calculate the internal rate of return, r, and compare it with the interest rate, i. The internal rate of return (IRR) is the discount rate, r, at which the NPV is zero. Hence, the equation becomes
Vp = E0 + E1(1+r)-1 + E2(1+r)-2 + ... + En(1+r)-n = 0
The IRR, r, indicates the maximum rate of interest, i, that would allow investment to break even. If r exceeds the market rate of interest i, then investment is profitable. It is profitable to invest up to point at which i = r.
A number of generalisations and implications can now be made. First, the longer the expected working life, the more likely it is that the NPV of an investment in human capital will be positive, explaining why more young people than old people attend college. It is also a factor that explains the wage differentials between men and women, because female participation rates may be discontinuous on account of leaving the labour force to marry and raise children. Second, the lower the cost of investment in human capital, the larger the number of people who will find such an investment to be profitable. Thus, the lower the direct and indirect costs of attending college the higher the NPV of a college education, giving another reason why less older people attend college; the indirect costs (opportunity costs) of attending college are greater the older the individual. Third, the larger the college-high school earnings differential, the larger the number of people who will invest in college education, ceteris paribus. From the countless empirical studies which have estimated the returns on investment in human capital, there appears to be a general consensus that the rate of return (ROR) to an individual of a college education is between 5% and 15% above that of a non-college graduate.
There are certain biases in the estimated rates of return on education. This is because factors such as subsidies which are not paid by the individual are included in the SROR are not included in the PROR. Education benefits society in a number of ways. More educated workers tend to have lower unemployment rates and receive higher wages. Therefore society benefits by receiving more taxes (as more educated workers tend to work more continuously on average and the tax take from these workers are proportionally higher as they are taxed at higher marginal rates due to higher salaries). More poorly educated workers may also find crime an attractive means of supplementing their lower incomes. Society may benefit from investing in education by paying less for social welfare programmes and crime prevention/law enforcement. The children of more educated parents tend to receive better guidance and grow up in a more desirable environment.
Blaug (1972) concludes that if the SROR exceeds the PROR then more investment in human capital should take place and vice-versa. The SROR also provides us with the rationale that education should be subsidised with public funds. McConnell and Brue (1989)state that "the size of these public subsidies to education should be determined on the basis of the magnitude of the associated social benefit."
McConnell and Brue (1989) give three reasons why different people invest different amounts in human capital. First, consider two individuals, A and B, with different demand curves for human capital, DA and DB respectively, and a common supply curve, S, indicating that both individuals have the same access and terms to money capital for investment in education. DB is to the right of DA and this may be explained by B having greater natural abilities than A, where individual B can transform a given input of schooling into greater productivity, and hence greater earnings than A. Given the perfectly elastic supply curve of financial capital, individual B will invest in eB years of schooling. B's ROR on investment in education exceeds A's. The earnings differentials between B and A is further widened.
Figure 2.
Second, assume A and B are identical in terms of ability. However, their demands for human capital may not be the same due to discrimination. "A" may be black or female, for example, and therefore likely to encounter discrimination which reduces A's chance of transforming their human capital into incremental earnings. A's ROR on the same amount of education as B may be less due to this discrimination. A's demand for human capital is less than B's and this is why A invests less in education (eA) than B (eB). Discrimination in labour markets leads to less investment in education and further wage differentials.
Third, assume that A and B have identical abilities implying DA = DB. But now consider the situation where B obtains more favourable terms on acquiring money capital than A. B's superior credit rating may be explained if B is from a wealthier family than A and therefore has more collateral than A. The effect of this is shown below where SA and SB are A and B's respective supply curves for financial funds i.e. B obtains money capital at a lower rate of interest than A. Going back to the first equation, we find that the lower the rate of interest the greater the ROR. It is therefore rational for B to invest in more years of education (eB) than A (eA).
Figure 3.
Capital market imperfections have important consequences. First, due to the increased rates in lending to students (especially those who are young) financial institutions may choose not to make funds available for education. Students from better off families may still be able to afford a college education, while students from poorer families may not. The outcome of this is that the college/high school wage differential will tend to increase; the poor get poorer and the rich get richer. A second implication is that the government may attempt to offset capital market imperfections by subsidising education or by providing human capital loans [2].
Figure 4.
General training is paid for by the worker during the training period, where the worker typically receives a wage w<wu; wu is the wage of the untrained worker in the same firm. Training helps explain the convex age/earnings profile. Earnings rise quickly as the new skills are acquired, i.e. wt applies. Specific training is paid for by the firm since all benefits accrue to that firm alone. As in the above diagram, during training the worker receives a wage, wu, in excess of their Marginal Revenue Productivity (MRP). The employer is paying the worker more than his worth and is therefore losing out during the training period. However, once the skills have been acquired worker productivity is increased and the employer now gains as wu<MRPt. The employer may decide to pay an above competitive wage, w*t> wu, in order to reduce worker turnover.
The human capital model has been criticised by many economists on a number of points. First, the model assumes that all expenditures on education are investments. Blaug (1972) refutes this by saying that "a years schooling for someone, invariably shares both consumption and investment aspects."8 By ignoring these consumption aspects, empirical research underestimates the ROR on educational investments. Second, non-wage benefits are also omitted from the model. The fact that college graduates obtain generally more pleasant and interesting jobs than high school graduates is also omitted and again tends to underestimate the ROR on educational investments. Third, "screening" has become a contentious issue in educational economics. Proving what schooling actually does is very difficult indeed. It is not easy to distinguish between higher wages caused by increased education, or by the fact that by grading and labelling a student, it is easier and more efficient in finding jobs that are suitable to their skills. This means that the ROR is overestimated because higher earnings may be due to credentials rather than increased productivity. Fourth, the model does not deal with the tendency that people with more innate ability (higher IQs etc.) go to college more and that they tend to do better in labour markets. Again, the ROR is overestimated. The most fundamental problem lies in the assumption that human capital can be observed in measurable units. The standard used for measuring human capital is the number of years of schooling. This statistic is by no means conclusive due to the large variation in the quality of education. Wealthier families can afford better education and this quality can not be readily compared to the quality of education afforded by a poorer family. This along with the fact that students from wealthier families tend to invest more in terms of years of education , widens the gap between the average levels of human capital of the rich and the poor. This further increases the earnings differentials between the rich and the poor, distorting a nation's distribution of income.
Cultural differences are also used to explain human capital and earnings differentials. Different ethnic views and cultures can lead to systematic differences in utility functions that lead to behavioural differences among women. For instance, black wives have a higher labour force participation rate than white women due to greater marital instability among blacks, extended black family households, black husbands' lower wages and less stable employment. These are just illustrations to show the diversity of factors influencing human capital and earnings differentials.
The reality of the situation is that all individuals are not the same and do not possess the same skills and qualities. Individuals tend to be better at some activities than at others and it is these innate qualities that should be exploited. These innate qualities, such as cultural background, social class, religion and personal drive, are very hard to measure empirically, but it is these qualities that make separate us as individuals. Education, important as it is, is only one of a multitude of factors influencing the level of human capital.
[1] Note how the earnings differential between streams A and B increases over time.
[2] Siebert and Blaug (1985) in their studies of subsidies in education find that it is children from better-off families who tend to take up higher education subsidies. The targeting efficiency of subsidies is inadequate. Blaug is in favour of educational vouchers, whose value declines as parental income increases, as a means of evening-out the distribution of education among the different social classes.
[3] It is important to realise that pure general and pure specific training are only of theoretical use, as they are unrealistic in practice. Training of a mixed form is more likely.
Books
Blaug, M. (1972) : Introduction to the Economics of Education Harmondsworth, London.
McConnell, C. & Brue, S., (1989) : Contemporary
Labour Economics , McGraw-Hill, New York.
Mincer, Jacob (1974): Schooling, experience and earnings New York National Bureau of Economic Research, London.
Siebert, D. et al, (1985) : Surveys in Economics: Labour Economics.
Smith, A., 1776 : Wealth of Nations, Book I.