Notes on Human Capital

by Alan Hyde(1)

The theory of human capital conceptualizes information as the product of investments in individuals' schooling and other training. It is a generally-accepted economic model, in the precise sense that it has been announced by prestigious economists at prestigious schools, and it doesn't seem worth any one's time to challenge it, since there is more to be gained in career terms by fitting one's findings into human capital's entirely vacuous categories.

The influence of human capital ways of thinking has been entirely negative. Its sole alleged achievement within economics, on which it has totally failed to deliver, is to model the economic returns on years of schooling or training. This project has been a failure, for the simple reason that there is no simple relationship between schooling and earnings. The value of any additional year of schooling always depends on exogenous factors concerning the affected individual and the economic opportunities open to him or her.

Within economic discourse, human capital theory functions primarily as a justification for inequality. The soul unlettered in economics recoils from the sight of two individuals, working similar hours at the same firm, one of whom makes ten or a hundred or a thousand times the earnings of the other. Human capital justifies this patent inequality, or any other, which can always--always-- be said to rest on "unobserved" and "unobservable" differences in the two individuals' human capital.

As incorporated into legal analysis, human capital theory has been equally apologetic, while contributing no explanatory power. In law, it is invoked on a sporadic, ad hoc basis. Human capital is not part of accounting conventions, for example. When AT&T lays off forty thousand trained knowlege workers, it does not write down any corporate assets; it merely deducts their salaries from what would otherwise be business expenses.(2) Human capital theory plays no role in tax law or accounting conventions, and its absence facilitates downsizing and layoffs over other possible ways of cutting corporate costs. Human capital theory is also, curiously, absent without official leave from employment discrimination cases in which educational requirements for particular jobs are struck down because of their "disparate impact" on different racial groups. Such educational requirements cannot be saved as incentives for further individual investments in human capital.

On the other hand, human capital theory lurks in the background of the justification of some otherwise indefensible features of employment law, such as restrictive covenants, employee promises to compensate employers for training, and trade secrets. Human capital theory stands as a major obstacle to our most pressing legal and economic need: understanding the economics of information, and specifically, how information has value when it is nobody's property.(3)

I. Human Capital's Supposed Economic Achievements

The achievements of the human capital school, according to its adherents, are that it explains why people go to college; that it predicts the increased earnings associated with additional schooling or training; and that it has redirected economic analysis of education and earnings to life cycle models and away from current wages.(4) The last of these is indeed an achievement, and human capital theory deserves whatever share of the credit is rightly its. Of course, it is not news that education pays out whatever benefit it has over the course of a lifetime. I seem to remember this as a theme of commencement addresses long before anyone had heard of human capital theory. However, in the world of labor economics, there has been a fruitful turn over the past few decade "toward life earnings rthaer than current wage rates as the empirical constructs of primary interest," and, to the extent this is attributable to thinking about human capital, this is a genuine achievement.(5)

The first two alleged achievements, however, are frauds. The problem of why people go to college is a problem, if at all, only to an economist, and, to the extent it is a problem, is not solved by human capital theory. It's hard to believe that anyone was walking around before the late 1950s baffled as to why people went to college, the reasons for which are many and varied. However, an economist who had never heard of human capital theory would have posited a taste for educational services. There would have been nothing to explain.

Human Capital theory, of course, explains going to college as an investment that will pay off in future earnings. This is just the formal assumption of the theory, and is not subject to empirical demonstration. Human capital theory has a "standard" regression that states that increasing schooling completion by a year will raise wages about 8 per cent.(6) Such an average could be derived from any data set of education and earnings--Mincer's was of 1960 Census Data; "For no obvious reason, similar analysis of more recent data has not appeared"(7)--but would tell us nothing. The problem, which is always identified by human capital theorists and promptly ignored, is that additional years of schooling contribute very different amounts to earnings, depending on which year of schooling it is, who is the recipient, and what is the recipient's job.

Which year of schooling is it? An amusing discovery has been that the returns to the twelfth grade, that is, the final year of American high school, are more than three times greater than the returns to completing the 11th grade.(8) It is impossible to explain this result using human capital theory. One would have to explain why that senior year in high school contibuted uniquely to each individual's intellectual development. It would be difficult to find a teacher or student who would support this account. In fact, there is little observable difference in job performance between 11th grade finishers and 12th grade finishers. It seems evident that finishing 12th grade is a signal of some sort, perhaps a valuable signal: 12th grade finishers do indeed have less propensity to quit or be absent than 11th grade finishers. In this case, differences in earnings are associated with increased schooling but could not reflect increased human capital.

Who is the recipient? Human capital analysts measure the effect of schooling on earnings without making any attempt to determine recipients' "initial" endowments and capacities or individual investments in human capital.(9) Obviously, you have to be pretty well-off to go to college at all, to stay out of the workforce while attending college, and to attend expensive, "quality" colleges.(10) It is hard to imagine the analysis of any other investment that ignored earlier or contemporaneous investments. Of course, human capital analysts are aware of this problem, and every book on human capital struggles at length to explain the differential effect of education on the earning of different social or racial groups. Something like this must be the right approach: education should be associated with increased earnings, but the amount will vary quite a bit, and it's worth finding out the key variables. Still, it seems utopian that one could draw any conclusions about whether any individual, let alone society, was making the right level of investment in human capital without knowing so many other variables that one would be just as well off relying on one's intuitions.

What is the recipient's job? "[T]he essence of human capital theory is that the job is unimportant. Wages and wealth are determined by the individuals' skills. Occupation and industry variables are secondary and almost an embarrassment to the theory. The fact that occupation and industry often enter significantly into wage equations is something that human capital theory rarely explains....This view is quite different from the one held by the typical businessperson. Deans, department heads, and personnel managers alike are generally aware of their slot allocations. While there may be some flexibility in tailoring jobs to individuals within the firm, the fac remains that managers often speak of hiring workers into given jobs."(11)

II. Human Capital Theory's Actual Achievements

Justification for inequality. Whole thing rests on unobserved human capital--they recognize they can't observe it.

Also justifies inequality as alternative explanation for lifetime jobs.

III. HumanCapital Theory's Influence on Law

A. Unexpected absences

1. Downsizing and layoffs

2. Discrimination and educational requirements

B. Unexpected presences

1. Covenants not to compete. Rubin and Shedd assume employer pays for human capital, contrary to basic assumption of the Becker school that this is shared between employer and employee. Rubin and Shedd assume contra, to give employer ownership claim. Hyde: ownership should reflect comparative advantage at creating value, mostly incentives for investment.

2. Agreements to pay training

3. Individualized conceptions of information

IV Proposals

A. Alternative 1: Get serious about human capital

B. Alternative 2: Collective models of innovation and information. Wiener etc.

1. Professor and Sidney Reitman Scholar, Rutgers. The State University of New Jersey, School of Law, Newark, NJ 07102-3192. Telephone: (973) 353-5463. Fax: (973) 353-1445. E-mail:

2. Tom DeMarco, Human Capital, Unmasked, The New York Times, April 14, 1996, at F 13, col. 5.

3. Alan Hyde, Real Human Capital: The Economics and Law of Shared Information (unpublished ms).

4. Mincer in Kerr & Staudohar [modify as necessary]

5. Sherwin Rosen, Human Capital: A Survey of Empirical Research, 1 Research in Labor Economics: An Annual Compilation of Research 3 (Ronald G. Ehrenberg ed. 1977).

6. See, e.g., Edward P. Lazear, Personnel Economics 78 (1995), citing Jacob Mincer, Schooling, Experience, and Earnings (1974).

7. Rosen, supra n.5, at 5.

8. Andrew Weiss, High School Graduation, Performance, and Wages, 96 J.Polit.Econ. 785 (1988).

9. See, e.g., Mincer, supra n., at 2.

10. This is the gist of Samuel Bowles & Herbert Gintis, The Problem with Human Capital Theory--A Marxian Critique, 65 Am.Econ.Rev. Papers & Proc. 74, 75 (1975)("human capital theory formally excludes the relevance of class and class conflict to the explication of labor market phenomena.") One doesn't have to be a Marxist to share this critique, however, or to join Bowles and Gintis in their definition of "class", or any other definition of "class." I prefer the more general critique made in text, that human capital theory is blind to any factors that might explain why education does not contribute uniformly to earnings across individuals.

11. Lazear, supra n., at 78.