Human capital

Behind the veil of human capital
Professor of Economics, University of Amsterdam

Human capital is one of the buzzwords of the knowledge society. But how do we value it? And what does it tell us about education and future wages? The Observer invited a leading expert in labour economics, Professor Joop Hartog, to explain.

Human capital has become a household word. The concept lay essentially dormant for two centuries after Adam Smith first introduced it, but it was kissed awake in the 1960s by American economists, such as Gary Becker and Jacob Mincer, who then exported it triumphantly around the world. Human capital is a beautifully unifying idea, facilitating the quantitative analyses that economists are so fond of.It is an excellent vehicle for framing policy discussions too, whether on schooling, training and labour market performance. But with the term there is also a tendency to see the labour market rather like one sees financial markets, although for humans rather than financial capital. This seriously distorts perceptions and may lead to damaging political decisions.According to a recent OECD report, Human Capital Investment: An International Comparison, human capital may be defined as the knowledge, skills, competence and other attributes embodied in individuals that are relevant to economic activity. Human capital is therefore a notion which captures the valuation of the attributes people invest in. But the main problem is not so much how to define human capital as how to measure it. And for that, we have to look at financial returns, which are defined broadly as the increase in earnings for every year that was spent in school.Extensive work has been carried out to estimate the return on investment in human capital throughout the world. It has thrown up some interesting findings. Investment in formal schooling tends to have a rate of return of 5–15% in additional earnings for every year in school. The returns are higher for private individuals than for society, because schooling is mostly subsidised and hence individuals do not pay the full price for their education.The rate of return is different for each level of education: primary education generates the highest returns, secondary level has lower returns and third level education has higher returns than secondary education. Also, the rates of return on investing in human capital are higher in developing economies than in developed ones. In recent years, the relative wage of university graduates has risen strongly for several countries, particularly the US and the UK. Many studies suggest that this has to do with technological progress favouring the demand for higher education. Others suggest that the relative strength of the wage in these economies rose because the wage of the low skilled has fallen with the shift of low skill industries to developing countries.Looking only at earnings gaps generated by schooling hides the fact that individuals in different levels of schooling have different abilities. When the evidence is adjusted for this, it shows that about a third of the returns on investment in human capital is in fact a return on higher intellectual capacity. Despite difficulties in getting the data, there is some evidence that cognitive abilities make a contribution too, as do social abilities and manual skills. Research in psychology shows us that personality factors also have an effect in terms of additional earnings. But, when it comes to predicting an individual’s earnings, abilities like these play a minor role in generating a return compared with the actual years spent at school.When over-education is a good thingA clearer picture of the returns on schooling can be gained by looking at the type of job an individual takes up after graduation. Not so long ago the main policy concern used to be about ‘over-education’. The question everyone asked was whether there would be enough high quality jobs around to match the increased level of education in the labour force. That question drove research in several countries, and some consistent patterns have emerged.Suppose we characterise a job by the level of education required to get it. The level would be identified, for example, by referring to the qualifications specified by trained job analysts. We then can compare an individual’s education to the education required for the job, and also measure possible gaps: ‘over-education’ (more than required) and ‘under-education’ (less than required). Interestingly, according to the research, the years spent in ‘over-education’ always bring a positive return on investment. The return may be less than that on each of the minimum schooling years required, but it is a fair return nonetheless, of at least 4% a year in real terms. The finding emphasises the fact that the labour market will find a way to extract value from schooling years. ‘Over-education’ is not strictly speaking an inefficiency.Conversely, if an individual brings insufficient education to the job, there is a penalty, but the penalty is smaller than the returns the individual has already made on the years spent in education. This emphasises the importance of basic schooling. A recent study on Portuguese men will help to illustrate. For a job requiring exactly the number of school years that the individual actually has, the returns were estimated at 13% in additional wages per year of schooling. For the years above the required minimum the return is 10% per year. In other words, if an individual has 16 years of schooling and the job only requires 14, the returns are 13% for the first 14 years and 10% for each of the last two. If the individual has a job that requires more education, the penalty for not having the required amount is 8% per year. That means if you bring 12 years to a job requiring 14, you get 13% for each of the first 12 years of training and only 5% for each of the last two years (13% less the penalty of 8%).The relationship between wages and schooling is depicted in chart 1. The solid line gives the compensation for jobs matching a given education. The dashed lines link to a job requiring a particular level of schooling. Bringing more education to a given job moves you up along the dashed line to the right, giving you a higher financial return on your investment. Bringing less education into the job moves you down along the dashed line to the left, which illustrates a lower return. And under-education?Remember that this thinking developed out of a concern about ‘over-education’. But policy concerns today are more about under-education. The question now is whether people are sufficiently equipped for the knowledge society. A simple conclusion from the research findings presented above is that a lack of formal education can be compensated for later on, thanks to personal attributes, on-the-job training, experience and so on. As the chart indicates, a lack of formal education does not generate complete punishment in wages, and presumably not in productivity either. If, with a given education, you work in a job requiring more education, you get paid more than if you work in a job just matching your education. But, of course, this is only a crude picture, and perhaps the key worries concern the bottom end of the labour market. For the lower skilled at least, a good start in learning would appear essential to improving educational and job chances in later life (see Observer 214, Pathways to Lifelong Learning, October-November 1998).In many cases on-the-job training can compensate for any lack of formal education. It can also be applied to upgrade the workforce when technological change demands it. The returns of on-the-job training are commonly found to be quite high, at easily over 15%. That might suggest under-provision, since if it were so profitable, firms would devote more resources to it. It throws into question any need for fiscal incentives, since if there is such a high return on businesses investing in their own training, why subsidise? The problem with on-the-job training is precisely what generates those high returns. Perhaps it is not just the training, but its combination with innovations in work organisation and new technologies which is responsible for the high return. And training is often given selectively to individuals who are strong performers anyway. It is arguable that the same training would provide the same high return with lower skilled workers, although it would probably generate some return. These questions have not yet been answered.The US paradoxHuman capital is like a mystery bag of groceries: we know the price (what we paid for in earnings forgone when in school) and the rewards (higher wages), but we do not quite know what groceries are in the bag. In other words, we are not absolutely sure what it is we learn at school that makes us more economically valuable. Researchers have tried to find out.One of the more exciting projects is the International Adult Literacy Survey (IALS) which the OECD carried out jointly with Statistics Canada. It attempts to measure individuals’ information processing skills directly, by applying an identical survey to individuals in different countries. The skills are differentiated as digesting prose (newspapers), documents (government forms, maps and tables) and quantitative information (applying mathematical operations as instructed in printed material). The survey reveals some interesting results. For a start, the average skill level of the population varies very little across countries (with the exception of Poland and Sweden). But the dispersion is very unequal. Few would doubt the quality of the top universities and schools in North America as being among the strongest in the world. But a low education level in the USA and in Canada implies a much lower skill level than in other countries. European welfare and redistribution arguments might not be enough to explain the different income gaps. Indeed, the survey’s results indicate that a low education level in the United States, in particular, is simply a poorer bag of groceries than its counterpart in Europe. OECD Bibliography
  • Human Capital Investment: An International Comparison, 1998, Centre for Educational Research and Innovation, OECD 1998.
  • Literacy Skills for the Knowledge Society: Further Results of the First International Adult Literacy Survey, OECD/Statistics Canada, 1997
  • Literacy, Economy and Society: Results of the First International Adult Literacy Survey, OECD/Statistics Canada, 1995.
  • ‘Pathways to Lifelong Learning’, The OECD Observer, No. 214, October/November 1998
Further reading:
  • Hartog, J., ‘Over-education and Earnings: Where Are We, Where Should We Go?’, special issue of Economic Education Review, forthcoming.
  • Hartog, J., ‘On Returns on Education: Walking along the Hills of ORU Land’, forthcoming in Heyke, H. (ed.), Education, Training and Employment in the Knowledge-based Economy, MacMillan, London, forthcoming.
  • Hartog, J., ‘On Human Capital and Individual Capabilities, Universiteit van Amsterdam, FEE, 1997.
  • Leuven, E., H. Oosterbeek and H. van Ophem, ‘International Comparisons of Male Wage Inequality: Are the Findings Robusts,’, Working paper, Universiteit van Amsterdam, FEE, 1997.
©OECD Observer No 215 January 1999

Economic data

GDP growth: -9.8% Q2/Q1 2020 2020
Consumer price inflation: 1.3% Sep 2020 annual
Trade (G20): -17.7% exp, -16.7% imp, Q2/Q1 2020
Unemployment: 7.3% Sep 2020
Last update: 10 Nov 2020

OECD Observer Newsletter

Stay up-to-date with the latest news from the OECD by signing up for our e-newsletter :

Twitter feed

Digital Editions

Don't miss

Most Popular Articles

NOTE: All signed articles in the OECD Observer express the opinions of the authors
and do not necessarily represent the official views of OECD member countries.

All rights reserved. OECD 2020