Inequality in America: What Role for Human Capital Policies?

Business Economics, April, 2004 by W. Steven Barnett

Inequality in America: What Role for Human Capital Policies?

By James J. Heckman and Alan B. Krueger. 2004. Cambridge, MA: MIT Press. Pp. 370. $40.00 hardcover.

From time to time, the economics deities gather on Mt. Olympus and battle. This battle, the third Alvin Hansen Symposium at Harvard, is a debate on the role of human capital policies in response to inequality. Nobel Laureate James Heckman and Pedro Carneiro (both of the University of Chicago) cast thunderbolts from the sky--radically revising human capital theory and advancing a strong, new interpretation of the empirical evidence. Alan Krueger (Bendheim Professor of Economics and Public Affairs at Princeton) offers a strongly argued alternative that yields remarkably different policy recommendations given substantial agreement about both theory and the relevant empirical evidence.

Krueger's 75-page essay "Inequality: Too Much of a Good Thing" opens the book. He sets out the basic facts regarding changes in the distribution of income and earnings and discusses various perspectives on the nature, consequences, and causes of inequality. He evaluates evidence on the returns to investments in human capital from preschool to college and proposes policies to reduce though not eliminate inequality. His essay is admirably broad and well worth the time of anyone who wishes to become acquainted with contemporary research on inequality and human capital development. However, the 163-page treatise by Carneiro and Heckman that follows is the heart of this volume. Their tour de force opens the debate, attacking much of the conventional wisdom regarding human capital theory, research, and policy.

Carneiro and Heckman argue first that because "learning begets learning" investments in human capital are more productive both earlier in the lifecycle and for higher-ability individuals. By extension, returns to schooling are higher for children from higher income, more educated families. Second, human capital theory and policy must recognize the dynamic interactions among families, schools, and businesses in the development of human capital over the lifecycle. They state (313), "Schools work with what families give them. Job training programs work with what schools and families give them." Third, non-cognitive skills (such as motivation, leadership, honesty, and social skills) are at least as important as the cognitive skills that are most often the sole focus of human capital theory, research, and policy. Both types of skills are forms of human capital and not rivals of human capital, as models employed to analyze policy frequently assume.

Carneiero and Heckman begin by reviewing the data on inequality, but they focus more attention on inequality in schooling and skills, including the early origins of these inequalities. They also review evidence for the effectiveness of policies and programs across the age span. They conclude (90) that "at current levels of spending in the American economy ... the return to investment in the young is apparently quite high; the return to investment in the old and less able is quite low." Their graphics portray the return to human capital investments as falling below an acceptable rate just after the preschool years. In depth, their views have more nuances. Cognitive ability is much less malleable beyond the early years, while non-cognitive abilities can be significantly altered well into the adolescent years.

Reviewing specific programs, Carneiro and Heckman find that preschool education is highly effective, although with more impact on non-cognitive than cognitive abilities. Schools are much less productive, and returns are low to increased investments in K-12 education in the form of higher salaries, smaller classes, and so forth. They suggest, that structural changes that increase school choice and competition should have higher returns, but are careful to note that returns to increased investment in schools are limited by what families contribute to the production process. They also conclude that added investments in job training and higher education have low rates of return, particularly for lower ability adolescents and adults. Tax and immigration policies are deemed unlikely to reduce inequality in skills much.

Carneiro and Heckman note some exceptions to their general thesis. They find short-term credit constraints limit college attendance for up to eight percent of the population. Financial programs targeting these families could pay off. School-based interventions with adolescents still enrolled (but not dropouts) have substantial impacts, particularly on non-cognitive skills. Private job training has a relatively high rate of return but does not reduce (and may increase) inequality. Public policies providing training in the classroom have high rates of return, though other public job training programs do not.

The exceptions are taken to prove the rule, because they are relevant to so few people and have such small impacts on inequality. Carneiro and Heckman conclude (123): "the ability that is decisive in producing schooling differentials is shaped early in life. If we are to substantially eliminate ethnic and income differentials in schooling, we must start early." Thus, a "serious reformulation of human capital policy is needed." Yet, even with the benefit of a response and rejoinder, the reader is left with only hints as to what this might entail beyond rejection of most current policies and movement toward choice, competition, and local incentives in schools. Most intriguing are their statements that emphasize early family policy (135): "Good families promote cognitive, social, and behavioral skills. Bad families do not. The relevant policy issue is to determine what interventions in bad families are successful."