Earning and Learning: How Schools Matter

This useful edited volume is in two parts. One, summarized by Susan Mayer's Chapter 1, "From Learning to Earning," asks whether educational and economic success is determined primarily by student aptitudes, which are not under the control of the school, or by school-taught achievement. The second, introduced by Paul Peterson's Chapter 5, "School Reforms: How Much do they Matter?" explores the effects on academic achievement, of several well-known and controversial school reform strategies.

The chapters published in the first section were written partly in response to the publication of The Bell Curve (Free Press, 1994) by Richard Herrnstein and Charles Murray. The conclusion that those with more schooling tend to earn more is one of the strongest empirical findings in education research. There is much less consensus about why more educated people earn more. Herrnstein and Murray argued that for the most part, initial aptitude (or IQ) determined both success in school and in the workplace, thus equalizing schooling will not equalize income. The three substantive chapters in this section of the book all argue that what children learn in school does influence their subsequent earnings.

The fundamental methodological problem of research in this area involves sample selection. Comparisons of the earnings of individuals with different levels of education cannot measure the effect of that education because individuals with different underlying characteristics are selected or select themselves into different levels of schooling. Some of these characteristics cannot be measured, or are unknown, and therefore cannot be controlled for in regressions. Each of the authors uses a different approach to the selection problem. Christopher Jencks and Meredith Phillips measure the effect of learning that takes place between 10^th and 12^th grades, controlling for 10^th grade test scores. In their analysis, Christopher Winship and Sanders Korenman use a measure of cognitive aptitude as well as of the amount of schooling and find that schooling and aptitude both have independent effects on earnings, but they also reinforce each other. Therefore schooling also influences earnings by boosting aptitude which then influences earnings. Susan Meyer and David Knutsen argue that different state mandated school-starting ages create a natural experiment. Students are in effect randomly assigned to schools with different starting ages. They find that students who start school earlier eventually earn more both because they tend to accumulate more time in school and because, even for children with the same amount of schooling, those who start earlier apparently learn more. Thus schooling must influence earnings in addition to any aptitude effect since there are systematic differences in earnings for students randomly assigned to different schooling regimes.

The second section of the book is comprised of a series of chapters addressing the benefits of several controversial education reforms. These chapters measure the extent to which the reforms do or do not raise test scores. Mayer and Peterson then combine the two sections of the book by taking estimates of the effects of increases in cognitive achievement (as measured by test scores) on earnings and multiplying those by the effect of the reforms on test scores. This gives what they acknowledge is a very rough estimate of the economic benefit (as measured by wage increases) of the various reforms. Using estimates of the costs of those reforms they then calculate the ratio of the benefits to costs-this ratio can be referred to as the rate of return.

The most common education reform has been simply to increase the years of schooling. This will have a modest rate of return if it extends schooling at the upper end, but a very substantial return if it involves lowering the school-starting age. Requiring more rigorous math and science courses could potentially have a substantial rate of return depending on how much it would cost to hire the currently scarce math and science teachers. Research reviewed in this volume by Mosteller and by Hanushek suggests that a class-size reduction in kindergarten and first grade would have a substantial rate of return. The studies they cite have not found any additional benefits to class-size reduction in subsequent years so extending the policy throughout elementary school would only add costs and therefore reduce the rate of return. According to these authors, the introduction of rigorous statewide or national academic exams would have a substantial rate of return. Per-student costs would be low, and John Bishop argues that such tests create incentives for teachers and students that result in increases in academic achievement.

Chapters by Caroline Hoxby and by Peterson and coauthors argue that school choice also has the potential to raise test scores. This remains controversial, and Peterson has long been associated with a pro-voucher position. Nevertheless, Hoxby's chapter uses a novel approach to measuring the effects of choice by comparing school characteristics in metropolitan regions that have different numbers of school districts. If an area has many school districts, then parents can choose among a wider variety of schools when they decide where to live. Hoxby finds that in metropolitan areas where parents can choose more easily among districts, the schools tend to have more challenging curricula, stricter academic requirements, and more structured and discipline-oriented environments.

This collection is about important educational issues and uses a wide variety of data sets and sources of information. While it certainly will not resolve all of the controversies it addresses, it does present thoughtful analyses of the effects of schooling on earnings and of the effects of education reforms on educational outcomes. In some cases, the authors use imaginative approaches to overcoming difficult methodological problems. It will be a useful addition to the libraries of educators, parents, policy makers, and researchers, and it belongs on the reading lists of courses in educational policy and the economics of education.

Thomas Bailey is a professor at Teachers College, Columbia University.