Chapter 11 Public Goods in Action–Education Jonathan Gruber Public Finance and Public Policy Aaron S. Yelowitz - Copyright 2005 © Worth Publishers

Introduction The single largest expenditure item for state and local governments is education, consuming about 30% of their budgets. The U.S. spends more per-pupil than nearly every country in the world, yet U.S. students perform only around the international average on reading, math and science ability. Figure 1 compares the U.S. to other nations in terms of money spent per pupil, and 8th grade math scores.

Figure 1

Introduction Although there is agreement that our educational system has problems, there is less agreement on the solutions. Some blame overly bureaucratic systems that are not subject to enough financial discipline for the trouble in education. Others believe inadequate resources are the problem.

Introduction A recent piece of major legislation concerning education was the No Child Left Behind Act of 2001. It requires states to test students in grades 3 through 8; the results are used to judge school performance. Underperforming schools are subject to various sanctions.

Introduction This lesson addresses numerous public finance issues in education policy: Why should the government be involved at all? How does public provision of education affect private education? How does school choice or vouchers affect efficiency? What is the return to education? What is the government’s role in higher education?

WHY SHOULD THE GOVERNMENT BE INVOLVED IN EDUCATION? The involvement of the government in elementary and secondary schooling is extensive: 90% of students in those grades are in public schools. Is there an economic rationale? Education is not a pure public good because it is: Rival: more children in a classroom lowers the quality of the experience. Excludable: private schools can decide which students to accept.

WHY SHOULD THE GOVERNMENT BE INVOLVED IN EDUCATION? There are potential positive externalities, however. These are related to: Productivity Citizenship Credit market failures Failure to maximize family utility

Productivity In general, the extra productivity from education is largely internalized. However, social benefits from higher productivity occur in two ways. The first is “spillovers” to other workers: a person’s productivity could raise the productivity of their coworkers, thus raising their wages and well being. The second is through taxes: if higher productivity is reflected in higher pay, then the government collects more tax revenues as a result.

Citizenship Education may make people more responsible citizens, by participating to a greater extent in the democratic process and reducing the likelihood of criminal activity. These arguments are fairly compelling for basic education such as elementary school, but provide less rationale for secondary or higher education.

Credit Market Failures Another justification may be credit market failures. In principle, a family could borrow against a child’s future earnings to finance the education, but unlike a home purchase, there is no collateral. The creditor cannot really observe if the child is a good risk to pay back the loan or not, and will generally offer too little credit in a situation like this.

Credit Market Failures The solution to credit market failures would be to make loans available to finance education, but the government in the U.S. (and elsewhere) do not take on this role except in the financing of higher education. Rather, the government supplies a fixed level of publicly-funded education.

Family Utility Even with well-functioning credit markets, there may still be failures to maximize family utility. Parents may not choose appropriate levels of education for their children; even with loans, private education would still involve some sacrifice on the part of parents, such as paying the cost of schooling not covered by the loans. Without public provision, some smart children would be penalized for having selfish parents.

Redistribution In addition to motivations based on positive externalities, another key justification for public provision involves redistribution. Public education provides a level playing field that promotes income mobility.

HOW IS THE GOVERNMENT INVOLVED IN EDUCATION? Typically, to deal with positive externalities, the government uses either the price mechanism or quantity mechanism. In the context of education, the price mechanism would involve offering discounts off private educational costs. The quantity mechanism would be to mandate that individuals obtain a certain level of education.

Free Public Education and Crowding Out In practice, the government does neither. Instead, it provides a fixed level (12 years) of education at no cost. A problem with the system of public education provision is that it may crowd out private education provision.

Free Public Education and Crowding Out In fact, Peltzman (1973) argued that it is possible that providing public education could lower the educational attainment in society. In Peltzman’s model some parents choose lower-quality public schools over higher-quality private schools in order to take advantage of the free education entitlement.

Free Public Education and Crowding Out The key insight of Peltzman’s model is that the typical provision of public education creates a budget constraint with potentially perverse incentives. In order to consume higher quality private schooling, the consumer must forgo the entitlement to free public education. Figure 2 illustrates this.

The children in family X now consume higher quality education. A consumer must forgo the free education if they want a higher quality school. The slope of the budget constraint is determined by the relative prices of education and other goods. Initially, without free public education, parents choose between educational spending (educational quality) and other goods. Other goods spending Family X spends little on education for their children and send their kids to a low-quality school, and spends much more on other goods. Family Y spends more on education and sends their children to a medium-quality school. Finally, family Z spends even more on education and sends their children to a high-quality school. ICX ICY ICZ The children in family X now consume higher quality education. Now suppose that the government provides free education that costs F. The children in family Y now consume lower quality education. While the decision of family Z is unchanged. EF Education spending Figure 2 Effects of free public education

Free Public Education and Crowding Out On net, public educational spending could crowd out private spending on education as individuals reduce their own spending in response to this free public option. Moreover, if enough families were like family Y in figure 2, then total educational quality could actually fall rather than rise.

Solving the Crowd-Out Problem: Vouchers Educational vouchers give parents a credit that can be used toward the cost of tuition at any school, public or private. Imagine a voucher of EF dollars that could be used for “free” public school, or for more expensive private schooling. Figure 3 illustrates this.

With a voucher, the budget constraint is different. You no longer have to forgo the public subsidy to get higher quality schools. The children in family X consume the same amount as before. Other goods C The children in family Y now consume more education, not less. Y2 The children in family Z now consume more than before. ICX Z2 ICY ICZ EF Education spending Figure 3 Effects of a voucher for $EF

Solving the Crowd-Out Problem: Vouchers The type of analysis shown in Figure 3 motivates support for a voucher system. The arguments made in favor of vouchers are: Consumer sovereignty Competition

Solving the crowd-out problem: Vouchers The first argument in favor of vouchers is that they enhance consumer sovereignty, by allowing individuals to match their educational choices with their tastes.

Solving the Crowd-Out Problem: Vouchers The second argument in favor of vouchers is that competitive pressures will allow the education market to work more efficiently. For example, real per-pupil spending has doubled since 1970, yet test scores have hardly changed. In addition, the number of administrative staff has grown by 50%, while the number of students has hardly changed.

Solving the Crowd-Out Problem: Vouchers In theory, vouchers put discipline on public schools by making private schools a more realistic option. Inefficient schools will be driven out of the education market, just as competition forces out inefficient firms. In principle, there is already some degree of competitive pressure through the Tiebout mechanism.

Problems with Educational Vouchers There are problems with vouchers, however. They may lead to excessive specialization. They could lead to stratification or segregation. They may be poorly targeted. The education market may not be competitive. They do not account well for high cost students.

Problems with Educational Vouchers As schools try to specialize and attract customers, the benefits of a common program will be undercut. In principle, “football schools” or “art schools” could emerge to attract particular market segments. These schools could give less attention to what are viewed as the central elements of education. Accreditation and testing could ameliorate this problem.

Problems with Educational Vouchers Vouchers have the potential to reintroduce segregation along many dimensions. This is more likely if children of disinterested or uninformed parents end up in low-quality public schools, while children of motivated parents end up in higher quality private schools. On the other hand, vouchers allow motivated students to end the segregation imposed by location.

Problems with Educational Vouchers Another potential problem with vouchers is they are inequitable. In Figure 3, family Z used the voucher, yet education attainment rose only modestly. For families like family Z, who don’t change their educational attainment by much, the voucher is basically the same as a cash transfer.

Problems with educational vouchers To the extent that vouchers can be targeted to families such as Y, (rather than Z), then educational spending goes up by a larger amount. In practice, this is difficult to do.

Problems with Educational Vouchers Another problem is that the education market might be described as a natural monopoly. This is because there are fixed costs (such as the building) and economies of scale. In such a case, it is not efficient to have many small schools competing with each other.

Problems with Educational Vouchers Finally, some students are more costly to educate. Special education students, for example, cost more than twice as much as the average student. Unless a voucher was commensurate with the cost of a student, schools have an incentive to cream-skim and avoid these costly students.

EVIDENCE ON COMPETITION IN EDUCATION MARKETS This section reviews the empirical evidence concerning the effects of competition in the education market. Rouse (1998) studied the Milwaukee voucher program, and found that the treatment group saw an increase in academic performance.

Estimating the effects of voucher programs Empirical Evidence The most convincing, direct evidence on vouchers comes from a small-scale voucher program in Milwaukee. Starting in 1990, the state of Wisconsin allowed near-poor families in Milwaukee to apply for a voucher for any nonsectarian private school.

Estimating the effects of voucher programs Empirical Evidence Although a comparison between families who use vouchers and those who do not would certainly introduce bias, Rouse (1998) notes that oversubscribed private schools had to select voucher applicants randomly, using a lottery. The “treatment group” are students who applied to an oversubscribed school and were randomly accepted into it; the “control group” are applicants who were randomly rejected. The results suggest that the treatment group saw an increase in academic performance, with a rise in test scores of 1-2% per year relative to the control group.

Estimating the effects of voucher programs Empirical Evidence In many developing countries, public schools may be of particularly low quality. Angrist, et al. (2003) studied a Columbian voucher program that paid for somewhat more than half of the costs of private secondary school. The vouchers were distributed by lottery, an ideal empirical approach. The authors found that students who won vouchers were 10% more likely to finish 8th grade, primarily because they did not repeat as many grades. They also had higher test scores. Overall, the vouchers cost $24 per winner, yet the benefits were between $36 and $300 per person per year, making this an enormously successful program.

Experience with Public School Choice Instead of offering vouchers, some school districts have allowed students to choose freely across public schools. In other cases, magnet schools were set up to attract talented students. And in other instances districts allow charter schools that are not subject to the same regulations as traditional ones.

Experience with Public School Choice Cullen, Jacob and Levitt (2003) examine school choice in Chicago, which also used a lottery to determine admission to oversubscribed schools. Unlike vouchers for private schools, going to a more selective public school did not influence academic outcomes.

Experience with Public School Incentives While there is limited U.S. experience with vouchers and school choice, there is a much larger experience with school accountability. As of 2002, 25 states explicitly linked student promotion to performance on state or local assessment tests. 18 states rewarded teachers and administrators on the basis of successful student exam performance. 20 states penalized teachers and administrators for sub-par performance.

Experience with Public School Incentives This approach was codified into law with the No Child Left Behind Act of 2001. There is some evidence that strong measures of accountability had the intended effect of raising test scores.

Experience with Public School Incentives On the other hand, this sort accountability had two unintended effects: Schools and teachers “teach to the test.” Schools may try to manipulate test scores by reclassifying low-skilled students as special-education, changing their cafeteria menus, and even cheating.

Bottom Line on Vouchers and School Choice There is little evidence to support the notion that public school choice has major beneficial effects on outcomes. Vouchers appear to improve academic performance for those who move to private schools, but raise serious equity issues.

MEASURING THE RETURNS TO EDUCATION Government must undertake some sort of cost-benefit analysis when devoting resources to education. The costs are fairly straightforward, but the benefits are much trickier. The return to education is the benefit that accrues to society when individuals get more schooling, or when they get schooling from a higher quality environment.

Effects of Education Levels on Productivity The most prominent question in the literature is how education affects productivity, as measured by wages. The literature shows that more education leads to higher wages, with a roughly 7% increase for each year of schooling.

Effects of Education on Productivity Whether this effect is causal is open to some debate. Intuition might suggest that education raises human capital, making a worker more productive and getting rewarded for that productivity in the market. On the other hand, in the screening model, education acts only as a mechanism of separating high and low ability individuals, but not actually improving skills.

Effects of Education on Productivity The implications for government policy are quite different for the two models. In the human capital model, education raises productivity and could generate positive externalities. In the screening model, there are no positive externalities; rather the returns are purely private.

Effects of Education on Productivity In practice, it is hard to separate the theories out, because it is true, underlying ability is unobserved. Researchers have used innovative techniques to control for the bias from underlying ability. The overall message of the literature is quite consistent: each year of education raises wages by 7 to 10%, after accounting for unobserved ability. This is strong evidence for the human capital model.

Estimating the return to education and evidence for screening Empirical Evidence A simple approach to estimate the return to a year of schooling in terms of higher wages is to compare people with more education (the treatment group) to those with less education (the control group). This approach likely suffers from omitted ability bias, however–more motivated students end up getting more education, and earn higher wages.

Estimating the return to education and evidence for screening Empirical Evidence One approach to control for this bias is to directly control for underlying ability. A researcher could include the following in a regression: Standardized test scores IQ Yet these are surely crude measures of ability, and do not take into account all of the omitted factors.

Estimating the return to education and evidence for screening Empirical Evidence Another approach uses quasi-experiments, with plausibly identical treatment and control groups, where the only difference is the amount of education received. Duflo (2004) studied a large-scale public school construction project in Indonesia between 1973 and 1978. More schools were opened in some areas than others. The treatment group of students lived in areas with more school construction; the control group lived in areas with less construction. The treatment group saw a larger rise in educational attainment, and much higher wages as adults.

Estimating the return to education and evidence for screening Empirical Evidence Another quasi-experiment uses the passage of mandatory schooling and child labor laws in the late 1800s and early 1900s. Before that time, there was no requirement for children to attend school, and no limits on child labor. Afterwards, there was a minimum age at which children could drop out of school, and there was a minimum number of years of education required before children could engage in full-time work. The laws varied across states.

Estimating the return to education and evidence for screening Empirical Evidence These laws significantly increased the level of education in the U.S. The treatment group were individuals born in states where the laws required more education. The control group were individuals born in states where the laws did not change. Although all of the approaches have limitations, they tend to find each year of education raises wages by 7 to 10%.

Estimating the return to education and evidence for screening Empirical Evidence Other studies have found evidence for a certain kind of screening, called the “sheepskin effect.” This effect refers to the fact that getting a degree from high school, college, or graduate school has a particularly high rate of return relative to obtaining the same amount of education but no degree. For example, students who passed the GED exam earned wages 10-19% higher than those with comparable scores who did not pass the exam. Other work has shown that if two individuals both have four years of college, but one doesn’t graduate, the one with the bachelor’s degree will earn 25% more, despite similar human capital.

The Impact of School Quality A smaller literature has examined the impact of higher quality schools on the returns to education. “Quality” is often defined as class size and spending per student.

Estimating the effects of school quality Empirical Evidence Experimental evidence from Tennessee suggests smaller class sizes lead to much higher test scores. Yet, a recent attempt to reduce class size in California did not have the expected positive effects, perhaps because other factors were changing too. For example, the rise in number of classrooms may have led the state to hire new, unqualified teachers.

THE ROLE OF THE GOVERNMENT IN HIGHER EDUCATION There is an enormous higher education sector in the U.S. Spending on higher education is about 40% of total educational spending. In contrast to other levels of education, higher education in the U.S. is viewed as an enormous success, with the best universities in the world and a vast inflow of foreign students.

THE ROLE OF THE GOVERNMENT IN HIGHER EDUCATION There are major differences in the degree of private provision and competition between higher education and primary/secondary education. In higher education, 23% of students attend private institutions, and individuals have free choice on which schools to apply to.

Current Government Role Figure 4 shows government involvement in higher education through: State provision Pell Grants Student loans Tax relief

Figure 4

What Is the Market Failure? The returns to higher education are mostly private. Even though studies show there are some public returns to college education, they are not large relative to government expenditures. The major motivation for government involvement is failure of credit markets.

What Is the Market Failure? The rationale for other sorts of government involvement (besides loans) are less clear: Pell Grants could be rationalized by the concern that low income individuals will avoid loans because of short-sighted fears about loan repayments. The rationale for state-subsidized universities is far less clear. Many college students are mobile and do not stay in the state after they finish college.

Recap of Public Goods in Action–Education Why should the government be involved in education? How is the government involved in education? Evidence on competition in education markets Measuring the returns to education The role of government in higher education