October 2000
Federal Reserve Bank of Dallas

Competition in Education

Society clearly benefits when businesses compete. Competition forces firms to innovate and adopt least-cost methods of production. It rewards efficient producers and punishes inefficient ones. As such, competition is a key to economic prosperity in a market economy.

There is an increasing body of evidence that competition is also a key to improving our educational system. I'd like to talk today about that evidence, about policy initiatives that can foster increased competition, and about the complications that are frequently swept under the rug in the school choice debate.

Evidence
School competition has attracted a lot of attention from economists in recent years. Because choice cannot be bad in traditional economic models, researchers take it as given that increased choice benefits families that choose the newly available options. Rather than try to quantify these individual benefits of choice, researchers have focused instead on the question "Can competition do for schools what it does for business?" The answer is yes. Almost across the board, researchers have found that competition is good for public schools.

• Academic outcomes are better. Researchers found that students who attended school in communities where student enrollment is dispersed among many educational providers subsequently scored higher on standardized tests, completed more years of schooling, and earned higher wages. (For additional information, see "The Evidence on Government Competition" by Lori L. Taylor, Economic and Financial Review, Second Quarter 2000.)
• Average per pupil expenditures by the public school system are substantially lower in states and communities where there are more public school districts to choose from.
• The bottom line is that an environment where schools must compete with other educational providers to attract students forces school districts to get more bang for their buck, or in other words, to use their resources more efficiently. For example, my own work on Texas suggests that school districts in highly competitive parts of the state have been up to twice as efficient as other Texas districts.

Interestingly, the research suggests that the source of the competition doesn't seem to matter very much. In terms of providing market discipline, another public school can be just as effective a competitor as a private school. Furthermore, it is not necessary that the other school be exemplary to be an effective competitor. In fact, even mediocre competitors can induce dramatic changes. For a local example, consider what happened when the Sherman Independent School District decided to foster a little internal competition and brought in a private firm to run a few of its campuses. Many believe that the private firm did no better than the rest of the school district, and that it was not especially innovative in its practices. The management contract was not renewed. Yet, the district was dramatically changed by the experience. Sherman's assistant superintendent is on record as stating his belief that the much maligned competitor's arrival speeded the district's reform effort by sending "a signal to the whole community that things were going to be different." Referring to a list of recent improvements to the town's schools (ranging from new academic standards to renovations in all buildings) he said "I do not believe we would have that had Edison not come into our schools. We'd still be writing on chalkboards and showing old film strips."

Policy
If increased school competition could improve the public school system, then the next obvious question is "How do we do that?" Three basic strategies for increasing school competition are charter schools, vouchers and standardized testing. The common thread among all these strategies is that they make it easier for parents to make educational choices.

Charter Schools. Charter schools are the most common policy prescription for increasing competition within the public sector. Charters are public schools that are generally subject to fewer regulations than traditional public schools. Most, if not all, charter schools are open-enrollment schools, meaning that the schools accept students from a wide geographic area. In contrast, traditional public schools generally only enroll those students who live within narrowly defined attendance zones. Depending on state law, charter schools can be run by for-profit firms or non-profit agencies. For-profit firms generally hold their charters indirectly under the auspices of a university or similar institution. Charter schools have no tax base or local taxing authority and derive their funding from the state and/or from the school districts their students would otherwise attend. Frequently, charter schools receive less revenue per pupil than a traditional public school. In Arizona the state with the most extensive charter school program, charters are funded at about 80 percent of their district peers. In Texas the money follows the child so that the charter school receives the same funding per pupil as the district he would otherwise attend.

The charter school movement started very small in 1991 and remains very small in most of the nation. Seventeen states have no charter schools, and in only six states do charters comprise more than two percent of public school enrollment. Arizona and the District of Columbia are the only states where charters have double-digit enrollment shares.

Charters may be small but they are also increasingly influential. Although charter school enrollments still account for less than one percent of the Texas public school system, the enrollment growth has been explosive. In 1999-2000, Texas charters enrolled nearly 35,000 students. Early estimates for the current school year are just over 38,000.

Vouchers and Tax Preferences. Vouchers, tax credits and tax deductions are the most common policy prescription for increasing competition from the private sector. However, these strategies are more popular in theory than in practice. Only five states have operational voucher plans (Florida, Maine, Ohio, Vermont, and Wisconsin) and only four states offer tax credits or deductions (Arizona, Illinois, Iowa, and Minnesota). The Ohio plan has been declared unconstitutional, but continues to operate on appeal.

The major differences between the voucher plans and the tax preferences are the scope of eligibility and the magnitude of the subsidy. The voucher plans cover tuition, but eligibility is limited to targeted groups such as low income students in Milwaukee or students in parts of Maine where there are no public schools. The tax credits and deductions are available to all parents, but the dollar amounts are modest. The Minnesota tax preference is the most generous, offering a $1000 refundable tax credit to families with incomes below $37,500.

Of course, public voucher plans are not the only voucher-based mechanism for increasing private sector competition. A number of private foundations have set up scholarship programs across the county. The nation's first fully-funded voucher program that is offered to every family in a district has been up and running for a couple of years in San Antonio. Children First CEO America offered every low income child in the Edgewood Independent School District a voucher to attend the public or private school of his choice. During the program's first year (1998-99), roughly 800 of the 13,000 eligible students elected to leave Edgewood I.S.D. for private schools. The average voucher payment was less than one-third of the average Edgewood expenditure per pupil. As one would expect, Edgewood responded to the competitive threat (and the loss of 3.5 percent of its revenue) by changing its policies. In the wake of the CEO vouchers, Edgewood I.S.D. instituted an inter-district public school choice plan that attracted 200 students from other area districts, and commissioned a $120,000 management study to improve its administrative efficiency.

Standardized Testing. Although it's easy to go overboard on the issue, standardized testing clearly enhances school competition. The simple act of publishing information about student performance raises the competitive threat as the parents of current and prospective students discover that the neighborhood school isn't quite as strong as they thought.

There are a number of innovative people out there that can act like an educational version of Consumer Reports, informing parents about effective ways to extract information from reams of data about schools. For example, Tom Luce's Just-4-the-Kids organization lets parents in Texas see how their neighborhood school performs, not only in absolute terms but also in comparison to the best practice of peer schools with similar student characteristics. However, without testing, such comparisons are impossible.

In addition to its benefits to parents, testing is also a key to social accountability. To the extent that the government has money on the table, it also has a legitimate interest in monitoring the outcomes of the educational process. Without testing, the government has no mechanism for monitoring outcomes and may be tempted to regulate the educational process instead. Although one could argue that private information sources would crop up to serve the parents' need for information, society's need for accountability implies that schools receiving public funds must test and publish the results. The tests need not be national in scope, but they must make it possible to make local, apples-to-apples comparisons.

Complications
Of course, nothing is as easy as it looks. There are a number of complications that must be considered in pursuing a policy of increased educational competition.

Highly Competitive Markets. Despite the strong support for competition as a force for good in education, the researchers do sound a cautionary note. Not all markets suffer from a lack of competition. It appears that most educational markets would benefit from increased competition but some are already highly competitive and would be little changed by an increase in competition.

In Texas, most educational markets are highly concentrated and would benefit from increased competition. However, the Dallas and Houston metropolitan areas appear to already be highly competitive education markets. In 1999, the Dallas metropolitan area had 85 public school districts and 78 private school systems to choose from; Houston had 61 public school districts and 83 private school systems. Increased school competition in these metro areas is unlikely to create new incentives for change or to improve the quality of education in general. If having to compete for enrollments with more than 160 other educational providers hasn't already provided the Dallas Independent School District with strong incentives to behave efficiently, a few more competitors aren't going to do the trick.

Markets where a lack of competition is not the problem can contain a large share of the student population. Dallas and Houston are home to over one-third of the school children in Texas. Therefore, the analyses suggest that fostering increased competition is only a partial answer to the question of creating "schools that work."

Conflicting policy objectives. Enhancing competition is not necessarily the primary objective of charter schools or voucher programs. As such, it is not surprising that from a competitive perspective, not all plans are created equal.

For example, the plan may be focused on a narrow segment of the student population. In Texas and Arizona, the charter school programs favor schools that serve students who are at risk of dropping out of school. While such a focus meets the needs of the targeted population, it can diminish the competitive impact of the charter school by segmenting the market for students. A survey of Texas school superintendents indicates that, rather than seeing charter schools as competitors to whom they must respond, a number of superintendents see them as a benefit to the district because they provide an alternative for difficult to educate, disaffected, or disruptive students.

Alternatively, some plans have virtually no impact on competition because they have only a negligible impact on the relative price of schooling. Consider, for example, Iowa. All parents are eligible for a $250 tax credit per child for educational expenses at the public, private, or religious school of their choice. While such a plan is attractively equitable, it is unlikely to change parental behavior very much. With the exception of a few parents teetering on the brink of sending their children to private schools, the primary beneficiaries of the credit are parents of existing private school students. For these parents, the credit may be a welcome relief but it is unlikely to cause them to move their child to a different school. If the policy doesn't at least threaten to change enrollment patterns, it doesn't increase the competitive pressure.

Student segregation. When people talk about the great melting pot of American society, they frequently mean the public school system. Many argue that the traditional public school provides a commonality of experience that socializes children and creates a cohesive society. Economic models predict that increased school choice should lead to increasingly homogeneous classrooms, making many believe that increased student segregation, a loss of social cohesion and "cream-skimming" by selective private schools are major potential costs of increased competition.

The evidence is mixed, however. The "melting pot" virtues of the existing system are probably overstated. There is little commonality of practice among public schools in the United States, so it is hard to argue that the system provides a common educational experience. Furthermore, there is a lot of classroom homogeneity in the current system. The courts have tried to stamp out racial segregation, but any system of neighborhood schools clusters kids according to family income. Second, this argument heavily discounts the potential benefits from specialized instruction. The Texas school system has been deliberately set up to favor charter schools that serve students at risk of dropping out of school. The benefits of providing these students with an alternative educational environment should not be overlooked.

Finally, there is the fear that the best and brightest of the public school system will be siphoned off by vouchers or charters, leaving only the weaker students in public schools. This is a substantive concern, particularly given the apparent benefits that children receive from taking classes with high-achieving peers. However, at least in Texas, there is little evidence that charters and vouchers skim the cream out of the public school system. Voucher children in Edgewood I.S.D. were middle-of-the- road academically, low income and ethnically similar to the district as a whole. Charter school enrollments vary according to the mission of the charters, but as a whole charter school children in Texas are similar to traditional school children in term of socioeconomic background while disproportionately black and Hispanic.

Regulatory burden. Competition creates incentives for school districts to behave efficiently, yet those incentives are moot if the regulatory environment prevents school districts from responding to those incentives. Without judicious deregulation of the public school system, many of the social benefits from increased educational competition would be lost. In Texas the regulatory burden has lightened as the state has emphasized accountability and competition. However, regulations are still the most plausible explanation for a significant part of school district inefficiency.

Conclusions
The evidence is clear—competition makes public schools better. Furthermore, a well designed policy can greatly enhance the competitive environment. However, not all markets suffer from a lack of competition, and the benefits of competition can be lost to a poorly designed policy or a regulatory environment that prevents schools from responding to competitive pressure.

—Lori L. Taylor

About In Depth This article is based on a presentation by Lori L. Taylor, senior economist and policy advisor, Research Department, Federal Reserve Bank of Dallas. The views expressed are those of the authors and do not necessarily reflect the positions of the Federal Reserve Bank of Dallas or the Federal Reserve System.