This file is also available in Adobe Acrobat PDF format
IN THE UNITED STATES TODAY, many different education reforms compete for the attention of political leaders, policy makers, parents, and school officials. Charter schools constitute one of the most widespread and important of these. Since Minnesota passed the first charter-school law in 1991, forty-two states, including the District of Columbia, have passed similar legislation, and thirty-seven of these have operating charter schools (WestEd 2003, 1). As of April 2006, there were over 3,500 charter schools serving over 1 million students nationwide (Center for Education Reform 2006),1 up from only 100 schools in 1995 (Research Policy Practice International 2001).
There are many excellent explorations of the history and the politics of the charter-school “movement,” and there is no need for us to repeat that information in detail here.2 But there is a series of fundamental questions and definitions that we need to address before moving on to the core of our analysis. First, there is the question of exactly what a charter school is. One interesting characteristic of most extant definitions is that, while they identify the core characteristics of charter schools, they almost always include language about not only what charter schools are but also glowing language about what they will do—that is, they often portray the hope of charter schools, while adding a bit of hype. For example, according to the U.S. Department of Education’s Office of Innovation and Improvement:
The promise charter schools hold for public school innovation and reform lies in an unprecedented combination of freedom and accountability. Underwritten with public funds but run independently, charter schools are free from a range of state laws and district policies stipulating what and how they teach, where they can spend their money, and who they can hire and fire. In return, they are held strictly accountable for their academic and financial performance. (U. S. Department of Education 2004, 1)
Putting aside the value-laden aspects of this description, this excerpt helps to identify the key structural characteristics of charter schools. Broadly defined, charter schools are publicly funded schools that are granted significant autonomy in curriculum and governance in return for greater accountability. In addition, the charter establishing a school is, ideally, a performance contract that details the school’s mission, its program and goals, the population served, and ways to assess success (or failure). Charters are granted for fixed lengths of time (usually three to five years), at which time the body that authorized the charter reviews the performance of the school and decides whether or not to renew it.3 We should note that almost every aspect of the legislation governing charter schools—such as the length of charters, who can apply for charters, and the like—can and does vary widely from state to state, so there are exceptions to almost every general statement describing charter schools (Kucsova and Buckley 2004).
While there are many perspectives on charter schools and while specific practices may vary widely across the nation, it is important to note that there is general agreement about what charter schools are supposed to do. The theoretical or intellectual basis for these expectations stems from the pro-market orientation that has helped to shape education policy reform in the United States for the last fifteen or so years. From this perspective, charter schools are expected to generate competition among schools, a competition in which poorly performing schools disappear and good ones prosper. Moreover, by reducing the level of bureaucratic regulation and control, charter schools should be free to innovate and create more effective and efficient programs that serve their clientele better. According to the U.S. Charter Schools web site,4 charter schools are designed to:
- Increase opportunities for learning and access to quality education for all students;
- Create choice for parents and students within the public school system;
- Provide a system of accountability for results in public education;
- Encourage innovative teaching practices;
- Create new professional opportunities for teachers;
- Encourage community and parent involvement in public education;
- Leverage improved public education broadly.
Or, once again according to the Department of Education’s Office of Innovation and Improvement, charter schools are free to:
lengthen the school day, mix grades, require dress codes, put teachers on their school boards, double up instruction in core subject areas like math or reading, make parents genuine partners in family-style school cultures, adopt any instructional practice that will help achieve their missions—free, in short, to do whatever it takes to build the skills, knowledge, and character traits their students need to succeed in today’s world.
By allowing citizens to start new public schools with this kind of autonomy, making them available tuition-free to any student, and holding them account able for results and family satisfaction, proponents hope that this new mix of choice and accountability will not only provide students stronger learning programs than local alternatives, but will also stimulate improvement of the existing public education system. With charter schools, it is accountability that makes freedom promising. No charter is permanent; it must be renewed—or revoked—at regular intervals. Continued funding, which is tied to student enrollment, also depends on educational results. (2004, 1)
Note that in this vision, the charter schools themselves are not only better schools, they also “stimulate improvement of the existing public education system.” To the extent that these processes—innovation, competition, leverage, accountability—work, charter schools help create a “market” for schools that produces better schools and increases “productive efficiency.”5
The theory behind charter schools holds another promise: charter schools should also increase “consumer” satisfaction. By being more responsive to the needs of parents, students, and the community at large, and by allowing parents and students to choose schools that deliver the type of education that they feel best meets their needs, charter schools can improve the match between what schools offer and what “parent/ consumers” prefer. We can think of this tighter match as improving “allocative efficiency” (see Schneider, Teske, and Marschall 2000).
While charter schools have found fertile ground in the market approach to policy reform, support for charter schools as a form of school choice has also grown for political reasons. In the competition between vouchers and charter schools, two contending ideas that use choice as a means to reform schools, state legislators have usually chosen to adopt charter schools. As Henig et al. (2002) explain, “[F]rustrated with the poor performance of public schools, but unable to advance more radical notions of privatization, such as vouchers, advocates saw charter schools as the most effective means of building market incentives.” (Also see Mintrom 1997. For more on the difference between charter schools and voucher schools see Hassel 1999.)
Despite the strong forces propelling them forward and a historically high growth rate, charter schools appear to be moving from the growth spurt of adolescence to a more sedate adulthood. Nationwide, there are signs that, while the number of charter schools is still growing, this expansion is now occurring at a decreasing rate. This may show that charter schools are reaching the normal maturation point of the typical sigmoid curve describing the diffusion of many innovations (Porter 1980; Schneider, Teske, and Mintrom 1995, chap. 4). For charter schools, this pattern may be driven by a variety of factors, such as the depletion of the population of reformers with the skills and energy needed to pioneer charter schools (Wells 2002, 1–2). It may also be driven by a particular temporal sequence: charter schools in some states have now been operating for five or more years, and there is growing pressure for a critical assessment of their academic and financial performance, perhaps slowing down their continued expansion. For example, in Massachusetts, where charter schools have existed since 1995, the state legislature in 2004 threatened a moratorium on further charter expansion, reversing itself only after reducing state funding for the schools (Greenberger 2004a, 2004b). In Washington, D.C., the Board of Education, one of the two chartering authorities in the city, has been approving less than 20 percent of applications before it in the last few years, while in the early years of the city’s charter-school experience it approved all applications (see chapter 2).
THE RATIONALE(S) FOR CHARTER SCHOOLS
Despite this slowdown in the rate of expansion, charter schools are clearly a popular policy. But are they a good one?
Charles Lindblom once wrote that the test of a good policy is that diverse actors can agree on it even if they cannot agree on why (Lindblom 1959). By this criterion, charter schools seem to be good policy; indeed, there sometimes seems to be as many arguments in favor of charter schools as there are schools themselves. Amy Stuart Wells (2002) argues that this collection of rationales can be boiled down into three main categories: standards-based or systemic reform, local autonomy, and neoliberal market-based reform.
Central to the appeal of charter schools for the proponents of systemic reform is the promise of increased accountability (Miron and Nelson 2000). In theory, charter schools, in exchange for the ability to operate relatively unencumbered by the policies of the local school district, are held accountable for improving the academic performance of their students. Thus charter schools can be seen as part of the broader trend in the last decade to reconstruct educational policy centered on measures of academic outcomes (Elmore, Abelmann, and Fuhrman 1996, 65–98)—a trend that is perhaps best exemplified by the provisions of the national No Child Left Behind Act of 2001, which has affected educational practices throughout the United States.
In reality, however, this autonomy-for-accountability trade-off can be problematic, as administrators find themselves in the ambiguous position of holding schools accountable while not directly supervising them (Wells et al. 2002, 29–53). In addition, charter schools have clear incentives to seek a continuing stay of evaluation or to obfuscate evidence of poor performance (Henig 1994, 234). Moreover, state rules and regulations governing the release of school performance data are often lax, and this has allowed charter schools (and, to be sure, other schools of choice) far too much latitude regarding the disclosure of the information parents and students need to make choices and policy makers need to evaluate the schools. In a recent report on charter schools in Ohio, for example, the Legislative Office on Educational Oversight found that many charter schools “are not reporting data that allow them to be compared to their contracts” (Legislative Office Of Education Oversight 2003, 26). Consider, too, Van Dunk and Dickman’s analysis of schools of choice in Milwaukee:
There is no question that there are good private voucher schools here, just as there are good public schools. Unfortunately, however, no accountability data exist to prove this, much less to prove that these schools’ high performance was spurred by competition. . . . We have found that, overall, parental knowledge of specific schools tends to be low, and that parents face considerable barriers in their efforts to obtain information about schools. Thus parents are unable, by themselves, to hold schools fully accountable. (2004, 52)6
While Van Dunk and Dickman were specifically referring to voucher schools, there is no question that the same problems exist in connection with charter schools and other forms of school choice (see chapter 6; also see Schneider, Teske, and Marschall 2000).
The ability of charter schools to provide a space for locally autonomous, community-centered education is another reason charter schools enjoy broad support from diverse constituencies. Charter schools can provide a means for parents and students who perceive themselves as marginalized or disenfranchised to seek shelter from an indifferent or even hostile public school system. Groups as diverse as right-wing Christians in California (Huerta 2000, 177–202) to Afrocentrists in Michigan (Yancey 2000, 66– 97) have taken advantage of charter-school laws to create schools catering to their particular interests. Charter-school operators and their founding parent communities frequently use a broad set of strategies, including selective recruitment and explicit parent contracts, to create and maintain their distinctive identity (Lopez, Wells, and Holme 2002, 129–58). Insofar as they are successful, charter schools of this type may nurture and protect their constituent community, but some worry that by so doing they can exacerbate racial, ethnic, religious, and cultural segregation and inequalities (Fuller 2000a).
While many critics of choice fear the fragmentation of education and extol the virtues of the “common school,”7 others argue that homogeneity within schools coupled with a diversity across schools increases the opportunity for parents to become involved in their schools and learn more directly about the benefits (and costs) of participation in the provision of public goods and services. Moreover, many researchers have argued that parental involvement in schools is essential for improving academic performance and that the more supportive homogeneous environment of charter schools coupled with their commitment to parental involvement increases the chances for academic success. This argument is intimately tied into the idea of building what have been called “effective schools” (Coleman and Hoffer 1987; Henderson 1987; Bryk and Schneider 2002) and also to the idea of “coproduction” of public goods and services (Levine 1984; Marschall 2004)—points we explore in depth in chapter 11.
The Market for Schools
The third rationale for charter schools and school choice more broadly, and the framework in which both the most compelling and the most contentious arguments for and against such reforms are now framed, is found in the neoliberal theories of the market.8
The closing decades of the twentieth century saw a profound transformation in American public policy—the privatization of services once thought to be the exclusive province of the state. In the area of education, this was not a new idea: Milton Friedman (1955) proposed educational vouchers in the 1950s, and the intellectual history of the idea can be traced back to the political philosophies of John Stuart Mill and Thomas Paine (Walberg and Bast 2003), but the changing intellectual climate among policy makers and analysts moved such plans to the forefront in the 1990s.
While the idea of vouchers had been around for decades, instrumental in bringing the idea of educational privatization to a broader audience was John Chubb and Terry Moe’s Politics, Markets, and America’s Schools, a book that provoked fierce debate by proposing that government-run schools were destined to fail because of the very mechanisms of American democracy itself. According to Chubb and Moe, “the specific kinds of democratic institutions by which American public education has been governed for the last half century appear to be incompatible with effective schooling” (1990, 2). More specifically, the authors believe that in the American education system, where interest groups have easy and continual entry into the policy-making process, many entrenched organizations, particularly teacher unions, exercise too much control over decisions—and they skew decisions in favor of their own needs rather than the needs of students. Making the point even more explicit, they argue that poor academic performance is “one of the prices Americans pay for choosing to exercise direct democratic control over their schools” (2). Their solution was a market for education in which parents would be free to choose their child’s school and where choice “is a self-contained reform with its own rationale and justification.” For Chubb and Moe, choice “has the capacity all by itself to bring about the kind of transformation that, for years, reformers have been seeking to engineer in myriad other ways” (217).
Critics of school privatization were quick to point out that a system of school vouchers (or “scholarships” in the language of Chubb and Moe) was theoretically quite a distance away from the ideal market of neoclassical economics, and that transactions in such a “quasi market” were rife with unintended and possibly pernicious consequences, such as increased stratification across socioeconomic groups or further weakening of already struggling public schools. Henig (1994) suggests that the “market” in this case is better understood as a metaphor than as social reality, and a misleading metaphor at that (also see Fiske and Ladd 2000; Smith and Meier 1995; cf. Walberg and Bast 2003, 210–21).9 But despite these criticisms, Chubb and Moe succeeded in redefining vouchers from a conservative, esoteric microeconomic theory into a broad-based practical approach to fixing the ills of urban education that became an integral (if contested) item on many agendas for educational reform.
The central idea of a market (metaphorical or otherwise) for education is that once the government’s monopoly on public schooling is broken and parents and students become “consumers,” a host of new suppliers of education will enter the market and compete with existing schools and among themselves to provide educational programs that better meet the demands of parents and students than does the current monopoly provision of education. One key underlying assumption is that parents and students know what they want—and what they want is academically strong schools. Another key assumption is that schools (or some other organization) will provide information about their programs and quality and that parents will be able to find this information and use it to make choices and to hold schools accountable (Schneider, Teske, and Marschall 2000). In this idealized system of choice, programs and curricula will align with consumer preferences, and efficiency and academic outcomes will improve. Schools that do not improve or do not meet parental and student needs will lose students and be forced to reform or close.
Note that, according to proponents, not all these outcomes need be present for a market-based reform to be viewed as a success. For example, even if academic achievement (in today’s environment, almost always defined by standardized test scores) remains constant, but costs go down, this higher productivity would still be a victory for market-based policy (Hoxby 2003, 287–341). Similarly, an improvement in consumer satisfaction or the utility of parents, even without gains in achievement or productivity, would still represent a net gain in social welfare.
Different systems of school choice approach the ideal of this neoliberal model in various degrees. Clearly voucher programs are most closely tied to the precepts of the model—and given recent court decisions, the expansion of voucher programs to include parochial schools will increase the number and range of providers dramatically. However, even school voucher programs still fall short of the ideal market, and charter schools are arguably even further from this ideal. At best, according to education economist Henry Levin, charter schools “simulate some of the dynamics of a market by increasing the supply of alternatives to parents and by competing with existing public schools” (Levin and Belfield 2003, 10). Nevertheless, according to Levin, “as long as there is some choice and competition it is believed that outcomes will be better than when there are no choices at all” (15).
Or will they? Although proponents of privatization and other market-based reforms do not often acknowledge it, a theoretical result in welfare economics, the “general theory of second best” (Lipsey and Lancaster 1956, 11–32), holds that “if any aspects of the free-market ideal are fundamentally unattainable (as is of course the case), then incremental movement toward that ideal is not necessarily a welfare improvement” (Ackerman 2004, 2–7). In a complex system like education, partial or even full privatization may turn out to increase costs and reduce benefits (Sclar 2000).
To ensure a net increase in social welfare in the real world of imperfect markets, the task of the policy maker is to meet all the sufficient conditions for such an increase, not merely a few necessary ones piecemeal. Unfortunately, identifying the sufficient conditions for a successful market for education is still an elusive goal. And, without the identification (and satisfaction) of such conditions, the “magic of the market” may, like many magic tricks, go up in smoke, leaving the evaluation of charter schools (and schools of choice more generally) much more difficult.
What Limits the Market for Education?
Economists and public policy analysts frequently evaluate policy options with reference to the Pareto criterion (named for Italian economist Vilfredo Pareto, 1848–1923). This standard holds that a policy choice is preferred if it makes at least one person better off and no one worse off. The attraction of the market as a mechanism for the allocation of goods and services is that, under ideal conditions, through voluntary transactions the market will yield a Pareto optimum, a point at which no one can be made better off without making another worse off.10
What the theory of second best simply points out is that, under less than perfect market conditions, a Pareto optimum is no longer guaranteed—and getting closer to the ideal market in some dimension does not, a priori, predict an overall increase in social welfare. To return to the context at hand, two important questions should be posed: How might the implementation of a charter-school reform differ in the “real world” from implementing charter schools in an ideal one? And, after we take into account these differences, have we met enough of the sufficient conditions to increase social welfare?
Another critical issue in the evaluation of the “market” for education is the extent to which education generates externalities—social costs or benefits to a market transaction that are not properly accounted for by the price mechanism. The presence of externalities can lead to the wrong level of a good or service being produced, and the existence of externalities is one of the prime justifications for the public provision of public education. The example most often presented of externalities in education highlights the difference between the individual and the collective (societal) benefits of education. Put simply, when an individual invests in her own education, many of the benefits of that education accrue directly to her in terms of a better career and higher wages. But her community and the nation as a whole get an added benefit from having a more highly educated workforce—there is a positive externality to education that spills over from the individual to society.
Left strictly to individuals to decide, the level of education in a society will not reflect this “added” payoff, and all the societal benefits of education will not be reaped. The traditional economic solution calls for the public provision of education to reduce its cost to the individual consumer.11 In turn, this lower cost will make people willing to “buy” more education, and the public subsidy helps society reach the optimum level of education for the community and the nation.
According to Levin and Belfield, “[p]robably the greatest challenge to the view of market efficiency in education is created by the presence of externalities” (2003, 16). But, and this is crucial, the externalities Levin and Belfield have in mind are broader than the standard case of externalities presented above. Specifically, they link the provision of education to the very functioning of democratic society: “It has long been held that one of the central purposes of schools is to improve the cohesion and stability of society through the provision of a common experience that prepares the young for adult roles and responsibilities. Schools are expected not only to educate students as individuals, but also to contribute to the overall effectiveness of society through creating competent adult citizens. . . . The larger question is how to reconcile the private choices of families with the public requirements of education for democratic knowledge and values” (16).
Political theorist Amy Gutmann uses a similar argument based on the presence of externalities to build a case against elite private schools. She argues that while parents are choosing to spend their own money to boost their child’s chances at lifetime success, there is a social cost in that the public schools will be rendered less capable of fulfilling their purpose of perpetuating democracy through the loss of socioeconomically elite students and politically powerful parents. This cost is a negative externality flowing from individual choices and transactions in the market for private education (1987, 115–18).12
Not all externalities are bad, however. As we discuss in chapter 12, parents may choose a charter school in order to improve their child’s chances of academic success, but this individually motivated choice may help create a stronger, school-centered community. Or, returning again to theories of democratic education, parents enrolling their children in Catholic schools may find that they emerge better trained for the duties of citizenship (Bryk, Lee, and Holland 1993; Coleman, Hoffer, and Kilgore 1982), helping to improve their communities and society as whole. In short, some school-generated externalities can have important positive effects on civil society—but to achieve them may require action by government that transcends strictly market-based rationales.
Closely related to the idea of externalities is the fact that some, perhaps a large, part of education can be considered a public good. Once again using the case of democratic education, if there is little obvious return to individual students from civic education and engagement, and since such areas of a curriculum have both a real cost (the cost of teacher time, for example) and an opportunity cost (that is, a cost in terms of lost opportunity to cover other subjects or activities), in a competitive market we should expect democratic education to be driven out of the curriculum by lack of demand. If, however, society as a whole desires the production of good citizens through democratic education, then either traditional, government-run public schools subsidizing the provision of democratic education or else some regulation or limit on the market will be required.
The Flow of Information
Markets also fail to function optimally if participants make transactions under conditions of information asymmetry. As the term suggests, this sort of market failure occurs when one party in a transaction knows more about the quality of goods or services than the other.
There are at least two types of asymmetries that we need to consider. First, in the public policy and economics literature, there is the traditional difference in the quantity or quality of information between buyers and sellers—here between parents and students, the consumers who are “shopping” for schools, and the schools themselves, which almost certainly have more information about the true quality and nature of educational service provided. Levin and Belfield note, “[T]here has been considerable debate over whether school quality can be easily codified and quantified (or even manipulated), leading to the possibility that parents will be making choices based on false information” (2003, 15). Schneider and various coauthors have shown that many parents know very little about the range of schooling available to them (Schneider et al. 1997; Schneider et al. 1998). This asymmetry may be exacerbated by false advertising (Lubienski 2003; also see Kane and Staiger, 2002) or through the spread of bad information through low quality social networks (Schneider et al. 1999). The current drive to release test-score data may do nothing to solve this problem. Ladd argues that measures of average levels of performance provide misleading information to parents about how much the school is likely to add to student learning. Instead, average test scores provide information about the socioeconomic status of the average student at the school. Ladd notes: “While that information is not irrelevant to parents (and could be very important to some), it differs from information about the contribution of the school to student learning” (2001, 391).
The second asymmetry stems from differences among parents themselves. To ask a question familiar in the school-choice literature: “Who chooses and who loses?” While it may be possible in a given market for education that a small number of well-informed or “marginal” consumers increase social welfare for all through their shopping behavior and the response of providers (Schneider, Teske, and Marschall 2000), it is also possible that this information differential may exacerbate existing socioeconomic inequalities (Henig 1996).
To add an additional layer of complexity to the problem of information, schooling is an “experience good,” the quality of which can not necessarily be evaluated until the student has been enrolled for some period (or perhaps even a “postexperience good” that cannot be evaluated until many years after consumption).13 Thus parents may be more satisfied in the short run with a charter school’s focus on athletics, but ultimately regret the decision when their child gets his first paycheck.
Yet another way in which a market for education could fail is through imperfect competition. For example, there may be high entry barriers limiting the number of providers who step forward to create new schools. Finn, Manno, and Vanourek (2001) identify a host of such barriers, including political opposition from state or local boards of education; school district resistance or excessive regulations; state education department resistance; union or bargaining unit collective bargaining agreements; and trouble with federal regulations (also see Research Policy Practice International 2001, 44). Finn et al. (1997) also cite financial difficulties, including lack of capital financing, especially for facilities;14 little or no start-up money; inadequate per-pupil operating funds; and uneven cash flow. And finally, they highlight what they call “policy problems” embedded in “weak” charter laws (see also Kucsova and Buckley 2004).
Among the provisions of weak laws that can inhibit the creation of new charter schools are the following:
- Narrow eligibility. For example, existing schools may “convert” to charter status, but new schools may not be started;
- Too many rules and regulations restricting the school’s autonomy;
- Severe restrictions on how many charters may be issued or how many students may enroll in them;
- Granting school boards sole authority for issuing charters (see Teske, Schneider, and Cassese 2005).
There can be other “supply side” problems in the market as well. For example, small providers who run a single charter school may be unable to compete with larger “educational maintenance organizations” (EMOs) due to economies of scale or the ability of large educational conglomerates to expend substantial amounts in political lobbying at the local, state, and federal levels. This could lead to the creation of an oligopoly in which the education market is made up of only a handful of large providers, possibly restricting consumer choice or charging high per-pupil fees from the district, state, or even parents, even if they are not producing higher achievement scores or increasing parental satisfaction.15
Additionally, one of the prime justifications for charter schools is their role as “laboratories of reform,” that are free from the bureaucratic restrictions placed on traditional public schools by teacher unions and administrators (Nathan 1996; Kolderie 1990). However, empirical research has found little evidence of this (Rofes 1998; Teske et al. 2001) and the structure of real charter-school markets may actually act to inhibit programmatic competition (Lubienski 2003). Ironically, it may be that parents are risk-averse when it comes to their children’s education—they may choose schools that emphasize traditional values and educational approaches rather than “buy” innovative programs with a high degree of risk. This is quite reasonable from a parent perspective, but may create systemic problems in a system of schools that is designed both to innovate and to respond to parental preferences.
EVALUATING CHARTER-SCHOOL REFORM
Obviously, the study of charter schools is complicated. Although a strong neoliberal, pro-market theoretical foundation propels much of the discussion in the literature, there are many signs that mapping the logic of the market onto the real world of charter schools is less than perfect. Even assuming that we want more consumer choice and more competition among suppliers, we have no strong guidance as to whether these alone are sufficient conditions to create markets for schools that actually improve overall social welfare. We also know that since schools have a strong public good dimension and they generate many externalities, their nonmarket dimensions are as important as their more narrowly defined efficiency.
As if explicating these “big ideas” is not enough of a task, there is yet another challenge. As empirical social scientists, we must find a way to translate these ideas into questions that are tractable for analysis. To lay out a way of doing this, in the next section we rearrange these issues concerning the market for schools into five dimensions around which we structure the analysis we present in this book.
Five Dimensions for Evaluating Charter Schools
We argue that the ongoing debate over charter schools can be thought of as encompassing five dimensions, which we use to organize our study. These include “three C’s:”
and “two A’s”:
Several of these dimensions map onto those identified by Levin (2002; Levin and Belfield 2003, 19–20) and Gill and his colleagues (Gill et al. 2001), but, we believe, our specification is more tractable for empirical analysis. In the next few pages, we begin to explore each of these criteria. We then describe the structure of the book and identify how these themes are to be discussed in the chapters that follow.
THE THREE C'S
Competition clearly lies at the very heart of any market approach to educational reform. The idea here is quite simple and pertains to the “supply side” of education. In the traditional system of school assignment, a system in which students are assigned to a school based on where they live, schools are monopoly providers. Given this monopoly status, schools have no reason to compete with one another, and they have no incentives to create new programs or to improve their efficiency. The underlying argument is right out of Econ 101—without the salutary force of competition, monopolists charge too much, they produce inferior products, or they do both.
Milton Friedman’s classic argument in favor of vouchers placed the reform of the supply side to increase pressure on monopoly education providers in a central role. Friedman’s key argument is that competition will improve the price/performance ratio of inputs to outputs (productive efficiency). Friedman did not argue that competition would automatically lead to higher performance per se—rather, he argued that we could achieve the same level of outcomes at lower unit cost.16 One interesting outcome of this intellectual history is that voucher programs tend to set the dollar figure attached to vouchers below the cost of public education—embodying the belief that competition can produce more for less. However, perhaps because charter schools are not as closely tied intellectually to Friedman’s argument for vouchers, the per-student payment in charter schools is usually higher than in voucher programs—even in the same city. For example, in Washington, D.C., per-student payments for students in charter schools total around $10,000, but in the recently approved federally supported program, vouchers are set at only $7,500.
The second dimension, choice, is also central to our study, for two reasons. First, as noted above, choice by itself should improve the match between what parents and students want and what they get from their schools. This in turn should improve parent and student satisfaction with their schools. Second, choice is intimately linked to competition. It is the diverse preferences and the ability of consumers to choose among a range of products that provides the foundation for competition. However, there is a “flip side” to this dimension: equity is a primary concern of most stakeholders in American education policy (see, e.g., Levin 1991).
In any market, there is always what Okun (1975) called the “big trade-off” between efficiency and equity. This trade-off may be particularly important in evaluating school reform since education plays a central role in reducing socioeconomic disparities and providing equality of opportunity in American myth and social reality (Hochschild and Scovronick 2003). In turn, successful charter school reform must not exacerbate inequalities through the “cream skimming” of students of higher socioeconomic status—even if this might improve productive efficiency or increase parental or student satisfaction. Thus, there are questions about who chooses and who loses in a system of expanded choice—and these questions are central to our analysis.
Community is not part and parcel of the market approach. However, the importance of community emerges from the distinctive nature of schooling. A system of charter-school choice should not jeopardize the social and political foundation of our communities and nation. As Levin and Belfield (2003, 20) write, “a market system that bases its appeal on differentiation and choice must adopt a mechanism to ensure common experiences across schools to prepare students for their civic rights and responsibilities.” In turn, the public good or positive externality aspects of education must be considered as an important part of the final evaluation of a charter reform.
In addition, education is not a simple service. Rather, it is a production process that requires the continued input of a range of actors—that is, education is “co-produced” by students, parents, teachers, school staff, and administrators. There is clear evidence demonstrating the importance of the ties among these stakeholders in producing a high-quality education. Also, as alluded to above, schools provide a venue for citizenship or democratic education, and such education may be more effective in a well-functioning school community.
THE TWO A'S
The first “A” is accountability, which should increase in systems of choice. In marketlike approaches to education reform, the first line of accountability lies in consumer choice. If parents and students are not happy with the school they chose, they can exit.
Since schools of choice are directly dependent on the revenues brought in by each student, if enough students exit the school, the school either improves or disappears. Paul Hill, in one of the most comprehensive studies of charter schools to date, argues that charter schools, freed from many of the bureaucratic rules and regulations governing traditional public schools, have created new “accountability” relationships with the teachers, on whose performance the schools depend, and with families, whom the schools must attract and satisfy (Hill et al. 2001, 6). These relationships, according to Hill et al., transform the way in which teachers, administrators, parents and students deal with each other, creating strong, cooperative, working relationships leading to strong school communities and higher levels of academic success.
Second, and more specific to charter schools, is the fact that charters are granted for fixed periods of time and are subject to periodic and strict review at which time they can be revoked. Consider the following example. In January 2004, the State University of New York Board of Trustees, which has authority to charter schools, acted on the renewal of the first three charter schools established in the state. Two schools were approved without problem. However, the decision on the John A. Reisenbach Charter School in Harlem’s District 5 was more difficult.
New York’s Charter Schools Institute, created to administer the state’s charter-school law, had recommended that the Reisenbach School be closed at the end of the school year, due to poor academic performance, governance problems, and other concerns. However, the choice was not easy. On the one hand, District 5 was one of the lowest-performing in the city, and parents in that district were desperate for better schools. On the other hand, Reisenbach’s poor performance after five years and the demands of accountability argued for closing the school. As this debate unfolded, an op-ed piece by Peter Murphy, which appeared in the January 31, 2004 edition of the New York Post, summarized the issues at hand concisely:
Charters differ from traditional public schools because they are held accountable for academic results. Failure has real consequences (rather than rewards of more money). This accountability system is made clear to charter-school operators up front and they commit to abide by it.
To protect the integrity of all charter schools and ensure the success of this reform movement, the SUNY board needs to act on the exhaustive findings of its own institute staff and close schools, like Reisenbach, that have failed to operate in an educationally and fiscally sound manner.
Accountability has to mean something in public education—at least for charter schools.
On February 24, 2004, the State University Board of Trustees voted not to renew the charter and the Reisenbach School joined the list of charter schools held accountable for failure. But how widespread are cases such as these?
While data calculating actual closure rates are hard to find, Hassel and Batdorff (2004) studied decisions to renew school charters, looking at all 506 nationwide renewal cases through 2001, and then focusing on fifty randomly selected cases. They found that 16 percent of charter schools up for renewal were terminated (conversely, 84 percent of the charter schools were renewed). The Center for Education Reform reports that 429 charter schools have closed from the inception of charter schools through 2003.17 If we set the number of remaining charter schools in the nation at around three thousand, we can estimate a closure rate of approximately 13 percent, not much different than the Hassel and Batdorff finding of 16 percent (also see Teske, Schneider, and Cassese 2005, who look at the politics involved in authorizing and renewing charter schools).
There are two problems with these existing data (besides the fact that they are, at best, approximations). First, it is close to impossible to state precisely the reasons for nonrenewal—there is no central repository for information about these actions, and news accounts are unreliable. While Hassel and Batdorff (2003) suggest that many authorizers have been willing to close underperforming schools, many nonrenewals make the news because of financial improprieties or other newsworthy acts of malfeasance. For example, in 2004, the Washington, D.C. school board closed the Village Learning Center Public Charter School after a city audit found that the school spent hundreds of thousands of dollars on leased space it did not occupy, credit card charges for apparel and gifts, and loan repayments that lacked documentation. Similarly, in September 2004, just as the school year began, ten thousand students in the California Charter Academy’s four schools (spread over sixty campuses) faced serious disruption as the state closed the schools as a result of financial audits. Whether the high incidence of media reporting on charter school closings due to financial impropriety is just another example of the media’s love of scandals is an open question; we still lack convincing evidence on the extent to which authorizers are really holding charter schools accountable for low academic performance.
Second, we have no standard by which to judge whether or not this (approximate) 15 percent closure rate is high or low (public schools are rarely closed and no data on private school closings are kept as far as we can tell). Critics often argue that charter-school closures are a sign of failure in the charter movement—that the charter “movement” is creating a legion of poor schools that are failing to educate the children placed in their care (and absconding with scarce public-education money, to boot). In contrast, proponents argue that while these closures are undoubtedly a sign of failure in individual schools, they are a sign of success for the movement as a whole as bad schools are weeded out and good ones survive. For them, closures are clearly proof that accountability is taken seriously.
ACHIEVEMENT: THE ULTIMATE GOAL?
While many advocates of choice argue that competition, choice, community, and accountability are goals in themselves, others argue that they really serve as intermediate steps to the ultimate goal of higher school performance. Lubienski makes this point clear:
[T]hemes such as “innovation,” “choice,” “competition,” and “diversity,” although they all have some value in and of themselves, often serve as intermediate goals that structure opportunities for institutions to increase achievement. Even as intermediate goals, however, choice, competition, and innovation are cast as the necessary vehicles for advancing academic outcomes. Therefore, the extent to which school choice reforms succeed in securing these ideals indicates the likelihood of reaching the primary objective in this logic model. (2003, 397)
Thus, these structural reforms should lead the final dimension that we argue should be used to study schools: achievement.
Higher academic achievement is now the mantra of educational reform and the call for better test scores is heard throughout the nation. According to Amrein and Berliner (2002):
In recent decades, test scores have come to dominate the discourse about schools and their accomplishments. Families now make important decisions, such as where to live, based on the scores from these tests. This occurs because real estate agents use school test scores to rate neighborhood quality and this affects property values. Test scores have been shown to affect housing prices, resulting in a difference of about $9,000 between homes in grade “A” or grade “B” neighborhoods. At the national and state levels, test scores are now commonly used to evaluate programs and allocate educational resources. Millions of dollars now hinge on the tested performance of students in educational and social programs.18
Indeed, there are many ways in which charter schools should be able to take advantage of their autonomy while responding to the pressures of choice, competition, and community to achieve higher academic outcomes (Chubb and Moe 1990; Hoxby 1997; Hill, Pierce, and Guthrie 1997; and Brandl 1998). Wohlstetter and Griffin make this explicit. For them, choice, autonomy, and competition make schools more flexible and innovative “with the perceived assumption that such innovations will produce identifiable improvements in student achievement” (1998, 3).19
As the descriptions of the criteria suggest, it is likely that there will be tensions among the various measures in almost any real choice program— improving productive efficiency through a choice mechanism may compromise equity; increasing social cohesion through regulations requiring democratic education may reduce freedom to choose.20 Moreover, it is difficult to measure the size of the trade-off across the different criteria, as they clearly have different metrics. We try to elucidate some of these trade-offs while assessing the success or failure of the charter schools we study in the chapters that follow.
THE PLAN OF THE BOOK
The book is organized in four clusters, and we examine competition, choice, community, accountability, and achievement throughout, although with different emphases in each cluster and in each chapter.
In the first cluster, beginning with this chapter, we describe the theoretical issues that inform our work. In the present chapter, we have identified the five broad dimensions that structure our approach to studying school choice and placed these dimensions in the larger debate over school choice and charter schools. In the next chapter, we move from general themes that underlie every school-choice plan to an exploration of the venue in which we develop and test specific hypotheses that are derived from applying the ideas we explored in this chapter to the city in which we have been studying the evolution of charter schools—Washington, D.C.
Washington, D.C., has one of the strongest charter laws in the country and has one of the largest concentrations of students attending charter schools in the nation.21 At the time of this writing (April 2006), there are fifty-one organizations running sixty-three charter-school campuses, enrolling about 17,500 students. This represents almost a quarter of the total number of public-school students in the city and almost one-third of the schools. At least six new charter schools are scheduled to open in the school year beginning in September 2006. In chapter 2, we describe the development of the charter-school sector in D.C., touching specifically on how competition and choice have evolved since the city adopted its strong charter-school legislation in 1996.
With the themes and venue laid out, we then introduce the two research projects from which the data we use in this book are drawn. In chapter 3, we introduce the four-wave panel survey that we use to assess how parents shop for schools, how they feel about their schools, and how they participate in their school community.
A multiple-wave panel study is one of the most valuable, but least frequently created, social-science research tools. In the field of political science, studies that attempt to model the dynamics of political phenomena using such data are rare and in the sphere of charter schools—to the best of our best knowledge—are currently nonexistent.22 However, in chapter 3 we will also introduce some of the problems that plagued us as we tried to keep the panel “alive” in an inner city where there is high mobility and where people “disappear” from the sample for many reasons.
We follow that chapter with an examination of the students in charter schools and the traditional D.C. public schools and briefly touch on the issue of student achievement. A major theme in this section is the extent to which charter-school students are any different than students in the DCPS. One argument in the often-contentious evaluation of how well charter schools perform is that the students in them are different and “harder to educate.” We explore this issue in chapter 4.
In chapter 5, we introduce our second major research tool. In 1999, we created a web site called DCSchoolSearch.com. As described in that chapter, we collected a large amount of data about each and every school in Washington, D.C. and made it available on a user-friendly web site. As visitors navigated through the information presented on DCSchoolSearch.com, we tracked their movements to see what information they were accessing and in what order. In addition, we surveyed parents who gave us their e-mail addresses to obtain more information about their search procedures.
With the sources of our data established, in chapters 6–8, we use data from both the survey and from DCSchoolSearch.com to study fundamental issues pertaining to choice. One of the main issues concerning choice has to do with what parents want from their schools, with perhaps the most important consideration being the relative weight that different parents put on academic achievement versus factors such as proximity, the racial and economic background of other students, and other, perhaps more idiosyncratic preferences. We compare the findings from our telephone survey with the search patterns parents used while “visiting” schools using DCSchoolSearch.com to present some rather different conclusions about parental preferences over school attributes.
In the two chapters that follow the introduction of our data sources, we combine theories of information search drawn from behavioral decision theory with traditional concerns of political science and theories of consumer behavior to explore how different types of parents search for information. We have two major research themes in that exploration: first, given the generally low levels of information that have been well documented, can we expect that parents will choose the right school for their child? Besides this question at the individual level, we are also concerned with a systemic one: if parents as a population of consumers have low levels of information, can the school system overall feel any competitive pressure to improve? Here we look more closely at the search patterns of the “marginal consumer” of education to see whether in the face of low information the school system may still be pressured to improve.
In chapters 9–12 of the book, we look at the consequences of choice and charters. Here we address the fundamental “so what?” question— does all the time, energy, effort, and money spent on creating these alternative charter schools matter? In chapter 9, we explore parents’ satisfaction with their child’s school. As noted earlier, given choice, parents should be choosing schools that do more of what they want done in education. Therefore, satisfaction should be higher among choosers—if not, this is a pretty strong indication of a failing system of choice. In this chapter, we also begin to deal with some fundamental methodological issues to which all studies of school choice must pay heed. Of greatest concern is the problem of self-selection—parents who have chosen to enroll their children in charter schools may be systematically different than parents whose children remain in traditional public schools. This self-selection into the “treatment” (i.e., charter schools) can lead to biased estimates of its effects.
We explore several statistical approaches to control for self-selection to see how robust the effects of charter schools are. We also explore the possibility that parents who choose have a psychological stake in saying that what they chose is good (often called the “rose-colored glasses” effect). To “telegraph” one of our most interesting results, we find that rather than viewing schools through rose-colored glasses, charter school parents may in fact be wearing “grey-colored” ones: they may constitute a group of parents that are particularly hard to please.
In chapter 10, we take full advantage of our panel data to explore parental satisfaction over time. We look at changes from the beginning of the school year to the end of the year and we look at how well schools did over time, comparing the 2001/2 patterns with 2003/4 ones. We also cope with some fairly difficult problems caused by attrition.
In chapter 11, we look at the extent to which charter schools and traditional public schools differ in the strength of community and the creation of social capital. Using methods established in our previous chapters, we examine the survey data for evidence of any charter-school effect on attitudes thought to be the foundation of parental civic engagement, both at the school level and in a broader context.
In chapter 12, we turn our focus from parents to students. We examine key indicators of student civic education and participation in search of empirical evidence for the contention that charter schools, like other schools of choice, build better citizens.
Finally, we conclude with an attempt to pull together the disparate empirical findings of the earlier chapters and view them through the lens of our organizing framework. We return to our central question: are charter schools, on the whole, a beneficial policy reform, or are the hope and the hype unjustified by the facts?
Return to Book Description
File created: 8/7/2007
Questions and comments to: email@example.com
Princeton University Press