Part III
Strategies for Meeting the Goals

Education finance is only one part of a total system of education. Many of the concerns about the financing of education reflect large issues regarding the overall education system. Hence, proposals for changing the finance system can be presented in at least two ways: (1) as a menu of options for driving the education system in desirable directions or (2) as intertwined components necessary to achieve a given vision of overall education reform. Despite the conceptual appeal and logic of starting with a vision of overall reform and designing a finance system consistent with that vision, we structure the discussion of finance reform in line with the first approach; that is, as the options available for driving the education system in the directions embodied in our three goals: (1) promoting high achievement for all students in a cost-effective way, (2) reducing the nexus between family background and student achievement and, (3) raising revenue in a fair and efficient manner.

The reason for proceeding in this way is the absence of a single consensus vision of how the education system as a whole should be changed. Although many observers agree that the education system requires change, people disagree on the best way to change it. These disagreements have various and overlapping roots: people view school performance through different lenses; they hold different values, which influence their views of desirable methods of change; and they can reasonably draw different conclusions from patchy and conflicting empirical evidence about the effectiveness of specific reforms.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 163
Making Money Matter: Financing America's Schools Part III Strategies for Meeting the Goals Education finance is only one part of a total system of education. Many of the concerns about the financing of education reflect large issues regarding the overall education system. Hence, proposals for changing the finance system can be presented in at least two ways: (1) as a menu of options for driving the education system in desirable directions or (2) as intertwined components necessary to achieve a given vision of overall education reform. Despite the conceptual appeal and logic of starting with a vision of overall reform and designing a finance system consistent with that vision, we structure the discussion of finance reform in line with the first approach; that is, as the options available for driving the education system in the directions embodied in our three goals: (1) promoting high achievement for all students in a cost-effective way, (2) reducing the nexus between family background and student achievement and, (3) raising revenue in a fair and efficient manner. The reason for proceeding in this way is the absence of a single consensus vision of how the education system as a whole should be changed. Although many observers agree that the education system requires change, people disagree on the best way to change it. These disagreements have various and overlapping roots: people view school performance through different lenses; they hold different values, which influence their views of desirable methods of change; and they can reasonably draw different conclusions from patchy and conflicting empirical evidence about the effectiveness of specific reforms.

OCR for page 163
Making Money Matter: Financing America's Schools The following chapters examine the major policy options relevant to each goal. The options are arranged within the framework of the four generic strategies for altering finance systems introduced at the end of Chapter 2: Reduce funding inequities and inadequacies. Invest in capacity. Change incentives to make performance count. Empower schools or parents to make decisions about public funds. Although we examine policy options separately by strategy, in many cases the strategies will be most effective if they are combined in a coherent way. In general, the case for combining strategies is most compelling when policy makers are trying to work within the existing system of school governance. Some proponents of major change in the governance system (e.g., through vouchers allowing parents to choose public or private schools) see little reason to combine finance strategies. To them, the introduction of more choice and competition among schools will provide whatever incentives are necessary to induce schools to make the types of investments needed to improve student learning. The importance of making major changes in the governance system takes on special urgency in the context of goal 2, because many decades of attention to the educational problems facing at-risk children and urban schools in the framework of the existing educational system have so far resulted in improvements that are marginal at best. The seriatim discussion of individual policy options for goals 1 and 2 arranged by finance strategy may mask an important overall question, which we as a committee wish to highlight specifically, although we are not of one mind about how to answer it. The question is whether it is more important to focus on finance changes that leave the structure of American education basically intact (as the first three strategies assume) or to explore options that would constitute a strong break with past practices (as policies emphasizing school and parent control over education dollars might do). Many reforms are occurring in schools and districts serving high proportions of at-risk students. It is still unclear, though, whether the current round of reforms will be more successful than previous ones. At the same time, how effective new structures would be is not yet knowable either, since many are largely untried. Thus policy makers face fundamental choices for which we cannot provide scientific solutions. These choices will rest on individual conclusions about the prospects for meaningful change within the current educational structure and on values in addition to those of enhancing fairness and productivity.

OCR for page 163
Making Money Matter: Financing America's Schools 6 Achieving Goal 1: Promoting Higher Achievement in a Cost-Efficient Way Promoting higher achievement for all students in a cost-efficient way poses fundamental challenges for the education system and will require significant changes in education finance. Among the finance options to be explored in this chapter are placing greater emphasis on the concept of educational adequacy; investing in the capacity of the system, particularly through improved professional development of teachers; changing incentives for teachers and other key participants to focus more on student achievement through new salary structures and accountability systems; and significant restructuring to give more authority over spending to schools and to parents. No single strategy or policy option is a panacea, all involve trade-offs of various types, and some of the strategies or individual options are likely to be most effective if combined with other strategies or policy options. Instead of laying out a blueprint for specific change, our aim in this chapter is to provide policy makers with information that will allow them to weigh the benefits and costs of each policy option. REDUCING FUNDING INEQUITIES AND INADEQUACIES As applied in school finance court cases for the past 30 years, equity concerns typically apply to the distribution of funding across districts. Equity defined in this way—but extended to the patterns of funding across schools and across states—continues to be an important concern that we address further in connection with goal 3, raising revenues fairly and efficiently. However, a con-

OCR for page 163
Making Money Matter: Financing America's Schools tinued focus on this definition of equity is not likely to promote the goal of raising student achievement across the board. More promising is the new legal focus on adequacy. Because the move from equity to adequacy shifts attention away from the distribution of funding levels across districts to the adequacy of funding for desired outcomes, the strategy of trying to ensure that school funding is adequate is potentially crucial to the goal of increasing achievement for all students. In the absence of adequate funding, it will be difficult for states, districts, or schools to generate high and ambitious levels of student achievement for all students. Policy makers, however, face the vexing problem of determining how much money would be required for true adequacy. This difficulty arises in part because there is so little firm knowledge about how school spending or inputs translate into school outputs given the current nature of teaching and learning. We know even less about what would be required to reach more ambitious achievement standards or what would be required if education were delivered in a more efficient manner. Will a push through the legal system to increase the adequacy of funding help to achieve the goal of higher achievement for all students? Given that the outcome is uncertain, the best we can do is to identify some of the issues that will affect it. One issue is whether the courts are likely to be more amenable to adequacy than to equity complaints. Notably, as discussed in Chapter 4, not every high court has been receptive to an adequacy argument. For example, high courts in Illinois, Rhode Island, and Florida all rejected adequacy-based claims on the grounds that it is the responsibility of the legislature, not the courts, to decide on the quality of education. Thus, the main uncertainty here is the willingness of the courts either to specify what educational adequacy entails or to require that state legislatures specify those desired outcomes. Without a relatively explicit statement of desired educational outcomes, the court will not be able to determine whether a state is providing adequate funding for education. A second issue is how state legislatures are likely to respond to adequacy judgments from the courts. State politics played a large role in how states responded to court decisions related to financing equity. On the surface, an adequacy approach (with its focus on the level of outcomes) may be more broadly appealing than the redistributive remedies that typically emerge from cases based on equity considerations. Such an approach may be easier to sell to a public that wants more accountability from government (Carr and Fuhrman, 1999). In practice, however, adequacy, too, is likely to call for politically controversial redistributive remedies, in that additional funds are most likely to have to be directed to those districts with the lowest-performing students. Moreover, lack of knowledge about the educational production function—and disagreement about whether it even makes sense to talk about such a function—adds a huge element of ambiguity and uncertainty to the concept of adequate funding, one that could well provide a basis for legislative stalling and inaction.

OCR for page 163
Making Money Matter: Financing America's Schools As documented in Chapter 3, court cases (those focusing on equity or adequacy claims or both) appear to have raised school spending in the low-spending districts without at the same time lowering it in the high-spending districts (Evans et al., 1999). Evidence from a few of those states suggests as well, however, that the additional revenue received by the recipient jurisdictions has been devoted primarily to the same activities as schools undertook before and hence may not have done much to increase student achievement (Goertz and Natriello, 1999). This evidence provides support for the committee's view that changes in funding alone are not likely to be sufficient to increase student achievement. Regardless of the success of adequacy as a legal strategy, it is the committee's view that adequacy of funding has a central role to play in any education reform strategy designed to increase student achievement. In particular, discussions of educational funding should include explicit considerations of what the public and policy makers want the educational system to accomplish and what kinds of educational opportunities must be provided to meet those objectives. However, the committee also concludes that the provision of adequate funding by itself will do little to foster significant improvements in overall student achievement. Thus, while improving the adequacy of funding may be a necessary part of any education reform effort—and is likely to be especially crucial for districts or schools serving disproportionate numbers of disadvantaged students—it is only part of an overall program for increasing student achievement in a cost-efficient way. INVESTING IN CAPACITY To many education reformers, especially those committed to standards-driven systemic reform, a central element of any finance reform program is investing in capacity. Such investments are necessary to ensure that the system can deliver the quality of product required to enable students to achieve to high levels. Thus, this strategy calls for strategic investments that will yield high rates of return in the form of student achievement. These might include investments in school inputs, such as the capacity of teachers, the quantity of teachers (to reduce class size), technology and school buildings, and investment in the capacity of students to learn, such as preschool programs and family support services. Developing the capacity of students to learn can be extremely important in some cases but, because the returns are likely to be greatest for students from disadvantaged backgrounds, we defer the discussion of that type of investment to the next chapter. In addition, we defer to that chapter investment strategies that focus (1) on reducing class size (because there is some evidence that the benefits of that strategy are greatest for low-performing students) and (2) on technology and facilities, because current policies for financing these investments appear to generate even greater inequities across districts than is the case for operational spending. In this chapter, we examine whether new approaches to paying for teachers

OCR for page 163
Making Money Matter: Financing America's Schools and teacher development could align investments in teacher capacity more closely to the goal of improving overall student achievement. Overview of Issues Related to Teacher Capacity The capacity of a teacher to be successful depends of a variety of factors, including his or her knowledge of the subject matter, of how students learn, and of methods for teaching. These types of knowledge are needed for a teacher to succeed in promoting student learning by, for example, selecting good teaching materials, making wise instructional decisions, and assessing student progress. Moving toward the goal of higher levels of achievement for all students will require two other types of teacher capacity as well. One is the ability and the skills to teach the kinds of knowledge and skills demanded by 21st century jobs and citizenship, and the other is the skills to teach highly diverse groups of students, with increasing proportions whose language and culture differ from their own, as well as increasing numbers of students with disabilities and other students with special educational needs. New developments in the science of learning, for example, are increasing awareness of how important it is for teachers to pay careful attention to the prior knowledge, as well as the skills, attitudes, and beliefs, that learners bring to school. Prior knowledge not only consists of the individual learning that students bring to the classroom but also knowledge that they acquire from their social roles, such as those connected with race, class, gender, and cultural and ethnic affiliations. Teachers must be helped to develop teaching practices that start from the structure of a child's prior learning and are sensitive to the cultural and language practices of students and the effect of those practices on classroom learning (National Research Council, 1999b). The committee is persuaded by the evidence reviewed in Chapter 5, including recent work by Hanushek et al. (1998) using Texas data and Wright et al. (1997) using Tennessee data, that teacher quality matters for student achievement. At the same time, we note that research has not been able to systematically link teacher quality to traditional teacher measures, such as experience or holding a master's degree. These measures are of particular interest for finance, since they are the ones that are linked to teacher pay and hence affect the cost of providing education. Ferguson (1991) and Ferguson and Ladd (1996) show that a different measure—but still an imperfect one—of teacher quality, namely teacher test scores, does emerge as an important determinant of student achievement in both Texas and Alabama. Other studies document the importance of teacher preparation. For example, Goldhaber and Brewer (1997) report positive effects on student math and science achievement of subject-specific training programs and, using a matched comparison design, Hawk et al. (1985) show similar results for mathematics achievement. Darling-Hammond's (1990) review of the literature provides support for the importance of teacher preparation and certification.

OCR for page 163
Making Money Matter: Financing America's Schools To summarize, there is little doubt that the quality of teachers matters. The challenge is to develop the best policies to enhance it. Perhaps the most frequently discussed policy change involves altering teacher education and induction practices, which in the first instance is not a finance option but has finance implications for teacher training institutions, would-be teachers, and school districts. Two other policy options, raising teacher salaries and expanding professional development, fit more closely with the concept of financial investments intended to build teacher capacity. We examine these three options, with more attention to the latter two, in the following sections. The challenge of enhancing teacher quality must be addressed in context. One aspect of the context is the rising demand for new teachers. The Department of Education estimates that the nation's schools will need to hire about 2.2 million new teachers during the next decade, a figure that reflects both a continuation of the growth in student bodies that started in the mid-1980s and a growing number of teacher retirements (U.S. Department of Education, 1998a). The fear is that this rising demand will put pressure on schools to lower their standards and to hire unqualified individuals. Given that about one-half to two-thirds of the new teachers hired will be first-time teachers, the rising demand also implies that attention needs to be paid to the quality of the training new teachers receive and the quality of professional support as they start their careers. Failure to provide that support could exacerbate the current situation, in which 22 percent of new teachers leave the profession during the first three years (U.S. Department of Education, 1998a). Another contextual aspect is the great variation across areas and in the quality and preparation of teachers and in the outlook for teacher shortages. For example, while 28 percent of high school mathematics teachers nationally lack as much as a minor degree in the subject, state levels range from a low of 9 percent for Missouri to over 45 percent for Alaska, Washington, and California (NCES data reported in Darling-Hammond, 1997: Appendix B, Table 3). Differences also emerge between central cities and other areas. For example, 21 percent of the public school teachers who teach science in grades 7-12 in central cities report no degree (major or minor) in their main teaching area, more than twice the 9 percent in other areas (Lewis et al., 1999: Table 5). In a 1997 report prepared for the National Commission on Teaching and America's Future, Darling-Hammond emphasized that shortages of well-prepared teachers are largely a problem of distribution rather than numbers. Demand for teachers is particularly great in the South and the West, and in port cities on both coasts, whereas slow-growing states have teacher surpluses. Wealthy districts with high salaries and desirable working conditions rarely experience shortages in any field, whereas districts and schools with large numbers of low-income and minority students are much more likely to face difficulty recruiting qualified teachers and to hire unqualified teachers or to use substitute teachers to fill positions (Darling-Hammond, 1997). A recent story in Education Week

OCR for page 163
Making Money Matter: Financing America's Schools provides further evidence that many regions have surplus teachers in some areas and shortages in others (Bradley, 1999). Teacher Education Programs and Licensing Standards Building teacher capacity by improving the education of new teachers responds to long-standing observations about the shortcomings of teacher education in the United States and to new standards in some disciplines, such as those developed by the National Council of Teachers of Mathematics, that clarify what teachers need to know and be able to do to promote all students' learning. The impetus to change licensing standards comes also from international comparisons—from U.S. students' poor showing on the Third International Mathematics and Science Study (TIMSS), coupled with evidence that other nations invest much more intensively in developing new teachers (Paine and Ma, 1993). Finally, it reflects concerns about the importance of assuring that the large number of new teachers who will be needed during the next decade are adequately prepared to teach. Research is emerging that clarifies shortcomings in existing teacher education programs and provides guidelines for improvement (Darling-Hammond, 1997; National Research Council, 1999b). Standards for teacher training programs are now emerging. However, quality control is quite limited, research on teacher learning is still relatively new and limited, and teacher education programs vary widely in how well their graduates are prepared to teach. National strategies to promote improvements in the quality of teaching include standard-setting for teacher education programs, for teacher licensing, and for teaching excellence. These include national professional standards for teacher education programs developed by the National Council for Accreditation of Teacher Education (NCATE), standards that are being developed by the Interstate New Assessment and Support Consortium (INTASC), and national standards for advanced certification developed by the National Board for Professional Teaching Standards (NBPTS). Such standard-setting initiatives are intended to provide frameworks within which state and local systems can develop quality controls that upgrade the preparation and support of teachers. As a committee on education finance, we were not able to thoroughly evaluate a strategy for enhancing teacher quality through the use of stricter accreditation requirements and higher requirements for teacher certification. However, we have two comments to make about this strategy. First, despite its apparent logic, some observers have questioned parts of it, such as the desirability of having education programs accredited by NCATE. Ballou and Podgursky (1998, 1999) provide evidence showing that teachers emerging from accredited programs are no more qualified than teachers emerging from other programs. These two researchers are also concerned that requiring accreditation may eliminate programs that currently provide teachers to districts that serve disproportionate

OCR for page 163
Making Money Matter: Financing America's Schools numbers of disadvantaged students.1 Second, raising entry barriers to teaching (through, for example, extending the amount of time it takes to get certified or through limiting the number of ways that one may enter the teaching profession) could be undesirable to the extent that it disproportionately discourages the most able students—those who have the greatest alternative opportunities—from making the investment to enter the teaching profession. Improving teacher preparation and developing professional standards for teachers are fully consistent with the goal of raising achievement for all students. Whether teacher quality will best be enhanced through teacher licensure exams, holding teaching colleges and universities more responsible for the preparation of the teachers they prepare, or other mechanisms is beyond the scope of this committee. We simply urge that policies aimed at improving teacher preparation and certification be evaluated at least in part for their impacts on overall student achievement. Teacher Salaries Several arguments are frequently offered for raising teacher salaries, including the fair treatment of existing teachers, the need to keep teacher salaries in line with the salaries of other college graduates so as to attract qualified people into the teaching profession, and as a way to increase teacher quality. Of most interest to us here is the extent to which raising teacher salaries across the board is a good way to raise the quality of teachers. Before we turn to that issue, however, we briefly look at the trends in teacher salaries relative to those in other occupations. Table 6-1, using data based on the decennial censuses of population for the period 1940–1990 (reported in Hanushek and Rivkin, 1996), shows percentages of male college graduates earning less than the average male teacher and similar percentages for women. The lower the percentage, the less attractive is teaching as a profession. The table shows a decline in the attractiveness of teaching as a profession for men from 1940 to 1970, with a partial turnaround since then. For women, the drop is more precipitous and has no recent turnaround. As of 1990, only 36.5 percent of college-educated men were in jobs that paid less than that of the average male teacher, and 45.3 percent of college-educated women were in jobs that paid less than that of the average female teacher. A refinement of the analysis that isolates the patterns for 20- to 29-year-olds indicates that the relative position of teacher salaries for young female teachers 1    Information from Pennsylvania underscores these concerns. In that state, the districts with the highest proportions of NCATE-certified teachers have the lowest proportions of students going on to post-secondary education (Strauss, 1998:152). Although this observation provides no information on causal links, it is consistent with the two concerns raised by Ballou and Podgursky and calls out for more investigation of the reasons for the observed pattern.

OCR for page 163
Making Money Matter: Financing America's Schools TABLE 6-1 Position of the Average Teacher in the Nonteacher Earning Distribution, 1940–1990 Year Percentage of Male College Graduates Earning Less than the Average Male Teacher Percentage of Female College Graduates Earning Less than the Average Female Teacher 1940 52.5 68.7 1950 36.2 55.0 1960 28.7 52.7 1970 25.7 47.1 1980 31.0 50.1 1990 36.5 45.3   SOURCE: Hanushek and Rivkin, 1996. Data obtained from the U.S. Decennial Census of Population, Public Use Microdata, 1940–90. (the women who have the most lucrative opportunities outside teaching) is approaching that for men. On the basis of that convergence, Hanushek and Rivkin (1996:29) conclude that schools will not be able to count on a continual supply of high-quality female teachers in the future and that the rising opportunity costs for women will put upward pressure on school budgets given that women made up 68 percent of all teachers in 1990. Overall, these trends based on average salaries suggest that teaching is becoming less financially attractive for college graduates compared with other occupations. While the average salaries mask what has been happening to the structure of salaries, a more detailed analysis of salary trends by teacher experience or education level is not likely to change this basic conclusion. But will higher salaries increase teacher quality? One reason they may not emerges from empirical observations from New York, Michigan, and other states that salary increases are often disproportionately directed toward the more experienced, senior teachers, an outcome that results from the contract bargaining between local teachers' unions and school district officials (Lankford and Wyckoff, 1997; Monk and Jacobson, 1985; Murnane et al., 1987). The backloading of salary increases onto veteran teachers means that salaries for entering teachers are kept relatively low. The low salaries for entering and inexperienced teachers may well interfere with the ability of the system to attract high-quality new teachers into the profession, and those how enter may well leave before they are eligible for the higher salaries available to more experienced teachers. Thus, the salary bargaining process in which veteran teachers exert a lot of power means that money devoted to increases in teacher salaries is not being used as effectively as it could be toward the goal of increasing the capacity and quality of teachers.

OCR for page 163
Making Money Matter: Financing America's Schools A 1997 book by Ballou and Podgursky using national data provides further support for the view that rises in teacher salaries may not increase teacher quality. In particular, they convincingly document that increases in salaries during the 1980s did not increase the quality of the new teachers attracted to the profession. They explain this finding in part by the possibility that, despite the higher salaries, highly qualified college students may have been discouraged from making the investment in teaching by the decline in openings for teachers that occurred as veteran teachers responded to the salary increase by staying in the profession longer. Thus implicitly these researchers draw attention to the importance of the structure of salary increases. Two aspects of the Ballou and Podgursky study could limit its applicability to the potential efficacy of future increases in teacher salaries. First, it is a national study that abstracts from the tremendous variation across the country in the market conditions for teachers. Indeed, a recent study by Loeb and Page (1998), based on a state-level panel dataset, supports the view that teacher salaries can affect the quality of schooling and student outcomes. Using school dropout rates as their primary measure of student outcomes and hence of school quality, the authors conclude that, holding all else equal, raising teachers' salaries by 10 percent would reduce dropout rates by 3–6 percent. Their careful attention to the specification of their equations and in particular their focus on changes rather than on levels makes the analysis convincing and worthy of attention. Their rough calculations suggest that the cost of raising teacher salaries by 10 percent would slightly exceed the benefits as measured by the present value of the increases in individuals' future salaries associated with their higher educational attainment. However, the authors note that their measure of the benefits of an increase in teacher salaries probably underestimates the true benefits, since it is based on a single measure of student outcomes. They note further that if salary increases were better targeted, they are likely to be more cost-effective than the across-the-board increases they examined. Second, the Ballou and Podgursky study applies to a period of declining enrollments and a general teacher surplus, a market context that the authors used to help explain their findings. In a market with excess supply, an across-the-board increase in teacher salaries provides an incentive for the existing stock of teachers to remain in the profession longer than they otherwise would, which in turn limits the number of new openings. Given the outlook for a tighter market for teachers in the future, and a very tight market in such fields as mathematics and science and in some areas of the country, salary increases could potentially be more successful in attracting higher-quality teachers in the future than they were in the 1980s. Nonetheless, this different market context does not negate the conclusion that when or if they are planning to raise teacher salaries, policy makers trying to increase the quality of teachers entering the profession will need to pay more attention to the structure of salaries and in particular to the level of entering salaries than has been the case in the past. In addition, they may want to

OCR for page 163
Making Money Matter: Financing America's Schools school may become a more coherent community with a shared mission and shared set of values than is the case when children are assigned to schools. The hope is that this shared sense of community would make each school a more effective organization and hence able to provide higher-quality education. As Brandl (1998) notes, these two mechanisms of competition and community through which parental choice might improve schools need not be in conflict and would complement each other in a well-designed system. Much has been written on these strategies for improving schools. In a book of this type, we obviously cannot do justice to all the literature. Our more limited intent in the following sections is to rely primarily on empirical studies to determine how likely it is that strategies of this type will increase student achievement or, alternatively, will generate any given level of achievement in a more cost efficient manner. We start with the school-oriented strategies of site-based management, charter schools, and contracting with private firms. We then turn to various policies that enhance parental choice. Effects of School-Oriented Strategies on Student Achievement and Efficiency Although we begin this section with school-based management (SBM), it is worth noting that most so-called SBM programs in the United States to date have included only limited financial autonomy. Hence, we have little direct U.S. evidence on the effects of giving schools more financial autonomy. Nonetheless, the findings from the literature are suggestive. Charter schools, in contrast, provide a better example of meaningful devolution of financial authority to the school level, but as the evidence on the effectiveness of charter schools is still quite limited and the jury is still out on their effectiveness in promoting high achievement and encouraging innovation. Moreover, limitations on capital funding for such schools have put them at a disadvantage. Finally, the limited experience with contracting between school districts and private firms provides more insight into the nature of contracting problems than it does about any potential of that strategy. School-or Site-Based Management Studies that have examined the decentralization of authority to schools have raised questions about the logic of the theory (Hannaway, 1993, 1996; Wohlstetter and Odden, 1992). One expectation, for example, is that school personnel and parents, as opposed to administrators in the central office, will focus more directly on teaching and learning. Studies show, however, that the participation of school-level participants, in particular parents and teachers, is weak (e.g., Malen and Ogawa, 1988; Easton and Storey, 1994; Hess, 1993) and that decisions do not

OCR for page 163
Making Money Matter: Financing America's Schools seem to focus more heavily on teaching and learning concerns (e.g., Hess, 1993; Weiss and Cambone, 1994). Consistent with this finding, a review of 18 studies by Summers and Johnson (1996) found not only little effect on student achievement but also little effect on the expected behavior of schools and school participants. Few of the studies even attempted to estimate impacts on achievement (suggesting that higher student achievement may not have been the goal) and those that did estimate impacts on achievement found no statistically significant impacts, although some found that school-based management programs had a positive, but small, impact on student attendance. The most dramatic U.S. experiment with decentralization of authority to the school level is the 1988 Chicago reform that set up 550 local school councils in which parents had a statutory majority and which had the authority "to hire and fire the school principal, determine the school's educational priorities, and approve the spending of discretionary funds, eventually amounting to half a million dollars or more" (Shipps et al., 1998:1). The effects on student achievement were mixed at best. Some schools appeared to have improved student performance; others remained unchanged; and some performed worse (Bryk et al., 1998). About half the schools did not seem even to take advantage of the freedom and resources offered under the law to try to change their schools (Sebring et al., 1996). Some people have interpreted the fact that in 1995 Chicago increased central control in a striking way as an indication of the instability of decentralization reform efforts, which calls into question whether this form of school governance can exist in the turbulent and politically charged environment of big U.S. cities. However, others have interpreted the change as the logical extension of a decentralization strategy that is designed to increase student achievement (see Hess, 1999). Odden and Busch (1998), drawing on an extensive body of recent research studies, conclude that newer strategies of SBM point to a series of organizational conditions that must exist at the school level for SBM to lead to improved student achievement. In their view, there are nine key steps that must characterize SBM: center change on student learning and a rigorous instructional program; involve all teachers in decision making; allow schools to recruit and select staff; invest in training and professional development; create a professional school culture; create a comprehensive school-based information system; provide rewards and sanctions; select principals who can facilitate and manage change; and give schools control over their budget. Charter Schools Charter schools represent a more complete form of decentralization than school-based management, in that schools are exempted from much district and state regulation in exchange for the accountability that comes from the possibility

OCR for page 163
Making Money Matter: Financing America's Schools that their charter will be revoked or of other sanctions imposed by the state. Charter schools are a relatively recent phenomenon, and few systematic studies about them are available. While no information is available yet on their contribution to student achievement, some information is available on whom they serve, limitations related to accountability, their effect on public schools, and the problems they face in obtaining funding. Contrary to opponents' predictions, the early experience with charter schools does not support the view that they disproportionately serve white and economically advantaged students (U.S. Department of Education, 1998b; Vanourek et al., 1997). The second-year report of a national study of charter schools showed that, while in 1996–97 they enrolled a smaller proportion of students with disabilities than other public schools in their state, their racial composition, the proportion of low-income students, and the proportion of students with limited English proficiency were similar to statewide averages. Moreover, when the analysis was extended to the district level, it showed that about 60 percent of charter schools were not racially distinct from their districts and another third enrolled a distinctly higher percentage of minority students than the district. Racial/ethnic enrollment patterns differ, however, across states with charter schools, with some states (California, Colorado, and Arizona) having a somewhat higher average percentage of white students in charter schools than in all public schools, whereas others (Massachusetts, Michigan, Minnesota, Texas, and Wisconsin) had lower average percentages of white students in charter schools (U.S. Department of Education, 1998b). With respect to accountability, Wohlstetter and Griffin (1998:14–15) found that in practice "the myth of greater accountability for charter schools far exceeded the reality." Although the 17 schools in their study reported that they appreciated the value of a sound accountability system, not one had such a system in place and the schools derided the use of externally imposed standardized tests because they would not assess accurately what the school was trying to accomplish. Similarly, a study of charter schools in ten California districts (UCLA Charter School Study, 1998) reported that, in most instances, charter schools were not yet being held accountable for enhancing the academic achievement of their students; they were more likely to be held fiscally accountable. One of the rationales for charter schools is that they will promote greater effectiveness and efficiency in regular public schools. In the only study of this issue of which we are aware, Rofes (1998) conducted case studies of 25 randomly selected school districts in which charter schools were operating. Although more work on this issue would be desirable, his findings are suggestive. Based on interview data, he concluded that about one-quarter of the school districts responded to charters in an energetic way and had significantly altered their educational programs. In one district, for example, the formation of a local charter served as a catalyst for improving the district's middle school. Other responses included opening schools organized around a specific theme, setting up pilot

OCR for page 163
Making Money Matter: Financing America's Schools schools, creating add-on programs such as an after-school program or an all-day kindergarten, and offering more diverse activities or curricular resources. Another quarter of the respondents responded more moderately by making significant efforts to aggressively market their schools to the public or by becoming more receptive to community input. From the perspective of school finance, the most striking aspect is the financial disadvantage under which many charter schools operate. In many states, charter schools receive as operating expenses just the state share of operating revenues, not the combined state and local revenue that is available to the public schools. Even more burdensome is the lack of capital and start-up funds. A large percentage of charter schools, particularly those for which the charter was granted by nondistrict entities or those that are start-up schools, have no access to local district funds levied for capital improvements and do not have access to the capital market. As a result, most charter schools, are forced to use a portion of their operating funds or to seek funds from private sources to secure, furnish, and maintain facilities (Bierlein and Fulton, 1996). Furthermore, except in two states (Arizona and New Mexico), charter schools receive no extra state support for planning or implementation. Without minimizing the conceptual difficulties of determining a charter school's fair share of funding, the committee simply notes that charter schools in most states have not been put on the level playing field with the public schools that would allow them to compete effectively. At the same time, we note that fair treatment with respect to the financing of capital facilities could increase overall costs of providing education unless existing school facilities are turned over to the charter schools or are sold or rented out. The jury is still out on charter schools. Early interest and enthusiasm for them has probably been far greater than most policy makers expected, but the difficulty of establishing these schools is probably also far greater. Moreover, no information is yet available on their impacts on student achievement. Contracting with Private Firms Contracting with private firms would appear to have some important advantages over the alternative of enabling groups to set up charter schools. Unlike the charter schools for which start-up capital is a serious problem, private firms have access to capital that allows them to expand and make investments. In addition, because firms could run multiple schools, they are able to operate like a ''virtual" school district and can ensure that the schools they run have sufficient capacity to operate effectively. In addition, they have strong incentives to provide quality control to preserve the firm's reputation. Third, contracting gives the school district more control over the types of schools being provided without its being involved in the running of the schools. However, the well-publicized experiences

OCR for page 163
Making Money Matter: Financing America's Schools of Baltimore and Hartford with the private firm of Education Alternatives Inc. (EAI) did not give contracting a good name. Part of the problem in Baltimore was that the school district negotiated a bad contract with EAI that required the district to pay more to EAI to run nine elementary schools than it would have cost to run the schools themselves. This outcome occurred because the district agreed to pay the average per-pupil cost of running all schools in the district rather than the average cost of running the cheaper elementary schools (Walsh, 1995; Brown and Hunter, 1996). Despite its additional funds. EAI was unable to produce any positive effects on student achievement. Although EAI initially released test score results that showed improvements in the schools it managed, these scores were later discredited when the Baltimore school district released district scores that showed student performance in the EAI schools was lagging. A University of Maryland study also shows that EAI schools in Baltimore produced no real difference in student performance or in the impact of technology on learning (Molnar, 1996). For this reason, as well as concerns by the district about some of EAI's practices, such as moving students out of special education classes and the use of college interns as teachers' aides, Baltimore terminated its contract with EAI in 1995 (Molnar, 1996). The experience of Hartford was a bit different, in that EAI was hired to perform back office management functions in that city rather than to operate the schools, but that contract too was cancelled in early 1996. The Edison Project has had the longest history in operating contract schools, but data on its effects is limited and the number of schools still relatively small. Thanks in part to the growth of charter schools, however, the firm is now growing. The school year 1997–98 marked the third year of Edison's operation and at that time, the company had 25 public school partnerships serving nearly 13,000 students. For 1998–99, the number of schools was expected to increase to 75. The only data available on student test scores are reported by Edison itself (although they are compiled by independent researchers). Comparisons of changes in test scores between Edison's student and a control group indicated some positive effects overall (Edison Project, 1997). At least one school showed striking gains, but in others there were small declines. No public information is available on the other private firms, including the Sylvan Learning Systems, that have emerged in this market. In sum, solid information on the effects of these initial trials with private firms is limited. Moreover, little is known about how such firms will behave over time as their need to show profits increases and, indeed, it is not yet clear whether the firms will be able to provide a high-quality product and still make a profit. The one bright light for the Edison Project and other private firms is that the expansion of state charter school legislation has opened up a new opportunity to run schools as charter schools, which could give these firms a chance to grow enough to benefit from economies of scale.

OCR for page 163
Making Money Matter: Financing America's Schools Conclusions about School-Oriented Policies On the basis of the evidence, we cannot be assured that any of these approaches will by itself fulfill the goal of higher achievement in a cost-efficient manner. However, we note that the newness of the charter school movement makes it difficult to evaluate. Charter schools may well fulfill their proponents' predictions about exerting productive impacts on the rest of the public school system sometime in the future, and indeed preliminary evidence of these impacts is positive in some areas. In fact, in the committee's view, some of these options are likely to be quite promising for the goal of increasing achievement, especially if they are included as part of a larger education reform strategy that encompasses, for example, clear outcomes-based accountability standards. Additional flexibility at the school level may be crucial as a way to achieve greater achievement, given the absence of a clear and identifiable production function for education that applies to all students. However, that flexibility will be productive in increasing achievement only to the extent that higher achievement is the goal, that schools are somehow held accountable for achievement, and that teachers have the capacity to teach effectively and students the capacity to learn. Effects of Greater Parental Choice on Student Achievement and Efficiency Many parents already exercise school choice, through their choice of where to live or by electing to enroll their children in private schools. One problem with the first form of choice is that by bundling the choice of school with the residential choice decision, any tendencies toward residential segregation by income or race will be exacerbated. Another disadvantage to either form of choice is that not all families can exercise it equally. Rather, wealthier families have more options than poorer families as they can more easily move to another location, or alternatively they could send their children to private schools. Moreover, low-income families in big-city school districts have even fewer choices than others, since even if they move to another location within the city they still receive education services from the same district. We return in Chapter 7 to the implications of the constrained choice available to low-income families in large urban areas. Various strategies have been proposed, and implemented, to expand parental choice and to break the connection between residential location and choice of schools. Historically, the goal of many strategies that provide for more choice within a district, such as for example magnet schools and controlled choice programs, had more to do with reducing racial segregation than with increasing student achievement or making schools more efficient. Newer forms of parental choice include open enrollment schemes that allow children to choose schools in

OCR for page 163
Making Money Matter: Financing America's Schools other districts, charter schools, and vouchers that can be used in any public or private school. Of interest here is what we know about the effects of parental choice either on student achievement or on the productive efficiency of the education system. Economists have been particularly interested in the issue of efficiency, so we begin with some theoretical predictions about choice that emerge from the economics literature before turning to the empirical evidence. Our attention here to the predictions from economic theory are not intended to discount other mechanisms, such as the contribution of parental choice to the development of schools as organizations with shared values, which also could lead to greater productivity. Instead, we include the economic predictions here because of the insight they provide about different forms of choice programs. The point is that choice programs differ in their potential to increase the efficiency of the education system. Moreover, the impacts of the various choice programs are likely to vary by type of district. Economic Predictions about Different Forms of Choice Economic theory suggests that schools or districts will operate inefficiently unless they are disciplined by competitive market pressures. Given this perspective, not all forms of parental choice are likely to be equally effective in promoting production efficiency. For example, choice programs within districts may provide pressure for individual schools to become more effective as they attempt to retain students, but provide no pressure for the district administration itself to become more efficient. Choice programs between districts could provide some of that pressure, as districts would be in danger of losing some of their students to other districts. However, in an interdistrict choice program, some economists worry that an important link between education quality and housing values will be weakened. According to this argument, residents in a district have an incentive to care about the quality of education provided by the district to maintain housing values in the district. With an interdistrict choice program, students have access to the district's school without residing in the district, which breaks the link between the quality of neighborhood schools and house values, and thereby the monitoring function could be weakened. These two offsetting incentives could conceivably lead to a reduction of school quality in a district. Charter schools are, in principle, quite different from other forms of public school choice because of the potential for entry and exit. Unlike other forms of choice, charter schools have the potential to undercut the monopoly power accorded to a public school in a typical public system in the United States. A charter school might enter and supplant an ineffective public school. Similarly, a charter school is not endowed with the monopoly rights to a geographic area in the way that public schools typically are. So, in principle, the force of entry brings charter schools closer to the textbook competitive norm than other forms

OCR for page 163
Making Money Matter: Financing America's Schools of open enrollment. Similarly, voucher-like arrangements have the potential to encourage entry of schools that would potentially supplant ineffective public schools. The significant point here is that many economists believe that parental choice is likely to lead to the greatest gains in productive efficiency when there is the potential for a supply-side response in the form of the entry of new schools or the closing down of existing schools. Types and Quality of Empirical Evidence Empirical evidence on the effects of parental choice on achievement and productive efficiency comes in various forms, some of which are more reliable than others. Because there has been a large amount of choice within the education system for a long period of time, there are a huge number of studies of the effects of choice. Not all of these studies, however, shed light on the question in which we are interested: Would the introduction of significantly more parental choice into the education system increase achievement or the productivity of the system? Moreover, some of the studies are far better than others, either because of the study design or because of the quality of the data. The evidence, which is far from perfect, provides some limited support for the view that parental choice can lead to better educational outcomes. This evidence comes from a number of sources. Rouse (1997) and Greene et al. (1997) found that students in the Milwaukee voucher program showed achievement gains, especially in mathematics. Evidence from some of the privately funded programs has also begun to emerge: Peterson et al. (1998), for example, found that New York students who received vouchers from a private foundation scored higher on standardized tests. This study is noteworthy because it is the first one to be based on a true experimental design. Because the program was oversubscribed, those who were awarded a voucher were selected at random. Thus the evaluation is based entirely on the comparisons between those who were selected and those who were not selected for a voucher.2 In addition, there is some evidence that Catholic schools are more effective than public schools. Evans and Schwab (1995), for example, find that Catholic school students are more likely to finish high school and go on to college. But while this evidence is suggestive, it would be very difficult to argue that it constitutes an ironclad case in favor of more parental choice. In many cases, researchers looking at the same problem have come to very different conclusions. Witte et al. (1995), for example, conclude that the Milwaukee experience offers no evidence that private schools have been more effective. Neal (1997) and 2    Issues still arise about how to handle the applicants who were awarded a voucher but did not use it, especially given that the researchers had a lower success rate in testing those students than in testing the students who used the voucher.

OCR for page 163
Making Money Matter: Financing America's Schools Figlio and Stone (1997) conclude that while urban minority students often realize significant benefits from a Catholic school education, Catholic schools are no more effective than public schools for other types of students. The problem here in part is the ability of researchers to control fully for unmeasured differences among students and families. To be sure, the recent studies of Catholic school students have made great strides compared with earlier studies in controlling for the selection bias that arises because the students who choose differ from those who do not. Nonetheless, Catholic school students and students who use vouchers to attend private schools in Milwaukee differ from other children in at least one fundamental way that is difficult to control for using standard statistical procedures: someone was willing and able to take the necessary steps to take them out of the public schools. Research (e.g., Wells and Crain, 1997) has shown that parent-child relationships are very different in families that exercised their option to participate in a school-choice plan. Choosers (in all income groups) are more actively involved in their children's educations and more actively direct their children in important decisions. It is possible to mistakenly attribute the success of children whose parents choose private school for them to the school rather than to differences in parenting. Similar problems will arise if private schools are able to successfully screen potential students or expel students as a result of poor behavior and poor academic performance. (This problem, however, does not apply to the New York City privately funded voucher program described above.) Hence, evidence from the statistical studies is at best suggestive. Of interest is not only the effect of choice on the achievement of the choosers but also its effect, through the competition for students, on the traditional public schools. In addition to one interview study on this issue related to charter schools already cited (Rofes, 1998), more general research on this question is provided by Caroline Hoxby, who makes use of the variation across school districts in the amount of competition they face naturally from public schools in other districts or from private schools to draw inferences about the effects of a voucher program on overall student achievement. Hoxby (forthcoming) concludes that districts with less concentration of enrollment (more competition) have higher test scores and that competition makes a bigger difference in districts with less educated adults (where less than 20 percent of the adults have a bachelor's degree) and in districts in states with more local control. Because this study focuses on the form of choice that is most common in the United States, choice among public school districts, the evidentiary base is large. Moreover, the study is ambitious in that it uses data from a number of large U.S. databases and explicitly addresses the potential problem of reverse causation. In two other studies, Hoxby (1995, 1996) measures the effect of competition from private schools and concludes again that such competition increases the achievement of students in the public schools. If Hoxby's results are correct, they are potentially very important. However, technical criticisms have been

OCR for page 163
Making Money Matter: Financing America's Schools raised in connection with her papers on private schools as well as some questions of interpretation (Kane, 1996). In addition, in contrast to the Hoxby findings, a new study by Sander (forthcoming) that examines the effects of private schools on the achievement of public school students in Illinois finds no direct effects. Like Hoxby, Sander uses Catholic religion as a means of identifying the private school effect, and in particular, of addressing the statistical problem of simultaneity that arises because in areas with poor public schools more students are likely to opt out of the public school system in favor of private schools. Given these conflicting findings, the question of the effect of competition on public schools must be considered unsettled at this point. In sum, the U.S. research on parental choice, including choice that extends to private schools, is limited largely by the absence of experience on a large scale with voucher programs or with broad-based choice programs that break the link between residential location and school choice. Much remains to be learned about the potential for choice to increase student achievement. The most urgent context for that additional learning appears to be in urban areas, with their large concentrations of disadvantaged students. It is worth noting, however, that other countries have had more experience with parental choice programs that break the link between place of residence and schooling, including some public funding of private schools, than has the United States. The experience from Europe with those programs provides reasonably clear evidence of the potential for an undesirable side effect of some choice programs, namely that schools may become more socially stratified (Vandenberghe, 1996, for Belgium; Ambler, 1994 for France; and Karsten, 1994, for The Netherlands). The mechanism by which competition might increase stratification is intimately tied with how parents choose schools. For example, to the extent that parents' perceptions about the quality of a school depend in part on the socioeconomic characteristics of the students in the school, parents will tend to transfer their children into districts or schools with higher average incomes (Fossey, 1994; Armor and Peiser, 1997; Fowler, 1995; 1996). In Scotland, which includes state-subsidized private schools and public school open enrollments, most parents appear to choose schools based on the social status of the student body (Glenn, 1990). Similarly, Echols and Willms (1995) find that Scottish parents who choose are more educated, have a higher occupational status, and tend to select schools with high average test scores and socioeconomic status. School choice programs will not necessarily increase social stratification. Indeed, there are many examples in the United States (such as the experience with magnet schools and with some of the new voucher programs that are restricted to low-income parents) that do just the reverse. Such an effect can be avoided by careful attention to the design of a choice program, an issue to which we return in Chapter 9.

OCR for page 163
Making Money Matter: Financing America's Schools CONCLUSIONS Although this review of knowledge provides no silver bullet strategy for raising student achievement, it provides a variety of promising policy directions for designing an improved financing system, one that will harness school finance more closely to the goal of increasing achievement for all students. Provided the existing system of school governance is maintained essentially in its current form, the committee's main conclusions are that capacity building and incentive strategies must be designed with careful attention to their effects on student achievement and that they will need to be combined in thoughtful ways if they are to promote the goal of higher achievement for all students in a cost-effective way. For example, a policy of investing in the capacity of existing teachers is likely to be more effective if it is combined with a change in incentives that make performance count than if it is implemented by itself. Similarly, the success of a program designed to hold schools accountable may depend on the extent to which teachers have access to the skills they need to improve the performance of their students. That still leaves open for policy makers, however, the significant choice of whether to try to promote higher achievement and greater productivity of the education system primarily within the context of the existing system or to opt for major changes in the system of school governance.