Equity and Adequacy in Education Finance, Issues and Perspectives
reviewed by William H. Clune - 2001
Title: Equity and Adequacy in Education Finance, Issues and Perspectives
Author(s): Helen Ladd, Rosemary Chalk, and Janet S. Hansen, Editors
Publisher: National Academy Press, Washington
ISBN: 0309065631, Pages: 316, Year: 1999
Search for book at Amazon.com
This book is composed of eight background chapters by authors distinguished in the fields of equity and adequacy in school finance. The chapters were commissioned by the Committee on Education Finance of the National Research Council (NRC), under contract with the Department of Education, as part of a larger study requested by the United States Congress. The full report of the Committee has been issued as another book, Making Money Matter (1999), also available on the National Academy Press web site. The question posed to the Committee was: "How can education finance systems be designed to assure that all students achieve high levels of learning and that education funds are raised and used in the most efficient and effective manner possible?" If the question sounds different from the way the central question about school finance would have been posed twenty-five years ago, it is—as reflected in this statement in the Introduction:
Spending disparities, especially those within states, have inspired education finance reform efforts for decades. . . . In recent years, however, questions about finance systems have increasingly been linked to questions about improving student performance. . . . One manifestation of this broader concern has been the emergence of the comparatively new legal paradigm of educational adequacy . . . [which] has spread rapidly . . . and now serves as the foundation for many current court cases and deliberations. (pp. 1-2)
I have done major research on both equity and adequacy—work that is referenced frequently in the book. I consider each of the chapters of this book a genuine contribution to the field, although I believe that the picture of adequacy litigation, rights, and legislation presented can be sharpened and deepened. Part I of this review will survey the findings and conclusions of the book, with some editorial comments. Part II contains my more general reactions.
I. The Book’s Findings and Conclusions
The eight chapters address the following eight questions: 1) What are the meanings of the words "equity" and "adequacy" in the history of school finance and research? 2) What did the courts require in various kinds of equity litigation? 3) How much equality in dollar spending actually resulted from the equity cases? 4) How were these dollars used (or wasted)? 5) How do politics and state context affect legislative reforms, including responses to court cases? 6) What is the thrust of the new adequacy cases (which way do the cases push policy), and what is likely to happen as a result of the cases in the future? 7) How can the level of adequacy funding be calculated as one of the cost-specific educational inputs deemed necessary by professional experts? 8) How can the cost of adequacy be calculated statistically as the amount necessary to produce a minimum level of educational outcomes?
The Introduction indicates that the book stops short of examining how finance systems could influence the behavior and effectiveness of schools and omits the topic of the relationship of dollars to student performance, questions that evidently will be covered in the Final Report. But, as readers will see from the summary below, both kinds of questions are addressed to a certain extent in some of the chapters. I will discuss these in Part II.
In Chapter 1, Robert Berne and Leanna Stiefel introduce concepts necessary to the definition of equity and adequacy and describe the movement toward adequacy according to that concept map. They regard equity as the all-embracing concept, defined as fairness in the distribution of funds for public education. Depending on underlying value judgments, fairness can be defined in many ways—for example, as equality of dollars or resources for similarly situated students (horizontal equity), or sufficient (and therefore not always equal) resources for students with different needs (vertical equity). Equity may be ex ante (fairness of the rules for distribution), or ex post (fairness in the actual spending patterns after local decisions are made) and may concern inputs (e.g., dollars), process (e.g., coursework), or outcomes (e.g., test scores). On this map, the movement of courts toward adequacy can be seen as a movement toward vertical equity (student needs) and along the input-process-outcome continuum, all familiar ideas in school finance history, although adequacy’s right to minimum outcomes strikes the authors as novel. In Part II, I will make a somewhat sharper distinction regarding adequacy and differ with these authors slightly in their interpretation of my work; but I agree with and will expand on their idea that it is important to see adequacy as rooted in concerns about equity.
In Chapter 2, Paul Minorini and Stephen Sugarman discuss the evolution of school finance equity litigation. After the narrow defeat of a national standard in the United States Supreme Court in the Rodriguez case, numerous state courts did strike down school finance laws in their respective states under education clauses of their respective state constitutions. The first successful group of cases adopted the wealth neutrality theory of the "Coons team" (including, besides John Coons, both Sugarman and myself) that was advanced but rejected in Rodriguez, according to which the state must remove the influence of local wealth from spending but otherwise has great flexibility in choice of funding schemes (such as guaranteeing an equal tax base and allowing local choice, requiring equal spending, or adjusting for student needs). Adequacy litigation, with the 1989 Kentucky case as the watershed, is different in focusing on a right to outcomes and some absolute level of those outcomes, regardless of how the resulting spending compares to that of other districts. Adequacy has the same strength as the early "legal services" cases of the 1970s in focusing on outcomes and student needs, but a weakness, relative to wealth neutrality, in being harder to define and enforce. The authors perceptively note that Courts have defined adequacy in a negative sense by holding, in effect, that, regardless of how adequacy is ultimately defined, present systems obviously do not, in fact, result in adequacy for all students, and legislatures obviously have not consciously built the systems on adequacy principles. Cases are returned to the legislature, under this view, for a "good faith" determination of adequacy that will be honored by the courts, regardless of divergent definitions and actual outcomes, if such a process is faithfully carried out. I agree that adequacy must ultimately be defined by some kind of outcome-oriented process, but I believe there is a process, described in Part II, that is more faithful to the emphasis on minimum outcomes, clearer and move convincing in its results, and more clearly aimed at reversing a pattern of political discrimination, probably the strongest attribute of wealth neutrality.
Chapter 3, by William Evans, Sheila Murray, and Robert Schwab use a comprehensive methodology to analyze the impact of court-mandated school finance reform, comparing states with and without court cases. The good news is that, in court-mandated finance reform states, almost every goal of reform was accomplished in practice to a greater degree than in the absence of litigation—a success story rare in the world of implementation and even more amazing because litigation clearly stimulated voluntary equity-based reform in other states. Despite the experience in California, where equality was accomplished at the cost of lower spending, total spending in the nation as a whole (including California) increased. Poor districts got more aid than other districts; inequality declined; districts with higher costs, such as urban districts, got more, as did districts with higher percentages of children in poverty. On the subject of inequality, the authors say that without reform, inequality would have risen sharply. Specificity about poor children is important because doubt about this point was one factor mentioned by the Rodriguez majority in its refusal to intervene. Table 3.5 of the chapter shows that, as a result of reform, districts with the poorest 25 percent of families gained $804 per pupil in state aid, compared with about $380 for the middle fifty percent of districts and $155 for the richest 25%. Because the income-poor districts used a substantial amount of state aid for tax relief, their net revenues were about the same as the others. In other words, they suffered no discrimination in revenue even after lowering their taxes, thereby presumably making their districts more attractive for local taxpayers. The authors summarize by noting that "court-ordered reform has achieved its primary goal of fundamentally restructuring school finance and generating a more equitable distribution of resources." Of course, the key question for adequacy is whether poor children get enough extra aid to raise their outcomes, but the data in this chapter suggest that the equity movement in litigation states provided a good base for compensatory aid aimed at more equal outcomes.
Chapter 4 by Margaret Goertz and Gary Natriello analyzes the question, Where did the new dollars go? or, more precisely, How did spending patterns change in a selected small sample of low- and high-spending districts in states where the courts ordered major reforms (Kentucky, New Jersey, Texas). Confirming the findings for the nation that were reported in Chapter 3, the authors of this chapter find that most of the new aid went to low-wealth districts and that spending disparities statewide decreased substantially. Eighty-five percent of the new aid went to increased spending rather than to tax relief. Local tax rates actually increased for several reasons (required local effort, incentives in the aid scheme, large, previously unfunded enrollment increases). The main finding on spending patterns is that the use of new funds did not differ substantially between poor and rich districts nor did they differ from previous findings about the distribution of spending by budget categories across many kinds of districts and background conditions. Spending on instruction increased faster than spending on administration. New money was spent on both new teaching positions and higher salaries. Many poor districts had a backlog in maintenance projects, equipment (e.g., computers), and new classroom space. The authors regard the "no change" finding as good news and bad news. The good news is that predictions about massive waste flowing from a sudden influx of funds proved completely unfounded. The bad news is the lack of evidence that new spending was tied systematically to improved learning under new standards (or old ones, for that matter). Of course, generic program categories, like new positions, can be used to support new directions. But the impact of inertia in spending seems basically correct, as does the authors’ concern for how to design a tighter link between new spending and outcomes. My comments in Part II will address that issue.
Chapter 5 by Melissa Carr and Susan Fuhrman analyzes the politics of school finance and particularly whether successful reform following court orders is associated with identifiable characteristics of politics and context. Four characteristics seem to favor legislation that conforms to court decrees: fiscal surpluses, top level leadership (e.g., the Governor), consensus building and articulation of a broad public interest, and a demonstrated link between spending increases and improved student achievement. A common problem in reform is the perception that monies will be redistributed from middle class, white, and suburban areas to urban areas predominantly populated by the poor and minorities. The "decade of school finance reform" in the 1970s coincided with fiscal surpluses in the states, and almost every state (California excepted) funded reform with large amounts of new state revenue, leveling up rather than leveling down (giving at least some new money to almost all districts). The authors conclude that, because of budgetary constraints, taxpayer resistance, ideological resistance, and anti-urban sentiment, the immediate future is an unfavorable time for comprehensive school finance reform. I agree that the obstacles are formidable, but I believe that the same political analysis points the way to a solution: relatively low cost reforms tightly targeted on student outcomes, broadening the constituency for adequacy, and intelligent consensus building. I will explain this strategy in Part II.
Chapter 6, the second chapter by Stephen Sugarman and Paul Minorini on legal aspects of school finance reform, picks up where Chapter 2 left off and discusses the "promises and perils" of the new paradigm of school finance adequacy. Adequacy certainly has been a success in the courts. Ten states followed Kentucky in deciding cases based on this approach. In North Carolina, plaintiffs lost an equity case but won a subsequent case based on adequacy. A Wyoming court said reform legislation must not be driven by revenue availability and politics but rather a good faith attempt to define adequate educational opportunity for that state. The chapter discusses the advantages of adequacy over equity to account for this success in the courts. A main appeal of adequacy is the focus on educational and social opportunity rather than on dollars and tax resources. Equity cases had always drawn criticism because of their lack of specific focus on meeting educational needs and goals. Equal spending or resources by no means guaranteed equitable adjustments for extra costs and needs. As a legal concept, adequacy, as well as equity, can be seen as rooted in the jurisprudence of the United States Supreme Court. In denying relief under an equity theory, the Rodriguez case cited lack of proof that the plaintiffs were excluded from any defined educational opportunity. Some of the early race cases, as the authors point out, focused on intangible opportunities involved in law school and graduate school (e.g., reputation of the faculty, influence of the alumni), as well as on dollar equality. Tracing the theme of the race cases to the present, adequacy litigation might be seen as filling the void left by the withdrawal of the federal courts from desegregation remedies (which themselves often involved extra resources for poor and minority children).
What courts can do to advance the laudable goals of adequacy is less clear.
Judicial intervention of some kind is justified because courts can easily determine that existing aid schemes, like minimum foundation programs, have not been designed with adequacy in mind, and there is too little aid for special needs. The kind of remedy required is more difficult to ascertain. The first step is defining the goals and standards that will give specific meaning to the very general idea of educational opportunity. For this, courts legitimately may look to the states’ own legislation, policies, and standards. The harder part is specifying what kinds of resources and practices educational systems need in order to reach their goals, given that many different approaches to educating children can be effective. To illustrate this difficulty, the authors cite the wide disparity between the costs estimated for urban children cited by Guthrie and Rothstein in Chapter 7 and those presented by Duncombe and Yinger in Chapter 8—$500 and $5,000 per pupil respectively. On the limits of judicial intervention, the authors conclude that courts have only the power to say "no" to the existing law, thus requiring, but not defining, a good-faith response by the legislature and relying on educational experts to develop effective plans. The objective is estimable, but the road to success is unclear. Given this uncertainty, the greatest contribution of adequacy litigation, the authors say, may be that court intervention will serve as a spur to innovation and experimentation.
I agree with this chain of reasoning to a great extent. In a way that is reminiscent of the early race cases, the core problem addressed by adequacy is that children are excluded from educational opportunities and socially valuable outcomes. The basic reason for judicial intervention and a role for the courts is that the state has defaulted in its duty to raise educational outcomes, especially for those whose outcomes are the most deficient. The source of standards for educational opportunity must be the educational goals and policies of the state itself, as applied to all of its children. In the end, a court must rely on some evidence beyond judicial competence to define the level of resources and links with outcomes. I disagree only with the idea that courts are limited to striking down legislation and that they must accept some version of expert opinion. On the contrary, as a supplement to expert opinions, the courts can order social experimentation with the links between resources—such as reductions in class size—and improved outcomes. As I will explain, experimentation of this kind could fill an important gap, although no court to my knowledge has ordered a systematic program of experimentation as part of an adequacy remedy. By doing so, and structuring a process of social inquiry, courts can go beyond saying no. I agree with the authors that recent developments in New Jersey are especially interesting in this respect because the New Jersey Court went further than has any court so far in requiring a link between new dollars and outcomes. The court 1) required the state to show how new aid would address the problems of disadvantaged students, 2) ordered whole-school reform based on models with proven effectiveness, 3) required that standards be developed for schools to request extra aid for supplemental programs, and 4) ordered a comprehensive facilities plan. This kind of remedy is experimental in spirit because it includes programs that have worked in the past (to some extent, but not necessarily to the level required to meet standards) and because it allows adaptations based on school need. It is less than experimental because it does not include a range of interventions, some of which could meet standards in a short time (e.g., one year in the case of very small class sizes). Further, it does not establish a program of continuous experimentation maintained until the results required by the standards are actually reached on a small and then a larger scale.
Chapter 7 by James Guthrie and Richard Rothstein, addresses the issue of how to calculate the level of resources necessary to adequacy, or, in their words, making adequacy real. The chapter explains and advocates the professional expert strategy they used in advising the state of Wyoming about how to satisfy judicial requirements. They also criticize the econometric method used in the final chapter, which they characterize as a "black box." The method employed in Wyoming was a modification of the "resource cost model" developed by Chambers and Parrish, which consisted of asking experts to identify the necessary components of adequate education. The Guthrie/Rothstein modification advocated initiating the expert process with certain assumptions based on their own understanding of educational research on effective inputs (small class sizes in early grades, small elementary schools, the importance of professional development, extra aid only for concentrations of special needs students). The actual study conducted in Wyoming was truncated because of time limits and concentrated heavily on proper teacher salary levels and fringe benefits, bench marking with the pay received for comparable professional jobs in the private sector (the authors emphasize that school districts have monopoly power and related inefficiencies, so that it is important to estimate a true-market based wage). In spite of the skepticism built into the analysis, the resulting package was not skimpy: basic spending per pupil of $6,580, an average teacher salary of $41,433, and average class sizes of 16. To go beyond the limited scope of the Wyoming study, the authors recommend further study of extra plant and maintenance costs because of studies that show a backlog in these areas for many financially strapped districts.
The chapter gives little explanation of the method for calculating aid for special needs. A categorical grant would be available if the proportion of economically disadvantaged or limited English Proficiency students in a school exceeded 150% of the state average. The amount of financial aid that results from this grant is also unclear. Table 7-2 lists categorical grants in amounts of approximately $150,000 for school enrollments of about 300, which would yield $500 per pupil, the amount the Sugarman and Minorini chapter ascribes to Guthrie’s method. Apparently the amount received will vary depending on the number of special needs pupils, but details are not provided. Footnote 9 mentions that the amount might differ in other states, because only eleven percent of Wyoming children were characterized as poor and fewer than two percent had limited English proficiency.
The Guthrie/Rothstein method does not control in any empirically based way for the extra costs of achieving higher outcomes, nor do the other methods they mention and approve (Augenblick - see the discussion of efficiency in Part II). Recognizing the importance of this omission in any approach based on adequacy, they launch a barrage of objections to the final chapter of Duncombe and Yinger, which does estimate the cost of minimum outcomes under varying circumstances. Included among their objections are that such estimates are 1) unreliable (producing widely varying numbers); 2) obscure (cannot be understood by legislators as easily as factors like teacher salaries and class size); 3) too expensive (could never be adopted even if correct); 4) probably inefficient (based on existing spending patterns); and, 5) poorly specified in theory (for example, based on proficiency levels in standardized tests that are themselves unreliable). I agree that all of these difficulties must be confronted, but I will say below that the lack of a clear focus on outcomes in their own approach is even more damaging than the authors realize, and experimentation offers an answer to many of the technical and political problems of empirical analysis. One internal contradiction of their argument against adjusting for outcomes should be noted. The primary modification suggested in the simple resource cost model—a preference for small class sizes in the early grades—is justified by empirical research studies, like the Tennessee experiment, which the authors cite as compelling rather than unreliable and obscure. The contradiction is that these studies use statistical methods similar to those they disapprove, connect inputs to outcomes as defined by the kind of standardized tests that they reject, and omit the kind of explicit controls for efficiency that they require. Thus, the chapter slams the front door on empirical outcome studies like Duncombe and Yinger’s, but allows similar research to enter its own analysis through the back door. I regard this as testimony to the growing reliance on evaluation through standardized testing as the litmus test of academically successful programs. What else is really persuasive?.
Chapter 8, by William Duncombe and Donald Yinger, estimates the cost of adequacy from the actual expenditures of schools and districts, in light of two categories of the extra costs: higher input prices and higher environmental obstacles to producing the outcomes. The chapter begins with what the Introduction to the book calls a compelling metaphor: the input prices and environmental costs of keeping homes warm in cold weather. The adequate outcome in the experiment is a comfortable indoor temperature—for example, seventy degrees. An example of varying input prices is the cost of natural gas in different regions of the country. The greatest environmental cost factor, making an adequate indoor temperature more or less costly to reach, is exterior temperature. A warm home costs more in Minneapolis than in San Diego. Efficiency is introduced into the analysis by controlling for certain practices: for example, homes that have adequate weather stripping. Options that can be exercised by home owners must also be addressed: for example, typical owners rather than people who leave their doors and windows open all winter. Taking all of these factors into account, it is possible to calculate a cost index: the amount above or below the average cost necessary to obtain adequate warmth in different locations. In practice, two general methods are used to calculate cost indices of this kind: the engineering study, which uses a controlled experiment to measure the costs, and a statistical study, which calculates average costs from a regression analysis of data from real households that controls for cost variables and efficiencies.
The parallels between the heating cost index and an educational cost index are clear. Methods exist, such as those described in the Guthrie/Rothstein chapter, for measuring the varying costs of educational inputs, like teacher salaries (note that the same environment that makes the pupils more expensive to educate may also raise the price of inputs of hiring qualified teachers). A long research tradition (including work of the otherwise skeptical Rick Hanushek) has identified the environmental factors that tend to reduce school performance on educational outcomes (poverty household, minority/ racial status, low education of mother, non-English background). Some kind of target outcome (analogous to a 70º house) must be chosen and would logically be based on the politics and policy of a given state as goals for all students. Both input and environmental costs must be based on efficient practice and factors outside the control of schools and district (for example, salaries produced by effective rather than inept negotiation at a competitive rather than excessively generous level).
While the principles of developing an educational cost index are clear, the econometric methods for estimating those costs regrettably are not, even to someone with a solid but not advanced statistical background like myself. The authors, applying their study to New York State, present two methods, the indirect and the direct, which yield enormous differences in the cost estimates (despite the fact that both are described as well understood, sound, and based on long-established research traditions). Under the indirect method, the cost of education is represented indirectly by the level of public spending (here, educational spending) controlling for income, tax price (the tax rate required for higher outcomes), voter characteristics, input costs, and environmental costs. The basic idea is to calculate the spending for a district that is average on all the control factors, then add an adjustment for higher- or lower-than- average tax price, demographics, and income. This method yields a cost adjustment for New York City of about 10% over the state average, or $500 per pupil.
The authors prefer the direct method as intuitively more obvious to policy makers (and more obviously parallel to the heating-cost example). The method is termed direct because it makes an explicit choice of the targeted educational outcomes rather than using the proxy of actual educational spending. The first step is to choose both the type and level of outcomes aimed for. The authors choose a basket of outcomes that their previous research has shown to be valued by voters: third and sixth grade minimum proficiency scores, share of students receiving New York’s rigorous Regents’ high school diploma, and graduation rate (representing a range of grade levels, test scores plus educational level attained, and the high and low ends of the performance continuum). The target is to bring low-achieving districts up to the state average. A regression analysis calculates the costs of reaching the target relative to the average cost in the state (either higher or lower), by controlling for input costs and environmental costs. Efficiency is estimated by a process called "data envelope analysis," which estimates whether a given district has higher than average costs for reaching a given level of outcomes. This method yields an estimate of extra costs for large schools in New York City of two to three times the state average, an increment of $5,000 to $10,000 per pupil (as do estimates for other states by other authors based on the same kind of analysis). The authors calculate that, while the existing foundation system of state aid in New York closes thirty-six percent of the performance gap between New York City and the rest of the state, the larger amount specified by the direct cost index would close eighty-four percent of that gap.
I share the uneasiness expressed by Guthrie/Rothstein and Minorini/ Sugarman about the variability of results yielded by these quantitative methods. Although I agree with Duncombe and Yinger that ad hoc adjustments for costs like those of Guthrie/ Rothstein are ungrounded and unconvincing (having no method for connecting costs and outcomes whatsoever), I also agree with the reservations regarding the statistical methods of this chapter. Part of the obscurity is due to the use of regression analysis, which will always be obscure to policy makers. Here, however, it is fair to point out that the methods used in the example above for computing the heat index are based on regression analysis, which causes no difficulty to policy makers. But the authors do not explain why two widely respected methods reach such different results, while the heating indices produced by different methods are presumably very close to each other. Another difference, as the Introduction suggests, is that the technology of heating and the results themselves are well understood and familiar, whereas the technology of producing adequate education outcomes and the cost of doing so are poorly understood and politically disputed. What is the raw energy of educational outcomes, similar to natural gas, and what is the "furnace" that produces it? Based on research, I would make the argument that individual attention to students by qualified teachers is the natural gas, and class size is the furnace. But the econometric methods are not tied to any such commonly available understandings. Here again, I see a role for experimentation.
II. Some General Reactions
In this section, my purpose is to explain how the collected findings and insights of the book can be pulled together to yield a sharper and more useful understanding of adequacy in school finance. I do so by discussing a series of fundamental analytical questions that cross cut across the chapters.
What is the moral and legal imperative that adequacy must address?
Defining the problem to be solved is the most important step in sharpening our understanding of adequacy, and the essential insight is that adequacy is primarily a remedy for those suffering exclusion of one kind or another, rather than an equalizer for every child and every district. True, in one sense, the right to an adequate education is universal: every child has the entitlement. But not every child is denied it. The heart of the adequacy problem is exclusion of large numbers of children from minimally acceptable outcomes and, consequently, from minimal participation in society. This kind of exclusion also represents implicit discrimination and violation of equity, because the state tolerates these sub-par outcomes in massive numbers of its least favored children whose minority status is reflected in minimal political representation and power. My position is not, as Berne and Stiefel suggest, that the only the poor have a right to adequacy: rather, my point is that minorities and the poor are the only group for whom large numbers of low outcomes are tolerated. A perfect embodiment of the adequacy problem for me is that many states and districts set performance standards for promotion and graduation (which I regard as generally a good incentive for higher performance), but then fail to vigorously experiment with extra resources aimed at bringing failing students up to the standard that has been set. In this case, the state or district is basically watching children fail and willfully doing nothing about it in a manner that would not be tolerated for the more affluent children of the majority. (The example of children retained in grade is so compelling that it might well be the first target for experiments on raising outcomes; a possible danger is that these might be considered the only children whose outcomes are low enough to deserve relief, as opposed to all children with sub-par outcomes).
If this is the problem, the remedy obviously must be single minded in focusing on increasing outcomes rather than focusing, as Guthrie and Rothstein do, on the average district, setting ad hoc levels for children with special needs, and ignoring the results produced by those ad hoc estimates. But adequacy should also support three related claims that are not focused on outcomes: educational spending and resources far below the norm (the same type of claim that was the thrust of many of the older equity cases); drastically sub-par facilities; and, drastically incomplete educational programs. Resources equal to those of the wealthiest districts in the state cannot be guaranteed to all; but at some point, where they are represented by departures from substantial equality, the absence of dollar resources becomes exclusionary. Again, this claim for substantially equal resources may be brought under either an adequacy or equity theory. Shortages of adequate instructional space are common across the states, and sparsely populated rural districts are sometimes unable to afford to offer a minimally complete academic program. These circumstances, too, represent a total exclusion from important educational opportunities in ways that are undoubtedly connected to but lacking one-to-one correspondence with outcomes. Adequacy should not be defined in any single manner, but should allow for different standards directed at different problems.
How are equity and adequacy related and distinguished?
In effect, a number of the chapters posed the question as to whether adequacy represents a break with the past of equity litigation. I think not. Adequacy and equity are both based on principles of equity in the general sense, because both respond to unfair discrepancies in educational services based on some kind of political discrimination. Equity litigation (in the narrow sense) corrects for substantial unfairness in the base of raw inputs—resources and dollars. Outcome-based adequacy does use an absolute standard, which could yield great disparities of dollars across districts and, hence, may appear inconsistent with equity. But the extra dollars are themselves adjustments for a different kind of inequality (e.g., of outcomes) and, fundamentally, are no more in conflict with the equity of the base than any other kind of compensatory aid, such as Title I (or adjustments in building funds compensating for rapid enrollment growth, or program funds compensating for sparse population). Put another way, adequacy is vertical equity—adjustments for special needs—and, as such, coexists peacefully with horizontal equity of base spending (horizontal equity being, as Berne and Stiefel explain in Chapter 1, equality of similarly situated students, that is, those without special needs).
In fact, logically and pragmatically, adequacy should be built on a pre-existing base of horizontal equity. Suppose that students in an urban center need $10,000 extra per pupil to reach minimum outcomes. If the district is also a poor district, and $3,000 of the $10,000 would already be available under a fair equity adjustment, other districts of comparable wealth lacking the special needs students should also be receiving an extra $3,000. The Wyoming adequacy approach described by Guthrie and Rothstein, which sets a high but fiscally disciplined base for all students, may not differ very much from the kind of traditional equity litigation that culminates in a new, high-foundation program. (If the adequacy level is higher, it must be because the court thinks, as the Kentucky court apparently did, that educational standards should be raised throughout the state). In this sense, one can clearly see that the ad hoc, small, and philosophically stingy adjustment for special needs in the Wyoming approach leaves the main job of adequacy unfinished. On the other hand, the technique of setting a high, adequate base for all students may be ideal for reaching substantial equity in the dollar base while avoiding unrealistic hopes of equalization with the wealthiest districts.
How should the outcome standard be defined?
The issue of the standard for adequacy is different from and logically precedes the calculation of the amount of resources necessary to reach the standard. As several of the chapters point out, for educational outcomes, the philosophical difficulty of choosing an adequate outcome standard is almost self-evident in framing the issue— adequate in what kind of knowledge and at what level of proficiency? For example, a standard for minimal proficiency in language skills, mathematics, science, and social studies, leaves out art, music, and athletics, and, more important, stops short of higher levels of proficiency, such as advanced, that would certainly be useful to students in today’s world. Although state constitutions and courts articulate types of adequacy based on the social functions of education, precise standards depend on unexplained policy judgments. For example, why should not every student in the state perform at the advanced level, given global competition in the work place? The answer may be that achieving such a high goal would cost a state ten times its annual budget, a pragmatic consideration strong enough to influence courts in their legal definitions of adequacy.
The intrinsically judgmental nature of setting educational goals leads me to believe that the proper standard must be set by the legislature in balancing factors such as goals and costs. The standard for all students then can be set at around the state average on the goals selected as important by the state, because outcomes of the average student are the best representation of state goals. Adequacy, and the judicial role, are necessary to protect minorities in the population from becoming the victims of a discriminatory double standard—for example, one set of outcomes for the average student and another for children in urban areas. Conceptually, we would ask how far the sub-group needs to move in order to equal the state average and spend the resources necessary to close the gap. (Technically, this approach seems to threaten a "Lake Wobegon" problem, where all children are above average, because raising the scores of urban children would raise the state average, which would then increase the gap; but, like the paradox of the rabbit chasing the tortoise eternally and closing the gap by one half without passing the tortoise, the gap would be smaller with each increase and eventually disappear).
Selecting the state average also solves another problem, which is why not every single child must attain minimum outcomes. Guthrie and Rothstein point out, in a list of "unsolved mysteries" about adequacy, that equalizing average outcomes still leaves many students below the minimum target. This is a problem for any absolute standard that is not based on discrimination; but it is not a problem for avoiding a double standard, because there always will be many students in the state below the state average. Interestingly, the state average target need not be limited to minimum proficiency. Should urban populations also be entitled to equal representation at the "advanced" level of achievement? The answer could be "no" if the moral and legal meaning of "adequate" had something to do with minimum functioning in society (again, as defined by the state). On the other hand, given the philosophically indeterminate nature of the meaning of adequacy, advocates for the poor and minorities will make a strong point in asking why an equal proportion of their populations should not be at the advanced level, where most of the prizes for excellence in our society are available (for example, admission to selective colleges and universities). A final interesting note is that a state like Wyoming that does not have a standardized test defining adequate levels of performance should be required to enact one, just as the Wyoming court required the legislature to define a level of adequate resources using the input standard.
How is the exact level of dollars and resources required calculated?
Once a standard like state average in achievement is designated, I believe that the greater part of the mystery surrounding choosing the level of spending evaporates. The simple response is, It costs what it costs. If urban areas can achieve parity with the rest of the state on $500 per pupil extra, there is no reason to ask for more. But the same is true for $5,000 or $10,000 per pupil. Given that we don’t know how much it would cost, we need a preliminary estimate. My sense of the literature is that the high estimates calculated by Duncombe, Yinger, and the other authorities cited (like Rechovsky) are the most unbiased estimates presently available. To me, these high estimates also have a considerable amount of face validity based on educational research and common sense. Creating very small class sizes of five or ten could double or triple total spending in a school (depending on the number of children who needed such classes). Reducing class size seems to me to be the most obvious way to raise scores. Guthrie and Rothstein actually do endorse class size as the single most effective educational input demonstrated by research but, then, inexplicably do not apply the same logic to underachieving students. Class size does have all the intuitive appeal that these authors criticize as lacking in the statistical estimates; for example, practically all teachers will say that they need more time with underachieving students, and every parent knows how much time it takes to tutor a child through a difficult lesson.
But an estimate is just an estimate, and intuitive validity is not the same as scientific proof. Given the high cost of reducing class size, it seems reasonable that the state should be allowed to experiment with its effects on achievement before allocating the full amount for every low achieving child. Conducting controlled experiments will be methodologically challenging, because of such problems as small classes drawing better students; but these, I believe, are surmountable. The mandate to experiment also flows in the opposite direction—of finding lower-cost methods. If outcomes can be raised to the desired levels by more efficient school designs and high stakes testing, obviously education should move in that direction. A right to adequate outcomes is not a right to the most expensive, inefficient methods of achieving the outcomes. Efficiency is further addressed below, but the one thing that is not permissible is to choose methods based on assumptions about efficiency and then ignore the logical implication presented when these methods do not succeed in raising outcomes. That kind of sitting on one’s hands is just another way of maintaining double standards: actually effective practices for most students and theoretically effective practices for the poor and minorities. Yet, collective hand sitting is exactly what is going on in most states.
What is the proper role for courts, legislatures, and educators?
The classic role of the courts in institutional litigation in education and other areas of social policy (the kind that requires a complex social remedy) is identifying the underlying legislative bias, breaking the logjam of institutional inertia, and stimulating aggressive problem solving by the legislative and administrative branches of government. Minorini and Sugarman make the excellent observation that courts do not know exactly what adequacy is but can recognize what it is not, and it is not the present system, because the present system obviously was not designed with adequacy in mind. The underlying reason for the lack of " adequacy design work," if you will, is the legislative logjam. If poor and minority parents ran the state legislature, the logjam would not exist, and the states would not be sitting on their hands watching children fail. If experimentation is needed, the role of the legislature is to authorize and fund the experiment, and the role of educators is to get the most out of the increased funding. That basic judicial role—identifying a missing legislative priority, requiring a good faith response, and insisting on adequate funding —is exactly the role played by the court in the Wyoming and in other adequacy cases, except for the experimentation requirement. In other words, there is no need for a new judicial role but a need only to apply the old judicial role to the neglected area of actually generating success in raising outcomes through social experimentation.
What political strategy and context for adequacy litigation and reform?
The Fuhrman/Carr chapter lists formidable obstacles to the success of comprehensive school finance reform, but I believe that a successful political strategy can be fashioned to support the right to adequacy described here. The greatest obstacle is finding large amounts of money in a time of fiscal constraints. Here, it is fortunate that the hard work of achieving basic equity already has been completed in many states (to the extent that it has not, educational taxes probably have not been raised significantly, and consequently a broad constituency is available to support increased spending). As for the extra spending needed for adequacy, the final price tag for achieving adequacy in target groups should not amount to anything like the cost of comprehensive equity reform. A common figure for the cost of comprehensive equity reform in the past (giving money to practically every district) has been about $1 billion per state. In contrast, suppose that per pupil spending in 100 high poverty schools (a reasonable figure in a large state) were doubled from $5,000 to $10,000. If the average school size were 500 pupils, there would be 50,000 pupils in the 100 schools. Assuming that all of these students were entitled to the high poverty adjustment, the total cost would be $250 million. That is a large sum of money, and it would go exclusively to the poor and minority students disfavored by voters. However, two additional factors improve the odds of political success: First, under the proposal to begin with experimentation, the initial cost would be quite small; and the only way that the spending could go to scale would be on convincing demonstrations of actual success. Political support for spending on the poor rises dramatically when a program actually produces success based on criteria aligned with American values (of which success through hard work may be the foremost). Second, the bonus for under-achieving students should go to all districts and schools that have such students. Even though schools with relatively few disadvantaged students would not receive a large amount of aid, the strength added to the constituency could be significant, particularly since the new aid would go part way toward meeting the resentment of many voters and educators over the "unfunded mandates" of special education. Third, the proposal to include claims for a backlog of school building and inadequate programs in sparsely settled areas would broaden the benefits of the litigation and aid in building a constituency.
What is the proper role for efficiency?
One of the common objections to higher levels of funding is that the extra money would not be spent efficiently. Guthrie/Rothstein object to the high estimates of Duncome/Yinger on grounds that the estimates do not make sufficient allowance for efficiency, especially given the presumed monopoly behavior of public schools. In their study of schools receiving large increases in aid, Goertz and Natriello found absolutely no evidence of waste, but they also found little evidence of careful targeting. In principle, the problem is serious. It is impossible to give enough money to help all students even in inefficient districts and also encourage efficiency in spending; if efficiency is encouraged, some students in inefficient districts will not reach the target outcomes. And it seems wrong for public policy to use the most inefficient schools as a model. We do not know how large a factor inefficiency plays in subduing outcomes. My sense is very little, simply because the background predictors of low outcomes (poverty, etc.) do explain so much of the variance (just as cold weather, not a lack of weatherstripping, explains most variance in winter heating costs). Also, unbiased empirical estimates, such as those presented in the Duncombe/ Yinger chapter, show relatively little effect from estimates based on the average efficiency in the state (as opposed to empirically ungrounded ideological assumptions).
The question nevertheless remains about how efficiency could be maintained in a regime of social experimentation. Would teachers who are allowed small classes unconsciously reduce their level of effort on student outcomes? Again, the problem does not seem insurmountable. Experimentation should be done with different class sizes and different incentives for efficiency. For example, if a class size of ten that incorporates rewards for success gets just as good results as a class size of five without the incentives, there is no reason to spend the extra money. There is, however, one approach toward encouraging efficiency that is contrary to a fair understanding of the right to adequacy. Guthrie/Rothstein write approvingly of the approach used by Augenblick in Illinois (co-authored by Guthrie), which derived estimates of adequacy only if the district spent less than the average in the state. If we had data on schools that achieved adequacy for low-achieving students, such a standard would be fair, basically requiring average efficiency in those kinds of schools. But the Augenblick study did not include any adjustment for students with special needs, which means that the implied standard for the most costly students would be the spending required for an unusually efficient district serving the average non-costly student, which translates into an astronomical level of efficiency for students with special needs (perhaps not even present in those schools led by visionary leaders and great principals who are sometimes held up as models for all low income schools—assuming, of course, that such schools succeed by raising outcomes rather than attracting better students). That approach is obviously just another version of the double standard that it is the main job of the adequacy court to prevent.
Indeed, a double standard is dimly visible in the way that the efficiency "problem" is usually raised with respect to urban districts that need money for poor and minority children. Wealthy districts that already have equivalent funds to educate privileged students almost never bear a high burden of proof of efficiency. True, part of the reason is that unnecessarily high salaries and a golf course on the public school property of a wealthy community might be seen as an extension of the higher standard of living in the community as a whole, a privilege inevitably not extended to other districts. But tolerance of inefficiency is probably a minor factor, because even rich public schools have come under increased fiscal scrutiny. The main reason for tolerance, I believe, is that these schools are spending money for functions that are facially reasonable for the success of their students, such as a diverse and advanced curriculum taught by highly qualified teachers. Almost all urban schools can easily pass the "facially reasonable" test (Goertz and Natriello have the data to prove it.), and I would argue that any stricter test is presumptively discriminatory. My sense is that the facially reasonable test and the average efficiency test, measured quantitatively, would not yield substantially different results.
Is a state, or a national, standard preferable?
A number of the chapters discuss whether adequacy might lead to a national standard for educational finance. My sense is that, if it does so, the mechanism will be indirect rather than resulting from a literal compliance with judicial decrees. The only direct route to a national standard would occur if the standard of adequacy in each state were defined relative to outcomes that would be nationally competitive—for example, if Louisiana spent enough to catch up with test scores in Connecticut in a short period of time. If the best approach to setting state standards were the legislative goals of each state, as I have argued, state standards would become national standards only through a gradual process of interstate competition and emulation.
The other mechanism that might favor a national standard is a response by the federal government to adequacy-type claims through categorical programs aimed at politically appealing areas of adequacy funding. I do sense that the appeal of spending for adequacy is rising, especially in what are perceived as the more efficient and targeted areas (where I would place all three categories recommended here—producing increases in outcomes through smaller class sizes, catching up with facilities backlogs, and enhancing curriculum in sparsely settled areas). On the other hand, a massive increase in federal spending is somewhat difficult to imagine, given that the federal government already spends most of its money for special needs students (Title I, Special Education), and strong political pressures resist increasing the relatively small federal share of total education budgets. In terms of national norms and constitutional standards, education still is mainly a state and local function; and there is no "national education clause" in the United States Constitution under which interstate disparities in the level of adequacy could be directly challenged (the so-called national standard that would have existed with a one-vote change in the Rodriguez case would have been a national standard of intra-state equity). In other words, it is the case that now as always, if you like the public schools in Connecticut, you need to move there to access them, an option that could become much more viable if the decades-long, litigation-spurred effort to raise standards in the poor cities of that state became more of a reality.
This book of background readings on equity and adequacy in school finance greatly advances our understanding of the issues, but the picture of adequacy presented could be sharpened and made more effective. I am concerned that adequacy should keep its roots in equity by addressing the political double standard of inequitable outcomes, just as equity addressed the double standard of resources. Equity and adequacy are mutually reinforcing because they address different claims, large disparities in general resources (a problem of horizontal equity), and large disparities of meeting special needs (the problem of vertical equity addressed by adequacy). Three types of adequacy claims should be recognized: funding necessary to produce minimum outcomes, funding necessary to meet drastic backlogs of capital needs, and funding needed to staff complete academic programs in sparsely settled areas.
The background standard for calculating the amount of extra resources needed under the outcomes claim (which is, by far, the most difficult of the three to quantify) should be the average outcomes of the state as a whole for those subject areas chosen by each legislature as essential for all students. Under this approach, adequacy is achieved, and the double standard eliminated, when high-poverty schools reach the same average outcomes on measures deemed important for all children as the state as a whole. The judicial role in adequacy is the same as it was in equity, breaking the political logjam and requiring a good faith and substantially successful remedy. The funding required for adequacy in high poverty schools must be tightly tied to outcomes, rather than computed on the basis of guesswork and available funds, with no subsequent adjustments when outcomes fall short of targets. The preferred empirical estimate presented by Dunbombe and Yinger for urban districts, of two to three times state average spending, may seem high on first impression, but it has high face validity because of two factors: what parents and teachers know about the costs of catching up and what independent research tells us about the reductions in class size that would be necessary to close the achievement gap. Efficiency in the use of new funds must be encouraged, which means that some schools probably will not reach their targets; but the efficiency standard should not be higher than for the average district, in order to avoid another kind of double standard (super efficiency for the disadvantaged, regular efficiency for everyone else).
Given the large size of the best estimates of the new resources needed and uncertainties about educational technology, new funds should not be given immediately to every child in the state. Instead, courts should order a program of careful, controlled experimentation, probably with reduced class size as the main intervention, because it is the most powerful and robust in implementation. Once increased outcomes are demonstrated, the political chances for the extra funds should improve because of the demonstrated link between dollars and effective education. Building lawsuits around three claims—outcomes, facilities, and programs in rural areas—as well as giving at least some aid for special needs to all or almost all districts should broaden the coalition and help build consensus.
What is most important is that the state be discouraged from having one set of standards for most children, and watching kids fail, while making no serious, realistic attempt to intervene. Indifference to this kind of exclusion from educational opportunity, presently tolerated for the poor and minorities, would not be tolerated for those with more political power. And time may be short. Voters will support moderately expensive school finance reforms if the reforms produce results. Judicially mandated controlled experiments can produce inexpensively the kind of success on which it is possible to build a broader consensus. We should ask ourselves collectively whether we avoid trying the more expensive interventions, like very small class sizes, because we are afraid that they will fail or that they will succeed, and we must then decide whether to pay for them. Let’s take the problem one step at a time and find out, for example, how far we can raise outcomes in a very short period of time, like one school year. That is a period of time that matters most for the children now in school.