4

Financing Schools: Analysis and Recommendations

chooling matters. Decades of research confirm that both the quality and the quantity of schooling are strongly associated with increased income, better health, lower levels of criminal activity, and less reliance on public assistance.1 The justification for public financial support of schooling is both civic and personal. Schools are expected to prepare children for the responsibilities of citizenship and to improve their individual economic prospects and quality of life.

Yet schooling costs a great deal. Americans are taxed more than $900 per capita annually to support public elementary and secondary schools.2,3 This amounted to $265 billion (excluding construction) in the 199394 school year to educate 43.5 million students.4 Of every dollar spent by a government agency in the United Stateslocal, state, and federal combinedalmost nine cents goes to elementary and secondary education.5 Public school expenditures dwarf all other state and local government expenditures for children, including expenditures on health and welfare.6 The cost of educating a single studentthe per-pupil expenditure has been climbing for decades. When dollars are adjusted using the Consumer Price Index, per-pupil spending increased from $1,299 in 194950 to $5,734 in 199394.7 Guthrie in this journal issue discusses at length some of the reasons for this increase in expenditures. A large share of the increase occurred during the baby boom years between 1950 and 1972, when per-pupil expenditures rose from $1,299 to $3,517.8 This increase was largely the result of teachers pay increases and improved student-teacher ratios. Expenditure increases in the past 25 years have been partly due to additional services provided by schools, such as compensatory education for disadvantaged students, special education for students with disabilities, and a longer school year. Lawsuits to eliminate segregation and to increase equity in state funding have also contributed to higher total costs.

The Future of Children FINANCING SCHOOLS Vol. 7 No. 3 Winter 1997

While costs are rising, academic results are also improving, especially for minorities. Since 1971, student scores at all grade levels and in most subjects on the National Assessment of Educational Progress (NAEP) have gradually risen, with the greatest increases shown by black and Hispanic students.9 Graduating seniors of all races in 1992 completed an average of two more year-long academic courses than their 1982 predecessors.10 Between 1982 and 1992, the number of seniors planning to enroll in college right after high school rose from 60% to 77% of whites, from 58% to 75% of blacks, and from 46% to 75% of Hispanics.11 Nevertheless, the media regularly publish accounts of school failures which raise questions about how well tax dollars are being used. For example, accounts abound about schools with dropout rates exceeding 50%,12 high school graduates unable to read,13 school buildings in disrepair,14 and gross disparities in services between neighboring districts.15 Accounts also highlight hard-won successes.16,17 Even in the most impoverished neighborhoods, some schools bloom. And across the nations 80,000 schools, even with their widely varying funding levels, many provide a good education to virtually all of their students. Again, some argue, insufficient funding is not the problem so much as poor use of available dollars.1 In short, the key questions concerning education financing are the following: First, is the right amount of money being spent on public schools? It is impossible to answer this question without deciding what schools are expected to accomplish. Second, is the money spent on public schools being distributed in a fair manner? This analysis concludes that some states distribute funding more fairly than others. It recommends establishing a reasonable basis for each states foundation funding level and for special needs, rewarding local tax effort above the minimum funding level , and removing inequities in school construction funding. Third, can the education dollar be spent more efficiently? Among the nations more than 80,000 schools, certainly there are instances where efficiency could be improved. In any event, an effective system should routinely investigate ways to improve efficiency. This analysis summarizes some major areas of effort to increase accountability and efficiency. As a starting point, it is important to consider what schools are expected to accomplish and who is responsible for the task.

THE FUTURE OF CHILDREN WINTER 1997

Goals and Governance

A 1984 survey of several thousand parents, teachers, and students indicated that the public expects schools to achieve goals in four broad areas: academic, vocational, social/civic, and personal.18 The academic area embraces all the intellectual skills and domains of knowledge. The vocational area prepares students for the expectations of the workplace. The social/civic area prepares students for social and civic participation and develops their awareness and appreciation of their own and other cultures. The personal area emphasizes individual responsibility, talent, creativity, ability to think independently, and self-discipline. The responsibility for the far-reaching task of educating children has traditionally rested in local hands. From the colonial period, locally elected school boards have governed schools in America, though the role of the state has grown substantially during the twentieth century. This is in sharp contrast to most industrialized nations, where educational expectations are defined by a centralized ministry of education or by a core national curriculum reinforced by a national examination system.19 No such system exists in the United States, and the U.S. Supreme Court has ruled that the Constitution does not require comparable education systems among or within the states.20 As Howell and Miller discuss in this journal issue, local communities provided about 80% of all financial support for schools until the Great Depression. After the Depression, states dramatically increased their financial contribution, providing about 45% of the education budget by 1950 and maintaining that level of contribution today. This, however, is a national average. Actual state expenditures range from 74% of school costs in New Mexico to 7% in New Hampshire.21 As states financial contributions have increased, so has their control over the schools. States have implemented regulations as broad as requiring statewide tests for high school graduation and as specific as Californias requirement that students who object to the dissection of animals in biology class must have the option of using a computer simulation instead.

The federal role in education finance, though highly visible, is relatively minor. The federal government contributes about 7% of the education budget nationally. Federal regulation of education, for the most part, is not mandated. This is true of the Goals 2000 program, the Title I program for disadvantaged students, and the Individuals with Disabilities Education Act (IDEA).22 The financing and governance of schools, then, is a responsibility shared more or less equally by local government23 and the states, with relatively small federal contributions. The division of responsibilities between state and school district varies by state. All states at least specify compulsory attendance requirements, and almost all specify high school graduation requirements.24 Some states provide detailed curriculum frameworks, which may specify expectations at every grade level for every core academic subject. And most states, over a period of decades, have produced volumes of regulations on items from playground equipment safety to class size to parental permission requirements for sex education classes. Within this web of state and federal regulations, the job of the local school board is to see that an education is provided to local students. The school board sets policy for the district (including spending priorities), hires the school superintendent, and enters into collective bargaining agreements with teachers. Given the shared governance of schools, it is easy to see how layer upon layer of regulations could evolve over time, with the outcome that some believe the financing system itself might create perverse financial incentives which discourage good management or good pedagogy. Thus, the primary tasks of the school financing system are (1) to ensure that all schools have adequate funding to succeed if properly managed and (2) to support the efficient and effective operation of schools. The next section discusses the challenge of adequate funding, including equity (fairness) issues, and makes recommendations regarding how states can ensure adequate and equitable funding.

Financing Schools: Analysis and Recommendations A critical issue is at what level funding will be compared. Equity could be measured by comparing per-pupil expenditures among states, among districts within a state, or among schools within a district. Inequality is common at all three levels. The greatest amount of funding inequality is apparent when comparing states, as illustrated by Augenblick, Myers, and Anderson in this journal issue. In 199394, average per-pupil expenditures varied from $3,439 in Utah to $10,180 in the District of Columbia.28 In fact, an estimated two-thirds of the variation in per-pupil expenditures nationally can be attributed to variation in funding levels between states. However, when dollar figures are adjusted to reflect cost differences between the states, these differences are attenuated, as Howell and Miller discuss in this journal issue. Currently, equity is virtually always measured by comparing the funding of various school districts within a state. As Augenblick and colleagues discuss, by this measure some states have greater funding equity than others, and in most states, school districts serving higher-income communities spend more per pupil. It is worth noting that such measures rarely account for cost differences within a state.29 For example, wages are often higher in cities than in rural areas. Variation in spending from school to school within a district is rarely measured but, where studied, has been shown to be a common situation. This is often due to the clustering of more experienced, higher-paid teachers in certain schools within a district, as Monk and colleagues discuss in this journal issue. Some experts have proposed that efficiency would be improved by requiring equal funding for each school within a district and allowing each school greater leeway in allocating its budget, including personnel decisions.30 Recently, the nations secondlargest school district, Los Angeles Unified, agreed to provide equitable funding for each school as part of a lawsuit settlement.31 Arguments that schools and students would benefit most if equity were measured and guaranteed at the level of individual schools are intriguing. However, there is not yet sufficient data to recommend dramatic changes in the relationship between school

Adequate and Equitable Funding

Most states have at some time been sued on the grounds that their system of school finance failed to provide adequate funding, or equitable funding, or both. In this section, the concepts of adequate funding and equitable funding are discussed, and then the current funding system in which they exist is examined.

Basic Schooling and Adequate Funding

Each state constitution requires a public education system, which may be described with terms such as ample, basic, general, efficient, thorough, or uniform.25 Because this obligation arises from the states constitution, the standards for schooling within each state are decided by its legislature. Lawsuits concerning school finance are brought in state court. The term basic schooling is used to refer to the states core requirements for all schools. In many states, basic schooling is defined primarily in the form of high school graduation requirements. Again, states vary widely in the amount of detail provided in their basic schooling expectations. At one extreme, Colorado is constitutionally prohibited even from establishing statewide graduation standards.26 In Kentucky, on the other hand, the state supreme court seized the initiative by defining an adequate education for the legislature.27 Adequate funding, then, is sufficient funding to produce basic schooling. The problem is that the cost of producing a given level of education is unknown.

Equitable Funding
There is no universally accepted definition of equitable school funding. Equity is a concept based in fairness, and most people agree that fairness does not require absolutely equal per-pupil expenditures. An equitable system may provide greater funding for school districts with greater needs. Some definitions of equity extend in the other direction, reasoning that local control over tax levels is a critical component of equitythat is, unequal school funding is equitable if the difference arises from the choice of local communities to adopt higher or lower school taxes.

8 districts and schools. In addition, by far the majority of available data on distribution of funding address funding at the school district level, and these data are the basis of this analysis. The currently available data suggest that equity at the school district level should be defined by employing three criteria. First, in an equitable system, all school districts within a state are guaranteed an adequate foundation level of funding, sufficient for an efficient school system to provide a basic education to all students. Second, the foundation funding level is adjusted for special needs, especially for the number of students in poverty, with disabilities, or with limited English proficiency. Third, local communities have an equal opportunity to increase their school budget by increasing local taxes. Each of these criteria will be discussed in greater detail below.

THE FUTURE OF CHILDREN WINTER 1997 Each type of tax has its own economic characteristics. Local property taxes have the advantage of being relatively stable under changing economic conditions.32 However, property tax funding, if not adjusted, can result in extremely unequal per-pupil funding; the amount of property wealth in a school district rarely bears any direct relationship to the number of children to be educated, let alone to those childrens special needs or to the costs of doing business in that district. Sales and income taxes levied by the state tend to drop more sharply than property taxes during recessions,32 and those taxes do not exist in every state. Most states fund schools out of general revenues, not dedicated funds, and during recessions, state tax revenues fall. If states are unable to cut school funding during recessions, the legislature may be forced to make massive cuts in other state-funded programs. State legislators who create school finance formulas must take into account the widely varying circumstances of each states many school districts. Only Hawaii, the District of Columbia, and outlying areas such as Puerto Rico and Guam administer their schools in a single district. The other 49 states each have anywhere from 17 (in Nevada) to 1,044 school districts (in Texas). The nations 14,772 school districts vary widely in size. Fully 30% of the nations schoolchildren are enrolled in the 207 largest districts, although more than 3,000 school districts have enrollments of fewer than 300 students each.33 School districts also vary widely in characteristics such as the extent to which their students are provided with transportation, subsidized breakfasts and lunches, remedial education (for students performing below grade level), compensatory education (for students with inadequate preparation for learning), special education (for students with disabilities), English language instruction (for students who speak other languages at home), magnet schools (which may be required under desegregation orders), and specialized vocational education or advanced placement classes. Some districts have great difficulty in recruiting and retaining teachers, while other districts have high percentages of teachers nearing retirement who are being paid at the top of the salary scale. Some schools have signifi-

The Current Funding System

At a recent national conference of state legislators, one participant commented that only three or four people in my state actually understand our school finance system. For the next two days, participant after participant prefaced his or her remarks with comments such as, Well, of the three people in my state who understand school finance . . . or Were lucky. We must have eight or nine people who understand our states system. Why is school finance so complex that even those who write the laws confess to limited understanding of the system? Funding for schools comes from multiple levels of government, as discussed earlier, and depends on multiple taxes, each of which reacts in different ways to economic fluctuations. Funding must be distributed fairly among many school districts, which vary widely in size and needs. Changes in the economy, political priorities, and court orders all necessitate frequent adjustments to the funding system. And states distribute funds to school districts through multiple mechanisms, each of which has a different impact on adequacy and equity. Funding for schools is derived from taxesprimarily property taxes at the local level, sales and income taxes at the state level, and income taxes at the federal level.

Financing Schools: Analysis and Recommendations cant costs related to vandalism and security. Schools with growing enrollments and overcrowded classrooms have financial needs that are different from those of schools with shrinking enrollments. All of these variations have significant implications for the amount of money a district might need. The mechanisms by which states distribute funds to school districts matter greatly because state dollars comprise a significant portion of most district budgets and because some distribution mechanisms enhance funding equity, while others exacerbate problems of inequity. State funds are distributed primarily through three mechanisms, used simultaneously in most states: a basic funding level, commonly called the foundation level of funding; categorical funding; and state equalization of locally generated tax revenues. (Other distribution mechanisms and variations on these three are described in the article by Augenblick, Myers, and Anderson in this journal issue.) These three basic mechanisms form the core of the school finance system, and each is important. Calculating the cost of a basic education is a difficult task. The strengths and weaknesses of several approaches are summarized by Augenblick and colleagues in this journal issue. The simplest approach is historical: spend at the same level as in the past, with cost-of-living adjustments. This approach, however, does not consider whether past expenditures were adequate or used efficiently. Other approaches rely on theoretical models of schooling: a model school may be theorized with 500 students, one principal, 25 teachers, so many textbooks, and so on. Such models may be simple or complex. These designs may omit some important costs, or they may be associated with legislative mandates of certain school structures. The approach that offers the greatest promise is the one currently being tried in Ohio, which Augenblick and colleagues call the successful schools approach. Analysts in Ohio identified schools with higher-thanaverage student scores (above the 70th percentile for the state) in core subject areas at several grade levels. An effort was made to ensure that selected schools represented the full range of demographics found in the state, looking at students, facilities, enrollment trends, and community variables. The average per-pupil expenditure in these districts then formed the states foundation level of funding. Similar work is being done in Mississippi and Illinois. In Illinois, the state board of education identified efficient districts, defined by high academic performance in comparison with other schools enrolling students of similar socioeconomic background, and with belowaverage per-pupil spending.35 There are limitations to this approach. While it calculates the cost of operating an efficient school, it does nothing to make inefficient schools more efficient. It does not ensure greater student achievement or improved district or school management. Like many finance changes, it may cause some districts to lose funding, which they can be expected to oppose. It may lead to higher overall spending if the identified successful schools are currently spending more than the states average. These levels of higher spending may not be sustainable in times of recession, disrupting school operations.

Foundation Level of Funding

Almost all states establish a foundation level of funding for all districts. In other words, every school district is guaranteed a level of funding which is presumed to be adequate to provide a basic education. A fundamental problem is that the states foundation funding level is usually a political calculation rather than an educational one. The state legislature calculates the amount of money it plans to allocate to schools (usually about the same as last years budget), subtracts the money it wants to restrict to specific educational purposes, and essentially declares the remainder to be the foundation level of funding.34 There is no reason to assume that the foundation level thus calculated is adequate to provide basic educational services. However, there appears to be a growing expectation that the cost of basic schooling must be more carefully calculated. State courts in Ohio, Oregon, and Wyoming have recently required their respective state legislatures to explicitly define a basic education and to cost out an adequate foundation level of funding in a logical manner.35

10 The major benefit of this approach is the assurance that all districts in a state are at least funded at a level at which student success has been shown to be possible. An additional benefit is that the successful schools approach avoids mandating a specific instructional program or management structure. Foundation funding based on successful schools, if clearly communicated, should help assure voters, the media, and the state supreme court that the legislature is providing adequate, equitable funding to all schools. Additional steps should strengthen that assurance, as is discussed next. Overall, the first, fundamental step a state can take to ensure adequate and equitable funding is to calculate a statewide foundation funding level on a rational basis.

THE FUTURE OF CHILDREN WINTER 1997 staff training, or food service health standards. Categorical funding may either promote or work against equality of funding among districts. For example, funding for textbook purchases is generally provided on a perpupil basis, without regard to district wealth. By contrast, federal Title I funding is based upon the number of students in poverty in the district and so will benefit poor districts more than wealthy districts. Pupil Weighting Funding based on student demographic characteristics may be distributed through the foundation funding program, rather than through categorical funds, by assigning additional weight to students with certain characteristics. For example, a student with profound hearing loss might count as 2.3 students for funding purposes, while a student who qualifies for a free lunch might count as 1.2 students. The use of pupil weights tends to equalize funding by providing more funding to school districts with more students in poverty. Funds distributed through a pupil weighting system are also less likely to be restricted, which may increase their impact, at least in an effectively functioning school district. Where this occurs, the funding mechanism itself promotes equity. However, there are some concerns. First, when weights are assigned to students with disabilities, they create incentives to group students in broad categories that may not reflect their true educational expenses. Two students with cerebral palsy may have dramatically different educational needs. Second, weighting based on student poverty generally reflects past funding levels. If previous programs provided a 20% supplement for poor students, those students will be weighted as 1.2 students in the future, usually without analysis of whether 20% is the true cost of an effective intervention. Student Poverty Perhaps the most challenged schools are those serving a concentration of students in poverty, here defined as a school in which more than 40% of the students qualify for free or reduced-price lunches. In fact, student outcomes are more strongly affected by

RECOMMENDATION
s To provide adequate funding for all schools, states should guarantee every school district a foundation level of funding sufficient to achieve state-specified basic academic goals. The foundation level of funding should be based on the actual expenditures of schools currently meeting or exceeding those academic goals.

Adjustments for Special Needs

Once a state has determined a foundation level of funding, that amount is usually adjusted to accommodate the special needs of the students or the district. Those adjustments are made through one of two mechanisms: categorical funds or pupil weighting. Both mechanisms are discussed here as are three student characteristics that commonly require additional funding: poverty, disability, and limited English proficiency. Categorical Funds The states and federal government provide categorical funds, which are funds restricted to specific purposes. Here, only funds related to student poverty, disability, and English proficiency are discussed. However, states may use categorical funding to address any aspect of schooling, such as building operation and maintenance, computer purchases,

Financing Schools: Analysis and Recommendations the concentration of poverty in the school than by the students own family income.36 A poor student in a school with few poor students has a better chance of good academic outcomes than does a nonpoor student in a school where most of the students live in poverty. Poverty is most prevalent in urban school districts. About 40% of urban students attend schools with high poverty concentrations, as compared with 10% of suburban students and 25% of rural students.37 Only one study, prepared for the National Center for Education Statistics (NCES), compares school funding with detailed data on household income by aligning census data blocks with school district boundaries. NCES findings are discussed at length in Appendix B in this journal issue. In brief, the study found that household wealth had a substantial impact on school funding only at the ends of the income spectrum. Schools serving communities in the wealthiest decile spent 21% more per student than the national average, while those serving the poorest decile spent 11% less than the national average. The districts serving the other eight deciles were fairly tightly clusteredall within 6% above or below the national average. Thus, any examination of equality in school funding should focus especially on the schools in very-low-income communities. Although school failure is common in low-income communities, schools serving a concentration of students in poverty have achieved success. Some of these successful schools have implemented instructional or management models such as those associated with Robert Slavin, James Comer, or Ted Sizer, while a smaller number of others have brought about success based on their own designs.16 These success stories almost always occur at the level of individual schools rather than throughout entire school districts. The question for school finance is whether additional funding is needed for schools in very-low-income communities to succeed. It is difficult to know whether successful schools in low-income communities spend more than their less-successful neighbors because school districts rarely keep accounts separately for individual schools. (Rather, districts keep accounts by line item, such as teachers salaries or classroom supplies, on a districtwide basis.) However, most successful school improvement efforts in low-income communities have required substantial resources for planning and staff development to implement new teaching strategies, new organizational practices (such as teacher problem-solving teams), or both.16,17 For example, the New American Schools initiative supports several school improvement models, each designed to be implemented with minimal additional outside resources but each requiring outside assistance and teacher staff time for planning and training.17 In some instances, these resources have been provided by reallocated regular school funds (for example, Title I funds, desegregation funds, or school district staff resources). In other instances, financial support has come from outside the school funding system (for example, grants from foundations or corporations, or free technical support from university researchers). Finally, planning and staff development should not be considered a one-time need; schools require periodic reinfusions of planning and development activities to maintain success.

11

RECOMMENDATION
s To improve student outcomes, schools serving concentrated populations of lowincome students should have access to highquality technical support and sufficient funding for planning and focused staff development.

To determine an adequate level of funding for schools serving a concentration of students in poverty, states should identify successful schools operating in challenging circumstances and analyze in detail their budgets and other factors such as staff stability, focused schoolwide staff development, or school size. Special Education Students with disabilities have the right to a free, appropriate public education provided

12 in the least restrictive environment.38 This right is protected through a complex set of procedures discussed at length in the spring 1996 issue of this journal. The costs of special education are extremely difficult to estimate because they are often integrated into regular expenses (for example, part of the regular teachers time). However, it is clear that special education is more expensive than regular education. Nationally, about 10% of public school students receive special education services, though the proportion varies by state. The process of identifying and diagnosing disabilities is somewhat subjective, especially for learning disabilities, behavioral/emotional disorders, and mild mental retardation. However, data from the National Longitudinal Transition Study (NLTS) of Special Education Students show that children living in poverty are more likely to be diagnosed with all types of disabilities, including clearly measurable disabilities such as hearing loss or impaired mobility. In addition, impoverished students with disabilities are less likely to have access to the advanced education and training that could enable them to break out of poverty.39 Schools need, and receive from federal and state sources, supplemental funding for special education services. The difficulty is that basing funding on a diagnosis creates a financial incentive for schools to overdiagnose disabilities, especially in marginal cases. In some states, inadvertent financial incentives encourage the use of certain diagnoses (for example, by providing more state support for a student with mental retardation than for one with a learning disability) or for restrictive placement (where the state provides a larger share of the cost of educating children in a separate classroom). Some experts encourage policies that allow schools flexible use of special education funding so long as it benefits the child. Flexibility in funding makes it easier for a school to design ways for a student with learning disabilities to receive supplemental tutoring in the regular classroom rather than being pulled out of class for tutoring in a separate setting. New Hampshire and Iowa have made major commitments to supporting flexible use of funding to promote

THE FUTURE OF CHILDREN WINTER 1997 greater integration of special and regular education. We do not recommend increasing existing state support based on the number of children identified as needing special education, because it can create an incentive to overidentify disabilities and may prevent the flexible use of funds. However, all types of disabilities are associated with poverty, and states could address this problem by increasing the amount of the supplement provided for student poverty in recognition of the increased special education needs of children in poverty. Limited English Proficiency More than 2.3 million students ages 5 to 17 have limited proficiency in spoken English. This does not include students who speak English well but have limited ability to read or write in English.40 As a group, these students require educational services beyond those of the average English-speaking child. In Dade County, Florida, the superintendent estimates that it costs $1,141 more per year to educate a foreign-born student than a native-born one.41 While research does not indicate whether bilingual education or English language immersion produces better outcomes, it is clear that teachers require additional support to meet the needs of students with limited English skills. Teachers who are not fluent in the childs language can benefit from training and assistance in areas such as assessing the childs current academic skills; adapting classroom materials to make them more visually accessible; relying less on lectures to convey information; including cultural issues related to the childs background, especially for refugees; and training in the language acquisition process.

RECOMMENDATION
s To provide needed supplemental services, the foundation level of funding should be adjusted to account for the level of student poverty and limited English proficiency within each school district.

Financing Schools: Analysis and Recommendations

13

State Equalization of Local Funding

In most states, local communities have the option of approving higher taxes to fund schools above the level of foundation funding and special needs funding. The problem is that districts with strong tax bases can raise money relatively easily, and districts with limited taxable property cannot raise much per pupil, even at high tax levels. Property-rich districts often have two advantagesa larger tax base and fewer schoolchildren among whom to divide the proceeds. (Howell and Miller discuss this issue in some detail in their article in this journal issue.) For example, the wealthiest and the poorest districts in a state might each vote to raise property taxes by ten cents per $100 in assessed property value. However, District A has high property values and a small school-age population, resulting in $500,000 in assessed property valuation per child. A ten-cent property tax increase in District A would generate $500 in tax revenues per child. District B, with $30,000 in assessed valuation per child, would generate only $30 per child with a ten-cent increase.42 If the state equalized funding between the two districts, the state would contribute $470 per child to District B, to match the $30 contributed by local taxpayers. The concern then is how much of a burden this would place on the state. No state has yet been willing to fully equalize local taxes that is, to guarantee that all districts will be able to choose school funding as generous as that provided in the wealthiest communitiesbecause the costs are simply too high. However, it seems reasonable that every community should be able to increase funding within a reasonable amount over the bare-bones budget guaranteed for a basic education. As Adams discusses in this journal issue, this problem was addressed in Kentucky by creating two optional tiers above the foundation level. Tier 1 local tax increases are equalized by the state, but Tier 2 increases are not. Tier 1 covers additional expenditures up to 15% over the foundation funding level. (The foundation level includes adjustments

for special needs.) Thus, if the foundation funding level in the two districts discussed above was $3,333, the next $500 (15% of $3,333) would be equalized. District A would raise the full amount through its ten-cent property tax increase. District B, which also raised property taxes by ten cents, would generate $30 per child in local taxes and receive $470 per child in state equalization funds. If both districts also elected to raise taxes to a still higher level, the next increment would be at the Tier 2 level and would not be equalized by the state.43 Kentuckys equalization of the first 15% increment above the foundation level of funding puts enrichment of the basic program within the reach of every community, and it rewards local tax effort without obligating the state to match the full level of expenditures that the wealthiest districts will choose.

RECOMMENDATION
s Foundation funding plus supplemental funding for poverty and limited English proficiency should make it fiscally possible to deliver a basic education. In addition, every community should have the option of providing an enriched, higher-quality school program at a reasonable cost to local taxpayers. When local communities voluntarily use this option to raise local taxes to benefit schools, states should partly subsidize low-property-value communities to reward them for their tax efforts, up to a reasonable limit.

Equalized Funding for Facilities Construction

No one wants to see children attending schools with leaky roofs, windowless classrooms, dangerously substandard electrical systems, or nonfunctioning toilets. Yet these and other sorry conditions persist in more than a few schools.44 Although inefficiency, deferred maintenance,45 or even fraud or graft are sometimes to blame, the lions share of the problem is caused by overreliance on local communities to pay for facilities construction.

14 Nationally, local taxes pay for 80% of new school construction (as discussed in Appendix A in this journal issue), yet as already noted, local taxes are the most unequal income source available to schools. Many states have passed statewide bonds to finance school construction, but the available funds address only a small portion of the need. Currently, state-level bonds and other state funding pay for only 20% of new construction, which is not sufficient to equalize the enormous disparity in local tax bases. Building new facilities is extremely expensive, with estimates ranging as high as $6.5 million per elementary school and $13 million per high school.46 While most school districts struggle to meet the cost of building and maintaining facilities, the struggle is clearly greatest in those districts with large school-age populations and limited tax bases. Many property-poor districts already tax themselves at a significantly higher rate than their property-rich neighbors, just to cover basic operating expenses.42 The nations time-honored practice of relying on local funding for facilities is a major contributor to the continuing fiscal and educational crisis in communities with inadequate tax bases. The problem is exacerbated by the cost of maintaining facilities, also usually a local responsibility. The most recent national analysis indicates an immediate need for $112 billion in school facility repairs, which does not include expansion to relieve overcrowding.44 In 1994, Arizona became the first state in which the court found the state school funding system unconstitutional solely on the ground that the state failed to provide for equitable funding of school facilities.47 There are, of course, understandable concerns behind the tradition of local funding for facilities. Many fear that, if states assume responsibility for school facilities, local districts will have no incentive to plan modestly or maintain facilities prudently, leading to considerable waste.48 Even those who agree that the state should play a greater role are concerned about what the total cost will be. Because the construction of facilities is so costly, it is important that districts closely examine all options. Some communities,

THE FUTURE OF CHILDREN WINTER 1997 given the choice, might opt to save money on facilities by renting inexpensive storefront or factory space and to shift those savings to alternative uses such as computers, counseling, or enrichment activities. Smaller schools, such as charter schools, have greater flexibility to experiment with this option. States should equalize local taxes for facilities construction. As discussed earlier, equalization provides equal reward (in revenues generated per pupil) for equal tax effort (in tax rates as a percentage of property value). Because local voters must contribute their own tax dollars to qualify, some incentive is created for prudent oversight and modest construction budgets. States may build in additional incentives for prudent construction and maintenance practices, as Florida recently has.49 Additionally, states can build in a per-pupil ceiling amount to avoid being called upon to contribute to construction of excessive or luxurious buildings. Equalization also keeps the responsibility for construction and maintenance at the local level, which avoids creating a state-level bureaucracy charged with owning or maintaining school buildings.

RECOMMENDATION
s To relieve school overcrowding or to pay for major building repairs or improvements, low-property-value school districts in most states currently must levy local taxes far higher than those in high-property-value school districts. When school districts raise local taxes (or issue bonds) for construction, states should equalize funds generated by low-property-value districts, up to a reasonable limit.

In addition, many states need to make up for the cumulative impact of decades of inequitable facilities funding. To determine the extent to which this is needed, each state should inventory its existing school facilities and analyze its history of funding for building construction.

Accountability
The mechanisms by which school funding is generated and distributed to school districts

Financing Schools: Analysis and Recommendations and schools have an important impact on the adequacy and equity of funding. To promote effective schools, it is important to integrate funding mechanisms with accountability mechanisms. An important example of a comprehensive effort to integrate funding mechanisms with programmatic school reform occurs in Kentucky, as Adams discusses in this journal issue. There, the state legislature increased funding in schools, transferred significant control and accountability to the school site level, and created school-site-level incentives and sanctions based on student outcomes. Here, four aspects of the overarching goal of designing financing mechanisms to support accountability are discussed. These four aspects are (1) the importance of a limited number of stable, measurable goals; (2) the role of incentives and sanctions; (3) the potential of vouchers, charters, and private contracts; and (4) the need for more detailed fiscal information to inform the accountability process. neously. The same conclusion was reached in a RAND study, which compared schools with focused missions against schools with diffuse missions. The authors concluded that schools with diffuse missions graduate a significantly smaller percentage of students, and a smaller percentage take the Scholastic Aptitude Test (SAT) and receive lower average combined SAT scores.50 Focus is a necessary, though not sufficient, condition for successful schools.

15

RECOMMENDATION
s To hold schools accountable for the efficient and effective use of funds, education authorities at the state and local levels should commit to a few highly stable, measurable academic priorities.

Incentives and Sanctions

A common complaint is that school finance systems hand out funding without regard to the quality of performance. The same perpupil funding and the same pay scales apply to the most effective and the least effective schools and teachers. Critics charge that this unbusinesslike arrangement impedes productivity in the schools.19 Can the school finance system contribute to school improvement through the use of incentives and sanctions? Incentive and sanction programs have been controversial and surprisingly complex to implement. Despite the challenges, however, at least six states (and some school districts) operate incentive programs for schools, based on student performance.51 Some of the major issues revealed through experience with incentives in South Carolina, Dallas,52 Kentucky,5355 and Mississippi,54 and with sanctions in Kentucky and Mississippi, are summarized below. It is, unfortunately, impossible to sift out the precise impact of incentives on student performance because these incentive programs were implemented simultaneously with other major reforms, including teachers pay increases.

Limited Number of Measurable Goals

The ultimate effectiveness of schools depends upon their ability to focus their efforts on specific objectives.50 Doing so is not easy. Battles erupt over the distribution of finite educational resources. A focus on reading at the elementary level requires resources which are then not available for advanced placement classes or athletic competition or replacement of aging textbooks. Establishing priorities is a difficult job. School boards are on the front lines of this issue, needing to create consensus about the goals of schooling in a political forum. In many states, the politics of setting standards for schools occurs also at the state level. In 1995, the Consortium on Productivity in the Schools19 noted that research in the commercial sector shows companies with a diffuse focus fail to raise productivity. Greater diversification reduces productivity; greater focus on core activities increases it. Because schools are governed through a political process, goals tend to multiply. This has the effect of disabling accountability mechanisms. It is simply impossible to hold a school accountable for achieving 200 different objectives simulta-

16 Most analysts recommend that incentive programs base awards and sanctions on the performance of entire schools, rather than of individual classrooms. As Clotfelter and Ladd note, Experience shows that financial incentive programs directed toward individual teacherslike merit payhave not worked well, and teachers have disliked them. The limitations of such programs are well known: the lack of consensus about what makes for effective teaching; the fact that gains in student achievement often reflect not just the actions of an individual teacher but also the more general environment for learning in the school; and the growing recognition that rewarding individual teachers encourages them to compete with one another rather than to work cooperatively.56 In contrast, school-based incentive programs are believed to encourage teachers to support and learn from one another and, potentially, to change schoolwide cultures so teachers demand more from students. To implement incentive and sanction programs, states must choose what aspects of student achievement to measure. The simple issue is which grade levels and academic subjects to test. More difficult is whether and how to assess important but difficult-tomeasure skills such as written expression, acceptable behavior, employability, and citizenship. Questions arise over how, if at all, the incentive system should take into account the fact that some schools operate in more challenging circumstances than others. Dallas implemented a complex system for assessing improvement which adjusts for factors such as student mobility and classroom overcrowding. However, these and other adjustments, made through complex mathematical formulas, were poorly understood. Teachers reported feeling that their students adjusted scores were not meaningful to school staff and yet were the basis of awards. All of the programs reported problems with inadvertent incentives not to test certain students, especially those who moved during the school year, were enrolled in special education, or were viewed as potential dropouts. Participants in Kentucky and Mississippi complained that the system rewarded schools for focusing teaching efforts primarily on top-performing stu-

THE FUTURE OF CHILDREN WINTER 1997

dents. In Dallas and Kentucky, where incentives took the form of cash awards to teachers (up to $1,000 in Dallas and $2,600 in Kentucky), there were numerous reports of cheating by teachers to improve student scores. Other major concerns were teaching to the test, a narrow focus on only those subjects expected to be tested, and morale problems among teachers who failed to qualify for an award. Sanctions, or negative consequences for poor performance, have been tried in several forms, including negative recognition, increased supervision, and takeovers. Negative recognition is the simplest and may possibly be the most effective sanction. In Kentucky, schools in crisis are publicized, including notifying students that they have the option of transferring to more successful schools. One survey of Kentucky teachers found that fear of this negative publicity was their strongest motivation toward greater teaching effort.16 Kentucky and Mississippi, among many other states, provide increased supervision of schools judged to be performing poorly. Several states have recently passed what are commonly called academic bankruptcy laws, allowing the state department of education to take over failing schools. In Chicago and in Cleveland, the mayors office has recently assumed control of the schools. A similar option has been tried at the school district level, commonly called reconstituting schools. In San Francisco, for example, the school superintendent has reassigned the entire staff in some failing schools. There is no definitive evidence yet that these severe sanctionswhether at the state, city, or district levelimprove student performance. Experience has, however, identified some concerns. One common concern is that [i]t is not clear that states have the technical capacity to improve low-performing schools and districts.57 In Mississippi, analysts note that the state department of education simply does not have sufficient resources to provide meaningful technical assistance to troubled schools. Officials involved in school takeovers in Chicago and New Jersey report problems establishing productive relationships with demoralized or hostile school staffs. In Kentucky, the

Financing Schools: Analysis and Recommendations poor reliability of student test data delayed the implementation of sanctions for two years. Public reaction to sanctions has, in some instances, been colored by fears that state assessment of schools represents a threat to family values.54 Although problems and concerns accompany programs that implement incentives and sanctions, these should not be taken as a judgment that incentives and sanctions will not work in schools. Accountability for results is important and must be meaningful. However, at this time, the analysis of existing incentive and sanction programs provides more insight into what problems to avoid than what processes result in improved student outcomes. dents with special needs without sufficient funding to meet these needs.58,59 Two critical questions, then, are: Do any of these models (charters, vouchers, and contracts) do a better job of serving the students enrolled? And does the existence of these models have significant consequences (positive or negative) for those students remaining in regular schools? These are extremely important questions for which existing evidence is not sufficient to provide answers. Charters Since 1991, more than half the states have enacted laws allowing charter schools, and at least 480 charter schools have begun operation. In many states, charter schools operate independently from the school district. Many of them have a special focus, such as a Montessori-style elementary program or an alternative high school program for students at risk of dropping out. Most charter schools have been in operation for three years or less, and student achievement data have not yet been collected on any wide scale. In addition, because charter school students are self-selected, comparing their performance to the performance of students in public schools will not be a simple matter. It is clear, however, that fiscal stability will be a major challenge for charter schools in almost every state. As Koppich discusses, most of the new laws grant charter schools only limited access (or no access) to start-up funds, categorical funds, or capital funds. Their access to day-to-day operational funds is often considerably less than that provided to traditional schools. Critics have also expressed concern about the fiscal impact of charters on regular schools. For example, the transfer of students to a charter school may decrease the regular schools income without being sufficient to cut some of its major fixed costs (salaries and buildings). Consequently, the impact would be disproportionately felt on the schools discretionary services, such as libraries, music, or teacher training, eroding the schools ability to support creative and innovative projects. Again, there is insufficient evidence to know how realistic this concern is.

17

Charters, Vouchers, and Private Contracts

In this journal issue, Koppich discusses the considerable interest today in charter schools (public schools which operate with minimal control and supervision from the school district), school vouchers (public funds for students to attend private schools), and contracted educational services (contracts with private agencies for services such as teaching foreign language classes or operating entire schools). Each of these models represents a new blend of reform in pedagogy, school management, and school finance. These experiments are controversial. While each model has its own controversies, there is also considerable overlap. Proponents of new models believe that decreasing regulation and increasing competition among schools for students will lead to greater efficiency in school spending. They argue that principals and teachers will be held accountable for student results to a greater extent, leading to improved student performance. Critics are concerned that loosening regulations may be harmful to students, especially those with special needs who are traditionally protected through regulation. In addition, they fear that selective admissions and competition for students will eventually lead to a two-tiered system in which the most advanced students enroll in selective schools at public expense and the public schools become heavily weighted with stu-

18 Vouchers Critics of public schools charge that private schools produce better student outcomes at a lower cost. Private school students are more likely than public school students to graduate from high school and apply to college, and they score higher on standardized tests of reading and math.60 And costs appear to be lower in private schools. In 199091, private schools charged an average tuition of $1,800 for elementary students and $4,400 for secondary students.60 However, tuition does not cover the full cost of private schooling. Most private schools also receive state and federal subsidies for compensatory education and special education, subsidies from affiliated churches,61 and/or income from endowments or alumni contributions. Appendix C in this journal issue compares public schools and Catholic schools serving low-income inner-city communities, but the studies cited could not control for the fact that neither the rules by which private schools operate nor their students are identical to those of public schools. Private school students must compete for admission and must pay tuition (a factor that presumably reflects family commitment to the school). Private schools have much wider latitude to invoke disciplinary actions and expulsions, they are rarely subject to collective bargaining agreements, their governing bodies are not subject to public election, and they are only minimally regulated by state and federal governments. Indeed, the differences are so great that it is impossible to draw any conclusions about the applicability of private school expenditures to the financing of public schools. However, there is considerable interest in (and also considerable opposition to) using vouchers as a way of making the presumed strengths of private schools available to a wider group of students. Vouchers are government payments to households which are redeemable for tuition payments at authorized private schools. Although the concept of vouchers is widely discussed, the only sizable experiments with publicly funded vouchers are in Milwaukee, Wisconsin, and Cleveland, Ohio. In both cities, courts have found the use of public funds for vouchers to religious schools to be unconstitutional, though both

THE FUTURE OF CHILDREN WINTER 1997

cases are being appealed.62,63 And, as Koppich discusses, two analyses of the same student data in Milwaukee reached opposite conclusions about the impact of the voucher program on student performance. Likewise, the impact of vouchers on public schools is, at this time, entirely speculative. Vouchers on a very small scale (for example, available to less than 5% of the student body) could have an impact completely different from that of universally available vouchers. However, no large-scale experiments have yet been conducted. Private Contracts Public school systems sometimes enter into contracts to purchase services from private for-profit or not-for-profit organizations, including universities or associations of teachers. These contracts may be for limited services, such as janitorial services or foreign language instruction, or may be to manage entire schools. Contracts to manage schools have been highly controversial and, in fact, have been deemed failures and terminated by school boards in Baltimore, Maryland; Miami, Florida; and Hartford, Connecticut. Part of the problem in these instances was the lack of clarity, and subsequent disputes, over what powers were delegated to the contractor and what powers were reserved to the school board.64 Other whole-school management contracts in several states appear to be less controversial, though they are still in the start-up stages. An estimated 10% of charter schools are managed by forprofit businesses through private contracts.65 Student performance data are not yet available. A recent book promoting the private contracting option recommends that all public schools be managed through contracts, which would require major changes in at least two common school finance mechanisms. First, the authors propose that school districts guarantee equal funding for each school (based on enrollment). As Monk, Pijanowski, and Hussain discuss in this journal issue, this funding approach would be a significant change from the current practice of allocating resources (one teacher for so many students) rather than cash to schools. Second, the authors recommend that contracts be negotiated between the district and each school, under which

Financing Schools: Analysis and Recommendations school site managers control all budgetary decisions, personnel matters (including hiring, firing, job assignment, and pay scales), and instructional decisions. These contracts would be renewed only if student outcomes were acceptable.30 Such major changes in finance mechanisms could, the authors suggest, lead to improved accountability, better incentives for teachers, and increased stability for successful schools, among other things. At a minimum, such proposals remind us that efforts to increase success in schools should carefully examine the role that financing mechanisms play in shaping instructional and personnel decisions, and in creating unintended incentives that either support or distort the schools mission. The U.S. Chamber of Commerce, in a recent effort to improve the fiscal management of schools, helped develop a new accounting model which allows school districts to routinely collect and analyze spending data by school, by grade level, and by academic subject.67 Comparable accounting models may be available from other sources. South Carolina is coordinating statewide adoption of a single, new and detailed accounting program, which is expected to have two benefits. Districts will have much more detailed financial information available for their own use, and the state will have comparable data from every district to improve financial analysis at the state level.

19

Ongoing Information Needs

To know which expenditures are more closely associated with improved outcomes, it would be necessary to obtain much more detailed information about spending than most school districts commonly have available. As Monk and colleagues discuss in this journal issue, much is known about how schools spend money within broad functional categories. Despite a wide variation in income and student needs, school districts spend money in surprisingly similar ways. Most school districts spend about 60% to 65% of their budget on instruction (teachers salaries, benefits and training, classroom supplies, libraries, and computers). Four other functions consume roughly 10% of the budget each: administration (including district staff, principals, and school secretaries); facilities maintenance and operation (including utilities);66 transportation and food service; and student services (for example, school nurses, psychologists, attendance officers, counselors, and speech therapists). These similar allocation patterns mean that wealthier districts simply purchase greater quantities or more expensive goods or services within the same categories. However, these data reflect broad functional categories for entire school districts. Until recently, virtually no school districts were able to analyze their spending by school, by grade level, or by subject. Without this important information, it is almost impossible to determine which expenditures are productive and which are not.

RECOMMENDATION
s To make ongoing, informed decisions about the use of funds, education departments at all levels should collect much more detailed information about how education budgets are spent and how resources are utilized.

Conclusions
Public schools are charged with a heavy responsibility: accept all of societys children, from every background, and give each one a fair start in life. Provide them with academic skills, job skills, citizenship skills, and self-knowledge. Recognize and develop the unique strengths of each individual. Motivate children when they are unmotivated. When they are difficult, do not give up on them, for without schooling their chances in life are greatly diminished. How much should it cost to do this? How much is society willing to spend? Desires must be balanced against pocketbooks. The dreams for our childrens futures are unlimited, but the resources are not. Ultimate responsibility for this difficult balancing act must rest at the state level. Even where local taxes supply more than half a schools budget, every state is required by its own constitution to ensure that all schools have at least adequate funding to provide a basic education. In making this

20 assessment, the critical problem of funding school construction must be faced. Beyond the bare minimum, every state has at least a moral responsibilityand, in many states, a legal responsibilityto promote equity. In school funding, that means providing assistance to meet the special needs of students in poverty, students with disabilities, and students with limited English proficiency. It should also include providing every community with the opportunity to choose to spend above the minimum funding level provided for a basic education at a reasonable cost to local taxpayers. To be effective, state education policies and school finance mechanisms must also be specific enough to hold educators accountable, yet flexible enough to address

THE FUTURE OF CHILDREN WINTER 1997 the unique challenges of each district and each school. Challenges are especially great for schools serving a concentration of students in poverty and students with limited English proficiency. These schools, more than any others, must have access to specialized expertise and sufficient resources for planning and staff development. Each mechanism for financing schools has the potential for either negative or positive impacts on the accountability, efficiency, and basic fairness of the system. Complexities notwithstanding, the design and details of each states school finance system are critical to the success of its school children. Donna L. Terman, J.D. Richard E. Behrman, M.D.

1. Hanushek, E.A. Outcomes, costs and incentives in schools. In Improving Americas schools: The role of incentives. E.A. Hanushek and D.W. Jorgenson, eds. Washington, DC: National Academy Press, 1996, pp. 2952. 2. Americans were taxed $933 annually to support public schools in 1993. See U.S. Department of Commerce, Bureau of the Census. Statistical abstract of the United States: 1996. Washington, DC: U.S. Government Printing Office, 1996, p. 298. 3. Throughout this analysis, the term public schools refers only to elementary and secondary schools, not institutions of higher education. 4. U.S. Department of Education, National Center for Education Statistics. Digest of education statistics, 1996. Washington, DC: U.S. Government Printing Office, 1996, pp. 5051. 5. This amount was determined by author calculation. Direct expenditures at all levels of government (excluding intergovernmental transfers and noncash tax expenditures) totaled $10,770 per capita in 1993. See note no. 2, U.S. Department of Commerce, Bureau of the Census. 6. Orland, M.E. State investments in education and other childrens services: The fiscal challenges ahead. Washington, DC: The Finance Project, 1995, p. 9. 7. U.S. Department of Education, National Center of Education Statistics. Digest of education statistics, 1994. Washington, DC: U.S. Government Printing Office, 1994, p. 163. 8. These figures are somewhat misleading because inflation should be calculated differently in service industries, such as education, which rely heavily on professional labor. Such calculations would reveal a slower rate of increase in educational expenditures. For further discussion, see Rothstein, R., and Miles, K.H. Wheres the money gone? Changes in the level and composition of education spending. Washington, DC: Economic Policy Institute, 1995. 9. Grissmer, D.W. Education productivity. Washington, DC: Council for Education Development and Research, 1997. 10. See note no. 4, U.S. Department of Education, National Center for Education Statistics, p. 132. 11. See note no. 4, U.S. Department of Education, National Center for Education Statistics, p. 137. 12. Rossi, R. Holding students back: Pass or fail? Chicago Sun-Times, August 25, 1997, at Section NWS, p. 12. 13. Unready students at college. The Washington Post. April 22, 1996, at A16. 14. Purnick, J. Old schools are unfixed, new unbuilt. New York Times. April 29, 1996, at B1. 15. Kozol, J. Savage inequalities: Children in Americas schools. New York: Harper Perennial, 1992. 16. Odden, A. Raising performance levels without increasing funding. School Business Affairs (June 1997) 63,6:412.

Financing Schools: Analysis and Recommendations

21

17. Stringfield, S., Ross, S., and Smith, L., eds. Bold plans for school restructuring: The new American schools designs. Mahwah, NJ: Earlbaum, 1996. 18. Goodlad, J.I. A place called school: Prospects for the future. New York: McGraw-Hill, 1984. 19. Consortium on Productivity in the Schools. Using what we have to get the schools we need: A productivity focus on American education. New York: Columbia University Institute on Education and the Economy, 1995, p. 37. 20. San Antonio Independent School District v. Rodriguez, 411 U.S. 1 (1973). 21. Hawaii, which has only one school district, has been excluded from this calculation of the range of state expenditures on public schools. 22. For the most part, the extensive regulations in the Individuals with Disabilities Education Act (IDEA) merely codify federal court rulings interpreting the constitutional equal rights of students with disabilities. If a state should choose not to accept IDEA funding, the states educational programs would still have to meet constitutional standards, which are largely the same as IDEA regulations. For further discussion, see Martin, E.W., Martin, R., and Terman, D.L. The legislative and litigation history of special education. The Future of Children (Spring 1996) 6,1:2539. 23. Historically, school boards had the ability to levy taxes, and most retain that authority today. However, some states, such as California, have virtually eliminated the ability of local school boards to increase property tax rates to benefit schools. And in several large urban school districtsfor example, New York City; Chicago, Illinois; Baltimore, Maryland; Philadelphia, Pennsylvania; Nashville, Tennessee; and Atlanta, Georgiathe city council or county board of supervisors controls property tax rates and, thus, a large share of the school districts budget. 24. See note no. 4, U.S. Department of Education, National Center for Education Statistics, pp. 14448. 25. Enrich, P. Leaving equality behind: New directions in school finance reform. Vanderbilt Law Review (1995) 48,1:10194. 26. See note no. 4, U.S. Department of Education, National Center for Education Statistics, p. 144. 27. Rose v. Council for Better Education, 790 S.W.2d 186 (Ky. 1989). 28. See note no. 4, U.S. Department of Education, National Center for Education Statistics, p. 144. 29. Mishel, L., and Rothstein, R. Measurement issues in adjusting school spending across time and place. Paper presented at the Annual Data Conference of the National Center for Education Statistics, Washington, DC, July 1997. 30. Hill, P.T., Pierce, L.C., and Guthrie, J.W. Reinventing public education: How contracting can transform Americas schools. Chicago: University of Chicago Press, 1997. 31. Rodriguez v. Los Angeles Unified School District, Los Angeles County Superior Court, No. C 611358, Consent Decree (1992). 32. Odden, A. School finance: A policy perspective. New York: McGraw-Hill, 1992. 33. See note no. 4, U.S. Department of Education, National Center for Education Statistics, pp. 9697. 34. Augenblick, J.G., and Myers, J.L. Determining base cost for state school funding systems. Denver, CO: Education Commission of the States, February 1994. 35. Education Commission of the States. Clearinghouse notes: Determining the cost of a basic or core education. Denver, CO: ECS, 1997. 36. Kennedy, M.M., Jung, R.K., and Orland, M.E. Poverty, achievement, and the distribution of compensatory educational services: An interim report from the National Assessment of Chapter 1. Washington, DC: U.S. Department of Education, 1986. 37. Parrish, T.B., Matsumoto, C.S., and Fowler, Jr., W. Disparities in public school district spending, 198990. NCES Research and Development Report 95-300. Washington, DC: U.S. Department of Education, National Center for Education Statistics, 1995. 38. See note no. 22, Martin, Martin, and Terman. 39. Wagner, M.W., and Blackorby, J. Transition from high school to work or college: How special education students fare. The Future of Children (Spring 1996) 6,1:10320. 40. U.S. General Accounting Office. Limited English proficiency: A growing and costly educational challenge facing many school districts. GAO/HEHS-94-38. Washington, DC: GAO, 1994.

22

THE FUTURE OF CHILDREN WINTER 1997

41. Visiedo, O. Something has to give. Education Week (January 18, 1995) 14,17:34. Available online at http://www.edweek.org. 42. In fact, for this reason, many property-poor districts have high tax rates. U.S. General Accounting Office. School finance: State efforts to reduce funding gaps between poor and wealthy districts. GAO/HEHS-97-31. Washington, DC: GAO, 1997. 43. In Kentucky, Tier 2 taxes can be raised up to a maximum of 30% of the sum of the foundation level and Tier 1. Using the same example, Tier 2 in Districts A and B could be a maximum of $1,150. Thus, the widest spread possible between districts would be $1,650. District A could have per-pupil funding of $4,983 ($3,333 foundation level funding plus $500 at Tier 1 and $1,150 at Tier 2). District B, if it chose not to raise taxes to Tier 1, would have the $3,333 of foundation level funding. 44. U.S. General Accounting Office. School facilities: Condition of Americas schools. GAO/HEHS-9561. Washington, DC: GAO, February 1995. 45. In many poor urban schools, building maintenance is routinely deferred to keep class sizes lower. See Appendix B in this journal issue. 46. See Appendix A in this journal issue. 47. Roosevelt Elementary School District No. 66 v. Bishop, 179 Ariz. 233, 877 P.2d 806 (1994). Figures cited by the Arizona court convey the urgency of the problem. The presence or absence of taxable commercial property led to assessed valuations per pupil ranging from $5,800,000 to $749. Even in two districts with comparable commercial property, differences in the value of residential property led to per-pupil assessed valuations of $130,778 and $18,293, respectively. A first glance shows Arizona had a normal level of commitment to funding equity: the state provided 45% of the total education budget, 90% of the states districts received some equalization funding, and the foundation funding level included a line item for capital improvements in every district. However, on closer examination, the court found that the foundation level of funding, and specifically the line item for capital improvements, was insufficient even to pay for long-term maintenance, let alone for new construction. In effect, all facilities needs had to be met entirely from local property taxes. The result was a gross disparity in local tax rates (with property-poor communities paying more) and the quality of facilities (with property-rich districts benefitting greatly). Although Arizonas court has taken a strong stand, the legislature has not yet created a funding plan acceptable to the court. See Johnson, R.C. Judge dismisses latest school finance plan in Arizona. Education Week (September 3, 1997) 17,1:22,28. Available online at http://www.edweek.org. The Ohio court has recently followed Arizonas example. DeRolph V. State, 780 Ohio St.3d 419 (1997). Unfortunately, the situation found in Arizona is common to most states. See Dahlkemper, L. Rundown schools: Whose responsibility? State Legislatures (September 1997) 23,8:1519. 48. Between 1985 and 1995, schools in Kansas City, Missouri, received more than $1.3 billion of court-ordered state support for desegregation, a substantial part of which was used to improve facilities, though critics charged it did not lead to better student outcomes. Walsh, M., and Schmidt, P. In K.C. case, court curtails Judges powers. Education Week (June 21, 1995) 14,39:1,23. Available online at http://www.edweek.org. 49. Floridas recently passed H.B. 2121 provides cash rewards to districts that practice frugal construction practices. See note no. 47, Dahlkemper. 50. Hill, P.T., Foster, G.E., and Gendler, T. High schools with character. Santa Monica, CA: RAND Corporation, 1990. 51. Stecklow, S. If their schools score well, state gives teachers cash. Wall Street Journal. September 2, 1997, at A1. 52. Clotfelter, C.T., and Ladd, H.F. Recognizing and rewarding success in public schools. In Holding schools accountable: Performance-based reform in education. H.F. Ladd, ed. Washington, DC: Brookings Institute, 1996, pp. 2363. 53. Miller, E. Early report from Kentucky on cash rewards for successful schools reveal many problems. The Harvard Education Letter (1996) 12,1:13. 54. Elmore, R.F., Abelmann, C.H., and Fuhrman, S.H. The new accountability in state education reform: From process to performance. In Holding schools accountable: Performance-based reform in education. H.F. Ladd, ed. Washington, DC: Brookings Institute, 1996, pp. 6598. 55. See note no. 51, Stecklow. 56. See note no. 52, Clotfelter and Ladd, p. 27. 57. See note no. 54, Elmore, Abelmann, and Fuhrman, p. 83.

Financing Schools: Analysis and Recommendations

23

58. DeSpain, B.C., and Livingston, M. Roads to reform. The American School Board Journal (July 1996) 183,7:1720. 59. Hawley, W.D. The predictable consequences of school choice. Education Week (April 10, 1996) 15,29:56. Available online at http://www.edweek.org. 60. National Center for Education Statistics. Private schools in the United States: A statistical profile, 199091. NCES 95-330. Washington, DC: U.S. Department of Education, Office of Educational Research and Improvement, 1995. 61. Eighty-six percent of private schools are affiliated with a religious organization. (Author calculation, based on note no. 60, National Center for Education Statistics, p. 11.) Subsidies may be in the form of free classroom space, free business or administrative services, or lower pay to nuns in Catholic schools or to other teachers. 62. Walsh, M. Voucher plan in Cleveland is overturned. Education Week (May 7, 1997) 16,32:1,31. Available online at http://www.edweek.org. 63. Walsh, M. Judge overturns expanded Wisconsin voucher plan. Education Week (January 22, 1997) 16,17:1,17. Available online at http://www.edweek.org. 64. Hill, P.T. Contracting in public education. In New schools for a new century. D. Ravitch and J. Viteritti, eds. New Haven, CT: Yale University Press, 1997. 65. Stecklow, S. Businesses scramble to run charter schools. Wall Street Journal. August 21, 1997, at B1. 66. These figures are for current expenditures only, not capital (that is, construction) expenditures. 67. Speakman, S.T., Cooper, B.S., Holsomback, H.D., et al. Tracking expenditures to the classroom. School Business Affairs (February 1996) 62,2:3444.