Skip Navigation
New Revenues for Public Schools: Alternatives to Broad-Based Taxes

Michael F. Addonizio
Wayne State University

About the Author



Although discussions of public elementary and secondary education in the United States often include calls for greater financial commitment to our schools, K-12 education enjoyed steady and substantial growth in real resources over the century-long period ending in 1990 (Hanushek and Rivkin 1997; Guthrie 1997). Nearly all of the school revenue of this period was raised from a set of broad-based state and local taxes, with the relative shares of state and local tax revenues changing substantially. Revenues from local taxes, almost exclusively property taxes, have fallen from more than 80 percent of the total in the 1910-30 period to about 45 percent by the mid-1990s, whereas state contributions, raised primarily by sales taxes and personal and corporate income taxes, have risen from less than 20 percent in the 1910-30 period to nearly 48 percent by 1995. Federal funding, never a substantial share of total public school revenue, has risen from negligible levels during the pre-1930 era to about 7 percent by 1995 (Howell and Miller 1997). In sum, varying combinations of broad-based local, state, and federal taxes provided U.S. public schools with steadily rising support over this extended period.

This century-long trend of steady revenue growth, however, came to an abrupt halt in 1990 (Hanushek and Rivkin 1997), although both school enrollments and expectations for academic achievement continued to rise. To meet their students' and communities' expectations in the face of essentially flat real revenues from traditional tax sources, local school districts in recent years have turned increasingly to nontraditional sources of revenue. These nontax sources of revenue, which are not consistently reported by local school districts in standard financial statements, include user fees; partnerships with postsecondary schools, government agencies, and private businesses; donations; volunteer services; interest earnings on investment of school resources; and the creation of educational foundations to promote giving from individuals and businesses. New sources of revenue may also come from new forms of school choice.

This paper will examine the sources of these nontraditional revenues, the institutional arrangements by which these revenues are raised, and the legal restrictions placed on these revenue-raising activities. This paper will also assess the extent of public reporting of such revenue and review the proposed reporting standards of the Governmental Accounting Standards Board (GASB) regarding these revenues. Finally, the activities and impact of local educational foundations in Michigan will be examined and comparisons will be drawn between foundation and nonfoundation districts in terms of educational and socioeconomic characteristics.

National trends in K-12 public school spending are summarized in the section titled "National Trends in Public School Spending." The rise of nontraditional revenue for the support of public elementary and secondary schools is discussed in the "Sources of Nontraditional Revenue" section, along with specific sources of such revenues. The work of the GASB regarding these revenues and comments on the extent to which states collect such revenue data from local school districts is reviewed in the "Donor Activities" section. In the "Enterprise Activities" section, public school revenue and expenditure trends in Michigan since the adoption of constitutional limits on taxes and expenditures in 1978 are examined. The "Shared or Cooperative Activities" section begins a detailed examination of nontraditional revenue for Michigan public schools, including trends in revenue collections since 1988-89 as reported in local district financial reports and key findings from a survey of local education foundations. This section also compares foundation and nonfoun-dation districts on selected socioeconomic and educational variables. A model of local education demand to test for behavioral differences between residents of foundation and nonfoundation districts in Michigan is presented in the "Reporting Nontraditional Revenue" section; empirical results are presented in the "Public School Revenue Trends in Michigan" section. Conclusions regarding the equity effects of nontraditional revenue, particularly local foundation revenue, and the extent to which such revenues are, and should be, included in standard school district financial reports are presented in the "Nontraditional Revenues for Michigan Public Schools" section.

back to top

National Trends in Public School Spending

For the past century, public elementary and secondary education in the United States has enjoyed remarkably steady revenue growth. Hanushek and Rivkin (1997) report that real expenditures per pupil increased at 3.5 percent per year over the entire period of 1890-1990, with total annual expenditures rising from $2 billion to more than $187 billion, in constant 1990 dollars, over this period. This nearly 100-fold increase is more than triple the growth of the U.S. gross national product (GNP) over this period, with K-12 public school expenditures increasing from less than 1 percent of GNP in 1890 to 3.4 percent in 1990. This increased spending resulted from a combination of falling pupil-staff ratios, increasing real wages paid to teachers, the expansion of educational services for special education students, and rising expenditures outside the classroom, including spending on central administration, plant maintenance, and pupil transportation (Hanushek and Rivkin 1997).

Since 1990, however, the growth rate in per pupil expenditures appears to have fallen precipitously. Although real spending per pupil grew at a rate of 3.75 percent in the 1980s, the growth rate from 1990 to 1993 was a mere 0.6 percent (U.S. Department of Education 1995). This lower growth rate is due, in part, to the return of growth in school enrollments, which have been rising nationally since 1981. Furthermore, resulting fiscal pressures on public schools are exacerbated by the steady growth of the special education population, for whom financial support is mandated by federal law. As noted by Meredith and Underwood (1995), cost containment is of only secondary importance in the special education paradigm. Under the Individuals with Disabilities Act (IDEA), school districts must provide every special education child with a free, appropriate education regardless of cost to the district. On average, per pupil expenditures for special education equal approximately 2.3 times per pupil expenditures for regular education (Chaikand et al. 1993). Moreover, the special education population continues to grow more rapidly than the general student population, rising from 11.6 percent of total enrollment in 1990 to 11.9 percent in 1992.1

These pressures on regular education funding are exacerbated by stringent tax and spending limits enacted in a number of states (Mullins and Joyce 1996; Mullins and Cox 1995).

As of 1994, 43 states specifically limited local revenues and expenditures by means considered more constraining than full disclosure—truth in taxation measures that require public discussion and specific legislative action prior to enactment of tax rate or levy increases (Mullins and Cox 1995). Twelve states have set overall property tax rate limitations. Thirty states limit tax rates levied by specific types of local governments. Twenty-five states limit local tax levies, 6 states limit the growth in assessments, 3 states limit general revenue growth, 8 limit expenditure growth, and at least 17 have some form of full disclosure requirement (Mullins and Joyce 1996).2 Mullins and Joyce (1996) examined the effects of tax and expenditure limitations (TELs) using pooled, cross-sectional, time-series models of state and local spending and observed a diminished use of broad-based taxes at the local level and a "dramatic increase in reliance on user charges and miscellaneous revenue sources from both state and local governments." As revenue growth from broad-based taxes slowed and enrollments grew, public schools increasingly sought revenue from alternative sources.

back to top

Sources of Nontraditional Revenue

Public school districts across the United States have long attempted to identify and tap into so-called nontraditional sources of revenue. The term "nontraditional" appears to stem not from a limited history of school revenue raised from sources other than broad-based taxes but from their relatively small magnitude. Research into the collection of these revenues dates to at least the early 1980s. Meno (1984) categorizes these efforts to augment traditional, broad-based tax revenues into three types of activities: donor activities, including the solicitation of goods, services, and money; enterprise activities, involving the selling or leasing of services and facilities; and shared or cooperative activities, whereby functions are pooled with other agencies or organizations to lower costs. Other nontraditional initiatives include the investment of school resources and the pursuit of new government funds through grant writing (Pijanowski and Monk 1996). Schools and school districts have enjoyed limited and uneven success in raising revenues from these sources. Potential budget impacts of 7 percent to 9 percent have been reported for public schools in regional studies of alternative revenues (Meno 1984; Picus et al. 1995; and Salloum 1985). Although the motivation of such revenue-raising efforts is often some degree of fiscal stress, some evidence suggests that relatively wealthy school districts enjoy greater success in tapping into these revenue sources than do their less-affluent counterparts. Thus, these revenues may exert a mild disequalizing effect (Addonizio 1997).

back to top

Donor Activities

Direct Donations. Meno (1984) characterizes donor activities as any activities intended to raise funds, goods, or services from nongovernment sources. These donations can take the form of direct district fund-raising from individuals or from corporations and foundations. Resources raised in this fashion may consist of large single donations for a specific purpose. For example, the Beloit public schools in Wisconsin received $440,000 from a local foundation to buy microcomputers as part of an experimental program in computer education (Meno 1984). Other examples of direct donor activity include an enrichment fund established by local businesses and community members in the Tuscon (Arizona) Unified School District, the funding of a dental prevention model in the Wichita city schools by the American Dental Association, and the support of a health education project in the North Glenn (Colorado) School District by the Gates Foundation (Meno 1984; Maeroff 1982).

Indirect Donations: School District Foundations.School districts in recent years have turned increasingly to an alternative type of donor activity—the indirect donation of funds through local district educational foundations, which are nonprofit organizations created to receive donations for the district. For example, in Michigan, 153 such nonprofit organizations have been established by local districts to raise revenue for curriculum improvements, enrichment activities, capital projects, and instructional materials, and also to strengthen links between schools and communities. Furthermore, this activity in Michigan appears to be part of a growing national trend. Although reliable national figures are not available, the National Association of Educational Foundations (NAEF) estimates that by the year 2000 there will be 4,000 public school foundations throughout the United States (NAEF 1996).

Districts may create foundations through which money can flow to fund a variety of school activities. Examples of large, urban districts taking this approach include San Francisco, Washington, D.C., Dallas, and Oakland, California (Meno 1984). Alternatively, foundations may be created for a single purpose. For example, the Escondido County Union High School District in California established a foundation following passage of Proposition 13 to support its interscholastic athletics program (Meno 1984). In New York City, parents in an affluent area raised money to retain a popular teacher whose job was threatened by budget cuts (Anderson 1997).

Although the scope of such foundation activity across the United States has yet to be accurately measured, the rise of these organizations is not surprising in light of the slowing of revenue growth for public schools. This development, however, has not been viewed with universal approval. Concern has focused on the possible disequalizing effects of foundation revenue. Virtually every state allocates school aid to local districts by means of equalizing formulas designed to offset disparities in local fiscal resources.3 Local education foundations have raised concerns that they may exacerbate fiscal disparities. For example, political economist and former U.S. Labor Secretary Robert Reich has characterized these organizations as "another means by which the privileged are seceding from the rest" (Pollack 1992). The impact of local education foundations on school finance equalization efforts in Michigan is examined in the section titled "Nontraditional Revenues for Michigan Public Schools."

Indirect Donations: Booster Clubs. In addition to school district foundations, schools rely on booster clubs to support specific activities. Club members develop fund-raising strategies, including networking with local businesses, and coordinate their efforts with the school activities they support. Club activities may focus on a single school or an entire district. School programs enjoying the support of boosters include athletics, band, orchestra, chorus, debate, and drama (Meno 1984). Booster volunteers, who are often school parents, frequently obtain donations of school supplies (e.g., equipment, uniforms) from local vendors who are then provided commercial access to the students through advertising in school venues and publications (Pijanowski and Monk 1996). In addition, members often make direct cash or in-kind contributions to support school activities and associated staff (e.g., end-of-season gifts or bonuses for coaches). Although anecdotal evidence suggests that booster activities are widespread across U.S. public schools, research in New York state revealed that many school officials were not familiar with booster club activities associated with their schools (Pijanowski and Monk 1996).

back to top

Enterprise Activities

User Fees.Under these arrangements, users of school-provided programs or services are required to pay for those services. Examples of fee-based arrangements are driver education programs, swimming instructions, school supplies, athletics, and pupil transportation. However, as a growing number of schools have considered the imposition of fees for educational supplies or services, these fees have been challenged on federal and state constitutional grounds and state statutory provisions.4 Restrictions on the enforcement of school district fee policies are largely a matter of state law, as federal courts generally defer to state authorities in these matters (Dayton and McCarthy 1992).

According to a 1991 survey of state departments of education, only eight states allow local public schools to charge fees for required textbooks (Hamm and Crosser 1991).5 Many more states, however, allow fees for general school supplies and services. This same survey found 29 states permitting equipment fees, 20 states permitting lab fees, and 20 states allowing fees for field trips. Other permitted fees included general supplies (12 states), workbooks (15 states), and pencils and paper (11 states; Hamm and Crosser 1991).

Tuition fees are generally prohibited for required courses offered during the academic year. Furthermore, although fees for elective and summer school courses have been allowed in the past, they have been subject to legal challenge in recent years (Dayton and McCarthy 1992).6 On the other hand, fees for extracurricular activities have become more widespread in recent years. A total of 23 states allow fees for participation in school clubs and 21 states allow fees for participation in interscholastic sports (Hamm and Crosser 1991). Thirty-four states permit fees for pupil transportation, although these fee revenues are relatively small (Wassmer and Fisher 1997). Many local school boards provide fee waivers for children of low-income families.7 Although many states permit the use of fees for "auxiliary" services, local school districts have used them only minimally. User charges provided only 3.2 percent of school district revenue in 1977, and then declined to 2.8 percent of revenue in 1991 (Wassmer and Fisher 1997).8

Leasing of Facilities and Services.Local school boards often raise revenue by leasing facilities to community organizations or private enterprises. In some instances of severely declining enrollments, districts have leased entire buildings to private tenants (Pijanowski and Monk 1996).9 Districts also lease excess space to public agencies in exchange for services to be provided to students, school staff, and neighborhood residents (Meno 1984). In addition to leasing property, some districts raise revenue by leasing services. Examples include selling food services or computer support (e.g., business services, test scoring) to private nonprofit organizations or private schools, and the sale of transportation services to public nonprofit organizations or government agencies (Meno 1984; Pijanowski and Monk 1996).

Sale of School Access.The sale of access to school markets, generally through advertising on school property or in school publications, is another means by which public schools generate revenue. Examples include the sale of advertising on school buses in New York City and advertising on homework handouts in California (Pijanowski and Monk 1996). School districts also sell concessions to businesses for various services such as student pictures and vending machine operations. Perhaps the most well-known example of the sale of school access is the arrangement between Whittle Communications Channel One and local school districts whereby, in exchange for about $50,000 worth of programming and equipment (including a satellite dish, recorders, and television sets), students are exposed to daily news broadcasts that include some advertisements. In 1995, over 8 million students in approximately 12,000 schools received daily broadcasts from Channel One. This audience comprises approximately 40 percent of the students in the 6th through 12th grades nationwide (Johnston 1995).10

back to top

Shared or Cooperative Activities

School districts sometimes seek to share operating costs by establishing cooperative programs with other governmental agencies, private nonprofit or community organizations, colleges or universities, or businesses.

Governmental Agencies. Examples of these activities include the use of public buildings for instruction, the shared use and cost of recreational facilities (e.g., pools, gymnasiums), and sharing transportation vehicles with local governmental agencies (Pijanowski and Monk 1996). According to Meno (1984), the most common shared activity between schools and governmental agencies involves the running of local parks and recreation departments, including the shared use and maintenance of playing fields and grounds. Although most of the arrangements are intended to be fiscally neutral for both parties, there are exceptions. For example, the Merced City (California) School District provides use of playing fields and grounds to the parks and recreation department. In return, the department makes a yearly contribution to the district's capital account for fields and grounds that exceeds the district's additional operating costs (Meno 1984).

Higher Education. These partnerships include opportunities for high school students to take courses at local community colleges or 4-year institutions in lieu of high school courses. Under such cooperative arrangements, students would not pay tuition, and the college would enjoy free use of school district staff.11 Meno (1984) identifies a number of school districts that participate in graduate student internship programs with local universities. For example, local colleges may place psychologist interns in public schools, where they perform standard school district functions under district supervision but at a substantially lower cost to the district as compared with regular staff costs.

Private Nonprofit Agencies. School districts often share excess space with local social service providers. Rather than charge the provider for a share of the cost of facility maintenance, the district makes the space available in return for social services provided to students at no charge. As Pijanowski and Monk (1996) note, the ability of local schools to negotiate such arrangements may assume more importance as greater demands are placed on local schools for social services.

Business and Industry Schools often rely on business partnerships to share operational, instructional, and programmatic costs. Businesses, in turn, are given an opportunity to enter schools and classrooms. Schools benefit by gaining access to the expertise of business officials who have the opportunity to shape educational programs to meet needs of the business community (Monk and Brent 1997). Such cooperative arrangements date back to at least the 1960s. For example, New York City schools have long maintained cooperative efforts with local businesses and industry to assist students as they enter the labor force. Activities include work-study, job placement, career guidance, basic skill training, remedial education, and curriculum development (Meno 1984).

A common result of school outreach to the private sector is school adoption. In return for donations of money or service, business employees receive training in teaching techniques, use of athletic facilities, and access to students for marketing research (Pijanowski and Monk 1996).

back to top

Reporting Nontraditional Revenue

Although revenue from enterprise, cooperative, and direct donor activities are generally reported in standard local school district financial reports, revenues from indirect donor activities are not. The apparent rise in the number of local education foundations and, to a lesser extent, booster clubs, and the dearth of information regarding revenue levels raised from these sources has been noted by the Governmental Accounting Standards Board (GASB 1994). GASB, established as an arm of the Financial Accounting Foundation in 1984 to promulgate standards of financial accounting and reporting with respect to activities and transactions of state and local governmental activities, has noted the rise of "affiliated organizations"; that is, organizations that are not themselves governmental entities but exist for the purpose of raising resources for such entities. According to GASB standards, affiliated organizations should be considered a part of the "financial reporting entity" and subject to the same public reporting requirements that apply to the governmental entity. Examples of such affiliated organizations arguably include school district foundations and, possibly, booster clubs. GASB Statement No. 14,The Financial Reporting Entity,defines that entity as consisting of not only the primary government but also "organizations for which the primary government is financially accountable" and "other organizations for which the nature and significance of their relationship with the primary government are such that exclusion would cause the reporting entity's financial statements to be misleading or incomplete" (GASB 1994). The statement cited a nonprofit fund-raising corporation affiliated with a college as an example of an organization that should be evaluated as a potential component unit subject to governmental reporting standards that apply to the financial reporting entity. However, the statement did not provide specific guidance for identifying these "affiliated organizations" (GASB 1994).12

In December 1994, GASB published a draft of a proposed statement that would establish a definition for affiliated organizations and financial reporting guidance for those organizations. According to the draft, an "affiliated organization" is one that meets the following criteria:

  1. The organization has separate legal standing, where neither direct association through appointment of a voting majority of the organization's governing body nor fiscal dependency exists.
  2. The affiliation with a specific primary government is set forth in the organization's articles of incorporation—for example, by reference to the name of the primary government in describing the purposes for which the organization was established.
  3. The affiliation with a specific primary government is set forth in the organization's application to the Internal Revenue Service for exemption from payment of federal income tax pursuant to Internal Revenue Code (IRC) 501(c)(3)—for example, by reference to the name of the primary government in response to any of the questions contained in the exemption application—and the organization has been granted that exemption.

According to the draft, the affiliated organization should be reported as a component unit of the primary government "if the primary government has the ability to impose its will on that organization or there is a potential for the organization to provide specific financial benefits to, or impose specific financial burdens on, the primary government." The draft also states in a footnote that an affiliated organization should be reported as a component unit of the primary government "if the nature and significance of the relationship with the primary government are such that exclusion would cause the primary government reporting entity financial statements to be misleading or incomplete." The draft would require that an affiliated organization component unit be included in the financial reporting entity "by discrete presentation" and provides guidance for reporting transactions between the primary government and the component units of affiliated organizations, based on the form of those transactions. In response to critical comments from public school booster clubs and parent-teacher organizations (PTOs), the exposure draft was withdrawn and, at the time of this writing, is being revised by GASB staff. Although the revised statement is expected to exempt small PTOs and booster clubs from the financial reporting requirements, local school district education foundations will likely be subject to new disclosure requirements. Such foundation activity has been particularly widespread in California and Michigan (Brunner and Sonstelie 1997; Addonizio 1997).

back to top

Public School Revenue Trends in Michigan

Prereform Period. Trends in state and local revenue per pupil from 1981-82 through 1992-93, in constant 1992-93 dollars, are presented in table 1.

As table 1 reveals, total per pupil revenue fell in 1982-83 and 1983-84, as Michigan (and the United States) weathered a recession that began in 1979 and persisted until 1983. Real revenue then rose slowly through 1985-86, and increased a robust 9.6 percent in 1986-87. Following a modest 1.2 percent increase in 1987-88, revenue rose by fully 14.5 percent in 1988-89. The rate of real growth then fell steadily from 1989-90 through 1992-93, turning negative in that year. This decline in real per pupil revenue growth, combined with flat or falling enrollments in many Michigan school districts and increasing academic expectations as reflected by more challenging state assessments of pupil achievement in reading, writing, mathematics, and science, and an achievement-based school accreditation program created by the legislature in 1994, led some districts to search for nontraditional sources of support.

Michigan School Finance Reform. In 1994, the Michigan legislature enacted the state's most sweeping fiscal reforms in more than 20 years, reducing property taxes, increasing the state share of school funding, and substantially reducing local discretion regarding school taxation and expenditure decisions. On the allocation side, the new legislation replaced a 20-year-old district power equalizing (DPE) school aid formula and numerous categorical grants with a foundation formula that closely regulated local per pupil revenue. Each district's 1993-94 combined state and local base revenue for school operations became the basis for determining its 1994-95 foundation allowance. The major components of a district's base revenues were localad valoremproperty taxes, DPE aid, and most state categorical aid.

The new state formula substantially constrained per pupil revenue growth for high-spending districts.13 Furthermore, the state-imposed constraint on per pupil revenue growth was designed to become binding on more local districts in the 1995-96 fiscal year and beyond. This constraint is imposed on local districts in the form of a state basic foundation allowance set at $5,000 for 1994-95 and indexed annually to nominal school aid fund revenue per pupil. This basic allowance has risen slowly, from $5,000 in 1994-95 to $5,153 in 1995-96, $5,308 in 1996-97, and $5,462 in 1997-98. Local districts at or above the basic foundation allowance receive an absolute dollar increase in their district foundation allowances equal to the dollar increase in the basic foundation allowance.14 Districts below the basic foundation allowance in 1995-96 and subsequent years receive increases up to double that amount. As the finance system is currently designed, the number of local districts subject to this constraint will rise each year, as relatively low-spending districts are boosted to the basic foundation allowance and then locked in at that level.

Aggregate Revenue Trends. The financial position and revenue levels of a local district also depend, of course, on its enrollment levels. Given the universal practice of allocating state aid on a per pupil basis, recipient local districts with excess capacity and rising enrollments enjoy positive marginal revenue and negligible marginal costs, whereas districts with falling enrollments face declining revenue and the need to lower variable costs, principally staff costs. Although aggregate school district revenues and expenditures will differ according to net changes in district fund balances, total operating expendi

tures provide some indication of the fiscal constraints facing local districts. Nominal and real total current operating expenditures (TCOP) for Michigan's local school districts from 1978-79 through 1996-97 are presented in figure 1. These data indicate a period of fiscal stress well before the implementation of Proposal A in 1994. Beginning in 1979-80, real TCOP declined 4 consecutive years and did not regain the 1979-80 level until 1991-92. Indeed, over the entire period examined, which begins with the first year of the implementation of Michigan's constitutional tax and expenditure limitation amendment, TCOP rose only about 1 percent annually in real terms.15

back to top

Nontraditional Revenues for Michigan Public Schools

Tracking the growth of nontraditional revenues in Michigan public schools is made difficult by the lack of complete and uniform reporting by local districts and consistent time-series data. One source of consistent, but historically limited, time-series data is Michigan's Common Core of Data (CCD) for school years 1988-89 through 1995-96. These data are summarized in table 2.

Nontraditional local revenue includes fees for transportation and student activities; investment earnings; direct donations; and revenues from food services, tuition, summer school, community service, and rentals. It does not include indirect donations, such as those from local education foundations. Note the reduction in "total local revenue" effected by Proposal A, beginning in 1994-95.

As table 2 indicates, reported nontraditional revenue for Michigan school districts has been fairly substantial, accounting for nearly 6 percent of revenue from all sources, down from nearly 8 percent in 1988-89, and more than 20 percent of all local revenue in the postreform period. Moreover, these reported revenues do not include indirect donations, consisting largely of revenue raised by local education foundations. The extent of such foundation activity and associated revenue levels are examined in the next section.

Local Education Foundations in Michigan. Generally, a foundation is a nonprofit, tax-exempt entity with a board of trustees engaged in raising, managing, and disseminating resources for one or more designated purposes, such as charitable, religious, literary, scientific, or educational. Foundation trustees are generally selected from the local community and focus on raising resources, whereas directors implement policies and programs.

Creating a local education foundation in Michigan is relatively simple. Organizers file a four-page Articles of Incorporation form, along with a $20 fee, with the Corporation Division, Corporation and Securities Bureau, Michigan Department of Commerce, as required by Michigan's Nonprofit Corporation Act (P.A. 162 of 1982). Foundations generally begin operations within 4 to 6 months of filing articles, and often exist alongside booster and parent groups that also raise funds for the local public schools. Although their fund-raising activities may overlap (e.g., raffles, sales, etc.), foundations often focus on developing partnerships with corporations, individual major donors and other foundations, and seek planned gifts through wills and memorials. Grants are often made to teachers for innovative instructional practices, visual arts, and technology, areas seldom supported by booster groups. Furthermore, education foundations usually limit grants to items not normally part of the local school district budget.

Surveying Foundations in Michigan.Local educational foundations in Michigan were identified through a key word search of files of both the Corporation Division, Corporation and Securities Bureau, Michigan Department of Commerce, and the Charitable Trust Division of the Michigan attorney general's office. A total of 153 local education foundations was identified. A questionnaire was then mailed to each foundation and, as a follow-up, to each associated local school district superintendent. A profile of the foundations and the respondents is presented in figure 2 and table 3.

As figure 2 indicates, the formation of local education foundations accelerated during the period of 1984 through 1993, a period marked by variable growth in real per pupil revenue from traditional sources. The greatest annual increases in the number of local foundations occurred in 1988 and 1990, when real state and local per pupil revenue rose 1.1 percent and 4.4 percent, respectively. Formation of new foundations slowed in 1994, when Michigan reformed its school funding system, and accelerated again in 1997 as the constraints on traditional source revenue imposed by the reforms became binding on more local districts. The survey data summarized in table 3 reveal that local education foundations are generally found in suburban and rural school districts. Annual foundation revenues, however, have been quite modest, averaging a mere $17,024 in 1994-95 among responding districts. Revenue levels have varied considerably across these districts, as indicated by the relatively large coefficients of variation for the district groups.

Comparison of Foundation and Non-Foundation Districts. Although total foundation revenues to date have been modest, the presence of a local education foundation provides a potential source of supplemental revenue for and suggests a heightened community interest in local public schools. To begin testing for educationally relevant differences between foundation and nonfoundation districts, one-way analysis of variance (ANOVA) was used to compare the mean values of selected district revenue measures, household economic characteristics, district size, and measures of student achievement of each district group. The foundation districts consist of all 153 districts identified through the state databases described above, not merely the survey respondents. These mean values and associated significance levels are presented in table 4.

As table 4 indicates, districts with educational foundations, on average, enjoy higher unrestricted public revenue per pupil, greater enrollments, higher household income, and higher student achievement than their nonfoundation counterparts. Foundation districts also allocate a lower proportion of their expenditures on general administration, whereas spending shares for instruction and school administration are roughly equal across the two district groups. The differences in the group means are statistically significant for all remaining variables except tax base per pupil and 10th grade reading achievement. Some differences are striking. For example, household incomes are more than 20 percent higher, on the average, in foundation districts as compared with their nonfoundation counterparts. Foundation districts also have a lower percentage of children eligible for free and reduced-price lunch under the National School Lunch Act and lower Federal Chapter 1 (now renamed Title I) expenditures than their nonfoundation counterparts. Furthermore, the average percent of students earning satisfactory scores on the Michigan Education Assessment Program (MEAP) are significantly higher among foundation districts on five of the six measures.

These results, although not unexpected, raise concerns regarding the equity in the distribution of educational resources across local school districts in Michigan. Michigan, along with virtually every other state, has adopted state school aid formulas designed to distribute more state aid to local districts with relatively low fiscal capacity, generally measured in terms of taxable property wealth per pupil. Furthermore, state categorical grant programs such as special education, compensatory education, and bilingual education are designed to target additional resources to local districts with relatively large concentrations of low-income children and other children who are educationally at risk. The rise of local educational foundations in relatively high-expenditure and high-income districts may offset, to some degree, the equity effects of the state's school aid system. Furthermore, students enrolled in foundation districts were overwhelmingly white, with an unweighted average of 91 percent among these districts, thus raising additional equity concerns. These concerns are mitigated, however, by the relatively small financial contributions of the local educational foundations, averaging $17,024 in 1994-95 among responding school districts. These effects may be further mitigated by the relatively large foundation contributions made to urban districts, which are generally property poor.

back to top

A Model of Local School District Spending

The demand for education spending is assumed to be derived from a median-voter, majority-rule model where it can be shown that, under certain conditions, a community's effective demand for education will be that of its median income voter (see Bergstrom and Goodman 1973).16

If the price of private goods x is denoted by p, the individual's budget constraint with private income Y is:

Y = px+ T(1-F)   (1)
where T = local property taxes
F = the proportion of local property taxes offset by the deductibility of property taxes from state and federal income taxes
Property taxes are supplemented by lump-sum and matching aid to cover the total cost of local public education. Furthermore, the median voter pays only a fraction of the total local cost, based on the voter's share of total taxable property in the school district. Thus, the tax obligation of the median voter is given by:
T = [c(1-s)-k](Vm/Vt)   (2)
where c = total cost of public education in district
k = lump-sum aid paid to district
s = state share of additional dollar of educational expenditures
Vm = median household property valuation
Vt = total property valuation of district
Substituting (2) into (1) and rearranging, the median voter's budget constraint becomes:
Y + k(Vm/Vt)(1-s) = Px+ [c(1-s)(Vm/Vt)](1-F)   (3)
Thus, the total income of the median voter consists of private income and the voter's share of lump-sum aid received by the district, while the voter's price of education is the marginal cost of increasing education expenditures per pupil by one dollar.

The median voter is assumed to maximize a utility function U = U (x, c) subject to the budget constraint given by (3). A demand function for local public education can then be derived in terms of price and income. A simple model of education demand is:

E = b0+ b1PRICE + b2INCOME + b3FREE + b4ENROLL + b5%INSTR   (4)
where E = educational expenditures per pupil as determined by local voters
PRICE = marginal tax price faced by the district's median voter
INCOME = median family income in the district
FREE = percent of children in district eligible for free or reduced price lunch under the national school lunch act (a proxy for educational need)
ENROLL = total district membership (to test for economies of scale in the supply of education)
%INSTR = percent of operating expenditures allocated to instruction
Marginal Tax Price. A district's marginal tax price of school spending is the cost to the district's median voter of increasing per pupil spending by one dollar. In a guaranteed tax base (GTB) aid system, used in Michigan in 1993-94 to establish foundation spending levels for 1994-95 and subsequent years, the matching rate (m)for a local district is the state share of an additional dollar in locally financed educational expenditures. This matching rate, in combination with district enrollment and the median voter's share of the local district property tax base, determines the marginal tax price:
PRICE = n(Vm/Vt)(1/(1+m))   (5)
where n = number of students in the district
Vm = average residential state equalized valuation (SEV) in the district (proxy for median household SEV)
Vt = total SEV of the district
m = {(V*—Vi)/Vi} if the district receives GTB formula aid, 0 otherwise
V* = nominal GTB formula SEV per pupil guarantee
Vi = district's SEV per pupil
Data. The data on local school district enrollments, expenditures, SEV, and free and reduced-price lunch eligibles were obtained from the Michigan Department of Education. The data on district average household income were obtained from the Michigan Department of Treasury.

back to top

Empirical Results

The model of school expenditures (equation 4) is estimated with tax price term PRICE calculated according to equation 5. Descriptive statistics for each variable are presented in table 5. To test for behavioral differences between residents of in-formula and out-of-formula districts, dummy variables are used for the intercept and for each independent variable. The equations are estimated by weighted least squares, where the weighting factor is the square root of the number of families in the school district.17

As shown in table 5, residents of foundation districts spend more per pupil from public (tax) sources than nonfoundation district residents, an expected finding in light of their lower tax price for school spending. Average household income is fully 19.5 percent higher in foundation districts, whereas the percent of pupils eligible for free and reduced-price lunch is 30.8 percent higher in nonfoundation districts. Mean enrollment is higher among foundation districts, whereas enrollments vary more among nonfoundation districts.

The regression results reveal structural differences in the demand for public school spending across the two voter groups. As noted in table 6, the coefficient on DUMMY has the expected positive sign and is statistically significant, indicating a preference for higher public school spending on the part of foundation district residents that is not explained by price, income, enrollment, or high educational need (i.e., FREE). The coefficient on PRICE has the expected negative sign but is statistically insignificant. The coefficient on D*PRICE, however, is negative and significant, indicating more price-elastic demand for school spending on the part of foundation district residents. Estimated point price elasticities of demand, calculated at mean per pupil expenditure levels and marginal tax prices, are -0.3097 for foundation district voters and -0.0049 for voters in nonfoundation districts.18

The coefficient on income has the expected positive sign and is significant at the 0.01 level. The coefficient on D*INCOME is insignificant, however, indicating that the relationship between income and desired school spending does not vary across district groups. The positive and significant coefficient on ENROLL (P-value of 0.000) and the insignificant coefficient on D*ENROLL indicate the absence of scale effects among both district groups. The coefficient on FREE is negative but insignificant (P = 0.607). In contrast, the negative and significant sign on D*FREE indicates a negative relationship between school spending and concentrations of low-income children among foundation districts. Within this district group, higher spending among high-income and high tax base (i.e., low PRICE) districts may swamp the effects of compensatory spending in less affluent foundation districts. Finally, the coefficient on %INSTR is negative and significant, whereas the coefficient on the associated DUMMY variable interaction term is insignificant, suggesting that high-spending districts in both groups allocate more resources to noninstruc-tional purposes at the margin.

back to top

Summary and Conclusions

Since the beginning of this decade, public schools in the United States have been faced with a dramatic slowing of per pupil revenue growth, although school enrollments and expectations for academic achievement continue to rise. To meet community expectations, school districts in recent years have turned increasingly to nontraditional sources to supplement revenues from broad-based taxes. Such revenues are raised through donor activities, enterprise activities, and cooperative activities. Indirect donor activities are undertaken by school booster clubs and, increasingly, by means of a new form of nonprofit organization, the educational foundation. In Michigan, 153 such nonprofit organizations have been established by local districts to raise revenue for curriculum improvements, capital projects, instructional materials, and enrichment activities, and to strengthen links between schools and communities. This activity in Michigan is representative of activity nationwide.

Although the rise of these organizations is not unexpected in light of the slowing of revenue growth and rising expectations for public schools, this development has not been viewed with universal approval. The equalization of educational opportunities for all children, regardless of the wealth of their respective local communities, has long been an important goal of education policymakers. Virtually every state allocates school aid to local districts by means of equalizing formulas designed to offset disparities in local fiscal resources. Local education foundations have aroused concern that they may exacerbate the very fiscal disparities public policy seeks to reduce. Moreover, state authorities are generally unaware of the scope of revenue-raising activities of foundations and booster clubs, because such revenues are rarely included in standard school district financial reports.

The Michigan research revealed that total foundation revenues to date have been modest, averaging a mere $17,024 per participating district in 1994-95. However, striking differences were found between foundation and nonfoundation districts, with average household income among the former group exceeding the latter by more than 20 percent. The foundation districts, as a group, also have a lower percentage of children eligible for free and reduced-price lunch under the National School Lunch Act, greater per pupil revenues from traditional tax sources, and higher measures of student achievement in reading and mathematics, as measured by the MEAP. Furthermore, students enrolled in foundation districts were overwhelmingly white, with an unweighted average of 91 percent across these districts. Again, however, these equity concerns are mitigated somewhat by the relatively small financial contributions of the local educational foundations.

In general, the demand for goods and services, including education, depends on price, income and tastes. An ANOVA found price and income to differ significantly between the foundation and nonfoundation district groups. Furthermore, the estimated school expenditure model revealed some difference in taste preferences for school spending between residents of the two district groups. The substantial per pupil spending differences across the groups were partially explained by differences in price, income, enrollment levels, and concentrations of low-income children.

In light of these findings, it appears that the rise of local education foundations in Michigan has not measurably negated that state's efforts to reduce interdistrict disparities through the reform of public funding mechanisms. This result could change, however, as state funding reform continues to constrain per pupil revenue growth in historically high-spending and high-income districts and as such districts seek additional revenue from nontraditional sources.

back to top


1 Because of the mandated status of special education, the expansion of special education in either scope or intensity would take a larger share of any new revenue in times of slow budget growth.

2 Mullins and Joyce (1996) note the difficulty in assessing the degree of constraint imposed by these limitations. Mechanisms such as local popular or legislative votes, authorization by state tax commissions and state legislatures, and charter and constitutional revisions are provided to suspend the provisions of these constraints and, depending on their comprehensiveness, circumvention is more difficult in some cases than in others. Comparisons across and within states are further complicated by variations in the definition of the property tax base, in assessment practices, and in the exclusion of various revenue and expenditure categories (e.g., long-term debt, fees, and charges) from the limitations.

3 Nationally, states provided 46 percent of K-12 public school revenues in 1993-94, with most aid distributed so as to offset differences among local districts in the ability to finance education. The sole exception is New Hampshire, where state aid comprises a mere 7 percent of K-12 public school revenue. Local property taxes, on the other hand, provide 90 percent of school revenue, while federal sources provide the remaining 3 percent (Gold et al. 1995).

4 For an analysis of the constitutional challenges to school fee policies, see Dayton and McCarthy 1992.

5 These eight states are Alaska, Illinois, Indiana, Iowa, Kansas, Kentucky, Utah, and Wisconsin.

6 Historically, courts have held that summer school fees were constitutional because summer school was not considered part of a student's entitlement to a free public education. However, as more states establish minimum competency testing programs for promotion and graduation and require summer school attendance for students who fail these exams, summer school may be increasingly viewed as part of a student's entitlement. Furthermore, although fees for elective courses have been upheld by the Supreme Courts of New Mexico and Montana, the Supreme Court of California held that all educational activities must be free (Dayton and McCarthy 1992).

7 As Dayton and McCarthy (1992) note, low-income families may choose to withdraw from user financed activities rather than face the potential embarrassment of seeking a waiver.

8 Wassmer and Fisher (1997) observe that, although U.S. school districts employ user fees only minimally, as much as $30 billion in expenditures on auxiliary services could be funded through fees.

9 Examples include a credit union in Southfield (Michigan) public schools, a dating service in the Hazelwood (Missouri) School District, and the rental of playing fields and locker facilities to professional sports teams for preseason training camp by the Phoenix (Arizona) public schools (Meno 1984).

10 Despite its broad list of subscribers, Channel One is not without its critics who cite its intrusive nature and the perceived school endorsement of advertised products. As of 1992, the highest subscription rates were found in Michigan, Ohio, Pennsylvania, and Texas. On the other hand, Channel One is banned in California, Massachusetts, New York, North Carolina, and Washington (Greenberg and Brand 1993).

11 Such arrangements, of course, may also be competitive. In Michigan, for example, high school students may enroll in courses at community colleges and public universities with tuition paid from a pro ratashare of state school aid, that is, in effect, a transfer from the local school district to the postsecondary institution.

12 This omission is explicitly noted in a subsequent GASB Proposed Statement: "Under the financial accountability criteria established in Statement 14, the inclusion of legally separate organizations in the reporting entity is based on either the appointment process or fiscal dependency. Certain entities, however, are affiliated with legally separate organizations, created for the specific purpose of providing financial assistance or other types of support to their programs without meeting the financial accountability criteria defined in Statement 14. This occurs particularly among colleges and universities; it also occurs among hospitals, museums, elementary and secondary education institutions, and other types of organizations. Because of the methods used to create and administer some of these organizations, the nature of their relationship is different from what has been set forth in the Statement 14 "financial accountability" criteria...The Board believes that, despite the absence of direct association through the appointment process or fiscal dependency, the relationships between the primary government and some of these organizations are such that either financial accountability exists through other means or exclusion would render the statements of the financial reporting entity misleading or incomplete...The Board concluded that in certain circumstances these relationships make an affiliated organization an integral part of the primary government reporting entity. The Board also concluded that financial reporting could best recognize the nature of this relationship (in the absence of direct association through the appointment process of fiscal dependency) through discrete presentation of the affiliated organization on the face of the financial reporting entity's financial statements" (GASB 1994, 7-8).

13 The foundation formula guaranteed each local district a per pupil allowance that ranged from the $4,200 minimum to a maximum of $6,660, provided the district levied a local property tax rate of 18 mills on nonhomestead property. Specifically, local districts with 1993-94 base per pupil revenue below $4,200 are raised either to $4,200 or to $250 over their 1993-94 level, whichever is greater. Districts between $4,200 and $6,500 in 1993-94 received a per pupil increase varying linearly from $250 at $4,200 to $160 at $6,500. Finally, local districts with 1993-94 base per pupil revenue in excess of $6,500 were allowed an increase of up to $160 per pupil if local voters approved hold harmless millage sufficient to raise the additional revenue. This local millage is levied against homesteads to a maximum of either 18 mills or the district's prior year millage vote, whichever is less.

14 The annual change in the basic foundation allowance is determined by a final index, which may be written as follows:

I = (Rt/Rt-1)(Mt-1/Mt)

where I = final index

Rt= total school aid fund revenue in current year

Rt-1= total school aid fund revenue in prior year

Mt-1= total pupil membership in prior year

Mt= total pupil membership in current year

The annual basic foundation allowance is determined by:

BFt= BFt-1* I

where BFt= current year basic foundation

BFt-1= prior year basic foundation

The local foundation allowance for an individual district is determined as follows:

LFt= LFt-1+ 2b—[(b-$50) * (LFt-1—$4,200) / (c—$4,200)]

where LFt= district's current year foundation allowance

LFt-1= district's prior year foundation allowance

b = BFt- BFt-1= current year increase in basic foundation allowance

c = BFt= current year basic foundation allowance

15 Popularly known as the "Headlee Amendment" after its author Richard Headlee, this constitutional amendment limited both local property taxes and total state tax collections.

16 From 1973 through 1993-94, Michigan required direct voter approval of local school taxes. Since 1994-95 district spending levels under the foundation system were linear transformations of prior year spending (see Addonizio et al. 1995) and local school districts serve a single purpose, 1994-95 district expenditures are likely to conform to the predictions of a median-voter model.

17 Because sampling theory reveals that the error term will be a function of the size of the population tested (heteroscedasticity), ordinary least squares would be an inappropriate estimation technique (see, for example, Kmenta 1971, 322-26).

18 The estimated price elasticity of demand for education spending for the combined sample obtained from a natural log form of spending model is approximately equal to the point elasticities reported above. This estimated expenditure equation is:

     ln E   =   ln 5.72  .1294 1n PRICE   +   .2699 1n INCOME
(.21) (.0130) (.0207)
Adj. R2  =  .278

The small standard errors indicate that the coefficients are statistically significant at the 0.01 level. This log form is a popular functional form for economic models because each slope coefficient may be interpreted as the (constant) elasticity of the dependent variable with respect to the independent variable (see, for example, Kelegian and Oates 1981, 102-4).

back to top


Addonizio, M.F. 1991. "Intergovernmental Grants and the Demand for Local Educational Expenditures."Public Finance Quarterly, 19(2): 209-232.

Addonizio, M.F. 1997. "Private Funding of Public Schools: Local Education Foundations in Michigan."Educational Considerations(forthcoming).

Addonizio, M.F., Kearney, C.P., and Prince, H.J. 1995. "Michigan's High Wire Act."Journal of Education Finance, 20(Winter): 235-269.

Anderson, N. 1997. "Going Beyond the Bake Sale."Los Angeles Times(December 2, 1997): A1-A3.

Bergstrom, T. and Goodman, R. 1973. "Private Demand for Public Goods."American Economic Review, 63(June): 280-296.

Brunner, E. and Sonstelie, J. 1997. "Coping with Serrano: Voluntary Contributions to California's Local Public Schools."Proceedings: National Tax Association 89th Annual Conference: 372-381.

Chaikand, S., Danielson, L.C., and Brauen, M.L. 1993. "What Do We Know About the Costs of Special Education? A Selected Review."Journal of Special Education, 26(4): 344-370.

Dayton, J. and McCarthy, M. 1992. "User Fees in Public Schools: Are They Legal?"Journal of Education Finance, 18(2): 127-141.

Gold, S., Smith, D., and Lawton, S., eds. 1995.Public School Finance Programs of the United States and Canada, 1993-94. Albany, NY: The American Education Finance Association and The Nelson A. Rockefeller Institute of Government.

Governmental Accounting Standards Board. 1994.Proposed Statement: The Financial Reporting Entity—Affiliated Organizations; Exposure Draft. Norwalk, CT: Governmental Accounting Standards Board.

Greenberg, B.S. and Brand, J.E. 1993. "Television News and Advertising in Schools: The `Channel One' Controversy"Journal of Communication, 43: 143-51.

Guthrie, J.W. 1997. "School Finance: Fifty Years of Expansion."The Future of Children, 7(3): 24-38.

Hamm, R.W. and Crosser, S. 1991. "School Fees: Whatever Happened to the Notion of a Free Public Education?"The American School Board Journal(June): 29-31.

Hanushek, E.A. and Rivkin, S.G. 1997. "Understanding the Twentieth-Century Growth in U.S. School Spending."The Journal of Human Resources, 32(1): 35-68.

Howell, P.L. and Miller, B.B. 1997. "Sources of Funding for Schools."The Future of Children, 7(3): 39-50.

Johnston, J. 1995. "Channel One: The Dilemma of Teaching and Selling."Phi Delta Kappan, 2: 437-442.

Joyce, P.G. and Mullins, D.R. 1991. "The Changing Fiscal Structure of the State and Local Public Sector: The Impact of Tax and Expenditure Limitations."Public Administration Review, 51(3): 240-253.

Kearney, C.P. 1994.A Primer on Michigan School Finance. 3rd ed. Ann Arbor, MI: Educational Studies Program, School of Education, The University of Michigan.

Kelejian, H.H. and Oates, W.E. 1981.Introduction to Econometrics. New York: Harper & Row.

Kmenta, J. 1971.Elements of Econometrics. New York: Macmillan.

Maeroff, G. 1982. "Schools Seek Private Funds."New York Times. (November 9, 1982): C1.

Meno, L.R. 1984. "Sources of Alternative Revenue." InManaging Limited Resources: New Demands on Public School Management, L.D. Webb and V.D. Mueller, editors. Cambridge, MA: Ballenger: 129-146.

Meredith, B. and Underwood, J. 1995. "Irreconcilable Differences? Defining the Rising Conflict Between Regular and Special Education."Journal of Law and Education, 24(2): 195-226.

Monk, D.H. and Brent, B.O. 1997.Raising Money for Education. Thousand Oaks, CA: Corwin Press.

Mullins, D. and Cox, K. 1995.Tax and Expenditure Limits on Local Governments. Washington, DC: Advisory Commission on Intergovernmental Relations.

Mullins, D. and Joyce, P. 1996. Tax and Expenditure Limitations and State and Local Fiscal Structure: An Empirical Assessment.Public Budgeting and Finance(Spring): 75-101.

National Association of Educational Foundations. 1994.Educational Foundations Focus, 1(Summer).

Picus, L.O., Tetreault, D.R. and Hertert, L. 1995. The Allocation and Use of Educational Resources in California(paper presented at the Annual Meeting of the American Education Finance Association, Savannah, GA, March 11, 1995).

Pijanowski, J.C. and Monk, D.H. 1996. "Alternative School Revenue Sources: There Are Many Fish in the Sea."School Business Affairs(July): 4-10.

Pollack, A. 1992. "With Budgets Cut, Public Officials Seek Private Money."New York Times(May 17, 1992): L40.

Salloum, K. 1985.Private Funding for Elementary and Secondary Public Education in British Columbia. Simon Fraser University: Unpublished master's thesis.

U.S. Department of Education, National Center for Education Statistics. 1995.Digest of Education Statistics, 1995. NCES 95-029. Washington, DC.

Wassmer, R.W. and Fisher, R.C. 1996. "An Evaluation of the Recent Move to Centralize the Finance of Public Schools in Michigan."Public Finance and Budgeting, 16(3): 90-112.

Wassmer, R.W. and Fisher, R.C. 1997.User Charges and the Financing of K-12 Public Education in the United States. Sacramento: California State University, Graduate Program in Public Policy and Administration, Working Paper 96-03.

back to top

Return to the Table of Contents