Page of

Allocation and Control of Public Funds

Allocation and Control of Public Funds

(p.89) 7 Allocation and Control of Public Funds
A Democratic Constitution for Public Education
Paul T. HillAshley E. Jochim
University of Chicago Press

Abstract and Keywords

This chapter describes a school funding system that, consistent with the principles of constitutional governance, funds schools and students equitably, gives schools a defined amount for every student that enrolls, and allow schools to set their own spending priorities. After reviewing the existing system for funding K-12 schools, the chapter describes a transformed funding system that would fund education not institutions; move money as students move from one school to another via a “student backpack” arrangement; pay for unconventional forms of instruction; pay for facilities without privileging conventional school models; withhold funding for ineffective programs without chilling innovation; and adapt funding levels and student weighting in light of experience. It explains how these principles would be put into practice, what problems might arise, and how these could be managed.

Keywords:   school funding, funding based on enrolment, student backpack, unconventional forms of instruction, facilities, innovation, student weighting

Governance and financing are not separable—a new system for one cannot operate as intended while leaving the other unchanged. There are two big problems with our current financing system. First, it tolerates a great deal of horizontal inequity, so that very different amounts can be spent on similar students in the same state and even the same district. This gives some schools big advantages over others and undermines any claim that schools can be fairly compared on performance.

Second, the system now does not provide real money to schools. It allocates money to district-wide programs, central office structures, administrative functions, a district-wide teaching force, a stock of buildings, and so forth. These are then distributed to schools via methods of the district’s choosing; schools make do with the teachers and other resources they are given. This method of resource allocation does not allow the kinds of flexibility needed for innovation and competition. Money does not move from one school to another when a student transfers, and schools cannot make internal financial trade-offs, for example, spending less on teachers and facilities and more on online resources or student internships.

(p.90) To support a diverse and innovative supply of schools, we must fund schools and students equitably, give schools a defined amount for every student who enrolls, and allow schools to set their own spending priorities. This chapter will show how these things can be done in the context of a constitutional governance system.

How We Got the Current System

The American approach to funding education, like its approach to governance, has emerged slowly and haphazardly, a product of politics and advocacy, not design. Starting in the 1950s as courts decided civil rights cases, Congress created new services and entitle ment programs for particular groups. State legislatures started small programs to solve emergent problems, school employees sought job protections and control over their work assignments, universities sought gatekeeper status for potential teachers and administrators, and families sought advantages in the competition for the best school placements. The sum of these actions was the inflexible financing system we have today.

For example, the federal Elementary and Secondary Education Act included only a few programs when first enacted in 1965, but over time new funding streams and sets of regulations (for purposes as diverse as teacher professional development and bilingual education) proliferated. These processes led to a labyrinth of rules and regulations connected to an evergrowing network of separate funding paths, each with its own “allowable uses” and reporting requirements.

Under the current system, people with ideas about how to do things differently—to introduce some actions that cost money and to substitute them for other current uses of funds—have great difficulty getting their ideas adopted or even seriously tested. This limitation applies to incumbent teachers and principals in traditional public schools. Outsiders, observing that U.S. schools have remained about the same, despite revolutions in technology and economic life, conclude that education stasis is due to (p.91) the lack of new ideas. But that is wrong. Individual teachers, principals, and technology innovators come up with new ideas about instruction all the time and often put them into small-scale practice as long as no one complains about it. But if a new practice generates imitators or publicity, it can be interpreted as something irregular and possibly threatening and be stopped.

Because funds cannot flow from established uses to new ones, good ideas cannot be fully developed or persuasively demonstrated. Imaginative people who have ideas about uses of technology will not expect that applications they invent for K–12 education can be funded, so they will likely apply themselves in other realms instead.

Today’s funding arrangements for public education were not designed with innovation or continuous improvement in mind. They assume that a student will attend a specific school, in which teacher salaries and other costs are paid by the school district. School leaders get very little money to spend on their own because money flows to schools not in the form of cash but in the form of people and resources sent by the district. A school can lose one student (or in a typical district as many as nineteen) and not lose any resources. In the rare case in which a student might take some courses at another school or outside the district (e.g., at an alternative school for gifted children or at a community college) the district, not the school, pays extra costs out of a central fund. These threshold effects governing the use of resources can also undermine the intention to target funds toward particular groups of students, as in some localities schools below an arbitrary threshold (e.g., 35 percent low-income for the federal Title I program) receive no assistance.

Our system is built to support schools that students attend all day every day and whose employees provide all the instruction a child needs to gain credit for a year’s learning. It is more difficult to get public funds to blended or hybrid schools that provide some instruction via computer online and some via in-person teaching. Even if states and school districts permit money to flow to virtual schools, the funding status of blended schools would be (p.92) precarious. States and districts could decide that blended schools were attracting too many students and drawing too much money out of their conventional schools and revoke policies that let funds flow to them.

How a Constitutional Governance System Would Fund Education

What would transparent and fair school finance system look like? It would need to

  • fund education not institutions,

  • move money as students move,

  • pay for unconventional forms of instruction,

  • pay for facilities without privileging conventional school models,

  • withhold funding for ineffective programs without chilling innovation, and

  • adapt funding levels and weights in light of experience.

Funding Education not Institutions

How can states and local CECs provide money for K–12 education in ways that allow continuous improvement?1 The answer is that states and localities should fund something that is permanent, not contingent, and tie money to the one element of the education system to which they are unconditionally committed—students.

This means designating a specific amount for every child in the state and distributing money to districts and schools solely on the basis of enrollment. Under a constitutional governance system, the state could opt to “weight” pupil-based funding by allocating more for one group of children than for another (e.g., providing more than the average amount of money to support the education of disabled or English language learners or those in high-risk communities).

States might let spending vary owing to variations in local taxable assets and tax effort. If local spending differences were correlated with local costs (e.g., low spending only in low-wage and (p.93) low property value areas), this would not undermine the “level playing field” principle. However, if some localities (e.g., large urban districts) ended up spending less despite high local costs, children in those schools would get less. Moreover, families’ abilities to enroll children in schools run by better funded districts would be limited by the fact that those children would bring less money than other children those districts might enroll. Thus, children in poorly funded districts would have fewer opportunities, and the schools in those localities would, ironically, experience less competition than schools in better funded localities.

The principles of a level playing field and funding students not institutions lead to the conclusion that the state should fund education uniformly, meaning that a child with certain characteristics would get the same amount of money no matter where in the state she lives.

The CEC, state, and federal governments would not mandate particular uses of funds or support particular programs, administrative structures, or salary scales. Of course, there would be some constraints on the use of funds—to support only schools that subjected themselves to CEC oversight and agreed to participate in the local enrollment system that ensures fair and equitable access to schools.

Districts could skim off very little money, and that would be used for central functions like data collection and analysis, administration of enrollment lottery systems, and decisions about what schools to open or close. The CEC would not be able to tax schools for services that other entities were also capable of providing (see chapter 6). The CEC, central office employees, and school leaders would all know exactly how much the central office spends and for what and understand that every dollar spent at the district level cannot be used by schools to educate students. Thus, all parties would have incentives to minimize and justify any district-level spending.

If states and localities would combine all the money they now spend on K–12 education and divide it up by enrollment, with the same or weighted fraction of the total assigned to each child, and (p.94) distribute dollars to schools in the same way, they could simultaneously eliminate the barriers to innovation and to improvement that are inherent in current funding systems. In a constitutional governance system where central administrative costs are very low, most schools would also get dramatically more money than they receive now. This would allow new uses of funds, an essential precondition to innovation.

Consistent with the discussion of the federal role in chapter 4, the federal government could reinforce the movement toward pupil-based funding by making its major grant programs pupil specific. It could do that by sharply defining student eligibility criteria for such programs as Title I and IDEA, then requiring that states divide the money received from any such program equitably among all eligible recipients in the state, and allocating it as extra money to the schools those children attend. Thus, federal programs could still increase spending on designated beneficiaries without privileging particular uses. Although there would still be some costs associated with counting eligible beneficiaries and distributing dollars to districts, this arrangement would greatly cut the share of federal funds used to ensure that funds are used only in specified ways, and in ways that cannot inadvertently benefit noneligible students.2

Move Money as Students Move

Funding students (and not programs) is a step in the right direction, but it is not enough. States must also make sure that all funds move from one district, school, or instruction provider to another as students transfer. Schools that lose students should lose the income associated with them in some defined period, say, by the end of the current semester, and schools gaining students should get the extra money as soon as it is taken from the school of origin. This makes family choice significant for everyone: students could move whenever their parents identified a more suitable school or set of instructional programs or as ineffective schools were closed and students shifted to more effective ones.

Charter school funding establishes a rudimentary version (p.95) of such a system. Funds follow students directly to the charter schools they attend and are taken away from schools that pupils transfer away from.

State charter laws are imperfect in many ways, including the fact that they only apply to students who transfer to such schools; all other students are still in district-provided schools where funds are tied up, hidden, and inflexible. Many states, moreover, leave some funding behind in district-run public schools even when pupils transfer to charter schools. Depending on state and charter laws and local policies, charter students either do or do not benefit from public funds allocated for federal programs, transportation, and facilities. Charter students often do not benefit from locally raised funds (e.g., from property tax levies).

However, the basic charter paradigm is a good frame for the funding mechanism to be attached to a constitutional governance scheme. Chartering ensures funding based on enrollment and school freedom over spending, staffing, and use of time. A funding system would have to apply to all students no matter what publicly funded school they attended. As with chartering, the school a student attended—whether it provided all the students’ courses or purchased some from other sources—would be accountable for results: test scores, credits, ability to take more advanced courses, and ultimately graduation and college eligibility.

Funds available for a child’s education must include all the taxpayer funds available to support a student’s education. To make this happen, some government entity, likely the local district central office, would need to assemble all of the funds available from all sources for K–12 education in a locality, keep an account for every student, and faithfully allocate its contents to whatever school or education program a student attends.

A new kind of public entity, possibly a county or regional finance office of the state government, could be an alternative. It could assemble and disburse all funds. It would also account for funds on a pupil basis. Every student would have an account that showed what funding from all sources was available for his education and to what schools and vendors it had been disbursed.

(p.96) Each student’s account would, in a sense, constitute a backpack of funding that the student would carry with her to any eligible school in which she enrolls. The contents of the backpack would be flexible dollars, not coupons whose use is restricted to a particular course or service.

Under a constitutional governance scheme, no school enrolling backpack-wearing students could charge tuition in excess of the full amount in a student’s backpack. Schools and other providers could also offer partial instructional programs, and students and their families could mix and match to the limit of the funds in the backpack. (See below for possible safeguards against misuse of backpack funds.)

Backpack-based movement of funds with students would impact existing schools’ budgets promptly, creating incentives for schools to avoid losing students to other schools. Innovators (educators and social service professionals with new ideas) would also be encouraged by the certainty that they could get full funding for every student enrolled in their school or program.

Allow Schools to Pay for Unconventional forms of Instruction

For free movement of funds to promote experimentation and innovation, it must be possible for students to enroll in schools that are configured in novel ways. Technology opens up the possibility of students learning in a one-to-one relationship with a computer-based system, or linking to a set of lectures and other presentation materials along with literally thousands of other students, or receiving their instruction through a mix of technology-and teacher-delivered approaches. These approaches are not certain to work in every case, but the system must be designed to allow such differentiation and track results closely.

Schools could pursue very different strategies—some spending all their money on teachers and a conventional building, others purchasing some instruction from other schools or online, and others serving only as brokers—assessing the effectiveness of various instructional providers and assigning individual students (p.97) to the ones most likely to meet their needs. Students would be able to join online courses provided by schools other than the one they normally attend and could take advantage of courses that combine experiential learning (e.g., participation in workplaces, arts events, or social services) with online materials that prepare students for those experiences and assess learning.

Such “broker” schools would buy most of the instruction their students receive from others: their contribution, and the justification for their keeping some of the funds that come in the backpack, would be identification of the best matches for an individual student. Schools that did this well would prosper, but those that did it poorly would fail quickly and disappear.

This would create strong incentives for local CECs to search for the best way to meet any student’s need and to be indifferent about whether a student takes a course delivered by his neighborhood school or some other source. Most schools will continue to offer adult supervision, counseling, and tutoring, and some will develop instructional specialties that both keep their students at home and draw students enrolled elsewhere. But to be competitive, especially under an accountability system that takes account of productivity, schools will need to organize themselves for nimbleness, directly delivering only those courses and other aspects of the student experience at which they are good and “buying” others from other schools or vendors. Thus, a school might employ teachers in English and the arts and buy physics and math instruction and extracurricular experiences like outdoor leadership and advanced sports instruction from others. In effect, schools will constitute a marketplace for instructional programs and other services, each trying to be an excellent provider (and thus seller) of some things and a buyer of others.

Bundle Facilities Funding into the Student Backpack

Schools and instruction providers should not be punished for being efficient. Blended schools might save money on teachers and/or facilities. Virtual and broker schools could, if they attract (p.98) enough students, pay far less to educate each additional student than they receive from student backpacks. These results are desirable because they can support investment. Even if providers are nonprofits, the low marginal costs of online instruction can enable continual investment in new and improved resources.

To create a fully level playing field for competition among all schools, states would need to add the funds now set aside for facilities construction and maintenance to the student backpack. States now fund separate line items for construction and building maintenance, and these are not counted in calculations of school operating costs. Thus, even if online schools were equitably funded on school operating costs, traditional schools would get a significant extra subsidy.

In a time when innovation is necessary, and the cost structures of different forms of schooling will differ dramatically, a special subsidy for bricks and mortar is counterproductive. It defrays cost borne by one kind of school, while ignoring the investment costs of online schools, which must pay for computers, methods development, and back-office administration out of their operating costs.

Under this arrangement, a school—whether traditional, hybrid, or online—would pay whatever facilities costs it used. Schools and instructional program providers with high equipment, research and development, and oversight costs could pay these from funds recovered from student backpacks, just as schools paying high rents could meet those costs. In the long run, this arrangement would probably lead conventional schools to reduce their costs by renting less space and refusing to pay for expensive amenities like theaters and swimming pools.

This neat solution might not work perfectly in cities where facilities costs vary sharply from one neighborhood to another (e.g., from a dense downtown area to a changing area with high vacancy rates). In such cases local CECs might need to add a weight related to a student’s neighborhood of residence to the backpack. These students would then become even more attractive (p.99) to schools that had figured out how to minimize their facilities costs.

Withhold Funding for Ineffective Schools

A funding system that is open to innovation must also have a mechanism for deciding which schools and instructional programs should be considered eligible to enroll students. A financing system must include arrangements to withhold or withdraw funds from ineffective providers.

For whole schools, including hybrid and virtual schools, chartering provides a useable framework for performance management. The CEC as a charter authorizer can close low performing charter schools or refuse to renew their authorization. Hybrid and virtual schools can be evaluated on the same standards, tests, and other measures of student progress as those used for conventional schools, and charters for low performers can be cancelled or not renewed.

Oversight of hybrid and virtual schools is a new demand that traditional local school boards are not organized to meet. CECs would be better designed for this function because they would not operate any schools or own any facilities and therefore not care what a school’s cost structure was, as long as it was effective.

Maximize Parents’ Options without Losing School Accountability

Even if parents chose the schools their children attended, the options available to students, and the opportunities for entrepreneurs, are still limited by the imaginations and tastes of current school operators. Of course, parents could be given full access to their children’s backpacks and allowed to purchase any combination of instructional experiences they wanted from any source. What children learn would then depend on the quality of their parents’ choices.

In the long run parents might learn to do this well, and the supply of good options could rise to meet the demand. However, (p.100) students or parents could make bad purchases, so that students did not learn what they needed to graduate or succeed in higher education and work. In those situations, would the children simply be out of luck? Could the parents be punished in some way that was not counterproductive for all involved? Or would the public be forced to pay again for instruction the student should have gotten the first time?

As the senior author has argued elsewhere, giving parents unlimited control of the student’s backpack could fatally compromise performance accountability.3 However, there are two ways that parents could have important choices, while maintaining the principle that one entity, the school, is responsible for a student’s overall learning:

  1. 1. limiting the amount of money parents can dispose of; and

  2. 2. limiting parents’ choices to supplementary or enrichment programs.

Parents could be required to choose a whole school provider, but each provider could set aside a limited amount from each student’s backpack that could be used to pay for tutoring and enrichment programs. Schools could offer cafeteria plans4 for extracurricular activities and supplementary learning (either online or in person). Students and families would then be free to shop for the best combination of courses and experiences their set-aside funds could cover.

Parents could control this limited amount, choosing the enrichment but not the core instructional programs. This would both increase families’ control over children’s educational experience and allow some public funds to flow to new and innovative programs. Providers of enrichment and support programs could receive some public funds, yet families could not be led into making choices that compromised their children’s core instruction. Providers would face competition, both on the quality and effectiveness of their services and on cost.

Schools could even compete on the quality of their cafeteria (p.101) plans and on the amounts of money available for families to control. Although schools that put too much money under families’ control might risk failure if their students then could not meet key performance standards, schools would have a strong incentive to let families control something important and help families choose effective programs.

This arrangement would eliminate the need for the CEC to vet every online provider or to negotiate with vendors about costs. Costs of supplementary services would be regulated by the amounts schools decide to make available, and parents would have incentives to avoid vendors that required all of their available funds for one service.5

Adapt Funding in Light of Experience

Understanding that what works best for one group of children might not be so for other groups (e.g., well-supported native-born middle-class children versus immigrants from war-torn countries who missed some years of school), the constitutional governance system would try to link spending levels to student characteristics. It would then employ a weighted student funding system whose weights would be determined by the estimated cost of delivering services to a given group of children.

Any resemblance between this approach to determining spending levels and the “adequacy” arguments made by some to escalate spending on U.S. schools is illusory.6 Spending levels would be based on the most efficient, not the average or the most expensive, approach to meeting the needs of a given group of children.

Weights for particular groups would be set according to demonstrated performance, not (as in the case of adequacy arguments) according to simulations of what programs of unknown character might achieve. Thus, weighting would be conservative. Today, weights for disadvantaged children could be based on the full cost of delivering simple but highly focused and disciplined instructional programs, resembling traditional Catholic parochial (p.102) schools or newer “no excuses” models like KIPP, the Knowledge is Power Program, a network of middle and high schools designed to meet the needs of disadvantaged minority youth.

The constitutional governance system would encourage trials of new approaches, especially for the education of children whom no existing set of schools serves well. If and when new approaches, for example, combinations of in-person and computer-based instruction with specific ancillary supports, prove uniquely productive for a particular group of students, the governance and financing system needs to adapt to the evidence. That can mean increasing spending for particular groups of students, at least until something equally effective and less expensive comes along.


A funding system cannot cause innovation or equity: it can only encourage it, or interfere. Whether innovation occurs, at what pace, and to what ultimate benefit depend on factors other than public funding. But a system like the one described here would make promising breakthroughs much more likely to scale rapidly. Likewise, no funding system can guarantee equitable opportunities or outcomes for children coming from different backgrounds, but it can certainly make them more likely. The next chapter describes the path toward enacting the system we have described into law and the likely challenges along the way.


(1) . The ideas in this section are developed at greater length in two recent papers: Paul T. Hill, “School Finance in the Digital Learning Era,” in Education Reform for the Digital Era, ed. Chester E. Finn and Daniela R. Fairchild (Washington, DC: Thomas B. Fordham Institute 2012); and Paul T. Hill, “Governing Schools for Productivity,” The Productivity for Results Series (Dallas, TX: George W. Bush Institute’s Education Reform Initiative).

(2) . The Hoover Institution’s Koret Task Force has recommended a transformation of the federal role in education that is consistent with this proposal. See Grover Whitehurst, Choice and Federalism: Defining the Federal Role in Education (Palo Alto, CA: Hoover Press, 2012).

(4) . For an explanation of the cafeteria plan idea, in this case applied to teacher benefits, see Noah Wepman, Marguerite Roza, and Christina Sepe, The Promise of Cafeteria-Style Benefits for Districts and Teachers (Seattle: Center on Reinventing Public Education, 2010).

(5) . Should parents be allowed to supplement the amounts allowed by the school with their own funds? Supplementation would allow experimentation with a wider range of services, but it would also allow some vendors to serve only the more affluent parents. It is, however, difficult to see how parents could be prevented from buying supplements with their own money.

(6) . For an explanation of the “adequacy” rationale now in use in school finance litigation, see Eric Hanushek and Alfred A. Lindseth, Schoolhouses, Courthouses, and Statehouses: Solving the Funding-Achievement Puzzle in America’s Public Schools (Princeton, NJ: Princeton University Press, 2009).

Copyright © 2021. All rights reserved.
Privacy Policy and Legal Notice