Charter Schools Have Rainy Days Too! But Are They Ready for When It Pours?

Abstract

Nearly 30 years since their inception in the United States, charter schools are now a well-established educational option for parents and students. Although they are an important education provider schooling more than 3.1 million students nationwide, we know little about their ability to accumulate fiscal savings for weathering rainy days and sustaining smooth service. Unlike most other fiscal savings studies focusing on the unrestricted fund balance, we examine both restricted and unrestricted fund balances across Pennsylvania charter schools, this study’s unit of analysis. Using a Newey-West regression and data spanning the years 2011–2019, we show that charter schools consider all fund balance classifications when making savings decisions; albeit the unrestricted was their primary savings vehicle. Given their limited revenue portfolio, they are left with only a few options for accumulating fiscal savings. Surplus from tuition payments and additional revenues from private funding sources appear as main fund balance boosters. Surprisingly, special education enrollment significantly increases the unrestricted fund balance, a finding that requires further attention from legislators and policy makers. Concerns are also raised about participation in the state pension system as it absorbs a significant amount of slack that otherwise could be used for other purposes. Overall, most charter schools retain inadequate fiscal savings not capable of insulating their operation from revenue volatility and other contingencies. Statutory fund balance minimums and the adoption of formal fund balance policies articulating how savings are accumulated, used, and replenished should, therefore, be considered.

Keywords

fiscal savings fund balance rainy day fund charter school

Introduction

Nearly 30 years since their inception in the United States, charter schools have become an entrenched educational option for parents and students. While their definition varies somewhat by state, charter schools are considered “public schools of choice, freed from rules but accountable for results” (Finn et al., 2000, p. 14). They borrow elements from both public and private schools: Like public schools, they are open to all who wish to attend and are paid for in tax dollars, but like private schools, they are more autonomous than their public-school counterparts. Still, they are accountable to an authoritative body, such as local, state, or university boards, even cities (Vergari, 2007).

In the U.S. education system, the charter school movement has been the fastest growing reform (Ericson, 2001). Charter schools now operate in 45 states and the District of Columbia (National Center for Education Statistics [NCES], n.d.), all promising to remedy education inequity between districts (Lay & Bauman, 2019), government waste, and red tape (Ravitch & Viteritti, 1997). As such, they have attracted the attention of scholars, who primarily study their governance and effects on student achievement. Governing boards,¹ commonly known as charter boards, could influence charter schools greatly by hiring (or firing) staff, even outsourcing their management to for-profit companies. Nonetheless, charter boards aim for efficient management and better student achievement (Lay & Bauman, 2019). But research on student achievement shows mixed findings, with charter schools underperforming traditional public schools in some learning areas and locations while overperforming in others (Betts & Tang, 2019).

About their financial management, specifically how charter schools accumulate and use fiscal savings, we know less. Since the 1990s when scholars first began to study fiscal savings, the focus has been primarily on states and municipalities (Gorina et al., 2019). School districts have also attracted some attention, but only recently (see Arapis et al., 2017; Barrett et al., 2019; Duncombe & Hou, 2014). There exists a consensus in literature that fiscal savings are an integral part of the budget process across states, municipalities, and school districts. States use formal budget stabilization funds (rainy day funds) for countercyclical stabilization; setting money aside in good times to manage fiscal stress and sustain service provision during economic downturns (Hou, 2003; Sobel & Holcombe, 1996). Municipalities and school districts build “informal” savings as part of their fund balance for purposes beyond countercyclical stabilization, from maintaining tax rates and supporting a steady cash flow, to paying for equipment or unexpected repairs, even funding capital projects and other public demands (Davare, 2018; Hendrick, 2006; Marlowe, 2005). Strategies states, municipalities, and school districts use to accumulate fiscal savings may differ as well given their unique operations, governance structure, and fiscal environment to name a few (Arapis et al., 2017; Barrett et al., 2019; Gorina et al., 2019).

Fiscal savings are necessary for charter schools too, which like school districts also have “rainy days,” unplanned expenses, and other contingencies that could add pressure to their operation and service provision. Charter schools, however, differ in many important ways from school districts. Charter schools, for instance, lack taxing authority. As a result, they maintain a severely limited revenue portfolio which is reliant on their state or sponsoring school district for tuition payments (Education Commission of the States [ECS], 2020). Charter schools and school districts differ in their costs as well, as in some states charter schools are free from collective bargaining (Miron, 2017), teacher certification requirements, and student transportation services (ECS, 2020). Depending on the state, they could also select whether to participate in the state retirement system or offer their own retirement plan (National Association of State Retirement Administrators [NASRA], 2017). Evidently, charter schools operate within a unique fiscal environment. As such, strategies school districts use to accumulate fiscal savings may not be relevant for charter schools.

Given this lack of attention, we offer first an exploratory analysis of fiscal savings trends across charter schools; then our attention shifts to identifying strategies of fiscal savings accumulation. In our study, we examine both restricted and unrestricted fund balance classifications reported by Pennsylvania charter schools, this study’s unit of analysis. Using a Newey-West regression and data spanning the years from 2011 to 2019, we show that charter schools consider all fund balance classifications when making savings decisions; albeit the unrestricted was their primary savings vehicle. Following our findings, surplus from tuition payments and additional revenues from private funding sources appear as main fiscal savings boosters. Surprisingly, special education enrollment significantly increases the unrestricted fund balance as well, a finding that requires further attention from legislators and policy makers. Concerns are also raised about participation in the state pension system as it absorbs a significant amount of slack that otherwise could be used for other purposes. Overall, most charter schools retain inadequate fiscal savings; statutory fund balance minimums and the adoption of formal fund balance policies articulating how savings are accumulated, used, and replenished should, therefore, be considered.

Our study advances the literature on three main fronts. First, we build upon the extant fiscal savings literature across general purpose governments and school districts and add new knowledge from charter schools. Second, our findings show that fiscal savings decisions are made not only for the unrestricted fund balance which is typically the focus of the literature thus far, but for the restricted as well. Such systematic investigations are necessary for broadening our understanding of fiscal savings behavior and advancing the literature further. Third, in addition to the aforementioned theoretical implications, our study also makes practical suggestions for prudent management of fiscal savings that charter school boards and legislators should consider.

Following this introduction is a discussion of the U.S. charter school movement and a review of the fiscal savings literature. The research methods section presents more details regarding the data, model, and estimations strategy. Next, the findings of the study are analyzed, followed by conclusions and policy recommendations.

Theoretical Framework

Charter School Basics: From Their Inception to Their Funding Challenges

As an idea, charter schools were first introduced in the 1970s when Ray Budde, an education professor at the University of Massachusetts, Amherst, described a new type of school that would provide teachers “increased responsibility over curriculum and instruction in exchange for a greater degree of accountability for student achievement” (Saulny, 2005). Others during the 1980s, including the American Federation of Teachers (AFT) union president Albert Shanker, saw charter schools as laboratories for developing ideas that could reinvigorate the entire public school system. Inspired by a school in Cologne, Germany, he advocated for more innovation and autonomy by empowering teachers in the school’s decision-making (Pendergrass & Kern, 2017).

For schools to be effective, economists John Chubb and Terry Moe (1990) argued, they need autonomy. Only then will they be able to develop clear objectives, ambitious programs, strong leadership, and high levels of teacher professionalism. A new institutional structure was then proposed in their book Politics, Markets, and America’s Schools (1990), where control would be transferred from the institutions to schools, parents, and students. Still, the state would determine the criteria for such schools and only those applicants that satisfied the requirements would be granted a charter to open a school, enroll students, and receive public money. School districts would have no authority over charter schools and would still be allowed to operate their own public schools. Students would then have a choice between schools, eventually driving school districts away from isomorphism (Arsen & Ni, 2012).

With support for charter schools rising across the country, Minnesota in 1991 became the first state to pass charter school legislation. Since then, another 7,200 charter schools have opened in 45 states and the District of Columbia educating more than 3.1 million students (NCES, n.d.). Their operation is established through a charter agreement; “a performance contract that details, among other things, the school’s mission, program, goals, and means of measuring success” (Miron, 2017, p. 224). Governing bodies (e.g., local or state school boards), universities, nonprofit groups, or other public entities are responsible to authorize charter agreements (ECS, 2020). Those failing to meet their goals could have their charter agreement, typically lasting between 3 and 5 years, revoked or not renewed (Miron, 2017).

Compared with school districts, charter schools have a limited revenue raising ability. Without taxing authority, they remain reliant on their state or sponsoring school district for tuition payments. Indeed, among the 45 states and the District of Columbia with charter laws, charter schools in nine states are strictly state funded while in the remaining 36 states they rely heavily on local funds passed through their sponsoring school district and considerably less on state and federal funds (ECS, 2020). Typically, tuition payments are calculated off the school district expenses, not the actual costs of educating a charter school student. Only in three states (Hawaii, Kansas, Wyoming) charter schools form their own budget requests (ECS, 2020).

Whether funding for charter schools is too little or too much has been one of the liveliest debates (Miron & Urschel, 2010). Critics, on one hand, argue that tuition rates lead to overpayments because they are calculated off school district expenses rather than charter school ones (Pennsylvania School Board Association [PSBA], n.d.). On the other hand, advocates such as the National Association of Charter School Authorizers consider charter school funding neither fair nor adequate (Zwara, 2020). Evidence suggests that charter schools receive less funding per pupil compared with traditional public schools. In 2018, charter schools in 18 cities from 16 states received on average US$7,800 less per pupil than traditional public schools, according to Charter School Funding: Inequity Surges in the Cities, a study conducted by the University of Arkansas. More worrisome, the funding gap across the eight cities which have been part of this study since its inception in 2003 has more than doubled (DeAngelis et al., 2020).

In all fairness, charter schools also face different costs which to some extent explains this difference in revenues (Miron, 2017). Overwhelmingly, charter schools are exempted from participating in school districts’ collective bargaining agreements. In fact, as of 2017, of the 45 states and the District of Columbia that have charter schools, only seven subjected charter schools to their authorizing school district’s collective bargaining agreements, and five of those offered avenues for exceptions or modifications. Altogether, teachers in only 11% of charter schools (781 out of 7,200 nationally) were unionized (National Alliance for Public Charter Schools, 2017). Unlike school districts which are legally enforced to participate in the state pension system, charter schools in some states are free to offer their own retirement benefit plan. As of 2017, charter schools in 13 states have the option to either participate in the state retirement system or offer their own retirement plan instead (NASRA, 2017).

Similarly, requirements for teacher certification are also quite different across states. While 36 states require certification, 13 of them only require a percent of teachers to be certified ranging from 50% to 90%. Not in all states charter schools are required to provide transportation either. In fact, only in seven states charter schools are legally enforced to provide transportation service to their students. In the remaining, responsibility falls to school districts or other providers (ECS, 2020). Charter schools, therefore, enjoy more autonomy than traditional public schools over certain budget decisions. Indeed, in a national study, charter schools were found to spend less on instruction but more on administration when compared with their public school counterparts (Miron & Urschel, 2010). That was also the case across charter schools in Michigan (Arsen & Ni, 2012).

What one considers fair “probably depends on within which sector one works or with which they otherwise identify” (Miron, 2017, p. 232). Regardless one’s opinion, charter schools’ limited revenue raising ability and high reliance for tuition payments from states or sponsoring school districts leaves them susceptible to political machinations and economic volatility. Take Pennsylvania as an example, where school districts created cash flow issues for their sponsored charter schools when in 2015–2016 they withheld charter tuition payments to counteract the state’s 9-month budget impasse (Palochko, 2015). Similarly, Chicago’s public schools withheld millions of dollars in charter school tuition payments until the school district reached a new agreement on how to calculate charter school tuitions (Kapp, 2019). Recently, Michigan’s Governor Whitmer’s decision to veto a spending increase for charter school students forced charter schools to dip into their reserves and postpone projects (Khan, 2019).

Further financial complications are now more likely than ever. Although the COVID-19 crisis is still unfolding, there is little doubt that school budgets have been affected. Potential disruption in revenue flow due to growing holes in state and local government budgets and expensive social distancing and other health and safety guidelines have put tremendous fiscal pressure on all schools, whether public or private (Barret, 2020). Therefore, developing a better understanding of how charter schools accumulate and use fiscal savings could prove essential for their financial management.

Fiscal Savings for State and Local Governments and School Districts

State and local governments have long realized the benefits of establishing a fiscal emergency mechanism to alleviate budgetary gaps and sustain service provision during periods of economic uncertainty. Today, all states but Colorado have established budget stabilization funds (rainy day funds) to balance their budget and support spending during downturn years. States used their rainy day funds in the 2001 and 2004 recessions to avert over US$20 billion in either cuts to service or tax increases (McNichol & Boadi, 2011). During the Great Recession, states drained their rainy day funds to deter draconian budget reductions (National Association of State Budget Officers [NASBO], 2013). Currently, states spend their rainy day funds once again for bridging budget gaps created by the COVID-19 pandemic.

But the effect of rainy day funds on state budgets differs considerably, depending on their level and withdrawal rules (Chapman et al., 2020). In 2020, for instance, Wyoming’s rainy day funds could support up to a full year of its expenses, while Illinois’s and New Jersey’s could not sustain their operations for 1 day (NASBO, 2020). Of the states with rainy day funds, only Kentucky, Maryland, Nebraska, Ohio, and Wyoming lack withdrawal rules. State legislatures can block or delay the use of rainy day funds too. Indeed, in Washington, Governor Jay Islee’s proposal to withdraw US$300 million in December 2019 did not find state legislature support. In Delaware, up until 2010, legislation did not clearly state the conditions that would trigger the use of the rainy day funds. Withdrawal limits also exist; in Mississippi, for instance, policymakers can withdraw up to US$50 million per year (Chapman et al., 2020).

Nonetheless, research has shown that state rainy day funds require well-designed rules to be effective. During the 1990–1991 recession, Sobel and Holcombe (1996) found rainy day funds to be more successful in limiting fiscal stress among states with mandatory deposit requirements. However, when and how much to save are often troubling decisions for state elected officials. Increasing rainy day funds when a budget surplus is achieved—a common strategy among states—could be problematic in some cases, considering a surplus could be reached through fiscal austerity, not just economic growth. Since the Great Recession, however, states increasingly link rainy day fund accumulation with revenue volatility. In 2019, among the states with formal rainy day funds, 20 have legally tied rainy day fund deposits to volatility; “saving more in high-growth years and less in leaner ones” (Murphy et al., 2018).

While rainy day funds at the state level are a result of policy design, at the local level, formal rainy day funds appear scarce, although this does not mean that local governments do not establish savings for addressing budgetary fluctuations. Overall, cyclical smoothing might be more of a necessity at the local level (Stewart, 2009; Wang & Hou, 2012), as local governments operate within a more restricted economic environment than states (Bahl, 1985). But rather than establishing formal rainy day funds, local governments build fiscal savings as part of their fund balance (Hendrick, 2006). They use their savings for purposes beyond countercyclical stabilization, including covering for contingencies, maintaining tax rates and cash flow, and funding capital projects and citizens’ or businesses’ demands (Hendrick, 2006; Marlowe, 2005).

Fiscal savings are also necessary for school districts and charter schools as they could provide a budgetary cushion against revenue volatility or unplanned expenses and other contingencies. Despite the benefits, school districts in some states face restrictions in how much savings they could accumulate. Seven states, for instance, place limits on the level of the unassigned fund balance while three states have introduced minimums,² the lowest level of unassigned fund balance a school district should retain. Surprisingly, these rules do not follow best practices; instead, they appear arbitrary. As a result, school districts might be unable to establish fiscal savings capable to sustain service provision when revenues fall short (Arapis & Reitano, 2017). By contrast, in all states but Connecticut,³ charter schools face no fiscal savings restrictions; hence they are free to build savings at their level of choice and ability.

Like local governments, school districts and charter schools build their savings within the fund balance. Depending upon the level of constraint placed on the use of funds, amounts reported in the fund balance are classified as nonspendable, restricted, committed, assigned, or unassigned. While funds in the nonspendable fund balance cannot be spent, most often due to their physical form (e.g., inventory), funds in the other classifications can be, but in some cases only for purposes determined by external actors, laws, or formal government actions. Amounts reported in the restricted fund balance, for instance, carry constraints on their use externally imposed by creditors, grantors, contributors, or laws of other governments. Other constraints could also be imposed by a government’s highest level of decision-making authority (e.g., council or board); such amounts fall under the committed fund balance. Funds spent for specific purposes (e.g., renovations, construction, repairs of buildings) as determined by a governing body (e.g., budget committee) are reported under the assigned fund balance. Only the unassigned fund balance is free of any restrictions and hence can be used for any purpose (Governmental Accounting Standards Board [GASB], 2009).

Evidently, all fund balance classifications play a role in financial management, but roles change depending upon the classification. That said, when devising our hypotheses in the next section, expectations on the use and accumulation of fiscal savings may change across the restricted and the unrestricted fund balance classifications.

Fiscal Savings Determinants

Fiscal sources

Fiscal savings, scholars argue, should reflect risk if they are to serve as buffers against budget shocks (Duncombe & Hou, 2014; Gorina et al., 2019; Stewart, 2006). State governments, for instance, with volatile revenues retain higher budget stabilization funds to mitigate revenue fluctuation (Pew Charitable Trusts, 2018). In this vein, local governments dependent on stable revenues (e.g., property taxes) should feel less pressure for building high savings (Gore, 2009; Hendrick, 2006; Stewart et al., 2013). Surprisingly, studies of school districts in Pennsylvania and New York found unrestricted savings being higher in districts with strong property systems (Arapis et al., 2017; Duncombe & Hou, 2014). School districts in Kentucky went even further, following a recent study from Barrett et al. (2019), as they purposefully overestimated expenses and underestimated revenues to build fiscal slack. Perhaps, school districts are risk averse. Given the opportunity, they will accumulate as much savings as feasible (Duncombe & Hou, 2014).

Like school districts, charter schools could also exhibit a similar opportunistic behavior. Given their lack of taxing authority, their ability to accumulate fiscal savings appears to be compromised. A window of opportunity, however, opens when tuition payments provided by the state or sponsoring school districts exceed charter school expenses. Because these tuition payments are based off school district expenses, not the actual cost it takes a charter school to educate its students, they could lead to drastic overpayments (PSBA, n.d.). In such cases, when tuition payments exceed expenses, charter schools could use the surplus to build savings in their fund balance.

Hypothesis 1 (H1): Tuition surplus is positively associated with both restricted and unrestricted fund balances.

The extent to which general-purpose governments, school districts, and charter schools have access to other financing sources could define their fiscal savings behavior further. Cities operating their own business type activities (e.g., electric or water utilities) tend to hold lower fund balances, which they boost when necessary through interfund transfers (Arapis & Reitano, 2018; Hendrick, 2006; Marlowe, 2005). While utilities appear to be a common source of fiscal slack across municipalities, school districts and charter schools find additional funds through other financing sources including private ones (Batdorf et al., 2015). For charter schools specifically, funding from private sources (e.g., private philanthropy) provides in some cases significant revenues. KIPP (Knowledge Is Power Program), for instance, a national network of charter schools, received in 2008 US$5,700 per pupil in private funding (Miron et al., 2011). According to another study of private funding sources across 15 states, charter schools receive up to 1.5 times higher private funding per pupil than traditional public schools (Batdorf et al., 2015). We expect charter schools generating additional revenue through private funding sources to retain higher savings.

Hypothesis 2 (H2): Additional revenue from private funding sources is positively associated with both restricted and unrestricted fund balance.

Reliance to intergovernmental aid (state or federal) could also influence fiscal policymaking. Such revenues typically carry restrictions and can only be used for specific costs. Charter schools, for instance, can apply for direct state support to cover start-up costs, or costs related to facility acquisition, lease rentals, even renovation of existing buildings (ECS, 2020). Federal aid supports specific groups of students (disadvantaged students, students with limited English proficiency [LEP], or with learning disabilities), professional development of teachers, and even curriculum design and facilities (National Association for Charter School Authorizers [NACSA], n.d.).

But future collection of intergovernmental aid entails some uncertainty as it is often subject to legislative or political discretion (Marlowe, 2005; Hendrick, 2006). State reimbursements of leases for charter schools, for instance, are determined by state law, which could change based upon political or economic conditions. Federal aid for students with specific socioeconomic backgrounds or learning needs could also fluctuate depending on enrollment or even arrive late. While constraints on state and federal aid should lead to higher restricted fund balances, charter schools should also retain higher unrestricted fund balance as a precautionary measure against fluctuation in aid or delays in payments.

Hypothesis 3 (H3): Receipt of intergovernmental aid is positively associated with both restricted and unrestricted fund balances.

Like revenues, expenses also influence fiscal savings. Indeed, when Arapis et al. (2017) examined fiscal savings of Pennsylvania school districts, they found school districts with high current expenses such as employment, transportation, and operation and maintenance costs to retain smaller unassigned fund balances. School districts, they noted, use their unassigned fund balance to support their spending. Compared with school districts, charter schools face different costs. In some states, for instance, they are free from collective bargaining agreements (Miron, 2017) and teacher certification requirements (ECS, 2020). If fewer laws and regulations constrain charter school management, charter schools should have greater control of their current expenses. Evidence, however, suggests that charter schools are most successful with controlling their instructional expenses (Arsen & Ni, 2012; Miron & Urschel, 2010). Therefore, charter schools with high non-instructional expenses might have a lesser savings ability or even face pressure to cover expenses through their fund balance, leading to lower restricted and unrestricted fund balances.

Hypothesis 4 (H4): Non-instructional expenses are negatively associated with both restricted and unrestricted fund balances.

Annual retirement obligations could also have a significant impact on fiscal savings behavior. As noted earlier, charter schools in some states are free to elect whether to participate in the state pension system or offer their own retirement benefit plan; 403(b) or 401(k) plans (NASRA, 2017). Given that state pension funds are largely underfunded across the country, an increasing number of charter schools are deciding to offer their own defined contribution plan (Podgursky et al., 2018). Regardless of the option, retirement costs reduce spending flexibility and absorb slack resources that could boost the unrestricted fund balance (Hendrick, 2006). Higher restricted fund balances, however, are necessary so that retirement contributions are made on time.

Hypothesis 5 (H5): Retirement cost is positively associated with the restricted fund balance but negatively associated with the unrestricted fund balance.

In addition to current expenses, long-term obligations could also play a key role in fiscal savings behavior. Debt, for instance, is considered another form of fiscal slack, albeit less liquid than the unrestricted fund balance (Hendrick, 2006). A trade-off, therefore, exists between debt financing and accumulating unrestricted fiscal savings (Hendrick, 2006). Indeed, research at the local level provides evidence that municipalities issuing debt maintain lower unrestricted fiscal savings (Arapis & Reitano, 2018; Hendrick, 2006; Stewart, 2009; Wang & Hou, 2012). Consistent with local government studies, Duncombe and Hou (2014) and Arapis et al. (2018) also showed that New York and Pennsylvania school districts substitute their unrestricted fund balance with debt. Following these findings, charter schools should use debt as an alternative source of fiscal slack and retain a lower unrestricted fund balance. But those with debt obligations should also retain high restricted fund balance to guarantee debt service payments.

Hypothesis 6 (H6): Debt is positively associated with the restricted fund balance but negatively with the unrestricted fund balance.

Education cost may differ significantly upon enrollment. Economies of scale could improve the potential for savings if expenditures per pupil decline as enrollment increases. Gains could be achieved if teachers in larger districts become more productive because “they can draw on the experience of many colleagues” or if “large districts [are] able to negotiate bulk purchases of supplies and equipment at a relatively low cost or use their monopsony to impose lower wages on their employees” (Duncombe & Yinger, 2008, p. 16). Despite the aforementioned, Duncombe and Hou (2014) found small New York school districts retaining higher fund balances than larger districts. Perhaps potential cost savings from economies of scale are seldom realized because school districts rarely lay off staff, salaries are often fixed, and transportation cost increases as enrollment increases (Lee & Smith, 1997). But for charter schools that enjoy more autonomy in employment decisions and are free from transportation service, economies of scale may be more easily achieved. High enrollment, therefore, leads to higher fiscal savings.

Hypothesis 7 (H7): High enrollment is positively associated with both restricted and unrestricted fund balance.

Student demographic characteristics could further influence fiscal savings. For example, disadvantaged students increase the cost of education because of their greater difficulty achieving academically (Duncombe & Hou, 2014). Indeed, both Duncombe and Hou (2014) and Arapis et al. (2017) in their studies of New York and Pennsylvania school districts found evidence for an inverse relationship between students receiving subsidized lunch and the unassigned fund balance. Similarly, students with LEP or with disabilities face significant challenges and require more resources to succeed in school (Duncombe & Yinger, 2008). Typically, funding for such students comes from the federal and state government along with service level mandates; hence charter schools enrolling low income or LEP students or students with disability should retain higher restricted fund balance. The increased cost of educating such students may need additional support, most likely from the unrestricted fund balance.

Hypothesis 8 (H8): Enrollment of low-income students, or with LEP, or with disabilities is positively associated with the restricted fund balance but negatively associated with the unrestricted fund balance.

Research Methods

Unit of Analysis—Pennsylvania Charter Schools

For this study, charter schools from Pennsylvania serve as the unit of analysis. In Pennsylvania, charter school legislation was passed in 1997 when Governor Tom Ridge found common ground with community activists who supported better educational options for historically underserved children (Behmarn, 2017). Shortly thereafter, Pennsylvania’s first brick-and-mortar charter school opened in Mercer County, with four more following closely thereafter in Philadelphia.⁴ Cybers were added to the mix in 2002 (PSBA, 2016). As of 2019, 165 brick-and-mortar charter schools operated in Pennsylvania, with an additional 15 cybers serving students statewide (Pennsylvania Department of Education [PDE], n.d.-b). They provided education for 146,556 (8%) of the state’s 1.8 million public school students (PDE, n.d.-c). Their vision, like that of their national counterparts, has been to provide an education choice free of bureaucratic intervention for low-income students and an alternative option when a public school closes, to instill cultural and ethnic perspectives, and to extend services that utilize a more procurable revenue stream (Miron et al., 2002).

In Pennsylvania, individuals, teachers, parents or guardians of students, colleges, universities, museums or other nonprofit organizations located within the Commonwealth, and even a corporation can establish a charter school (PDE, n.d.-a). The authorization process runs through the local school district for brick-and-mortar charter schools but is governed by the PDE for cyber charter schools. In both cases, the charter school applicant must submit an application describing the mission, grade levels, governance structure, curriculum, complaint review process, facility or lease arrangements, and financial plan. Applicants are guaranteed a hearing and are able to resubmit or appeal to the Charter School Appeal Board if denied (Public School Code of 1949, § 1717-A and § 1719-A).

As in all other states, Pennsylvania charter schools lack taxing authority. Their primary funding source is a per-pupil tuition payment from the school district of residence. It is based on the school district’s total budgeted expenditures per pupil net of costs related to special education, debt financing, and other services charter schools do not have to provide (e.g., transportation, adult education, community/junior college programs, nonpublic services, facility acquisition, and construction services) (PSBA, 2016). For special education students, charter schools receive an additional amount equal to the school district’s total special education funding divided by 16% of its average daily membership (Public School Code of 1949, § 1725-A).

In addition to tuition payments, charter schools could also apply for state and federal aid. The state of Pennsylvania, for instance, distributes aid to reimburse charter school leases, rentals, health services, and grants improving student achievement and school safety. Federal funding is used most commonly to support the academic achievement of disadvantaged students (Title I) and their nutrition, the recruitment and professional development of high-quality teachers and principals (Title II), as well as grants for 21st Century Schools (Title IV) and students with learning disabilities (Individuals With Disabilities Education Act [IDEA]); other grant funds through the Charter School program support planning and program design and even financing facilities (NACSA, n.d.). Charter schools also raise funds from other financing sources such as donations from private funding sources.

Differences exist in costs as well between Pennsylvania charter schools and school districts. Unlike public school teachers who must be certified, charter schools can hire teachers who are not. Each charter school, however, cannot hire more than 25% of non-certified teachers (ECS, 2020). Furthermore, while collective bargaining is allowed, charter school teachers cannot participate in the school district’s bargaining unit. As a result, collective bargaining remains limited with agreements formed in just 7.7% of the Commonwealth’s charter schools (National Alliance for Public Charter Schools, 2017). Pennsylvania charter schools face fewer transportation costs as well when compared with public schools, as their sponsoring school districts are required to provide transportation for resident charter school students traveling not more than 10 miles outside the school district’s boundary (ECS, 2020). Charter schools also have the option to elect whether to participate in the state public school employees’ retirement system (PSERS) or offer an approved alternative retirement plan. Surprisingly, as of 2017, all but one charter school have chosen to participate in Public School Employees Retirement System (PSERS) (NASRA, 2017).

Dependent and Independent Variables

For the purpose of this study, we use as the dependent variable all three fund balance classifications—committed, assigned, and unassigned—that Pennsylvania charter schools have reported since 2011 when GASB Statement 54 went into full effect. Of these three classifications, the committed fund balance is the most restrictive carrying restrictions placed by the highest level of decision-making authority, the charter school’s appointed board of trustees (Pennsylvania Office of the Budget, 2018). Amounts in the committed fund balance are generally set aside to cover anticipated future costs like debt payments, medical insurance premiums, or postemployment benefits. An authorized body (e.g., budget committee) could also restrict the assigned fund balance for specific purposes, such as covering operation or other education costs (PA Distance Learning Charter School, 2014). But those restrictions are not legally binding, and hence a vote of the governing board could lift them. In the unassigned fund balance, amounts deposited are free of any restriction and available to support cash flow and cover any kind of contingency, like a budget shortfall (Table 1).

Table 1.

Variables, Definitions, Data Sources, and Hypothesis.

Variables	Definitions	Data Source	Hyp. Committed Fund Balance	Hyp. Assigned Fund Balance	Hyp. Unassigned Fund Balance
Dependent variables
Committed Fund Balance pp	Committed fund balance per pupil	PDE; AFR data
Assigned Fund Balance pp	Assigned Fund Balance per pupil	PDE; AFR Data
Unassigned Fund Balance pp	Unassigned Fund Balance per pupil	PDE; AFR Data
Fiscal factors—Revenue
Tuition Surplus	Binary variable; 1 being for Charter Schools receiving tuition payments that exceed their total expenses, 0 otherwise	PDE; AFR Data	(+)	(+)	(+)
Private Funding pp	Funding from private sources per pupil	PDE; AFR Data	(+)	(+)	(+)
Intergovernmental Aid pp	State and Federal Aid per pupil	PDE; AFR Data	(+)	(+)	(+)
Fiscal factors—Expenditures
Instructional Exp. pp	Instructional expenditures per pupil	PDE; AFR Data
Non-Instructional Exp. pp	Current (due within 1 year)expenditures net of instructional per pupil	PDE; AFR Data	(–)	(–)	(–)
Retirement Cost pp	Social Security and Retirement Contributions per pupil	PDE; AFR Data	(+)	(+)	(–)
Debt Outstanding pp	Long-term debt per pupil	PDE; AFR Data	(+)	(+)	(–)
Demographics
Enrollment	Total number of students enrolled	PDE; Enrollment Reports and Projections	(+)	(+)	(+)
Free and Reduced Lunch %	Number of students receiving subsidized lunch as a ratio of total enrollment	NCES; ELSI data	(+)	(+)	(–)
English Proficiency %	Number of English Language Learners as a ratio of total enrollment	PDE; English as a Second Language	(+)	(+)	(–)
Learning Disability %	Number of special education students as a ratio of total enrollment	NCES; ELSI data	(+)	(+)	(–)
Control variables
Cyber Charter School	Binary variable; 1 being Cyber Charter School, 0 otherwise	PDE; AFR Data
Suburb	Binary variable; 1 being for Charter Schools located in suburban areas, 0 otherwise	PDE; School Locale Data
Rural	Binary variable; 1 being for Charter Schools located in rural areas, 0 otherwise	PDE; School Locale Data

Note. PA = Pennsylvania; PDE = Pennsylvania Department of Education; AFR = Annual Financial Report; NCES = National Center for Education Statistics; ELSI = Elementary/Secondary Information System.

In regard to independent variables, our model includes variables capturing major revenue sources—tuition surplus, intergovernmental aid, and private funding—and major expenses—instructional, non-instructional, retirement contributions, and outstanding debt. Demographic information is also included. Enrollment is used to capture the impact of economies of scale on fiscal savings, while the percentage of students receiving a subsidized lunch to capture a student’s socioeconomic status. We also include percentage of special education students and those with LEP as these groups impose higher costs on education providers. Furthermore, a binary variable is employed to examine whether cyber charters retain significantly different savings when compared with brick-and-mortar charter schools. Considering that charter schools may face conditions that differ between urban, suburban, and rural areas, we also use two binary variables to capture the difference in fiscal savings between suburban or rural charter schools and those located in urban areas.

Research Methodology

Drawing from the local government and school district fiscal savings literature, we use the following model to test each hypothesis:

\begin{array}{l} Y_{i t} = ϕ_{0} + B 1 {(R e v e n u e S o u r c e s)}_{i t - 1} \\ + B 2 {(E x p e n d i t u r e S o u r c e s)}_{i t - 1} \\ + B 3 {(D e m o g r a p h i c s)}_{i t - 1} \\ + B 4 {(C o n t r o l s)}_{i t - 1} \\ + u_{i t - 1} + T E_{t} + F E_{i}, \end{array}

(1)

where Y represents a charter school’s committed, assigned, or unassigned fund balance per pupil for charter school I in period t. B1(Revenue Sources)_it and B2(Expenditure Sources)_it represent all major revenue and expenditure sources per pupil; B3(Demographics)_it captures enrollment and student characteristics (socioeconomic status, type of student), and B4(Controls)_it reflects binary variables capturing whether a charter school is cyber or brick-and-mortar and whether charter schools are located in a suburban or rural area.

During the time of our study (2011–2019), some fluctuation occurred in the number of charter schools as new schools opened and others shut their doors⁵; hence an unbalanced panel is used. Both Breusch-Pagan and White tests suggested that equation (1) suffers from heteroscedasticity, while the Wooldridge test revealed the presence of autocorrelation. Given these concerns, a Newey-West regression is employed. All variables are lagged by 1 year to accommodate for endogeneity. Furthermore, considering Pennsylvania’s long history of delayed budgets, we added time-fixed effects (denoted by TE_t) to control for the impact of the state’s budget process on fiscal savings. Most charter schools are concentrated in urban areas, specifically in Allegheny, Chester, Erie, Dauphin, Lehigh, Philadelphia, and York County (see Appendix Figure A1). County fixed effects, therefore, denoted by FE_i are also included in the model to control for time-invariant unobservable factors. A constant is included in φ₀.

Findings

Descriptive Statistics

Table 2 provides descriptive statistics of this study’s dependent and independent variables. As displayed, Pennsylvania charter schools maintain fiscal savings in all fund balance classifications. On average, they retain an unassigned fund balance of US$1,608 per pupil, enough to support just over a month of the average expenses. Lower levels are retained in the committed and assigned fund balance. But the large standard deviation shows considerable fluctuation in all fund balance levels, with a few charter schools even holding fund balances that could cover the expenses of an entire calendar year.⁶

Table 2.

Descriptive Statistics.

Fund balance per pupil	Observation	M	SD	Minimum	Maximum
Committed	1,544	630.369	1,482.935	−491.233	15,107.910
Assigned	1,544	387.316	1,351.402	−895.980	18,978.490
Unassigned	1,544	1,608.163	2,845.689	−10,181.090	25,509.960
Revenues per pupil
Tuition Surplus (binary variable)	1,544	12.9	33.536	0	1
Private Funding	1,544	252.180	922.080	0	12,942.35
Intergovernmental Aid	1,544	1,561.061	1,477.793	0	14,558.18
Expenditures per pupil
Instruction	1,544	7,351.564	2,242.009	0	17,638.210
Non-Instructional	1,542	5,610.369	2,670.931	0	40,969.37
Retirement Cost	1,544	1,270.321	666.947	0	6,347.685
Debt Outstanding	1,544	7,859.549	10,731.460	−26,281.100	57,652.140
Demographics
Enrollment	1,544	728.130	1,156.185	29	10,559
Free and Reduced Lunch %	1,544	59.4	34.3	0	100
Limited English Proficiency %	1,544	3.6	7.2	0	62.9
Sp. Educ. Enrollment %	1,544	13.4	11.8	1	100

Of the three classifications, the unassigned fund balance is the most popular, the only classification free of any restrictions in purpose and use. In 2019, 84% of charter schools reported an unassigned fund balance; still, this percent is the lowest recorded since 2011 (see Table 3). On the contrary, the percent of charter schools reporting a restricted classification (committed or assigned) has increased over time. The level of the restricted classifications has improved as well; in 2019, both committed and assigned fund balances were 6 times higher than 2011. By contrast, the average unassigned fund balance has lost 28% of its value during the examined time period; from US$2,341 per pupil in 2011 to US$1,688 per pupil in 2019 (see Table 3). The aforementioned leads us to believe that charter schools’ ability to manage revenue fluctuations or unplanned expenses remains quite limited.

Table 3.

Percent of Charter Schools Retaining Fund Balance (2011–2019).

Year	Committed Fund Balance (%)	Assigned Fund Balance (%)	Unassigned Fund Balance (%)	No. of charter schools
2019	35	24	84	180
2018	32	21	95	179
2017	31	18	92	179
2016	27	16	93	175
2015	26	18	92	176
2014	22	18	94	176
2013	17	13	96	174
2012	14	14	98	161
2011	14	19	90	144

According to Table 2, on average, 12.9% of charter schools receive tuition payments from sponsoring school districts that are enough to cover their total expenses and generate a surplus. Some fluctuation exists as well; depending upon the year, the percent of charter schools achieving surplus ranges from a low of 7.6% in 2011 to a high of 20.5% in 2019 (see Appendix Table A1). Charter schools draw on average US$1,561.061 per pupil from intergovernmental aid, but our sample includes charter schools that received no federal or state funds either. Less revenues derive from private funding sources (averaging US$252 per pupil); but as with intergovernmental aid, significant variation is observed as well.⁷ On the expenses side, instruction (averaging US$7,352 per pupil) was the highest annual expense, followed by non-instructional (averaging US$5,610 per pupil) and retirement contributions (averaging US$1,270 per pupil). In regard to long-term liabilities, Pennsylvania charter schools owe on average US$7,860 per pupil in debt.

Following Table 2, Pennsylvania charter schools enroll on average 728 students, but our panel also includes schools with higher and lower enrollment. On average, 60% of students were of low-income, 13.4% identified as students with special needs, and 3.4% with LEP. In some schools, enrollment of low-income or special needs students reached up to 100%.

Regression Analysis

Table 4 shows the regression results grouped by fund balance classification: committed, assigned, or unassigned. Despite their limited revenue portfolio, our results show that charter schools find opportunities to accumulate fiscal savings. When charter schools, for instance, receive tuition payments from sponsoring school districts that exceed their costs, or have access to private funding sources, fiscal slack is generated. As expected in H1 and H2, tuition surplus and private funding developed positive statistically significant associations with both restricted and unrestricted fund balances. Charter boards as the ones responsible for the financial and operational management of charter schools, showcase financial maturity using tuition surplus and additional revenues from private funding sources not only to prepare for unexpected expenses but also to plan ahead for known educational or operational concerns. Notably, tuition surplus and funding from private sources improve the unrestricted fund balance the most. Charter schools receiving intergovernmental aid also retain a higher unrestricted fund balance, partially confirming H3. Charter schools appear wary of potential volatility with intergovernmental aid and hence retain higher unassigned fund balance to counteract payment delays or any fluctuation caused from state or federal legislative changes.

Table 4.

Results of Newey West Regression.

	Restricted fund balances		Unrestricted fund balance
	Committed	Assigned	Unassigned
	Coefficient (Panel Corrected SE)	Coefficient (Panel Corrected SE)	Coefficient (Panel Corrected SE)
Revenues per pupil
Tuition Surplus (binary variable)	511.289** (160.269)	524.583** (180.629)	999.637** (297.473)
Private Funding	−0.011 (0.049)	0.081* (0.044)	0.369** (0.152)
Intergovernmental Aid	0.019 (0.028)	−0.022 (0.030)	0.212** (0.088)
Expenditures per pupil
Instruction	0.097*** (0.026)	−0.035* (0.020)	−0.201** (0.085)
Non-Instructional	−0.105*** (0.027)	−0.054** (0.016)	−0.171*** (0.038)
Retirement Cost	0.605*** (0.116)	0.310** (0.094)	0.857** (0.246)
Outstanding Debt	−0.027*** (0.005)	−0.008 (0.005)	0.001 (0.013)
Demographics
Enrollment	0.112* (0.058)	−0.066* (0.037)	−0.177** (0.053)
Free and Reduced Lunch %	−3.058** (1.444)	2.936** (1.260)	−8.023** (2.593)
Limited English Proficiency %	−19.623** (7.277)	33.266* (19.811)	17.520 (12.056)
Special Ed. %	−4.094 (3.314)	0.245 (2.611)	18.770* (9.880)
Control variables
Cyber	−103.621 (231.362)	−470.436** (205.293)	255.157 (274.545)
Suburb	152.464 (137.579)	−242.987 (163.992)	−861.968** (287.121)
Rural	−1,764.234** (587.188)	−97.415 (381.300)	−77.877 (668.554)
Constant	1,036.248 (664.857)	−134.049 (688.403)	2,577.747** (788.367)
Obs.	1,338	1,338	1,338
Prob > F	.0000	.0000	.0000
Adj. R²	.2100	.2015	.3740

Note. The model also includes fixed effects for county and time effects not displayed in this table. The Adj. R² is from the ordinary least squares regression, as the Newey-West regression does not produce one. SE = standard error.

p < .05.**p < .01. ***p < .001.

In regard to expenditures, our results provide evidence of the charter schools’ ability to control their instructional expenses, but not their non-instructional. As hypothesized in H4, non-instructional expenses are significant cost drivers which require support from both the restricted and unrestricted fund balance. Interestingly, while charter schools use their fund balance to support their non-instructional expenses, they appear to achieve fiscal gains in their instructional expenses. Perhaps, being able to hire non-unionized and up to 25% non-certified teachers allows charter schools to move fiscal slack from their unrestricted to their committed fund balance. Charter boards, once again, show maturity in their fiscal decision-making and commit part of their fiscal slack for the most central expense: Instruction.

Annual retirement cost as a mandatory expense also developed positive statistically significant associations with both restricted and unrestricted fund balances; not all associations, however, were in the expected direction as noted in H5. The vast majority of Pennsylvania charter schools do not offer their own individual retirement plans. As of 2017, all but one charter school participated in the state’s PSERS instead (NASRA, 2017). Interestingly, during the time frame of our study, the state’s large unfunded pension liability brought sharp increases in the employer contribution rate for each participating employee which climbed from 5.64% in 2011 to 33.43% in 2019 (Pennsylvania Public School Employees’ Retirement System [PSERS], 2019). As such, charter schools plan ahead for funding their retirement commitments, making sure they hold enough funds into their committed and assigned fund balance. But alarmed from the steep changes in PSERS this past decade, they also retain considerably higher unassigned fund balance as another precautionary measure against new mandates. By contrast, charter schools with outstanding debt retain lower committed fund balance, challenging H6. Perhaps, low borrowing cost following the years of the Great Recession has undermined the importance of retaining high restricted fund balances.

In H7, we hypothesized that charter schools being free from employment regulations, collective bargaining, and transportation service could help them achieve economies of scale and hence further improve their potential for savings. Based on our findings, it is not clear whether economies of scale are achieved. While enrollment has developed a positive statistically significant association with the committed fund balance, directionality flips for the assigned and unassigned fund balance. Looking at the coefficients, it appears likely that higher enrollment improves savings potential enabling charter schools to transfer some of their slack from the unassigned and assigned fund balance to the committed fund balance.

Student characteristics, as expected in H8, play a significant role in fiscal savings use and accumulation. To some extent, our expectations for low-income students and students with LEP are met. According to the estimates, charter schools enrolling high concentrations of low-income and LEP students retain at least a higher assigned fund balance. Perhaps, charter schools move funds from the committed to the assigned fund balance, thus building some flexibility in the way they fund services designed to support their low-income or LEP students.

Surprisingly, in the case of high enrollment of students with disability, charter schools retain significantly higher unassigned fund balance. This could be a result of unfair distribution of funds as charter school funding in Pennsylvania for special education students is driven by the average cost of total special education enrollment of the sponsoring school district, not the true cost of each special education student enrolled in charter schools. Thus, charter schools receive an overpayment when they enroll below-average-cost special education students and are “shortchanged” when high-cost special education students enroll (Public School Code of 1949, § 1725-A; Special Education Funding Commission, 2013). For that reason, Pennsylvania charter schools have an incentive to target low-cost special education students. Indeed, for 2018–2019, PDE data show the share of special education students costing between US$0 and US$25,000 (the lowest cost category) was 89.9% for school districts, 93.6% for charter schools. As a result of this higher share, they are left with a significant amount of residual funds which become part of their unassigned fund balance.

Our control variables suggest that cyber charter schools retain lower assigned fund balance than brick-and-mortar; location matters; and delays of payments may significantly alter fiscal savings behavior. According to the time effects included in the model, the 2015–2016 budget impasse in the Commonwealth, which forced school districts to delay charter school tuition payments, changed significantly the way charter schools accumulate their savings. In fact, charter schools have made a clear effort to increase their committed and assigned fund balances, with increases being higher after the 2015–2016 budget impasse. Perhaps, this shows the intent of charter boards to establish formal savings used only for fiscal stabilization purposes.

Discussion of Findings and Conclusions

Since the 1990s, charter schools have spread across the nation; yet we know little about their fiscal savings and overall ability to weather rainy days or unplanned expenses and guarantee smooth service provision. Unlike most other fiscal savings studies focusing on the unrestricted fund balance, this study provides an examination of all fund balance classifications, both restricted and unrestricted. This allowed us a deeper understanding of fiscal savings behavior, as all fund balance classifications play a role in financial management. As such, fiscal savings decisions are made not only for the unrestricted fund balance but for the restricted as well. We hope this motivates future fiscal savings studies to pay attention to all fund balance classifications, not solely the unrestricted as typically is the case. Such systematic investigations are necessary for broadening our understanding of fiscal savings behavior and advancing the literature further.

Following the patterns of local governments and school districts, the most popular fund balance classification for Pennsylvania charter schools was the unassigned, the only classification free of any restrictions in purpose and use. Pennsylvania charters retain on average a rainy day fund of US$1,608 per pupil, enough to support just over a month of the average school’s operations. For this reason, we fear that charter schools are only prepared for drizzle, not for rain. Given their limited revenue portfolio, charter schools are left with few options for accumulating fiscal savings; tuition surplus and private funding sources appear as solid opportunities. Because tuition payments, the main revenue source for charter schools around the country, are a function of school district expenses, charter schools have an incentive to seek after efficiency gains in their service delivery. Doing so could maximize their potential for tuition surplus which then could be used for boosting both restricted and unrestricted fiscal savings. Our results also show that access to private funding sources provides significant benefits to fiscal savings accumulation. This adds to previous studies on the importance of private funding in charter school financial management (Batdorf et al., 2015; Miron et al., 2011) and offers an additional reason for charter schools to invest more time, money, and effort in fundraising.

Our results also showed that Pennsylvania charter schools might be receiving from their sponsoring school districts more than what they actually need (or spend) to serve students with learning disabilities. An urgent need exists, therefore, for revising the state’s charter school special education funding formula so that calculations involve the exact cost of educating students with different needs rather than each sponsoring school district’s average special education cost. Doing so would save Pennsylvania school districts millions of dollars, up to US$280 million a year according to the Governor’s office (Hanna & Graham, 2020), that they would otherwise have to allocate among their sponsoring charter schools. Such savings would provide much-needed budget relief to at least 70% of the Commonwealth’s school districts that now consider mandatory charter school tuition costs their biggest source of budget pressure (PSBA, 2020).

For charter schools, retirement costs appear central to their fiscal savings decision-making. According to our findings, charter schools excessively save slack in the restricted fund balances so that on-time payments are guaranteed. Surprisingly, retirement-related savings in the unrestricted fund balance appear even greater; perhaps signaling a fear for new state mandates. As noted earlier, the state’s large unfunded liability currently at US$44.1 billion has increased employer contributions rates by more than 720%. For 2021, the employer contribution rate is set at 34.51% or a third of every US$1 of eligible wages. With COVID-19 having accelerated retirement of public school teachers, some fear that the unfunded liability will grow further in the coming years and bring steeper increases in contribution rates (Charter Choices, n.d.). PSERS appears too expensive; perhaps, the Commonwealth’s charter schools should explore offering their own retirement plan as many others have already done around the country (Podgursky et al., 2018). Doing so could free fiscal slack for purposes beyond retirement contributions.

As evidenced by the state fiscal savings literature, rainy day funds are most effective when rules dictate their use and accumulation. Considering that all states but Connecticut lack rules on the accumulation of the unassigned fund balance, state statutory minimums could be beneficial for charter schools in Pennsylvania and elsewhere. State minimums could be modeled after Government Finance Officers Association’s (GFOA) recommendation, which encourages an unassigned fund balance of at least 2 months of operating expenses. Equally important to state minimum is for charter school boards to adopt formal fund balance policies articulating in detail how restricted and unrestricted fund balances are accumulated and used, and how they are replenished when their level falls below the state minimum. Taken together, state fiscal savings minimums and formal fund balance policies could facilitate charter schools in establishing rainy day funds capable of insulating their operation from unanticipated expenses or revenue fluctuation. Such rules could also lead charter boards toward prudent financial management and protect funds committed to serve specific type of students (e.g., students with disability) from being used for other purposes.

Our research does not come without caveats. It appears, for instance, that charter schools make some gains as enrollment increases; more research, however, is required before we claim that Pennsylvania charter schools have reached economies of scale. This would require fiscal and demographic information grouped by grades as cost savings studies show that economies of scale appear at different enrollment levels between elementary schools and high schools (Eberts, 1984; Lee & Smith, 1997). Unfortunately, splitting our data by grade was not possible for our study as only aggregate information is reported. Furthermore, because our findings are based on one state only, their generalizability might be limited or at least require caution. Surely, a national sample would be ideal; however, due to differences in charter school legislation across the country, a state-by-state analysis might still be the best avenue for scholars to discover best practices and ways to evolve charter school legislation.

Footnotes

Appendix

Table A1.

Tuition Surplus Over Time (2011–2019).

Year	No. of charter schools	% of charter schools	Total no. of charter schools
2019	37	20.5	180
2018	34	18.9	179
2017	25	13.9	179
2016	27	15.4	175
2015	19	10.7	176
2014	14	7.9	176
2013	16	9.1	174
2012	16	9.9	161
2011	11	7.6	144
2011–2019	199	12.9	1,544

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: One of the authors works as a Budget Analyst for PA House Democratic Appropriations Committee. The views expressed in this study are those of the authors and do not necessarily reflect the view of the PA House Democratic Appropriations Committee.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Theodore Arapis

Notes

Author Biographies

Theodore Arapis, PhD, is an associate professor of Public Administration at Villanova University where he teaches courses in financial management, public budgeting, and statistics. His research focuses on public budgeting and finance and has been published in such journals as the Journal of Public Administration Research and Theory, the American Review of Public Administration, Public Budgeting & Finance, and Public Finance Review, among others.

Sean Brandon works in state government as a budget analyst for the Pennsylvania House Appropriations Committee. His focus in education funding and policy issues. Sean is a graduate of Villanova University’s Master of Public Administration program.

References

Arapis

Reitano

(2017). School district fiscal savings and state budget rules. Government Finance Review, 33(3), 30–34.

Arapis

Reitano

(2018). A glimmer of optimism in government savings accumulation? An empirical examination of municipal unassigned fund balance in Florida. Public Finance Review, 46(3), 389–420.

Arapis

Reitano

Bruck

(2017). The fiscal savings behavior of Pennsylvania school districts through the great recession. Public Budgeting & Finance, 37(3), 47–70.

Arsen

(2012). The effects of charter school competition on school district resource allocation. Educational Administration Quarterly, 48(1), 3–38.

Bahl

(1985). Financing state and local governments in the 1980s. Land Economics, 61, 332–335.

Barrett

(2020). What charter schools need to know—COVID and education finance: Acting during the impending fiscal downturn. National Alliance for Public Charter Schools. https://www.publiccharters.org/sites/default/files/documents/2020-04/napcs_fiscal_downtrun_rd3.pdf

Batdorf

Cheng

Maloney

L. D.

May

J. F.

Wolf

P. J.

(2015). Buckets of water into the ocean: Non-public revenue in public charter and traditional public schools. https://scholarworks.uark.edu/cgi/viewcontent.cgi?article=1004&context=scdp

Behrman

(2017). Pa. Charter School Law Still Reshaping Education Years Later. The Philadelphia Tribune. https://www.phillytrib.com/news/pa-charter-school-law-still-reshaping-education-20-years-later/article_0e935820-43f6-5997-8144-4242437afaa5.html

Betts

J. R.

Tang

Y. E.

(2019). The effect of charter schools on student achievement. In Berends

Waddington

R.J.

Schoeinig

(Eds) School choice at the crossroads: Research perspectives (pp. 67–89). Routledge.

10.

Chapman

Loiaconi

(2020). State rules can complicate rainy day fund withdrawals. The Pew Charitable Trusts. https://www.pewtrusts.org/en/research-and-analysis/articles/2020/05/12/state-rules-can-complicate-rainy-day-fund-withdrawals

11.

Charter Choices. (n.d.). Pennsylvania’s Public-School Employee Retirement System (PSERS)—Then and now and what to expect with COVID-19. https://www.charterchoices.com/pennsylvanias-public-school-employee-retirement-system-psers-then-and-now-and-what-to-expect-with-covid-19/

12.

Chubb

J. E.

Moe

T. M.

(1990). Politics, markets, and America’s schools. The Brookings Institution.

13.

Connecticut Official State Website. (n.d.). Charter schools: Laws/regulations. https://portal.ct.gov/SDE/Charter-Schools/Charter-Schools/Regulations

14.

Davare

D. W.

(2018). Explaining school fund balances: Are PA local education agencies, with a $4.9 billion fund balance, holding too much in reserve? An update for FY 2016–17. Center on Regional Politics. https://drive.google.com/file/d/1cOCrpu5arH_un8sLUKg5MHfRzqgZpXCE/view

15.

DeAngelis

C. A.

Wolf

Maloney

May

J. F

. (2020). Charter school funding: Inequity surges in the cities. https://scholarworks.uark.edu/cgi/viewcontent.cgi?article=1080&context=scdp

16.

Duncombe

Hou

(2014). The savings behavior of special purpose governments: A panel study of New York school districts. Public Budgeting and Finance, 34(3), 1–23.

17.

Duncombe

Yinger

(2008). Measurement of cost differentials. In Ladd

Fiske

E. B.

(Eds.), Handbook of research in education finance and policy (pp. 238–256). Routledge.

18.

Eberts

R. W.

(1984). The effect of school size on student outcomes. Final Report. Eugene, OR: Center for Educational Policy and Management, College of Education. https://files.eric.ed.gov/fulltext/ED245382.pdf

19.

Education Commission of the States. (2020). 50-state comparison: Charter school policies. http://www.ecs.org/charter-school-policies/

20.

Ericson

(2001). Challenge and opportunity: The impact of charter schools on school districts. Office of Educational Research and Improvement, US Department of Education. https://www2.ed.gov/rschstat/eval/choice/district_impact.pdf

21.

Finn

Manno

Vanourek

(2000). What’s a Charter School? In Finn

C. E.

Jr. Manno

B. V.

Vanourek

(Eds.), Charter schools in action: Renewing public education (pp. 13–22). Princeton University Press.

22.

Gore

A. K.

(2009). Why do cities hoard cash? Determinants and implications of Municipal Cash Holdings. The Accounting Review, 84(1), 183–207.

23.

Gorina

Maher

Park

(2019). Toward a theory of fiscal slack. Public Budgeting & Finance, 39(4), 48–74.

24.

Governmental Accounting Standards Board. (2009). Statement No. 54 of the governmental accounting standards board. https://gasb.org/resources/ccurl/313/494/GASBS%2054.pdf

25.

Hanna

Graham

K. A.

(2020). Philly schools overpaying charters millions of dollars for special-education students, district says. The Inquirer. https://www.inquirer.com/education/charter-schools-philadelphia-special-education-costs-20200221.html#loaded

26.

Hendrick

(2006). The role of slack on local government finance. Public Budgeting & Finance, 26(1), 14–46.

27.

Hou

(2003). What stabilizes state general fund expenditures in downturn years—Budget stabilization fund or general fund unreserved undesignated balance? Public Budgeting & Finance, 23(3), 64–91.

28.

Kapp

(2019). Chicago Public Schools withholding millions from charter schools in spending standoff. NPR. https://www.npr.org/local/309/2019/04/18/714914706/chicago-public-schools-withholding-millions-from-charter-schools-in-spending-standoff

29.

Khan

(2019). Budget delay hurting charter schools. WILX10. https://www.wilx.com/content/news/Budget-delay-hurting-charter-schools-564872372.html

30.

Lay

J. C.

Bauman

(2019). Private governance of public schools: Representation, priorities, and compliance in New Orleans charter school boards. Urban Affairs Review, 55(4), 1006–1034.

31.

Lee

V. E.

Smith

J. B.

(1997). High school size: Which works best and for whom? Educational Evaluation and Policy Analysis, 19, 205–227.

32.

Marlowe

(2005). Fiscal slack and counter-cyclical expenditure stabilization: A first look at the local level. Public Budgeting & Finance, 25(3), 48–72.

33.

McNichol

Boadi

(2011). Why and how states should strengthen their rainy-day funds: Recession highlighted importance of funds and need for improvements. Center of Budget and Policy Priorities. https://www.cbpp.org/sites/default/files/atoms/files/2-3-11sfp.pdf

34.

Miron

(2017). Description and brief history of charter schools. In Fox

R. A.

Buchanan

N. K.

(Eds.), The Wiley handbook of school choice (p. 224). John Wiley.

35.

Miron

Nelson

Risley

(2002). Strengthening Pennsylvania’s charter school reform: Findings from the statewide evaluation and discussion of relevant policy issues. Year Five Report. https://eric.ed.gov/?id=ED480988

36.

Miron

Urschel

(2010). Equal or fair? A study of revenues and expenditures in American charter schools. https://nepc.colorado.edu/sites/default/files/EMO-RevExp.pdf

37.

Miron

Urschel

J. L.

Saxton

(2011). What makes KIPP work? A study of student characteristics, attrition, and school finance. Western Michigan University. https://www.scribd.com/document/51998560/What-Makes-KIPP-Work-A-Study-of-Student-Characteristics-Attrition-and-School-Finance

38.

Murphy

Loiaconi

Bailey

(2018). How states can balance saving for a rainy day and other priorities. The Pew Charitable Trust. https://www.pewtrusts.org/en/research-and-analysis/articles/2018/12/04/how-states-can-balance-saving-for-a-rainy-day-and-other-priorities

39.

National Alliance for Public Charter Schools. (2017). Unionized charter schools. https://www.publiccharters.org/sites/default/files/documents/2018-02/Unionized%20Charter%20Schools%202016-17_0.pdf

40.

National Association for Charter School Authorizers. (n.d.). Federal policy. https://www.qualitycharters.org/research-policies/archive/federal-policy-104-oversight-of-federal-funds-in-charter-schools/

41.

National Association of State Budget Officers. (2013). State budgeting and lessons learned from the economic downturn. https://higherlogicdownload.s3.amazonaws.com/NASBO/9d2d2db1-c943-4f1b-b750-0fca152d64c2/UploadedImages/Reports/State%20Budgeting%20and%20Lessons%20Learned%20from%20the%20Economic%20Downturn-final.pdf

42.

National Association of State Budget Officers. (2020). Fiscal survey of states. https://www.nasbo.org/reports-data/fiscal-survey-of-states

43.

National Association of State Retirement Administrators. (2017). State policies governing pension plan participation by charter school employees. https://www.nasra.org/files/Compiled%20Resources/charterschoolpolicies.pdf

44.

National Center for Education Statistics. (n.d.). Fast facts: Charter schools. https://nces.ed.gov/fastfacts/display.asp?id=30#:~:text=During%20this%20period%2C%20public%20charter,increase%20of%202.7%20million%20students.&text=The%20average%20enrollment%20size%20of,%E2%80%9301%20to%202017%E2%80%9318

45.

Palochko

(2015, October 21). PA to halt charter school payments amid budget impasse. The Morning Call. https://www.mcall.com/news/local/mc-pa-senate-charter-school-payments-20151021-story.html

46.

Pendergrass

S. A.

Kern

(2017). The case for charters. In Fox

R. A.

Buchanan

N. K.

(Eds.), The Wiley handbook of school choice (pp. 237–251). John Wiley.

47.

Pennsylvania Department of Education. (n.d.-a). Charter applications. https://www.education.pa.gov/K-12/Charter%20Schools/Pages/Charter-Applications.aspx

48.

Pennsylvania Department of Education. (n.d.-b). Charter schools. https://www.education.pa.gov/K-12/Charter%20Schools/Pages/Annual-Reports-Data-and-Resources.aspx

49.

Pennsylvania Department of Education. (n.d.-c). Public school enrollment reports. https://www.education.pa.gov/DataAndReporting/Enrollment/Pages/PublicSchEnrReports.aspx

50.

Pennsylvania Distance Learning Charter School. (2014). Fund balance policy. https://www.padistance.org/wp-content/uploads/2017/05/Tab-H-Fund-Balance-Policy-06242014.pdf

51.

Pennsylvania Office of the Budget. (2018). Manual of accounting and financial reporting for Pennsylvania local educational agencies. https://www.education.pa.gov/Documents/Teachers-Administrators/School%20Finances/Comptrollers%20Office/Manual%20of%20Accounting.pdf

52.

Pennsylvania School Boards Association. (2016). The critical need for charter school reform. https://www.psba.org/wp-content/uploads/2015/07/PSBA-Special-Report-Charter-School-Reform.pdf

53.

Pennsylvania School Boards Association. (2020). Charter school tuition now surpassing pension costs as the biggest challenge for district budgets. https://www.psba.org/2020/01/charter-school-tuition-surpassing-pension-costs-district-budgets/

54.

Pennsylvania School Boards Association. (n.d.). Meaningful charter school reform. https://www.psba.org/issue/charter-school-reform/

55.

Pennsylvania School Employees’ Retirement System. (2019). Fiscal year 2020/2021 employer contribution rate. https://www.psers.pa.gov/About/PFR/Documents/20191206%20FY%202020-2021%20ECR%20fact%20sheet%20FINAL.pdf

56.

Public School Code of 1949.24. P. S. § 6-688. https://www.legis.state.pa.us/WU01/LI/LI/US/HTM/1949/0/0014..HTM#:~:text=Relating%20to%20the%20public%20school,changing%20the%20laws%20relating%20thereto

57.

Pew Charitable Trusts. (2018). State revenue volatility and optimal reserve size are directly linked. https://www.pewtrusts.org/en/research-and-analysis/articles/2018/11/14/state-revenue-volatility-and-optimal-reserve-size-are-directly-linked

58.

Podgursky

Pendergrass

S. A.

Hesla

(2018). Pensions under pressure: Charter innovation in teacher retirement benefits. Education Next, 18(2), 8–16.

59.

Public School Code of 1949, § 688; § 1703-A; § 1717-A; § 1719-A; § 1725-A. https://www.legis.state.pa.us/WU01/LI/LI/US/HTM/1949/0/0014..HTM#:~:text=Relating%20to%20the%20public%20school,changing%20the%20laws%20relating%20thereto

60.

Ravitch

Viteritti

(1997). New York: The obsolete factory. In Ravitch

Viteritti

(Eds.), New schools for a new century: The redesign of urban education (pp. 17–36). Yale University Press.

61.

Saulny

(2005). Ray Budde, 82, first to propose charter schools dies. The New York Times. https://www.nytimes.com/2005/06/21/us/ray-budde-82-first-to-propose-charter-schools-dies.html#:~:text=Ray%20Budde%2C%20an%20education%20professor,He%20was%2082.&text=The%20charter%20arrangement%20could%20result,new%20type%20of%20school%2C%20Dr

62.

Sobel

R. S.

Holcombe

R. G.

(1996). The impact of state rainy-day funds in easing state fiscal crises during the 1990–1991 recession. Public Budgeting & Finance, 16(3), 28–48.

63.

Special Education Funding Commission. (2013). Special education funding report. http://pasenategop.com/browne/wp-content/uploads/sites/37/2015/05/Special-Education-Funding-Commission-Report-121113.pdf

64.

Special Education Funding Commission. (2013). Special Education Funding Commission report. https://www.senatorbrowne.com/special-education-funding-commission/

65.

Stewart

L. S. M.

(2009). Examining factors that impact Mississippi counties’ unreserved fund balance during relative resource abundance and relative resource scarcity. Public Budgeting & Finance, 29(4), 45–73.

66.

Stewart

L. M.

Phillips

J. N.

Modlin

(2013). An Evaluation of the Connection Between Economic Volatility and Governmental Savings Levels for Illinois, North Carolina, and Mississippi Counties. Public Finance & Management, 13(4).

67.

Vergari

(2007). The politics of charter schools. Educational Policy, 21(1), 15–39.

68.

Wang

Hou

(2012). Do local governments save and spend across budget cycles? Evidence from North Carolina. American Review of Public Administration, 42(2), 152–169.

69.

Zwara

(2020). What we’re watching: Trends from NACSA’s 2020 Charter School Advocacy Survey. National Association of Charter School Authorizers (NACSA). https://www.qualitycharters.org/2020/02/what-were-watching-trends-from-nacsas-2020-charter-school-advocacy-survey/