Constructing An Adequate School Funding Formula

In 1991, while I was a faculty member in the School of Public Policy & Management at Ohio State University (now known as the John Glenn College of Public Affairs), I was commissioned by Governor George Voinovich to conduct a comprehensive study of K-12 education funding in Ohio.  My report, which was released in November 1992, was titled “Equity, Adequacy and Reliability in Ohio Education Finance”. In the context of my 1992 report these terms can be defined in the following way:

Adequacy – An adequate funding formula is one which is based on the cost of serving students with different educational needs and adjusting funding to reflect uncontrollable cost differences across school districts (i.e. size, geography, sparsity of population).

Equity – An equitable school funding formula adjust for the disparities in local wealth and revenue raising capacity across Ohio’s 600+ school districts. The primary drivers of equity in the Ohio school funding formula are the State/Local Share calculation which directs more state aid to lower wealth districts and less state aid to higher wealth districts, and Targeted Assistance which provides supplemental funding to lower wealth school districts which lack the ability to raise revenue at the local level to the same extent as wealthier school districts.

Reliability – A reliable funding formula is one where the state foundation formula effectively works over time to provide funding to Ohio’s school districts in an equitable and adequate fashion as district conditions change and evolve.

Now, more than thirty years later, equity, adequacy and reliability over time remain the three objectives necessary to constructing a school funding formula that meets the needs of 1.6 million school children and complies with the “thorough and efficient” clause of the Ohio Constitution.

At a practical level, adequacy of the funding formula has two main components. The first component is known as the “base cost“ which is best interpreted as the cost of educating the typical student in the typical school district. Furthermore, the concept of adequacy also extends to the recognition of the additional costs above and beyond what is needed to educate a “typical” student. These costs are often referred to as “categoricals” and refer to additional funding provided for students with disabilities, English language learners, economically disadvantaged students, gifted and talented students, and career technical education students. Categorical funding also includes funding for transportation, the cost of which varies based on the pupil density and geography of Ohio’s school districts. In order for the state funding formula to be considered “adequate”, the base cost and categoricals must all be based on objective analysis of the cost of meeting the educational needs of different types of students as well as on the cost of transporting them to and from school each day.

Prior to the March 24, 1997 DeRolph I ruling by the Ohio Supreme Court, the base cost per pupil amount that was the starting point for Ohio’s school funding formula was determined based on the judgment of the legislature regarding the appropriate amount of funding to allocate to K-12 education. While Ohioans elect their legislators precisely to make such policy decisions, the specific mention of primary and secondary education in the Ohio Constitution indicates that it is the state’s responsibility to provide a “thorough and efficient” system of common schools, requiring a higher standard be met than legislative judgment.The issue of adequacy was addressed explicitly on pages 14 and 15 of the initial DeRolph I decision. Citing a passage in my report for Governor Voinovich, the Ohio Supreme Court stated, 

“The ‘formula amount’ has no real relation to what it actually costs to educate a pupil. In fact, Dr. Howard B. Fleeter, Assistant Professor at the School of Public Policy and Management at Ohio State University, stated that the foundation dollar amount ‘is a budgetary residual, which is determined as a result of working backwards through the state aid formula after the legislature determines the total dollars to be allocated to primary and secondary education in each biennial budget. Thus, the foundation level reflects political and budgetary considerations at least as much as it reflects a judgment as to how much money should be spent on K-12 education.’ (Emphasis sic.).”  (Italics added by the Court.)

The significance of this passage is that the ruling clearly stated that in order to be adequate, the state must utilize objective cost-based methodologies for determining the per pupil foundation level (aka the “base cost”, aka, the cost of educating the typical student) rather than the legislature setting that amount based on budgetary considerations. This conclusion was consistently reiterated in each subsequent DeRolph ruling and as mentioned above, extends to the categoricals as well as to the base cost per pupil.

The aftermath of the DeRolph I ruling saw the state employ a variety methods for determining an adequate base cost per pupil. Unfortunately, the state stopped employing any such methodology after the FY11 school year and returned to the “legislative choice” approach taken prior to the DeRolph rulings.

Ohio’s current school funding formula is based on the funding formula constructed by the Cupp-Patterson school finance workgroup of school superintendents and treasurers and is commonly referred to as the Fair School Funding Plan (FSFP).  The FSFP was introduced as House Bill (HB) 305 in 2019 and passed overwhelmingly by an 87-9 vote of the Ohio House of Representatives in December 2020. However, the Ohio Senate declined to take action prior to the end of the 2020 calendar year and the FSFP was reintroduced in February 2021 and then incorporated into the House version of HB 110 (the FY22-23 biennial budget bill) in April 2021. The final version of HB 110 included the main features of the Fair School Funding Plan, one of the most prominent of which is an inputs-based approach to determining adequacy which results in a base per pupil amount which can vary across districts based on the number of students and their distribution across grade levels. Fiscal year (FY) 2022 and FY 2023 represent years one and two of a planned six-year phase-in for the FSFP.

Without explaining all of the details of the Fair School Funding Plan’s approach to adequacy, it is important to note that the FSFP in both FY22 and FY23 is based on FY18 salaries and benefits for teachers, administrators and other staff included in the base cost calculation. Because of the timing of data availability, the salary data should lag two years behind the year of the funding formula, as is the case with the property valuations used in the calculation of the state and local share. This means that for FY22, FY20 salaries and benefits should have been used, and in FY23, FY21 salaries and benefits should have been used.

Updating Ohio’s funding model to reflect current input costs will increase the overall cost of the funding formula by raising the base cost per pupil at the heart of the model. However, there are two fundamental points on this issue that must be understood.  

1)    For the school funding formula to have any integrity as an accurate reflection of the adequate levels of school funding required in Ohio it must be based on current and appropriate data. 

2)    Annually updating the underlying data will not automatically lead to excessive cost increases for the state.

The first point above should be completely straightforward – it is just not defensible to claim that Ohio’s funding formula is adequate if the underpinnings of the formula are not the most current data available.

The second point requires a bit of explanation, however. Ohio’s funding formula can be thought of as having two main parts, with Part 1 being the formula amounts for the base cost and categoricals (adequacy) and Part 2 being the state/local share calculation which determines the share of funding for each district that should come from the state and the share which is expected to come from local resources (equity). Ohio’s current funding formula suffers from two deficiencies. First, as mentioned above, the adequacy aspect of the formula is constructed from out-of-date data. However, the second deficiency is that the state/local share calculation – which does appropriately rely on current data for both district property wealth and income levels – is updated each year whereas the adequacy parameters for the base cost and categoricals in the funding formula are the same in FY23 as in FY22.

According to the Ohio Legislative Service Commission (LSC), the overall state share of funding in FY22 was 42.2% which then fell to 39.8% in FY23. The reason for the decrease in the state share from Year 1 of the formula to Year 2 is that the property value and income data which determine the state share for each district were updated. The FY22 state share calculation is based on tax year 2018, 2019 and 2020 data and tax year 2017, 2018 and 2019 income data. The FY23 state share calculation is updated by one year and is based on tax year 2019, 2020 and 2021 data and tax year 2018, 2019 and 2020 income data. Because property values and income generally increase from one year to the next, updating the data makes districts appear richer to the funding formula and results in an increased local share of funding and a decreased state share of funding.

It is important at this point to clarify that updating the property value and income data each year is the correct thing to do. The problem is that updating the data on one side of the formula (the state/local share side) while not updating the data on the other side of the formula (the funding adequacy side) leads to an imbalance in the formula. The effect of this imbalance can be easily understood by looking at the Ohio Department of Education (ODE) School Finance Payment Report (SFPR) data for FY22 and FY23. ODE SFPR data shows that in FY22 the cost to the state of fully implementing the FSFP (i.e., no phase-in of the new formula) would have been $7.815 billion prior to the application of the guarantee. However, ODE SFPR data for FY23 shows that the cost to the state of fully implementing the new formula in FY23 was $7.442 billion (again prior to the guarantee). This means that the cost of the funding formula to the state will have decreased by $373 from FY22 to FY23 even though the adequacy parameters in the funding formula are exactly the same in each year. The reason for this is that the state share is lower in FY23 than in FY22. Furthermore, estimates by LSC show that the state share will continue to decline – to 37.4% in FY24 and to 35.5% in FY25 – as property value and income data is updated.

Again, the issue here is not that the state/local share calculation should not be updated annually, it is that the two sides of the funding formula must move in parallel with one another. Data must be as current as possible on both sides of the funding formula, and data must be updated on the same time frame on both sides of the funding formula. If these two steps are taken then whether the cost to the state goes up or down from one year to the next will depend on how input costs change as compared to the state and local share calculation. Furthermore, updating the data for all parts of the formula in parallel is the best way to ensure that long-term reliability of the formula is realized.

As a final point, concerns from some observers that annually updating the salaries used in the Fair School Funding Plan will necessarily lead to out-of-control cost increases can be easily seen to be completely unfounded. The most definitive way to demonstrate this point would be for LSC to simulate the FY22 funding formula using FY20 salaries and the FY23 funding formula using FY21 salaries. This would allow a calculation of both what it would cost to correct salaries to their appropriate year and also provide an indication of how the model functions over time.

Further indication of the fallacy of the “out-of-control costs” assertion comes from examining the cost of the Ohio House of Representatives FY24-25 school funding formula included in the House version of HB 33 (the FY24-25 budget bill). The House funding proposal updates salary data from 2018 in the FY22 and FY23 school years to salary date from 2022 in the FY24 and FY25 school years. The LSC comparison document indicates that the updated inputs would cost $8.05 billion in FY24 and $8.27 billion in FY25 (which reflect years three and four of the six-year phase-in). When compared to the FY23 cost of $7.40 billion for year two of the phase-in, the funding increases in the House budget are $652 million in FY24 and $217 million in FY25.

For context, the state of Ohio finished FY22 $3.775 billion in the black with General Revenue Fund (GRF) tax revenues $2.738 billion above estimate. Additionally, through the first nine months of FY23, the state is $1.775 billion in the black and GRF tax revenues are $805 million above estimated levels which were revised upward by $1.655 billion (6.3%) last July. The House budget also includes a state personal income tax cut which has been reported to be $1 billion. In light of these facts, it is difficult to argue that school funding increases of $652 million in FY24 and another $217 million in FY25 are not affordable to the state, especially when these increases would allow the state to plausibly claim that the school funding formula is taking important steps to reach compliance with the state Constitution.