Assessed Valuation Per Pupil

It is the metric that makes equity empirical. The equity claim is that per-pupil resources should not be explained by local property wealth; this is the measure of local property wealth, and the association between it and a resource metric is the claim’s entire content. inference

Its limitations are the reason the Fair School Funding Plan does not rely on it alone. Assessed valuation in Ohio is 35 percent of market value for real property, so the figure is not market value and should not be described as such. verified More importantly, valuation measures property, not the capacity of residents to pay tax on it — a district whose housing has appreciated faster than its residents’ incomes appears wealthy by this measure and is not. inference That divergence is why the FSFP local capacity measure adds income, and it is why this metric should be reported alongside a median income figure rather than alone.

The denominator is also contested: enrollment, average daily membership, and formula ADM differ, and for districts with substantial community school or scholarship participation they differ materially. verified Which denominator is used changes the answer.

What this repository computed # ↑ Contents

Not what Ohio publishes. Figures derived here from committed fixtures, each one citing the test that reproduces it.

How much it changes the answer, measured. Two arms of the state publish this metric for every district and they do not agree. The Department of Taxation’s Table SD-1 and the Department of Education’s District Profile Report carry identical numerators — multiply the profile’s valuation per pupil by its enrolled ADM and SD-1’s total taxable value comes back exactly, on all 606 districts that join. The two agencies value the same property at the same dollar, so the whole gap between their per-pupil figures is the denominator and nothing else. verified crates/dispersion

They divide it by different pupil counts, on tax year 2023:

district        Education (enrolled ADM)   Taxation (SD-1)    ratio
Columbus City   43,019                     71,947             1.6725
Dayton City     11,744                     22,628             1.9267
Youngstown      4,322                       9,655             2.2337

Taxation counts the children who live in the district; Education counts the ones it teaches. The gap is community school, scholarship and open-enrollment-out students — so it is widest in exactly the districts where this metric does the most work in the aid formula, and it runs the same direction every time: the urban districts with the largest scholarship populations look poorer on Taxation’s count and wealthier on Education’s. Youngstown is $77,451 per pupil by one table and $172,999 by the other. verified

Only 63 of 606 districts have the two within 2% of each other. The statewide medians are $248,097 on the enrolled count against $233,320 on the resident one — six per cent apart where the districts themselves are two to one apart, because the divergence is concentrated rather than general. That makes it more dangerous and not less: a page comparing one table’s district figure against the other’s median looks right for most of the state and is wrong by a factor of two for the districts a reader is most likely to look up. This site did exactly that until the divergence was measured. verified

And this node did the year version of the same thing # ↑ Contents

The table and the statistics above were published as [verified] (TY2023) and were not all TY2023. Table SD-1 gained a fourth tax year; Columbus’s row was TY2024 and the other two rows were TY2023; the “78 of 606” and one of the two medians reproduce only on TY2024; and the numerator identity the whole finding rests on holds only on TY2023. Nothing was invented — every figure was somewhere in the fixture — but no two of them had to be from the same year, and none was bound to a figure that would have said so.

It is the trap the SD-1 catalog entry records as having caught four callers, and this node is the fifth. Everything above is now computed from one constant, dispersion::valuation::TAX_YEAR, and pinned. verified

There is a third count. Base cost ADM — funded rather than enrolled — is what base cost per pupil divides by, and it differs from enrolled ADM by a median of 1.6% and by 27% at the extreme. Any per-pupil subtraction that takes a local share from a cost is crossing that seam. verified

Where it crosses, and where it does not. The Fair School Funding Plan’s own subtraction does not: R.C. 3317.017 divides base cost per pupil and local capacity per pupil by the same count, so the seam is at the multiply rather than the subtract, and the pupil count cancels out of the dollars entirely. Where the denominator does reach money is the four categoricals R.C. 3317.022 multiplies by the state share percentage, which is a ratio and cancels nothing. The measurement is at the local capacity measure. verified crates/project

Statewide distribution, FY2023, 606 traditional districts verified: 5th percentile $129,894, median $248,097, 95th percentile $465,615. A district at the top of the range has about 3.6 times the taxable base per student of one at the bottom.

And a case that shows why this measure alone misleads. Northern Local — the DeRolph plaintiff district, the archetype of property poverty in Ohio school finance — reports $279,983 per pupil, above the statewide median and well above Cleveland Municipal’s $184,904. verified

Neither figure is wrong and neither district has changed character. Valuation per pupil is a ratio, and a rural district with agricultural and utility property divided among 2,194 pupils scores well on it while a city dividing a large base among 32,744 pupils does not. The measure captures taxable base per student and says nothing about whether residents can pay tax on it: Northern Local’s economically disadvantaged share is 38.8% and Cleveland’s is 100%. verified This is precisely the failure mode the FSFP local capacity measure was built to correct by adding resident income — and it is why a wealth-neutrality test run on this metric alone understates how much targeting Ohio’s formula does. inference

  TY2022   $197,326   over 610 districts
  TY2023   $233,320   over 609
  TY2024   $251,661   over 609