Toledo City School District

FY2025 in one line: 21,160 pupils, $440.2 million of operating expenditure, $20,805 per headcount pupil, a Performance Index of 60.4, and a Progress effect size of −0.02. verified computed; see crates/dispersion/tests/expenditure_functions_fy25.rs

Where its extra spending goes, which is not where the argument assumed # ↑ Contents

Toledo spends $6,172 more per pupil than Perrysburg and a markedly smaller share of it in a classroom: verified

                              Toledo     Perrysburg
instruction                    51.3%          62.6%
classroom instruction (CRI)    63.4%          73.1%
operations and maintenance     14.8%           7.7%
school administration           8.5%           4.8%
pupil support                   8.1%           8.5%
pupil transportation            1.6%           3.4%

The function-level gap is an order of magnitude more informative than the special-education share the circulating claim happened to name, and it comes from the same file that fact-check already cited. Plant and building administration are where an old, large, dispersed physical estate shows up in an operating budget; whether that is inefficiency or the cost of the buildings Toledo actually has is not a question a share can answer. inference

What this repository computed # ↑ Contents

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

It is the corpus’s clearest case of the denominator problem in the wild. The same operating total divided by weighted ADM gives $14,312 — which is what the report card prints as expenditure per equivalent pupil, and what OCG White Paper 013 analyzed. Divided by headcount it gives $20,805, which is what OCG fact-check RL-2026-021 reported three weeks later. A 45% spread for one district in one year, in two publications from one author, neither noting the other’s basis. verified Toledo’s weighted ADM is 30,758 against 21,160 enrolled — a weight ratio of 1.45, the consequence of a 98.9% economically disadvantaged share and a 21.9% disability share.

Normalized by the population actually served # ↑ Contents

It reports Toledo spending $2,766 per pupil on special-education instruction against Perrysburg’s $2,032–$2,363, and reads the difference as “consistent with a modestly heavier student-need profile.” Toledo’s disability share is 21.9% against Perrysburg’s 11.3% — nearly double, not modest. Normalized by the population actually served, Toledo spends roughly $12,600 per student with a disability against Perrysburg’s $18,000–$20,900: about a third less, not somewhat more. inference the special-education dollars are the fact-check’s, from audited statements this corpus does not hold; the shares are the report card’s, so this is an order-of-magnitude normalization rather than an audited per-pupil cost. A 1.94x population difference against a 1.36x spending difference leaves no room for the sign to return.

This is the corpus’s denominator rule arriving a third time and from a new direction. The first two were about dividing spending by a need-weighted pupil count; this one is about dividing special-education spending by a population that is mostly not receiving it. inference

The years before FY2024, which the survey holds and this node did not read # ↑ Contents

The Census F-33 Ohio panel runs from FY2009 and carries Toledo in every year it holds. verified crates/dispersion

Toledo’s state revenue per pupil lost 15.1% against prices between FY2010 and FY2024 — measured net of the community-school deduct in both years, which is the only way the survey’s state column is one quantity across that span. Its local revenue per pupil fell 8.14% in real terms, while its enrollment fell 21%. Both of Toledo’s revenue sources lost ground against prices for fourteen years. Beside it, Perrysburg lost 20.3% of its state revenue on the same basis and gained 22.3% of its local. verified crates/dispersion

The state figures here replace a published pair — Toledo 34.3%, Perrysburg 22.75% — that read the column as the survey prints it and so charged each district its own deduct. Correcting it halves Toledo’s fall and reverses the comparison: on one basis the wealthy district’s state revenue lost more ground against prices than the high-need one’s, and it is the local column, not the state one, that separates them. verified

The apparent step at FY2016 is the survey changing what it counts # ↑ Contents

The published series does not decline smoothly: it drops about 30% in a single year at FY2016 and never returns. That drop is not Ohio’s. From FY2016 the U.S. Census Bureau subtracts a district’s payments to community schools from the state revenue it credits the district with, and before FY2016 it does not. The Bureau’s own state notes give the reason — Ohio’s deduct puts the same state dollar in the community school’s revenue and in the resident district’s, and the subtraction removes the double count — and the note is absent from the FY2015 documentation and present from the FY2016 documentation. Ohio is the only state it is written for. verified crates/dispersion

The data locates the same year without the note. Correlating each district’s deduct per pupil against its change in state revenue per pupil, the deduct predicts the change at one transition in eleven — FY2015 to FY2016, at −0.2506 — and the largest reading in any other year is 0.1026; putting both years on one basis takes it to 0.0183. Statewide, districts’ state revenue falls $849.9 million across FY2016 and those same districts paid community schools $920.0 million that year, which is the same quantity seen from the other side. verified

So the tail was never there, and what is left is one mechanism rather than two # ↑ Contents

On one basis thirteen of Ohio’s fourteen largest districts gained across the step. Columbus goes from a published −29.22% to +14.93%, Toledo from −20.49% to +13.88%, Dayton from −3.50% to +32.14%. The 27 districts falling more than a fifth become 19, the pupils they hold go from 6.7% of the panel to 0.0166 of it, and the median district’s gain rises from 9.56% to 13.74%. Toledo’s own deduct was 35.3% of the state revenue the survey credited it with in FY2015, which is most of the 30% fall this node recorded. verified

What survives is capacity aid. Across 604 districts the step’s strongest correlate is still the log of a district’s total assessed value, at −0.2691, and the gradient is still monotone: the smallest-valuation fifth gains 26.79% against the largest fifth’s 10.60%, and now every fifth gains. FY2016 is H.B. 64’s first year, and LSC’s greenbook describes its new component in these words: capacity aid “targets funding to smaller districts with relatively low total property valuation”, “based on the amount a district can raise with one mill”, “provided to districts that raise less than the median amount”. One mill’s yield is a measure of total valuation, and capacity aid is the only Ohio component that prefers a district’s total valuation to its valuation per pupil — which is the comparison the correction leaves standing, standardized −0.1994 against −0.1204. verified crates/xcheck

It is not the only thing in the step, and the published version said it was. On one basis a disadvantage term reaches t = +3.92 where the published column had it at +1.38, third of three terms that all clear two. A year that raised capacity aid and the money that follows need should look like this; the published column could not show it because the adjustment had taken the largest deducts — which belong to the poorest districts — out of the outcome. verified

The published local share carries the same break in the other direction, because the adjustment reaches total revenue too. Ohio’s districts appear to go from 50.19% locally funded in FY2015 to 52.84% in FY2016; on one basis they go from 52.39% to 52.84%. Five sixths of a jump that could be read as a shift onto the property tax is the Bureau’s bookkeeping. verified

What removed a fifth to a third of the state’s contribution to Ohio’s high-poverty urban districts between FY2013 and FY2016 was the question this node left open. Nothing did. The tangible personal property phase-out this node first named is not in it either, on any basis: the step correlates +0.1175 with a district’s industrial property share, which is the wrong sign for that reading. verified

And that holds class by class, which the summed measure it was first taken on could not show. Table SD-1 carries business property in three columns — industrial, mineral and public utility — and they are three different populations, which the withdrawal below did not separate. Added together at TY2021 they give +0.0164, and that near-zero is opposite signs canceling rather than an absence: the step runs −0.0321 on mineral and −0.0159 on public utility against the positive industrial reading. Separated, no class separates its concentrated districts from the rest of the frame at two standard errors — the mineral one included, so the oil and gas districts that dominate the same three columns a decade later carry no FY2016 signature either. verified crates/xcheck

The equalization series was checked against the published step and is untouched by the correction as well: it cuts by quartile and takes an unweighted mean over about 152 districts, so the poorest quartile’s state revenue per pupil rises across FY2016 either way and the band Equity states holds. A quartile mean is one tiny district wide, and by the same arithmetic it is six enormous districts deep. verified

Three cautions on the series. The survey’s state revenue column is every dollar the state sent, capital included, so one district-year of it is not that district’s foundation aid — a district in a build year reports state revenue above its whole current spending. FY2014 is absent from the archive. And the deduct column that makes the correction possible is itself reported as a zero rather than as a missing value in FY2009 and FY2022, so the corrected series runs FY2010 to FY2024 and no earlier. verified

Not yet held: audited-statement figures, and the scholarship deduction that a district of this size and poverty carries. The community-school half of it is now held, as the survey’s V92 column — $86.0 million in FY2015 — but only in the aggregate the survey reports, not by receiving school. open

The district dates itself to May 8, 1849. Its 175th-anniversary timeline gives “May 8, 1849 Toledo Public Schools is established with the election of school board officers”, and counts the years “from its inception on May 8, 1849”. That is the district’s own account, published for the anniversary, and neither the election return nor the board’s first minute was read; the Auditor of State’s audits, which recite an organization date for some districts, give none for this one. inference