Local Share Charge-Off Millage

The rate is uniform statewide, which is the point — it expresses a policy judgment that every district should contribute the yield of the same effort. inference But it is applied against valuation alone, and it presumes the district can actually levy at that effective rate. Under H.B. 920, a district whose reduction factors have driven effective operating millage below the charge-off rate is charged for revenue it is legally unable to collect at that level. verified The shortfall is phantom revenue, and it is a direct arithmetic consequence of two provisions that were written independently.

The parameter is retired for city, local and exempted village districts: the Fair School Funding Plan replaced their charge-off with a local capacity measure. verified It is not retired for joint vocational districts, whose local share the plan still computes as a half-mill charge-off — see the findings. The node also remains because it governs every state aid figure from FY1992 through FY2021, and because the historical series is required for any regime-diff between the two approaches.

What this repository computed # ↑ Contents

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

The series, sourced # ↑ Contents

rate        when                                    authority
20 mills    in force when DeRolph was filed, 1991   144 Ohio Laws, Part III, 3987, 4122
20.5 mills  raised during the litigation            Am.Sub.H.B. 152, Sec. 36.12,
                                                      145 Ohio Laws, Part III, 4432-4433
23 mills    in force at DeRolph I, March 1997       R.C. 3317.022
23 mills    FY2002-FY2007, by court order           DeRolph III, 93 Ohio St.3d 309, para 62
23 mills    FY2008                                  LSC, School Funding Complete Resource
22 mills    FY2010-11, on two bases                 LSC, H.B. 1 greenbook, 128th G.A.
none        FY2012-FY2021                           LSC, H.B. 59 greenbook, 130th G.A.
0.5 mill    JVSDs, throughout and still             LSC, H.B. 110 greenbook, 134th G.A.

verified DeRolph I at ¶97 recites the progression with its session-law citations; the LSC Complete Resource gives the rate and base still operative in FY2008; the last three rows are LSC’s own budget analyses, committed at lsc-education-greenbooks.txt and read by crates/project. Carried in crates/regime-diff.

The fourth era does not exist # ↑ Contents

The corpus held three of four eras and declined to guess the fourth. There is no fourth entry. The Evidence-Based Model kept a rate and forked its base; the Bridge formula replaced the rate with an index yielding a different implied rate per district, and LSC says so in as many words. The “20 mills applied to total valuation later” of secondary reporting is the statewide average of that spread. verified The caution was right for a better reason than the one it gave.

The rate stopped moving because a court stopped it # ↑ Contents

Three changes in six years, then nothing for the remaining life of the mechanism, is a shape that reads as a policy nobody revisited. It is not. DeRolph III ordered, at ¶62, that “in determining future biennial budgets through fiscal year 2007, the rate of millage charged off as the local share of base cost funding under divisions (A)(1) and (2) of R.C. 3317.022 may not be changed from twenty-three mills, irrespective of the language of R.C. 3317.012(D)(4) suggesting such a methodology.” verified For six of the years this parameter was flat, the General Assembly was enjoined from moving it.

The injunction was itself vacated fifteen months later by DeRolph IV, which erased DeRolph III entirely. verified So the rate held at 23 mills through FY2021 under a court order that had ceased to exist by the end of 2002 — the constraint outlived its authority, and no bill restored the methodology R.C. 3317.012(D)(4) had provided for. Why the General Assembly never returned to it is not established here. open

The base changed while the rate stood still # ↑ Contents

DeRolph I describes the charge-off as total taxable value times a percentage. By FY2008 the Legislative Service Commission describes the same 23 mills applied to recognized valuation. verified

valuation used in calculating a district’s local share of the base cost “recognizes” the district’s inflationary increase in carryover real property (property that was taxed in the year before) in the reappraisal or update year evenly over three-years instead of all at once.

recognized, reappraisal or update year = actual - 2/3 x inflationary increase
recognized, second year                = actual - 1/3 x inflationary increase
recognized, third year                 = actual

It is a three-year phase-in of reappraisal growth. H.B. 920 holds a district’s revenue flat by cutting the rate; recognized valuation holds its state aid flat by deferring the base. Two mechanisms answering the same event from opposite ends of the same multiplication — and LSC discusses them under one heading, “Provisions that Soften the Effect of H.B. 920 Tax Reduction Factors”, which is the likeliest way the conflation happened. verified LSC Complete Resource, pp. 19 and 82

The correction changes which districts are affected, not only by how much. Under the old reading the districts overcharged were those with long-bitten reduction factors: a wealth and tax-history story. Under the real rule they are the ones whose county reappraised recently. Those sets are close to unrelated, and the second is an administrative accident.

Computed rather than missing # ↑ Contents

The recorded blocker said tax-abstract was expected to supply recognized valuation and did not. Both halves of that were beside the point: the figure is not published per district by anyone, and it does not need to be. The rule is arithmetic, and Ohio’s staggered county calendar supplies the identification —

• the Department of Taxation publishes each county’s reappraisal and update year, and every one of the 88 has exactly one valuation event in TY2022–TY2024;
• Table SD-1, extended from two tax years to four, then gives each district one event year and two quiet ones, so its own ordinary growth measures what the reappraisal added.

The separation is not marginal: the median county’s real property rose 28.6% in its event year against 1.5% in its quiet ones, and all 88 counties’ largest jump falls in the year the calendar says. verified crates/regime-diff/src/recognized_valuation.rs and tests/recognized_valuation_against_the_abstract.rs

At TY2024 recognized valuation is 8.20% below total taxable value statewide — $34.5bn, or $793m of charge-off at 23 mills. It falls entirely by calendar: 12.97% of value for the 184 districts whose county revalued in TY2024, 7.04% for the 304 in TY2023, and nothing for the 123 in TY2022. verified

This is a high point of the cycle and the node should not be read as if it were a constant. LSC put the long-run effect nearer $125m a year, about 2% of valuation; TY2024 is four times that because Ohio’s 2023–24 revaluations were four times the size LSC’s own worked example assumes. What is not cyclical is the direction — total taxable value always overstates the charge-off. verified and the long-run figure is LSC’s

One component is still not modeled. LSC records an exempt property adjustment to recognized valuation for about 13 districts holding large state-owned exempt property — $836.4m in FY2008, $19.2m of local share. It needs the department’s own list of those districts. About 0.3% of statewide valuation, and absent. open

What the rate was doing to the system. At 23 mills against a statewide operating rate of 32.9 mills in TY2006, the base cost formula equalized about 69.9% of local operating revenue; a higher rate equalizes more and a lower rate less. verified LSC And the phantom revenue the mechanism generated was large enough to need its own program: the charge-off supplement, known as gap aid, reimbursed the difference and cost about $73.5 million across 145 districts in FY2008. verified LSC A mechanism requiring a supplement for a quarter of the state is the argument for replacing it.

The phantom revenue, counted — and it is not a minority case # ↑ Contents

The FY2008 comparison is the one worth holding beside it. Gap aid then covered 145 districts at $73.5 million. Sixteen years of reduction factors later the same rate against the same kind of base would reach twice as many districts, because that is what H.B. 920 does with time: every year of valuation growth above the floor moves another district further below any fixed assumed rate. The mechanism did not fail once; it was failing progressively, and the supplement’s cost was a lagging measure of it. inference

A further 65 districts have valuation high enough that 23 mills would exceed their entire base cost, leaving them no state aid at all — the charge-off had no minimum state share, which is a Fair School Funding Plan invention. verified

The comparison against the plan, on the corrected base # ↑ Contents

The incidence claim moved further than the level did. The corpus published that every wealth quintile does better under the plan, most of all the richest. Only the top quintile still gains: the bottom four lose by $118 to $365 per pupil, and the poorer the quintile the less it loses. The shape survives — the plan is genuinely more generous to property-rich districts than the charge-off was — but “every quintile gains” was an artifact of charging every district about 8% too much on the counterfactual side. verified crates/regime-diff/tests/the_charge_off_against_local_capacity.rs

A correction of this size from a definition is the argument for the node carrying the base beside the rate, which it did, and for it being right about what the base is, which it was not.

Joint vocational school districts were charged off at 0.5 mills, not the district rate. verified LSC This is the one place the “uniform statewide” description is false, and a comparison run over JVSDs at the district rate is wrong by a factor of forty-six.

What happened after FY2009, from a source the node had not considered # ↑ Contents

The statutory_basis below says the operative text for this era “must come from the opinions or from session law”, because Ohio Laws’ version archive for R.C. 3317.022 begins on 1 July 2014. There is a third route, committed here the whole time and never read: LSC’s education analysis of every enacted budget act from the 124th General Assembly through the 135th — 12 documents covering FY2002 through FY2025 without a gap. One test opens the file to check that each act the budget Catalog cites has a node; nothing had read a sentence of it. verified crates/project An analysis is not an act, and where the two differ the act governs; what an analysis gives that an act does not is a rate and a base stated in prose, which is the shape a parameter series needs.

The Evidence-Based Model kept the mechanism and forked its base. Districts whose class 1 effective current expense rate is 20.1 mills or lower — at or very near the twenty-mill floor — were charged 22 mills of total taxable valuation; every other district was charged 22 mills of recognized valuation. verified The node held the recognized-valuation base as a single fact about FY2008 and could not have anticipated a rule that assigns bases by floor status.

There was a second charge-off the series never held. Under the law the model replaced, districts contributed 23 mills toward base cost and up to 3.3 mills toward special education, career-technical education and transportation. verified The counterfactuals below are about base cost and are right to use 23 mills; what is understated is the total local contribution the mechanism asked for, which was up to 26.3 mills of the same valuation. inference

And the exempt-property adjustment has a stated rule. It is recorded below as open for want of the department’s list of affected districts. The list is still absent; the rule is not. Above 25% of potential taxable value exempted from taxation, the excess over 25% comes off the base before the charge-off is applied. verified

The Bridge formula has no charge-off rate, which is why none could be found # ↑ Contents

LSC, analyzing H.B. 59 of the 130th:

the state share index used for the opportunity grant does not result in a uniform charge-off rate. Rather, if a local share in FY 2014 were to be derived from the state share index and expressed as a charge-off of adjusted valuation, the charge-off rate would vary from 11.3 mills to 22.9 mills (excluding several outlier districts) with the statewide average charge-off being 20.6 mills.

verified Three things follow. The “20 mills applied to total valuation later” the node could not source is 20.6 mills and a statewide average, so entering it as a fourth row would state a distribution’s mean as a legislated rate. The spread is a factor of two across districts before outliers. And uniform — which this node’s own description calls the point of the parameter, the policy judgment that every district contributes the yield of the same effort — is exactly what LSC says the index does not produce.

The word leaves the analyses a biennium before the reader would expect. Counted across all twelve: 35, 25, 3, 3, then zero in the Evidence-Based Model’s own greenbook, then 11 in H.B. 153 — of which every one is “charge-off valuation”, the EBM’s valuation measure carried forward as an index, and not one a rate. H.B. 59 is the last analysis to write “charge-off rate” at all, and writes it to say the formula no longer has one. verified That is weak evidence about a mechanism and strong evidence about a search: the documents that would have carried the number stopped discussing the subject, and a reader looking for a fourth rate was looking for the wrong shape of thing. inference

For joint vocational districts it was never retired # ↑ Contents

The Fair School Funding Plan replaced the charge-off with local capacity for city, local and exempted village districts. It did not for JVSDs. LSC, analyzing the plan itself:

Instead of a per-pupil local capacity amount, the local share of the base cost for JVSDs is calculated by multiplying a 1/2 mill by the lesser of the district’s three-year average valuation or most recent valuation. This is consistent with the prior formula for JVSDs.

verified So the mechanism this node describes in the past tense is how a whole class of district was funded under the plan as enacted, and series below reaches further than “the observed range” suggests. What did change is the base — the Bridge formula took the three-year average, the plan takes the lesser of that and the most recent valuation, which is a guard against a district’s valuation falling. The same shape as everywhere else in this node: the rate holds and the thing it multiplies moves. verified