The corpus › Formula Parameter

Transportation Cost Rates

$1,337.175 per rider and $6.867 per mile — the two rates in the plan that nobody chose. The section names no dollar figure: both are trimmed means of what districts themselves reported spending last year, so the rate moves when districts' own behavior moves.

parameter/transportation-cost-rates · 2 nodes point here

Every district’s school bus payment is the greater of two bases: weighted ridership times a statewide cost per student, or bus miles times a statewide cost per mile. In the FY2027 model those are $1,337.175 and $6.867. verified

R.C. 3317.0212 contains no dollar figure at all. Division (C) tells the department to divide each district’s total school bus costs in the previous fiscal year by its qualifying ridership, then — after excluding districts that provide no bus service, the ten with the highest cost per student and the ten with the lowest — divide the remaining aggregate cost by the remaining aggregate ridership. Division (D) does the same per mile. Both rates are trimmed means of what districts reported spending. verified the section

So these are endogenous. A year in which districts spend more on buses raises the rate they are paid at the following year, and the twenty-district trim is the only thing bounding that. Nobody decides them and no bill is required for them to move. verified the mechanism inference the feedback

The rider weights are the General Assembly’s, and the mass-transit factors are the department’s. A community or STEM school rider counts 1.5 and a nonpublic rider 2.0, on both the rider base and the mile base — division (E)(1)(a) and (b) state both. Transport by means other than a school bus falls under division (G), which pays “in accordance with rules adopted by the department” and states no factor, so the 0.35 for mass transit and 0.50 for other types are the department’s rules rather than the statute’s. verified

What this repository computed Contents

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

What the state pays to carry children to private and charter schools Contents

$82.2m of the $608.9m school bus payment, 13.5%. The mile base carries more of it than the rider base does — $49.6m against $32.6m — which is the reverse of what the rider counts suggest, and the reason is a column that was not committed until now.

The department’s reported-inputs sheet splits miles by school type as well as riders, and [e] is a weighted total exactly as [b] is: public + 2 × non-public + 1.5 × community, on all 611 districts. The parts are not proportional to the riders. A non-public rider takes 1.72× the miles of a public one and a community or STEM rider 1.76× — these children are dispersed, so carrying one goes further — and the statute then weights those miles again. The effective multiplier on the mile base is the product, roughly 3.4× a public rider, which no source states.

So the three shares differ and are not interchangeable: non-district riders are 5.82% of heads, 10.42% of weighted ridership, and 16.70% of weighted miles. verified the FY2027 model, tested in crates/project

And the fuel arithmetic is not the payment arithmetic. Miles are what burns diesel, so non-district riders account for 9.64% of the fuel and 13.5% of the payment; the weights are what open the gap. inference

The two rates disagree about which way costs went Contents

Across the one interval the corpus can observe, the per-rider rate rose 4.88% and the per-mile rate fell 2.73% — $1,275.000 to $1,337.175, and $7.060 to $6.867. They are computed from the same districts’ returns for the same year under the same trimming rule.

Between the two cost years they are drawn from, Midwest diesel rose 17.5%. The rider rate rose about a quarter as fast and the mile rate moved against it, which is evidence that the mile base does not track fuel in any direct way — and the mile base is what 350 districts and $308.8m of school bus payment are actually paid on. A per-mile rate is a ratio whose denominator moves too, so districts driving further push the statewide figure down while their own costs rise; that is the obvious mechanism and two observations cannot establish it. verified the rates — the FY2026 model and the FY2027 one, tested in crates/project inference the mechanism

The 17.5% is the July-to-June mean of the monthly series, and the choice is worth half the figure: the June-point alignment of the same monthly series over those two years gives +39.6%, under which the rider rate rose about an eighth as fast as fuel rather than a quarter. The convention is the right one — fuel is bought all year, so a flow is averaged where an index is read at a point — but a comparison that rests on it should say so. verified crates/dispersion and crates/connect

Fuel is a cost the rates cannot be shown to track Contents

The rates are trimmed means of what districts spent, so the obvious question is how much of a fuel movement reaches them. Fifteen years of the F-33’s transportation column cannot answer it, and that is a measurement rather than a gap. Real per-pupil transportation spending was regressed on the Midwest diesel series, collapsed to fiscal years, under eight specifications — two alignments, the thirteen NCES years against all fifteen, with a linear trend and without — and none reaches |t| = 1.5. The slope does not even hold its sign: over all fifteen years with a trend it is -$2.49 per dollar-per-gallon under the June point and +$9.66 under the mean.

What the panel does support is a trend, significant in every specification that includes one at t = 2.1 to 3.6: real per-pupil transportation drifts up about $4 to $5 a year whatever diesel does. A pass-through estimated here would be reporting that drift.

This is not evidence that districts are indifferent to fuel, and the bias runs the other way — every slope is fitted to retail diesel, and Ohio refunds districts the 47-cent motor fuel tax, so a district’s bill moves further than the series does. The true coefficient is larger than any of these and still not identified. It is a statement about what fifteen annual observations of a survey can carry, and it is why the pass-through is carried here as unquantified rather than as a number with a wide interval. verified crates/dispersion

The appropriation began binding, and the year is known Contents

Special education transportation prorated at 1.0 in FY2026 and 0.917459740976215 in FY2027. It is the first year the line was cut to fit. The department’s own FY2026 explanation had added the same constraint to the density supplement a year earlier — “prorated to remain within amounts provided in the biennial budget” — so the pressure reached two transportation lines in two consecutive years. verified the two models and the explanations

The state share floor moved twice, and it is a floor on a floor Contents

The school bus payment is multiplied by the greater of the district’s own state share percentage and a fixed minimum: 45.83% in FY2026 and 50% in FY2027, both written into division (E)(1)(c). H.B. 96’s own summary records the sequence as 41.67% → 45.83% → 50%. verified the section for the two current values

That is a different floor from the minimum state share’s 10%, five times higher, and it binds far more widely. The committed 609-district panel puts 438 on the transportation floor against 138 on the formula’s; the department’s own 611 give 440, and both are right about their own population. Transportation is therefore the component where the state pays the largest share of cost for the largest number of districts, and it is usually discussed as a small line. verified crates/project

Two constants in the calculator are not in the section Contents

The per-mile base is bus miles × $6.867 × 180, and 180 does not appear anywhere in R.C. 3317.0212. The section speaks of “miles driven for school bus service as reported”, with no annualization.

It is an annualization, and the department says so in five consecutive editions of its own line-by-line explanation. The method is to “calculate the statewide average annual per-mile expenditure for each district based on 180 days of service”, and “the mileage is based on the average number of miles traveled during the rider count week”. Reported miles are a daily figure — one week’s average, taken in the first full week of October — and 180 is the school year it is multiplied out over. So the multiplier is a convention of the department’s rather than a constant of the statute’s, which is what this node suspected and could not show. verified the department’s own explanation, FY2022 through FY2026, tested in crates/project

The 0.91746 factor on special education transportation is likewise absent from the section, and behaves like the enrollment and performance supplements: alive in the calculator and not in permanent law. verified that it is absent inference the parallel

The transportation guarantee is not. The line that holds 38 districts at their FY2021 payment was recorded here as absent from permanent law on the same ground, and it is one section away, at R.C. 3317.019(A)(2) — which the corpus has held in its committed statute extract the whole time. The search missed it because the statute does not call it a guarantee: it is “temporary transitional transportation aid”, and the word guarantee appears nowhere in the section. verified crates/project

What it pays is the FY2020 amount under H.B. 166’s Section 265.220(A)(2) — “prior to any funding reductions authorized by Executive Order 2020-19D” — less the district’s FY2019 payment under a division of R.C. 3314.091 repealed in 2021, less its current-year payment under R.C. 3317.0212, floored at zero. verified Three subtractions reaching three years, two of them through provisions that no longer exist.

The calculator labels the line FY2021 and the statute anchors it to FY2020. Under H.B. 166 both years were held at FY2019, which would make the two labels the same number; the corpus holds no per-district figure that would confirm it. inference

This closes an open question the component carried Contents

fsfp-transportation recorded that “whether the rider weights and the guarantee are statutory has not been separated out here”. They separate: the weights are statutory, in division (E)(1); the guarantee is not in the section.

What the floor is worth, and which districts it pays Contents

The floor is not a detail at the edge of the program. Paying every district its own state share instead would cost $289,611,593.21 less — two fifths of Ohio’s transportation aid is paid by the floor rather than by the share. verified crates/project

And it runs up the wealth distribution. A floor binds where a district’s own share is low, and a low state share is what high local capacity produces — so the payment is monotone in wealth: $3.18 per pupil in the least wealthy fifth of districts by assessed valuation against $405.92 in the wealthiest, and every district in the top two quintiles sits on it against twelve of 120 in the bottom. verified crates/project

That is the opposite shape from the rest of the formula, and it is the fact any proposal to move this floor needs stated beside it — raising it pays the districts with the most local capacity, and lowering it falls on them.

Properties Contents

NameTransportation cost rates
Unitdollars-per-pupil
KindMeasured — and this parameter is the corpus's clearest example of the category.

The two rates are statistics. Division (C) and (D) specify a computation over reported data and the value falls out of it, so the answer to "what would have to happen for this to change" is: districts spend differently on buses. No act, no departmental decision, no negotiation.

The rest of the section is not measured, and the mixture is the point. The rider weights, the efficiency thresholds, the density supplement's 28 and 0.55, and the state share floor are legislated. The mass transit and other-types factors are delegated — division (G) hands them to departmental rule. The 180-day multiplier and the FY2021 guarantee are in neither and behave as uncodified.

Four kinds in one section, which is what the four kinds of parameter argued would happen and what a single kind enum would have flattened.

And, like almost every other section in the plan, it stops: eight divisions read "for fiscal year 2028 and each fiscal year thereafter" and hand the value or the definition back to the General Assembly. Division (E)(2) does it for the base payment itself.
Values over time
FY2027 rates, from the department's model:
  $1,337.175   statewide transportation cost per student   [measured, R.C. 3317.0212(C)]
  $6.867       statewide transportation cost per mile      [measured, R.C. 3317.0212(D)]
State share floor on the school bus payment, from R.C. 3317.0212(E)(1)(c):
  FY2026   45.83%   "forty-five and eighty-three hundredths per cent"   verified
  FY2027   50%      "fifty per cent"                                    verified
  FY2028   —        an amount determined by the general assembly        verified
H.B. 96 raised it 41.67% → 45.83% → 50%; the FY2025 value is carried by Am. Sub. H.B. 96 (2025) — FY2026-27 Budget and not read from the section here. inference

Legislated constants alongside them, all FY2026-27:
  1.5    community and STEM riders and their miles      (E)(1)(a)(ii), (b)(ii)
  2.0    nonpublic riders and their miles               (E)(1)(a)(iii), (b)(iii)
  1.5    efficiency index at which the supplement caps  (F)(3)(a)
  0.15   efficiency supplement rate                     (F)(3)(a)
  28     rider density the supplement tapers from       (H)(1)(a)
  0.55   density supplement rate                        (H)(2)
Departmental rule, not in the section:
  0.35   mass transit riders                            (G), by rule
  0.50   other types of transport                       (G), by rule
Statutory basisR.C. 3317.0212, read. (C) and (D) define the two rates as trimmed means; (E) assembles the base payment and states the rider weights and the state share floor; (F) the efficiency adjustment; (G) hands non-bus transport to departmental rule; (H) the density supplement. verified ohio-laws
Simulation keyproject::panel::supplements::TRANSPORT_PER_RIDER and TRANSPORT_PER_MILE, with TRANSPORT_SCHOOL_DAYS, TRANSPORT_NONPUBLIC_WEIGHT and TRANSPORT_COMMUNITY_WEIGHT beside them.

The rates are constants, not levers. The floor beside them is one — Policy::transportation_floor, the minimum state share the department applies to the calculated amount whatever the district's own share, which H.B. 96 moved 41.67% → 45.83% → 50% inside one act. It is the only lever in this model that touches nothing in core foundation funding: it moves Outcome::transportation, which is carried beside realized aid rather than inside it, because the department publishes transportation as its own line.
Written in formulas as
  • per rider base
  • per mile base
Sensitivity$726.1m of transportation and $182.6m of special education transportation. 350 of 611 districts are paid on the mile base rather than the rider base, so the two rates do not move the same districts: raising the per-rider rate reaches the 261 on that base and nothing else, until a district crosses over. verified

The max of the two bases makes the response to either rate piecewise. A district near the crossover is insensitive to an increase in the rate it is not paid on, and then abruptly sensitive. A fiscal note that scales both rates together will not show that. inference

The measured nature is the sensitivity that has no lever. If districts' reported bus costs rise 10%, the rates rise with them the following year and the state's bill rises without anything having been decided — bounded only by the twenty-district trim and by the appropriation, which is what the proration factor exists to handle. inference

What this node does not hold Contents

whether the calculator's FY2021 label and the statute's FY2020 anchor are one number — The guarantee itself is no longer unfilled — it is R.C. 3317.019(A)(2), read and pinned, and the department's own line-by-line explanation now states the same two terms in five consecutive editions, glossing them as FY2020 every time and never as FY2021. That is a second definition rather than a settlement. Under H.B. 166 both years were held at FY2019, so the column heading may name the same dollar the section reaches back for, and confirming it still needs a per-district FY2020 or FY2021 transportation figure that no committed source carries.

the rates before FY2026 — FY2026's are now committed, from an archived copy of that year's model — $1,275.000 per rider and $7.060 per mile. Earlier years are not recoverable. The department replaces the calculator rather than archiving it; the Internet Archive holds FY2025 only as a file truncated at exactly one mebibyte, and FY2022 to FY2024 not at all. The line-by-line explanations give the method for every one of those years and the rates for none of them, so what is missing is four values that no surviving document states.