School funding analysis · Draft for discussion
A school authorized by the Charter School Institute in Park County RE-2, moving through the phase-in of Colorado's new school finance formula. This page shows what the formula pays in each year, which of the two formulas is setting the number, and why special education, not the at-risk weight, carries this school's gain.
01
Two formulas run at once until FY32, and the split between them moves every year.
Through FY32, every Colorado charter school is funded under two formulas at once. The 1994 formula, the one the state has used for thirty years, pays a district per-pupil rate plus at-risk and ELL add-ons. The new formula, created by HB24-1448 and rewritten by HB25-1320, pays a statewide base rate plus separate 25% weights for at-risk, multilingual and special education pupils, plus district factors for cost of living, size and locale.
The state is not switching over all at once. Each year it pays the 1994 amount plus a rising share of the difference between the two: 30% in FY27, then 45, 60, 75, 90, and 100% in FY32. That share is the phase-in.
Underneath the phase-in sits a hold-harmless floor, a promise that a school will not fall behind. The floor is not the same thing in every year, and section 05 sets out how much room this school has above it.
Share of the gap between the 1994 amount and the new-formula amount that is actually paid, by budget year. C.R.S. 22-54-103.3(3).
FY27. The floor is last year's dollars, whatever the school received in FY25-26 total program plus one percent. The floor itself is a fixed dollar amount that enrollment doesn't change. Enrollment still moves the formula amount, though, so growth can lift a school off the floor and a decline can put it on.
FY28 through FY31. The floor is recomputed live each year as that same year's 1994-formula amount plus one percent. It moves with enrollment, with demographics and with the district's factors, and floor status can flip from one year to the next.
FY32. No floor exists in statute. The phase-in provisions stop at 2030-31, and SB26-023 repealed the old 100%-of-district-PPR minimum that used to backstop charters. At full implementation a charter takes its new-formula calculation and nothing else. We read this as a drafting gap rather than a settled outcome, and treat it as a legislative watch item.
FY27 floor. C.R.S. 22-30.5-112(2)(c.5)(III). The district distributes the greater of (A) the amount of district total program received by the charter for the 2025-26 budget year plus one percent, or (B) the charter's phase-in amount under (2)(c.5)(I), which runs 22-54-103.3(3)(b) with four charter-level substitutions: pupil enrollment for funded pupil count, and the charter's own at-risk, ELL and special education counts. Subsection (2)(c.5) is repealed effective 1 July 2028.
FY28 to FY31 floor. C.R.S. 22-30.5-112(2)(c.7)(IV). Greater of (A) the phase-in amount for the applicable year under (c.7)(I), or (B) the charter's same-year 1994-formula amount under (c.7)(II) plus one percent of that amount. The 1994 leg carries only three substitutions: enrollment, at-risk and ELL. There is no special education substitution on that leg, which is worth a third of this school's gain and is section 04's subject.
FY32 forward. No floor. (c.7) stops at 2030-31, 22-54-103.3 is captioned "2025-26 through 2030-31 budget years, repeal", and SB26-023 §6 repealed the former (2)(c)(II) minimum.
The FY27 negative phase-in bar. 22-54-103.3(3)(b)(II), added by SB26-023 §16, stops the 30% step subtracting where the new-formula calculation sits below the 1994 amount. The model applies it; this school's new leg sits above its 1994 leg in every year, so it does not bite here.
Rural and locale add-ons sit outside the greater-of. (c.5)(II) and (c.7)(III) each open "in addition to the distribution described in" the greater-of subsection, so both are paid per pupil on top of whichever leg wins: a share of the $100,000 remote grant on the new leg, and the district's own rural funding per pupil on the 1994 leg. Both reach this school, and both are easy to miss in a distribution built from the greater-of alone.
In FY28 through FY31 both the floor and the phase-in are built from the same 1994 amount. Write N for the new-formula total, O for the 1994 total and p for the phase-in share. The floor is 1.01 O; the phase-in is O + p(N − O). The floor wins when N/O < 1 + 0.01/p. Enrollment scale cancels out entirely, and so does inflation. What decides floor status in those years is the ratio between the two formulas, a function of demographic mix and district factors, not of how many pupils are enrolled. The FY27 floor is the only one enrollment can push a school onto.
02
Gross total program, before any buyback, purchased service or administrative retention the district may apply. Enrollment is held flat after FY27 so the table isolates the formula's own effect. Base-rate inflation runs 2.0% a year, editable year by year below.
FY26 is the FY25-26 total program that serves as the FY27 hold-harmless base. It is a reference figure, not a modelled year, and it is a League statutory estimate rather than a verified district distribution: see check item 1. Amounts are gross total program, the figure the formula produces before any retention or fee arrangement with the district. Source: C.R.S. 22-54-103.5 and 22-54-104 as amended by SB26-023, on the counts entered in section 03.
The bar is total program as paid. The two lines are the underlying legs: the new formula above, the 1994 formula below. The gap between them is what the phase-in share is applied to, and it is narrow here, which is what section 05 turns on. The dashed line is the FY26 figure. Source: as above.
Per pupil, FY27 to FY32
Which formula is setting the number
03
Every count below is editable. FY27 holds the funding-basis counts; FY28 onward repeats them. Enter your own enrollment plan and the whole page follows.
Park County RE-2 · Authorized by the Charter School Institute · CDE district code 2610
Funded FTE, membership and the three weighted counts are the funding-basis counts from CDE's FY26-27 Charter Pass-Through Model. The funding basis and the membership basis are different things and both are correct: funded ELL runs below identified multilingual enrollment because of the five-year eligibility window in SB21-268, and the special education count is the funded count rather than every pupil on an IEP. Gifted-and-talented, K-3 headcount and READ percentages are not total-program inputs at all.
Counts read off a published file are used as filed, and the model does not rescale them for you when funded FTE is changed, because a rescale is an assumption and should be a visible one.
04
At 21.3% at-risk there is not much for the at-risk weight to work with. The gain comes from a category the 1994 formula never funded at all.
The 1994 formula funds no special education inside total program. Not a reduced amount, none: the leg substitutes only enrollment, at-risk and ELL. So the new formula's 25% weight on 11 funded special education pupils is a straight addition with nothing on the other side of the comparison.
The at-risk weight matters much less here. At 21.3% this school sits below the statewide average of 46.75%, and with 18 at-risk pupils the line is worth , against per pupil on the new formula and on the old one.
That makes the funded special education count the number to watch. It is the October count, it moves year to year, and here it carries more of the gain than any other single input.
The base rate moves the same way. Park County RE-2's cost-of-living factor of 0.224 lifts the new formula's base above the 1994 per-pupil rate of $12,739.67, so base funding adds to the gain rather than offsetting it.
Component-level comparison of the two formula legs in FY27, before the phase-in share is applied. Base and cost of living are combined because the 1994 formula folds cost of living into its per-pupil rate rather than paying it separately. Both legs are computed on the ASCENT-netted count. Source: C.R.S. 22-54-103.5 (new leg) and 22-54-104 (1994 leg).
One at-risk pupil, FY27
New formula against 1994. The gap is wide precisely because this school is below the statewide average and the 1994 formula never lifted it off the flat rate.
The base and cost-of-living line
What base funding adds, because Park County RE-2's cost-of-living factor of 0.224 lifts the new formula's base above the 1994 per-pupil rate.
The new formula carries a separate at-risk concentration factor, an extra 7% of the base rate on every at-risk pupil, and SB26-023 writes the test into statute at (2)(c.5)(I)(B) as three conditions that must all hold: the district's funded pupil count below 7,000, the district's at-risk rate at or above 70%, and the school's at-risk rate at or above 70%. Park County RE-2 has 573.5 funded pupils and a district at-risk rate of 40.45%, and this school is at 21.3%. The conditions fail, so the factor pays nothing and CDE's published tool reaches the same answer here. Nothing turns on the difference between the two readings at this school.
05
On the counts currently entered the hold-harmless floor does not set the number in any year, and the margin is not close.
Phase-in amount minus the hold-harmless floor. A positive number means the phase-in is paying. FY32 is omitted: no floor exists in that year. Source: 22-30.5-112(2)(c.5)(III) for FY27 and (2)(c.7)(IV) for FY28 to FY31.
New-formula total divided by 1994-formula total, against the threshold each year's phase-in share implies. Below the threshold the floor pays; above it the phase-in does. Because both sides scale with enrollment, this ratio does not move when enrollment does, and because both legs run on one inflation factor it does not move when the rate changes either. It moves only with demographic mix and the district's own factors. Source: 22-54-103.3(3)(c) and 22-30.5-112(2)(c.7)(IV).
06
The first two are the ones that could move the number.
The base used here is . The FY27 floor lands at against a phase-in of , so it would have to be understated by more than before it changed the FY27 answer, and it plays no part at all from FY28 because the floor is rebuilt from that year's own 1994 calculation. Worth confirming against the school's own gross receipt for the record.
It moves from 0.224 to 0.23 under the 2025 rebase, and this model holds that value flat past the biennium as a stated assumption. Here it is worth roughly a year, and it moves in the school's favour. A further rebase reaches this school directly, because the district is not at the cap.
This district is rural and remote, so both legs carry an add-on paid per pupil on top of whichever leg wins rather than folded into the comparison: a share of the $100,000 remote grant on the new leg, and the district's own rural funding per pupil on the 1994 leg. Confirm the district is passing both through, because they are easy to miss in a distribution built from the greater-of alone.
The authorizer's own published FY27 estimate for this school is . This model computes , a difference of . Where the two disagree the question is usually which rule was applied rather than which arithmetic is right, and the Reconciliation tab of the exported workbook shows the components side by side.
This page applies the statute as written. Ask the district which file it is running and on what basis before the distribution is set, rather than after. Where its reading differs from this one, the Reconciliation tab of the exported workbook prices the difference year by year.
The base is actual gross total program received, escalated once by one percent. Gross means before the district's administrative retention, and it excludes mill levy override revenue, purchased-services buybacks and any fees the district charged. A net figure has to be grossed up before it is used. Where a school's records and a modelled figure disagree, the school's own audited receipt is the better evidence, but the district has to agree to use it, so that conversation starts early rather than at true-up.
07
Every number on this page is one of four kinds. If you want to argue with a figure, the kind tells you who to argue with.
Carried forward from this school's earlier analysis page. Anything marked Input is editable in section 03 and in the yellow cells of the exported workbook.
08
It applies the statutory calculation to one school rather than reimplementing a bespoke engine. Inputs are the FY26-27 funding-basis counts from CDE's Charter Pass-Through Model for school code 3947, together with Park County RE-2's district row. Both legs are computed from published inputs: the new formula from HB25-1320 as amended by SB26-023, the 1994 leg from C.R.S. 22-54-104 including the at-risk banding and the minimum-per-pupil test. Phase-in shares come from 22-54-103.3(3). The hold-harmless floor follows the year-specific rules in 22-30.5-112(2)(c.5) and (2)(c.7), computed per year rather than escalated forward.
FY28 forward inflates the base rate, the 1994 per-pupil rate and the minimum per pupil on a single shared cumulative factor anchored to FY27's appropriated base of $8,900.40. The default is 2.0% a year, editable in section 02. Because one factor drives both legs, the ratio between the two formulas, which decides hold-harmless status from FY28 on, does not move when the rate does. The cost-of-living factor is 0.224 in FY27 and 0.23 from FY28, held flat past the biennium as a stated assumption. The rural and remote add-ons are paid on top of the greater-of on both legs.
The engine was gated against the figures this school's earlier analysis page published: every dollar cell and every binding label reproduced exactly, as part of a run covering 420 cells across the single-site portfolio with no differences.
The engine was then reproduced independently in a second language and the two implementations agree to the cent across both formula legs, the phase-in, the floor test and the totals, in all six years.
The exported workbook was opened, recalculated and compared cell by cell against this page: zero formula errors and agreement to the cent. It was then edited to confirm it is live, that every downstream figure moves, and that the published-versus-live check on its Read me tab flags the change. The model was also run on a second, deliberately different year window to confirm nothing is pinned to a six-year FY27 to FY32 shape.
1 · Enrolled bill text. HB24-1448, HB25-1320 (signed 23 May 2025), SB26-023 (FY27 school finance act) and the codified sections at C.R.S. 22-54-103.3, 22-54-103.5, 22-54-104 and 22-30.5-112.
2 · CDE FY26-27 SB26-023 Charter Pass-Through Model, published 23 July 2026. Source for this school's funded pupil count, membership and weighted counts. Authoritative for what a district is likely to distribute, though not for what is correct.
3 · CDE District Funding Calculation Worksheets (FY2025-26 January update; FY2026-27 signed 28 May 2026) and the District Runs Memo of 13 May 2026. Park County RE-2's per-pupil rate, funded count, at-risk rate, cost-of-living factors, and size and locale designations.
4 · 2025 cost-of-living study, as carried in VAO's FY28-forward factor table. Source for the FY28 rebase from 0.224 to 0.23.
5 · CSI PPR Estimates FY27, July 2026 update. The authorizer's own published estimate for this school, with the line-by-line breakdown of both formula legs behind it.
6 · Correspondence with CDE and CSI. Agency guidance on the treatment of the hold-harmless base, on flat-rate pupil netting, and on defects in CDE's published tool. Primary as to intent, secondary as to numbers.
6 · Colorado League of Charter Schools template. Source of the FY25-26 total program figure carried as the FY27 hold-harmless base, and of the at-risk count this page has replaced. Context and cross-reference only, never a funding baseline, which is why the CDE at-risk count is used and why the FY25-26 figure is flagged as unverified.