The Board has identified $12 million in reductions as the FY27 target to restore structural balance and rebuild fund balance to 15%. The deficit is not a Covid hangover alone — it is the cumulative result of four structural shifts visible in the audited financials. Here is what the Board's own presentations and the audited ACFRs show.
The FY25-FY26 General Fund budget has been amended twice. Each amendment has expanded expenditures and reduced the projected fund balance percentage from the 15.05% originally adopted by the Board in June 2025 to 13.59% under March 2026 Amendment #2.1
The March 17, 2026 Amendment #2 presentation states the targets explicitly: structural balance and a 15% fund balance. To get there, the Board has set a $12 million reduction target for FY27, with cuts spread across four categories: salaries & benefits, purchased services, departmental adjustments, and discretionary spending.1
The Board's January 13, 2026 Budget Planning workshop offers important state-funding context for why the FY26 amendments have been needed:2
The March 17 presentation also documents the Board's formal Budget Reduction Rubric & Scoring system — a six-criteria weighted framework (Impact on Student Learning, Equity Impact, Legal/Compliance Risk, Cost Savings Amount, Feasibility/Timeline, Stakeholder Support) used to rank candidate reductions as Tier 1 (Strong Candidate), Tier 2 (Consider), or Tier 3 (High Risk).1
Areas under Board consideration for reductions:1
Final RIF targets and FY26-27 budget adoption come at the June 2026 Regular Meeting.1
Cost growth at TSD has been steady and structural, not Covid-shock-and-recover. The FY25 Annual Comprehensive Financial Report — audited by Plante Moran and accepted by the Board on November 18, 2025 — contains a ten-year statistical section covering FY16-FY25; combined with the prior-year FY24 ACFR it lets us see the full arc from FY15 forward.1
The single cleanest measure is operating cost per student. It grew from $10,924 in FY15 to $15,624 in FY25 — a 43% increase over ten years, with the steepest acceleration FY20→FY25 (about 5.4% per year compounded over the latter half).1 Enrollment over the same span moved 12,563 → 12,393 (-1.4%), but Covid took out a layer of about 680 students between FY20's peak (13,073) and FY22 (12,519) that has not returned.1
The FY25 audited results also show a notable one-year shift in compensation: average teacher salary jumped from $78,145 in FY24 to $83,536 in FY25 — a +6.9% single-year increase, well above the prior trend (~1%/yr).1 Over five years this metric had moved only $3,703 (FY19 → FY24); in one year it added another $5,391. That single-year change is large enough to be a real cost-driver going forward.
Total expenditures across all governmental funds grew from $181 million in FY15 to $308 million in FY25 — up $127 million, or 70% over ten years.1 That total is broader than what's usually called "the budget" in board meetings: in FY25 it bundles the General Fund ($193.6M actual) with the Capital Projects Fund ($57.9M, mostly 2023 bond construction draws) and the Nonmajor Governmental Funds ($56.9M — debt service, food service, athletics, community service). When the rest of this analysis discusses "the operating budget" it means the General Fund. The FY24 → FY25 jump in the all-funds total ($269M → $308M) is dominated by bond-funded capital outlay, not operating cost growth; the General Fund moved much less ($191.5M → $193.6M).
The General Fund itself grew from $185.0M in FY22 to $193.6M in FY25, the period when the post-Covid Edustaff buildup, MESSA premium inflation, the MPSERS UAAL stabilization rate true-up, and the FY25 teacher-salary step all landed in sequence.1
The single most consequential line in TSD's budget that very few residents know about: the General Fund subsidizes Special Education by about $9 million a year, and that subsidy has grown $2.7M since FY21.1
Per the Special Education Update presented to the board on October 7, 2025, TSD now spends $34.0M annually on special education programs and services — for 1,367 students with IEPs.1 State and federal sources (the 51c foundation, PA-18 county millage, IDEA flow-through, and Medicaid reimbursement) cover only $25.0M of that.1 The remaining $9.0M comes from operating dollars that would otherwise go to general instruction.1
TSD is not unusual in this. According to a state-commissioned report released in October 2025, Michigan reimburses only 28.6% of district SpEd costs — among the lowest rates in the country.2 Every Michigan district fills the gap from its operating budget. The proposed fix is a six-year transition to a tiered-funding system at an estimated cost of $4.55 billion statewide.2 Until that passes, the $9M annual subsidy in Troy is a feature of the system, not an anomaly.
"The district has the desire and ability to serve students with severe needs in-district with its broad range of supports and services… very few students are required to attend out of district programs." — New Solutions K12 (Levenson) special-education review, December 2023, Commendation #53
That December 2023 outside review — commissioned by TSD and authored by Nathan Levenson's firm New Solutions K12 — actually recommended fewer paraprofessionals over time, achieved through "natural attrition."3 The check register tells the opposite story: the district has added approximately 100 contracted Health Care Aides since FY22, sourced through Edustaff.3 The trajectory of those costs is the next chart.
Section 3 introduced the General Fund subsidy to Special Education. The single largest new line within that program is the contracted Health Care Aide (HCA) layer — third-party staff supplied through Edustaff, billed against object code 3116 in TSD's accounting system. From FY22 to FY26, this single line grew from effectively zero to $2.74M per year.1
FY22 (July 2021 - June 2022). The HCA program was effectively non-existent — total Edustaff HCA charges of $1,586 for the year.1
FY23 (July 2022 - June 2023). HCA spending jumped to $1.33M as the contracted-aide model was implemented at scale, distributed across approximately a dozen distinct GL lines (ASD, AI, EI, ESY, MICI, MOCI, etc.).1
December 2023 (mid-FY24). Nathan Levenson's firm, New Solutions K12, delivered the report "Improving Outcomes and Equity for Students with Disabilities and Other Students who Struggle" — a 32-page review based on 130+ stakeholder interviews and commissioned by the district itself. Page 11 explicitly recommended that the paraprofessional layer should shrink over time: "fewer paraprofessionals or student support aides will be needed," achieved through natural attrition rather than layoffs.2
FY24 (the year Levenson delivered). HCA spending grew to $2.57M — up another $1.24M over FY23, in the same year the recommendation was received.1
FY25 (July 2024 - June 2025). HCA spending reached $2.74M, a further $170K increase.1
FY26 (in progress). Year-to-date HCA spending of $1.94M as of the source file's generation (approximately mid-FY26). The full-year figure will not be available until the FY26 ACFR.1
Two important caveats:
Normalized against IEP enrollment (the FY25 figure of 1,367 students with IEPs from Section 3), the HCA layer alone runs about $2,000 per IEP student per year.1 That is on top of the in-house aide layer, the teacher cost, and the other SpEd program costs that together produce the $9M general-fund subsidy described in Section 3.
One way to understand the deficit is to ask which categories of district spending have grown faster than others. Over the seven years from FY18 (the proper pre-Covid baseline) to FY25 (the most recent audited year), four categories stand out — and the FY25 audit adds two important nuances to the FY24 picture: an OPEB actuarial recovery that lowered required retiree-healthcare contributions sharply, and a one-time MPSERS rate step-down.
Retiree benefits (MPSERS pension + OPEB retiree healthcare) added the most absolute dollars at the FY23-FY24 peak — about $16-21M per year above the FY18 baseline — but the FY25 audit shows a meaningful recession: combined retiree benefits ran $40.3M in FY25, down from $43.9M in FY24 and $48.8M peak in FY23.1 MPSERS alone grew from 30.3% of payroll in FY18 to 41.4% in FY24, then eased to 40.4% in FY25.1 Most of the MPSERS dollars are now state pass-through under Section 147c, but they inflate the expense base regardless. OPEB recognized a one-time accounting recovery in FY25 (statutorily required contribution dropped from $7.26M to $2.77M) as actuarial assumptions revalued the obligation favorably — a real cash-flow easement but largely non-recurring.1
Active employee health insurance (MESSA) added $4.6M FY18→FY25, growing at about 5.1% per year compounded over seven years — and there is no near-term sign of decline; healthcare inflation continues.2
Outsourced services (the combination of Edustaff contract staffing, DM Burr custodial, and First Group transportation) added $4.2M FY18-FY25 and continued growing in FY25 — though if you start the clock at the FY19 vendor-consolidation trough rather than FY18, outsourcing grew at 12.1% CAGR through FY25.2 The choice of baseline matters; both views are honest.
Salaries (covered payroll) grew from $73.15M in FY18 to $92.92M in FY25 — a 3.5% CAGR over seven years.1 What changed in FY25 is the per-person figure: average teacher salary went from $74,442 (FY19) to $78,145 (FY24) — about 5% over five years, well under inflation — then jumped to $83,536 in FY25, a single-year +6.9% increase.1 The "teacher pay is breaking the budget" framing was not supported by the FY18-FY24 data, but the FY25 step is large enough to change the trajectory going forward.
Between FY20 and FY24, the federal government sent Troy School District a one-time stream of pandemic-relief money — the Elementary and Secondary School Emergency Relief (ESSER) fund. Total federal revenue ran $15M to $21M per year during the peak ESSER years, well above the ~$6.5M pre-pandemic baseline.1 Stacked end-to-end, that's about $40M of one-time money above the long-run trend.1
ESSER was allowed to fund "extended learning," staff, materials, and facilities. Districts had until September 30, 2024 to obligate ESSER III dollars. The FY25 ACFR shows federal revenue settled to $8.7M — about $2M above the pre-pandemic $6.5M baseline, reflecting tail-end ESSER III draws and ongoing non-ESSER federal grants, but well off the peak. The cliff is real. Any positions or programs ESSER had been paying for either ended, or transferred to the operating budget.
What did ESSER actually fund? The popular framing — "ESSER hired a layer of staff that's now stuck on the operating budget" — does not hold up well against the audited Instruction FTE series. The count was already climbing pre-Covid (840 in FY18 → 887 in FY19 → 925 in FY20); the FY20 peak occurred when only ~$0.3M of ESSER had been deployed.1 During the biggest ESSER years (FY21-FY24), Instruction FTE was actually below that FY20 peak — FY21 dropped 63 positions to 862 (pandemic attrition, not ESSER hiring), then partially recovered to 879 by FY24 and 869 in FY25. Net positions still on the books above the FY18 pre-Covid baseline: +29 in FY25, not +85.1
Federal program rules allowed ESSER dollars to fund a wide mix: staff and programs, materials, technology, summer/extended learning, mental-health supports, HVAC and facility work, assessments. Single audits report ESSER expenditures by federal CFDA program code, not by position. We cannot, from public records, claim the $40M cushion specifically funded a staffing layer. What we know with certainty is that the FY20-FY24 period saw both the temporary federal cushion and the structural cost increases that the FY26 budget now has to absorb without that cushion.
The pattern is a familiar one in school finance: temporary federal aid gets absorbed into recurring commitments — staff, programs, contracts — faster than those commitments can be unwound when the aid ends. Reconciling that gap, position by position and line by line, is the substance of the FY26 budget conversation now in front of the board.
Michigan's Public School Employees Retirement System (MPSERS) is the second-largest cost line in TSD's general fund — $37.5M in the FY25 audit and roughly $29.3M in the FY26 projection.1 The contribution rate is set in Lansing, not Troy. As a share of covered payroll it grew from 17.6% in FY15 to a peak of 48.6% in FY23, then eased to 41.4% in FY24 and 40.4% in FY25.1 Roughly $2.1M of the FY25 figure was a one-time Section 147c(2) contribution funded by the state and remitted directly to the System.1
Starting in FY22, Michigan shifted the funding of MPSERS UAAL (unfunded actuarial accrued liability) toward direct state payment via Section 147c of the State School Aid Act. The dollars still show up on TSD's books — but they also show up as revenue, dollar-for-dollar.2
Per the FY25 ACFR Management's Discussion and Analysis: in FY25 the district's actual pension contributions included "an allocation of $9,297,729 in revenue received from the State of Michigan and remitted to the System to fund the MPSERS unfunded actuarial accrued liability (UAAL) stabilization rate" plus the $2,113,411 one-time Section 147c(2) contribution.2 That's about $11.4M of pass-through revenue in FY25, all routed specifically to absorb pension expense. Revenue-neutral, expense-inflating.
The implication: when board materials show revenue growth into FY25, a substantial slice of it is MPSERS pass-through, perfectly matched by corresponding expense growth. The free operating revenue — money the district can actually choose how to spend — grew much less.2
The check register lets us watch non-payroll spending vendor-by-vendor. Comparing FY19 to FY25, just six vendors account for +$14.3M of recurring annual cost growth.1 That's roughly two-thirds of the total non-payroll cost growth in the operating budget over the same period.
The pattern is mixed-cause:
Note what's not in this list: there is no "we spent too much on technology" line, no consulting feast. The vendor growth is concentrated in healthcare insurance, contracted SpEd staffing, and managed-services outsourcing — three things that are individually defensible and collectively heavy.
Michigan funds school operations on a per-pupil basis. The state-set foundation allowance — roughly $10,836 per pupil in FY262 — is multiplied by enrollment to produce the bulk of TSD's revenue. Lose a student, lose a foundation. Gain a student, gain a foundation.
Between the FY20 enrollment peak of 13,073 and FY25's 12,393, TSD lost 680 students net — roughly 5.2% of its enrollment base, and about $7.4M of annual foundation-allowance revenue.1 The students did not all leave Troy. The city's population grew +5.5% from FY15 to FY24 (83,496 → 88,051). The school-enrolled share of the population dropped from 15.0% to 14.1% — Troy is aging, and a higher share of school-age children are enrolling outside TSD (private schools, charters, school-of-choice).1
But the buildings stayed open. The teaching staff barely changed. The MPSERS rate didn't fall. Healthcare premiums kept rising. When the same cost base is divided across fewer students, per-pupil cost goes up by arithmetic. Cost per pupil rose from $10,924 in FY15 to $15,624 in FY25 — up $4,700 per student, or 43%, in ten years.1 And the FY25 audit surfaces a notable new wrinkle: revenue per pupil dropped year-over-year for the first time since the FY18-FY19 baseline period, from $15,680 (FY24) to $15,377 (FY25). For the first time in the modern data, audited rev/pupil was lower than audited cost/pupil ($247 below).1
Read by sub-period, the state's per-pupil contribution moved very slowly through the pre-Covid years. Foundation allowance grew only $310 from FY15 to FY19 — about 0.9% per year, below general inflation.4 Acceleration came later: +$1,199 from FY19 to FY24 (about 2.5%/year, partly Covid-era state aid increases). Then the state held the allowance flat in FY25 at $10,394 — the per-pupil contribution did not increase at all year-over-year — before the FY26 amended figure stepped up to $10,836.4 Even with the FY19-FY24 acceleration, foundation growth has trailed per-pupil cost growth in every period.
Reading Section 2's chart, both per-pupil cost and per-pupil revenue appear to grow at roughly the same rate. That picture is misleading: a meaningful share of the per-pupil revenue growth is one-time and pass-through money the board cannot redirect or repeat:
Strip those two layers out and the free per-pupil revenue — the dollars the board can actually direct — has lagged per-pupil cost growth materially. That gap, multiplied by enrollment, is the structural deficit.3
A natural question: Troy property values grew enormously — $3.49B in FY15 to $6.19B in FY24, up 78%.1 Doesn't all that new tax base feed the schools? Under Michigan's Proposal A framework (passed by voters in 1994), school operating revenue is decoupled from the local property tax base. Three constraints lock TSD in:
The practical result: TSD's operating millage and foundation allowance can only move when the state legislature acts. The board cannot vote itself more operating revenue. It can put bond questions for capital projects to voters (Troy did, and they passed) — but bond proceeds cannot be spent on operating expenses.
The compound problem in one line: revenue capacity scales with enrollment, but cost capacity barely scales at all. Lose enrollment, lose revenue; cost stays. The squeeze is mechanical, not the result of any one decision.
Troy School District's structural deficit is not the result of any single decision or bad actor. It is the compounding of four things happening at once:
| Driver | Approx incremental annual cost | Comment |
|---|---|---|
| MPSERS pension expense | ~$15-20M | Mostly state pass-through; visible on expense, offset on revenue |
| Healthcare (MESSA + BCBS) | ~$5M | Active employee insurance, structural inflation |
| Salary inflation (CBA + step) | ~$5M | ~3% CAGR; average teacher salary +5% over five years |
| SpEd non-payroll (HCAs, etc.) | ~$4M | $0 → $5M Edustaff line, partly offset by reduced sub usage |
| Outsourced services | ~$5M | DM Burr + First Group + Sec Shield + MASB-SEG combined |
| Other operating | ~$4M | Supplies, utilities, materials inflation |
| Total expense growth | ~$38M |
Revenue grew roughly $32M over the same period. But ~$15-20M is MPSERS pass-through (revenue and expense both up, by the same amount), and an additional ~$5M was ESSER pulled forward (one-time). The net "free" operating revenue growth — money the board could actually choose how to spend — was perhaps $12-15M.1 Free expense growth outran free revenue growth by roughly $20M. That gap is the structural deficit.
Changed: Support Services FTE collapsed from 276 to 175 over nine years as facilities and transportation were outsourced. The SpEd staffing model added ~100 contracted Health Care Aides on top of in-house aides. Instruction FTE has run roughly +29 above the FY18 pre-Covid baseline in FY25 (840 → 869), with a FY20 peak of 925 and a Covid-era FY21 dip to 862; ESSER may have supported some of the FY22-FY24 partial recovery, but the audited series does not show ESSER funding a clean new staffing layer.1
Did not change much: total FTE is roughly flat (1,072 in FY15 → 1,072 in FY25, with FY16-FY18 swings in between); average teacher salary grew at roughly half the rate of inflation through FY24 before the FY25 +6.9% step; operating millage is at the constitutional cap; foundation allowance grew at statewide inflation and was held flat between FY24 and FY25. This is a composition story — different mix of staff, more contracted services, more benefit cost — not an "everybody got rich" story.
A few things this analysis cannot answer from public records, and would benefit from board or administration clarification:
The honest summary: the district is structurally short on operating revenue, the largest individual driver is healthcare-and-pension cost inflation it does not control, the second-largest is a deliberate SpEd staffing choice that benefits Troy students but is under-reimbursed by the state, and federal ESSER money that was masking the gap is gone. The cuts conversation now in front of the board is the first one that has to be made without that cushion.