Table SD-1 — School district taxable value and taxes charged

Source. Ohio Department of Taxation, Tax Analysis Division. SD1CY21.xlsx through SD1CY24.xlsx (the last dated 3 April 2026), from the Tax Data Series. Type. Primary source — compiled from the DTE-13 and DTE-14 abstracts county auditors file with the department. Location. tax.ohio.gov, Researcher → Tax Analysis → Tax Data Series → Property Tax.

What it contains. Every Ohio school district, once per tax year: taxable value broken out by class (agricultural, residential, mineral, industrial, commercial, railroad, public utility), real property taxes charged for current expenses, calculated Class I and Class II tax rates, average value per pupil, and the fiscal-year ADM the per-pupil figure divides by.

This is the local half of the funding picture, and until it was retrieved the corpus had no per-district measure of it except the one-year millage and valuation columns of the District Profile Report.

Why four tax years and not two. Two give a level and a change, which is what the tax page reads. The fourth year is here for a different reason: Ohio’s counties reappraise or update on a staggered three-year cycle, so TY2022–TY2024 contains exactly one valuation event for every one of the 88 counties, and TY2021 makes the earliest of those a change rather than a level. That window is what lets each district’s reappraisal be measured against its own quiet years, which is how recognized valuation is reconstructed without a source that publishes it. The separation is stark — a median 28.6% jump in the event year against 1.5% otherwise.

A trap the extra years set, recorded because it caught five callers. With two tax years, first() and last() are adjacent, so every consumer that wanted a year-over-year change took the ends. With four they are three years apart and span a reappraisal. Nothing failed loudly: a rate is still a rate and a page still renders. Anything reading this fixture as a change must take the last two.

The fifth caller was a node rather than a function, and it failed differently: metric/assessed- valuation-per-pupil published a three-row table with one TY2024 row and two TY2023 rows, a count and a median that reproduce only on TY2024, and the numerator identity the finding rests on, which holds only on TY2023 — all labeled [verified — TY2023]. No figure was wrong about the fixture and no two of them had to be from the same year. Anything reading this fixture as a level must name a tax year in a constant, which is what dispersion::valuation::TAX_YEAR now is.

Two worksheets per workbook, differing in one thing. SD1DAT… counts the joint vocational school district’s operating levy in the taxes charged; ExJVS… removes it. 501 of 611 districts are in a JVSD, and the difference is $513 million in TY2024. A figure quoted without saying which basis it is on is ambiguous by roughly 4% of the state total, so the fixture carries both.

Three things “taxes charged” is not. Each has produced a wrong number somewhere, and each is recorded on build_sd1_extract as well as here.

• Not net of state reimbursement. The department’s own note: the figures “include taxes that have been reduced under various property tax programs that are reimbursed to local school districts by the state” — the non-business credit, owner-occupancy credit, and homestead exemption. About a tenth of what this column calls property tax is state money. Using it as the local share overstates local effort, and by more in districts with more owner-occupied housing.
• Not a fiscal year. A TY2024 charge is collected across calendar 2025 — half early in the year, half in July — so it straddles FY2025 and FY2026. Any ratio against a single-fiscal-year denominator counts money the district had not yet received, and the error is largest exactly where a levy has just passed.
• Not collected. Delinquency puts receipts below the charge by a district-specific amount this table cannot see. The five-year forecast’s line 1.010 is the receipt side, and the two are worth differencing rather than substituting.

Access constraints. Freely available, ~470 KB XLSX. tax.ohio.gov answers any non-browser user agent with a 403 page served under a 404 status, including for URLs that resolve perfectly well in a browser — so the connector fetches from dam.assets.ohio.gov, the asset host the site itself links to, where the same bytes are served without inspection. The mapping is tax.ohio.gov/static/X → dam.assets.ohio.gov/raw/upload/tax.ohio.gov/X; note raw/upload for workbooks against image/upload for the older PDFs.

Caveats.

• The worksheet names drift and the layout does not. ExJVS and SD1DATWK23 in the TY2023 workbook against ExJVS24 and SD1DAT24 in TY2024, over an identical 28-column table. The banner above the header is also one row shorter in TY2023. The connector matches sheets on their stem and finds the header by content for that reason.
• The IRN is stored as a number, so Manchester Local’s 000442 arrives as 442. Joining that against the department of education’s zero-padded keys matches nothing, and fails as an empty join rather than an error.
• N/A in TY2024 and NA in TY2023 mean the same thing — two spellings of not-applicable from one publisher two years apart. Both appear only in the value-per-pupil rank column, on the two districts with no ADM, and neither is extracted.
• The workbook is revised in place. The TY2024 file carries an internal date of 3 April 2026, later than the analyses circulating from it in August 2025. The digest manifest pins which bytes the fixture was built from; a mismatch means the department restated, not that the fixture drifted.
• 611 districts, against the 609 in the department of education’s FY2027 model and the 607 in the report card files. The extra rows are the island districts, which have valuation and no ADM.

What it settled # ↑ Contents

Policy Matters Ohio’s August 2025 brief on property tax repeal compared each district’s TY2024 real property taxes charged to its FY2025 operating expenditure. Reproduced from this table and the report card spending file, the published figures come back exactly: statewide median 36%, fiftieth-ranked district 74%, Columbus 47.8%, Cleveland 38.8%, Cincinnati 50.4%, and $513 million of JVSD operating levy. See crates/dispersion/tests/sd1_district_taxes.rs.

Also read by # ↑ Contents

• crates/dispersion/fixtures/sd1-district-taxes.csv

Feeds connector # ↑ Contents

tax-abstract