Midwest No. 2 diesel retail prices — the EIA monthly series

Source. U.S. Energy Information Administration, No. 2 Diesel Retail Prices, Midwest (PADD 2), monthly, dollars per gallon. Series EMD_EPD2D_PTE_R20_DPG. Type. Primary source — the publishing agency’s own machine-readable series. Location. https://www.eia.gov/dnav/pet/hist_xls/EMD_EPD2D_PTE_R20_DPGm.xls, about 47 KB.

What it contains. One observation per month from April 1994, on a two-column sheet: an Excel date serial and a price. The workbook’s other sheet is a table of contents. 389 observations as committed.

Why it matters here. R.C. 3317.0212(C) and (D) set both of the state’s transportation rates — $1,337.175 per weighted rider and $6.867 per mile — as trimmed means of what districts themselves reported spending in the prior fiscal year. Neither rate is chosen; both are statistics over district behavior. Diesel is the part of that behavior that moves fastest and the only part with a published monthly price, so this is the series that says what the rate will do a year before it does it.

What it shows on arrival. Ohio is in a compounding fuel shock. On a July-to-June fiscal year: FY2025 averaged $3.558, FY2026 $4.181 (+17.5%), and FY2027 through August 2026 averages $5.121 (+22.5%). The FY2027 rates were computed from FY2026 reported costs and the calculator is dated 16 December 2025, so the rate districts are reimbursed at this year reflects a fuel price roughly a fifth below what they are paying. That is the lag working exactly as the section specifies.

Access constraints. Free, no registration, no API key, no User-Agent requirement — unlike the Bureau’s flat file. The EIA’s v2 JSON API does require a key; the hist_xls path does not, which is why it is the one used.

Caveats:

• It is a legacy BIFF8 workbook, not an XLSX. Saved by Excel in 2004 and still served in that format. spreadsheet::ole2 reads it natively; nothing here needs LibreOffice.

• Retail, not what a district pays — and the difference runs the other way from the obvious guess. Ohio levies 47 cents a gallon on diesel under R.C. 5735.05(E)(2), “forty-seven cents on each gallon of motor fuel other than gasoline”, and school districts claim it back through the department’s Motor Fuel Tax Refund program. So the retail level is wrong for a district budget, which is expected.

  What is not expected is the direction. Subtracting a fixed per-gallon wedge from both ends of a rise widens the proportional change rather than narrowing it. FY2025 to FY2026 is +17.5% at retail and +20.2% net of the Ohio tax. The retail series understates what districts experience, and an elasticity fitted to it without the adjustment is biased low.

  (An earlier revision of this entry said the wedge “compresses the proportional swing”. That was wrong, and wrong in the direction that flatters the series. The arithmetic is unconditional: for p1 > p0 > w, (p1−w)/(p0−w) > p1/p0.) verified for the rate and the arithmetic; open as to the federal excise, which districts may also be exempt from and which is not sourced here

• PADD 2 is not Ohio. The Administration publishes no Ohio retail diesel series; the Midwest region is the smallest published geography containing it.

• The first row carries no value. March 1994 is written with an empty price because the weekly survey began part-way through that month. The builder skips valueless rows rather than failing, so an absent observation stays absent instead of becoming a build error.

• Fiscal-year aggregation is a choice, and a different one from the CPI’s. Fuel is bought across the whole year, so a July-to-June mean is the natural alignment — where the deflator takes a June point observation because an index level is not consumed. The committed fixture stays monthly so either can be computed.

A committed extract is at crates/connect/fixtures/midwest-diesel-monthly.csv.

Feeds connector #

eia-diesel, built by crates/connect/src/fixtures/diesel.rs.