Multifamily Property Tax Underwriting: The Year-Two Reset

Published September 27, 2026

Property taxes are roughly 30% of apartment operating costs, per RealPage Market Analytics data as of the first quarter of 2025 reported by CRE Daily — generally the largest single line on a multifamily operating statement. Most acquisition models treat it as an escalator: take the seller's trailing figure, add 3% a year, move on. In much of the country it behaves as a step function instead, and the step frequently lands in year two, after the model has already declared the deal underwritten.

The error is structural rather than careless. The tax line on a seller's statement is not a market figure that a buyer can trend. It is the output of an assessment the seller may have been sitting on for a decade, and in a long list of states the act of selling is what removes it.

Why Multifamily Property Tax Underwriting Breaks in Year Two

The timing is what makes this line different from every other expense. Most jurisdictions value property as of January 1 and bill against that roll for the year, so a buyer closing in, say, June generally inherits the seller's assessment for the balance of that tax year. The new number arrives on the following January 1 — which sits inside year two of a standard five-year hold.

That sequencing is why the problem survives diligence. Year-one actuals often come in close to the model, which reads as confirmation that the tax assumption was sound. Multifamily property tax underwriting that validates itself in year one and breaks in year two is a particularly expensive failure mode, because the refinance or supplemental-debt test is frequently scheduled somewhere near that same point in the hold.

The fix is not a bigger escalator. It is recognizing that the seller's trailing tax figure and the buyer's tax figure are two different numbers connected by a statutory event, in much the same way insurance resets at closing rather than escalating.

The Sale Itself Is the Reassessment Trigger

Florida is the cleanest illustration because the mechanic is written down. Non-homestead property, including apartments, is subject to a 10% annual cap on assessed-value increases — but per the Pinellas County Property Appraiser, "the market and assessed values are equal the first January 1 following a change of ownership or control," and the cap only begins again the year after that. A seller who held through a decade of appreciation accumulated a gap between capped assessed value and market value, and that gap does not transfer.

Two details compound it. The cap never applied to school district levies at all, per the same county guidance, and school millage is a large share of a Florida bill — so the capped portion was always smaller than the headline suggests. And a transfer of more than 50% of ownership or control can trigger the reset even where the deed does not change hands in the way a buyer might expect.

Florida is not unusual, only explicit. California reassesses to current fair market value on a change in ownership under Proposition 13, with roughly 2% annual growth permitted thereafter. Michigan uncaps taxable value in the year following a transfer of ownership. Several other states apply a similar change-of-ownership mechanic with different names and different carve-outs. These provisions are technical and amended periodically, so the specific treatment is worth verifying with local tax counsel for the actual jurisdiction rather than assumed from a neighboring state.

What the Step Costs: A Worked Example

Example — illustrative. A 200-unit Class B property in a Florida secondary market. Purchase price $36M at a 6.0% going-in cap rate, so year-one Net Operating Income (NOI) of about $2.16M. Debt at 60% Loan-to-Value (LTV) — $21.6M at 6.0%, 30-year amortization — for annual debt service of roughly $1.55M. The seller has owned since 2014; capped assessed value is $19M against a just value near $34M. Combined county, municipal, and school millage of 19 mills, or 1.9% of assessed value.

The step is approximately $285,000 of annual expense, or about 13% of year-one NOI. Capitalized at the going-in rate it is close to $4.8M of value — meaningfully more than the negotiating range on a $36M asset. On coverage, Debt Service Coverage Ratio (DSCR) falls from roughly 1.39x to about 1.21x, which is below the 1.25x covenant many lenders carry.

Millage rates, just-value determinations, and holding periods vary substantially by county and by asset, so these figures are illustrative rather than indicative of any particular market. The structural point is the one that travels: a line worth roughly 30% of operating expenses is being carried at the prior owner's basis, and the correction is a single-year step rather than a trend.

Annual-Reassessment States Hide the Problem Differently

Where there is no sale trigger, the exposure does not disappear — it changes shape. Texas reassesses commercial property annually, with appraisal districts valuing as of January 1, issuing notices generally in April, and setting a protest deadline of May 15 or 30 days from the notice date, per DMA's 2026 Texas property tax guidance. There is no capped basis to inherit, which sounds protective. It also means the assessment can move every single year, in either direction, independent of anything the new owner did.

Administrative changes are the underappreciated risk in these states. Marion County, Indiana added nearly $2B of assessed value across approximately 1,000 apartment complexes of 20-plus units, per a February 2026 analysis by Faegre Drinker, after the Indiana Department of Local Government Finance removed an adjustment that had been suppressing base rates. That same analysis describes first-floor apartment base rates rising about 32% for 2025 and a further 20% for 2026 — a compounding of roughly 58% over two years, driven by methodology rather than by any market event a rent comp would capture.

No underwriting model contains a line for "the state removes a base-rate adjustment." What a model can contain is a tax assumption wide enough that a two-year administrative reset does not break the covenant, which is the same discipline that applies when normalizing every other line on a messy T-12.

The Appeal Is an Assumption, Not a Plan

Appeals are real and frequently successful, and they are also the most common way an aggressive tax assumption gets defended. The problem is sequencing: the appeal outcome is generally unknown at closing, the filing windows are short, and the reduction — if granted — may apply to one tax year rather than permanently.

Texas offers an equal-and-uniform remedy alongside market-value protest, and several states allow appeals grounded in the actual income the asset produces rather than a cost-based estimate. Those are genuine tools. Treating an expected reduction as base-case NOI, though, puts the deal's coverage test downstream of a decision made by an appraisal review board. Underwriting the assessment the jurisdiction is likely to set, and carrying any reduction as upside, keeps the covenant math independent of that outcome.

Where This Analysis Breaks

The step is not always as large as the purchase price implies. Appraisers often set just value below a transaction price, particularly where the buyer paid for a business plan rather than in-place income, so the reset may land well under the sale figure. In markets where values have softened, a 2026 assessment could even come in below a seller's capped value, in which case there is no step at all.

Millage is the other half of the equation and it is not fixed. Taxing districts set rates annually and sometimes roll them back when assessed values rise sharply, so assessed value climbing 40% does not mechanically mean the bill climbs 40%. Modeling assessment and millage as one blended percentage obscures that offset in both directions.

There is also a reasonable argument that the market already prices this. If every bidder faces the same reset, the step is embedded in what the asset trades for, and the exit cap rate reflects a buyer who will face it too. That argument holds better in transparent, annual-reassessment states than in capped states, where the size of the step depends on how long the specific seller held — which is deal-specific information that a cap rate comp cannot carry. It also does nothing for a sponsor whose year-two covenant test is measured against actual coverage rather than against the market's average experience.

The Underwriting Takeaway

Property taxes are the largest expense line in most multifamily models and the one most often carried at the prior owner's basis. Three things follow. The tax assumption belongs on its own schedule with an explicit assessed value and millage rather than a single trended figure. The reset year is worth modeling where it actually falls, which in capped states is typically the January 1 after closing rather than the closing date. And the covenant test in the reset year deserves a look at the stepped number, not the trailing one — the same year-by-year discipline that underwriting the 2026 turn demands everywhere else in the model.

A deal that works on the seller's tax bill and fails on the buyer's was never a deal. It was an arithmetic gap that took about 18 months to show up.

UWmatic is an AI-powered underwriting platform built for multifamily and mobile home park investors. Model the assessed-value reset on its own schedule and see what the year-two step does to NOI, DSCR, and your covenant headroom. Try the free underwriting calculator →

This analysis reflects current market interpretations as of the publication date and may evolve as new data becomes available. Figures cited are drawn from public sources including the Pinellas County Property Appraiser's non-homestead cap guidance, RealPage Market Analytics operating expense data reported by CRE Daily, DMA's 2026 Texas property tax guidance, and a February 2026 Faegre Drinker analysis of Marion County, Indiana apartment assessments, and are subject to revision. Nothing in this post is investment advice; readers should conduct their own diligence and consult qualified professionals before making investment decisions.