Richmond's $1M+ Homes Are Assessed Furthest Below What They Actually Sell For

Every home that sells in Richmond gets compared against a number nobody chose: its tax assessment. Most of the time, that number runs behind the sale price — that's old news, and we've written about the metro-wide version of it before. What's new in this week's data is where the gap gets worst. It isn't evenly spread across price tiers. It concentrates hard at the top.

In 270 Richmond-metro sales that closed between August 24 and September 4, 2026, homes that sold for less than $1 million closed a median 15.2% above their tax-assessed value. Homes that sold for $1 million or more closed a median 32.2% above assessment — more than double. Fourteen of the sixteen $1M+ sales in this window had usable assessment data, and the pattern held across nearly all of them, not just a couple of outliers dragging an average.

VERIFIED — The Core Finding

Sales under $1M (n=226, 2026-08-24 to 2026-09-04): median gap vs. tax assessment = +15.2%. Sales at $1M+ (n=14): median gap = +32.2%. Source: data/sources/rva-recent-sold-2026-09-05.csv, Zillow scrape, scraped 2026-09-05.

This Isn't a One-Off — A University Study Already Found the Mechanism

A University of Chicago Harris School of Public Policy analysis of Richmond's property tax rolls found the same directional pattern using an entirely different method and an earlier year of data: the city's least expensive homes were assessed at 97.4% of their actual value, while its most expensive homes were assessed at just 71.9% — a 1.35x regressivity ratio between the two ends of the market. In plain terms: cheap homes get assessed close to what they're worth. Expensive homes get assessed well below it.

That study used assessment data, not two weeks of live closings. This week's sold data is a different measurement, closer to the ground, and it reproduces the same shape: the $1M+ tier's assessment gap runs roughly double the gap on everything below it. Two independent methods, two different time windows, same direction. That's what makes this worth reporting now rather than waiting for a bigger sample — the finding already has outside corroboration, and this week's data is a live, current data point on the same trend.

INFERRED — What This Does and Doesn't Prove

This week's 14-home $1M+ sample is real but modest — enough to show a pattern, not enough on its own to establish a precise number. Treat 32.2% as directionally consistent with the UChicago finding, not as a replacement for it. The mechanism most consistent with both datasets: assessors update high-value properties less aggressively than the market moves them, whether because reassessment cycles lag fastest at the top of the market or because appeal activity and comp availability differ by price tier. This piece does not attempt to adjudicate why regulators under-assess luxury homes — only that the pattern shows up, again, in fresh data.

A Data-Cleaning Note We're Not Going to Bury

Twenty-one of the 261 sales with both a sale price and a tax-assessed value in this scrape had to be excluded from every calculation in this piece — their assessed values were implausibly low relative to sale price (gaps above 100%, several above 400%, one listing an assessed value of $20,000 against a $377,934 sale). A few of these records carried placeholder-looking street names. The most likely explanation is that these are newer parcels still carrying land-only or pre-construction assessed values in the county's data feed, not evidence of a 400%+ real-world assessment gap. We're not reporting those numbers, and we're telling you why: reporting them would have pulled the metro-wide mean from a defensible 15-16% up to a meaningless 54%, while barely moving the median. We used the median throughout this piece for exactly that reason — it doesn't get hijacked by a couple of bad records the way a mean does.

The $1M+ Sales, One by One

Here's every $1M+ sale in the window with usable assessment data — no names redacted beyond what's already public record (sale price and county assessment are both public), no listing photos, nothing beyond the numbers a buyer could look up themselves.

ZIP	Sale Price	Tax Assessed	Gap
23226	$2,700,000	$1,933,000	+39.7%
23103	$2,250,000	$1,891,100	+19.0%
23229	$1,800,000	$1,451,200	+24.0%
23226	$1,430,000	$881,000	+62.3%
23233	$1,380,000	$1,159,300	+19.0%
23225	$1,300,000	$932,000	+39.5%
23059	$1,270,000	$855,500	+48.5%
23103	$1,190,000	$930,400	+27.9%
23226	$1,150,000	$858,000	+34.0%
23229	$1,100,000	$839,200	+31.1%
23113	$1,100,000	$906,100	+21.4%
23226	$1,050,000	$742,000	+41.5%
23229	$1,020,000	$830,100	+22.9%
23229	$1,000,000	$750,500	+33.2%

Every single one of these fourteen sales closed above assessment. The tightest gap in the group (19.0%) is still above the metro-wide median for homes under $1M (15.2%). The widest (62.3%, a 23226 sale) is more than three times the tightest gap in the $1M+ group itself.

What This Means If You Own — Or Are Buying — At the Top of the Market

If you already own a home that would sell for $1M+ today, the practical read is straightforward: your county's assessed value is very likely running further behind your home's real market value than a typical Richmond-area homeowner's is. That's a lower property tax bill than a fully current assessment would produce — a bigger version of the same quiet windfall we described in our June piece on the metro-wide assessment gap, concentrated at the top instead of spread evenly.

If you're buying at that price point, budget your ongoing tax exposure using your purchase price, not the current assessment. Reassessment cycles eventually catch up — when they do, the jump in assessed value (and the tax bill) for a $1M+ property tends to be larger in dollar terms than for a $400K one, even if the percentage catch-up is similar. Ask your agent or a local tax attorney what your specific county's reassessment cycle and timeline look like before you assume this year's tax bill is next year's tax bill.

This Is About Assessment Mechanics, Not Neighborhoods

This piece measures a relationship between sale price and tax-assessed value. It says nothing about, and should not be read as saying anything about, the people or communities in any ZIP code. The same price-tier pattern — larger gaps at higher price points — showed up across multiple counties and multiple ZIPs in this dataset, which is consistent with an assessment-cycle mechanism, not a neighborhood-specific one.

Data sourced from 270 Richmond-metro home sales closing 2026-08-24 through 2026-09-04, scraped via Zillow on 2026-09-05 (data/sources/rva-recent-sold-2026-09-05.csv). Of 261 sales with both a sale price and a tax-assessed value, 21 were excluded from all calculations for implausible assessed values (gap magnitude over 100%, likely land-only or pre-construction assessment records); all figures in this piece reflect the remaining 240. External corroboration: University of Chicago Harris School of Public Policy, "An Evaluation of Property Tax Regressivity in Richmond city, Virginia" (propertytaxdata.uchicago.edu), cited for its city-wide 71.9%/97.4% assessment-ratio finding — a separate dataset and methodology from this week's sold-home analysis. For the metro-wide version of the assessment gap across all price tiers, see our earlier piece, Virginia's Tax System Can't Keep Up With RVA Home Prices. Analysis by Raam RVA Research Team, September 12, 2026.