FHFA May 2026 House Prices: An Uneven Rebound

U.S. house prices rose 0.3% in May 2026 from April and 2.2% from a year earlier, according to FHFA’s July 28, 2026 release. The seasonally adjusted purchase-only index rebounded after April’s 0.1% decline, which was unchanged in this release.

Prices nevertheless fell month to month in two census divisions. For agents, the national gain describes appreciation within a defined mortgage sample through May. It cannot establish a neighborhood’s price trend or an August listing price.

May house prices across the nine divisions

The July 28 report’s comparison table, on printed page 3, shows seven monthly increases and two declines. All figures below are seasonally adjusted and nominal, meaning they are not adjusted for inflation.

As with other home price reports, the comparison period matters: East North Central prices fell from April but remained substantially above their May 2025 level. The Pacific division recorded declines over both periods.

FHFA combines state price trends using housing-stock weights to construct division and national indexes, as described in its methodology note. Each area therefore has a different influence on the aggregate, so gains in some places can outweigh losses elsewhere. This explains how the national and regional results coexist, without identifying what caused prices to change in a particular community.

What a repeat-sales house price index measures

Interpreting those changes starts with understanding home price measures. A median sale price is the middle value when a period’s sale prices are arranged in order. Its movement can reflect a change in which homes sold.

Consider a hypothetical market with five sales each month. In April, three smaller homes sell for $300,000 each and two larger homes for $500,000 each. In May, those counts reverse, with prices for each type unchanged. The median rises from $300,000 to $500,000 even though neither type becomes more expensive.

FHFA’s repeat-sales approach compares successive transactions involving the same properties. Its statistical model uses those pairs to estimate appreciation over time, reducing the influence of a changing sales mix. The homes need not sell in consecutive months; FHFA’s technical description explains how the method handles transactions separated by different intervals.

A property can still change between sales. Renovations are a potential source of bias, as FHFA researchers examined in a March 2017 working paper. Comparing the same properties improves comparability, but the index remains an estimate of market appreciation.

Which transactions enter the purchase-only index?

The headline series uses single-family purchase transactions involving conventional, conforming mortgages bought or securitized by Fannie Mae or Freddie Mac. Cash purchases, jumbo loans above applicable conforming limits, and FHA- or VA-backed purchases fall outside that sample. A conventional loan retained outside the two enterprises also does not enter merely because it meets conforming requirements. FHFA’s August 2025 methodology FAQ defines this coverage.

Other FHFA indexes draw on different evidence. The all-transactions index adds refinance appraisals, which are valuations without a sale. The expanded-data index adds FHA transactions and county recorder sales records. FHFA’s dataset descriptions distinguish these series. Using one in place of the monthly purchase-only index changes the coverage and requires identifying that difference.

Seasonal adjustment and revisions affect comparisons

Seasonal adjustment removes recurring calendar patterns to improve comparisons between adjacent months. It does not remove inflation or make May transactions a measure of current prices. FHFA also revises historical estimates as additional repeat transactions and older mortgage records arrive. Both practices are described in the methodology FAQ.

The July release revised March’s national monthly gain from 0.2% to 0.1%. The Mountain division’s April decline deepened from 0.8% to 1.0%, although April’s national figure stayed unchanged. The July tables place these revisions beside the previous estimates.

For a reproducible comparison, record the release date, index type, geography and adjustment basis. Use one release’s historical series throughout; mixing an older April estimate with a newer May estimate can distort the apparent change in momentum.

Evidence for an August pricing decision

A listing recommendation needs local housing market research: recent comparable sales, competing listings and pending activity. Check property condition, concessions and the dates when comparable transactions went under contract. These details address questions the national and division indexes leave open.

Fannie Mae’s appraisal guidance, dated June 4, 2025, permits house price indexes to support time adjustments while requiring evidence for the relevant period between a comparable’s contract date and the appraisal’s effective date. An annual regional gain alone cannot justify a particular adjustment to one home.

FHFA’s next release is scheduled for August 25, covering June and the second quarter. In an August pricing memo, show what has changed among comparable properties since May and how that evidence supports the proposed asking price.