Institutional SFR Ownership and Market Share: National Footprint vs. Local Geographic Concentration
Update (2026-10-05 cycle): A new GAO audit — GAO-26-108675, the second in a Congressionally-mandated series, published just after the ROAD Act's July enactment — is now the most authoritative property-level benchmark in the market-share debate, and it lands squarely on the "low overall share, high local rental concentration" reconciliation.
GAO used property-level ownership and mailing-address data from Intercontinental Exchange (ICE) to track institutional investors — defined as owners of 5,000+ single-family rentals nationwide with homes in 5+ metros — across six metro areas (Cincinnati, Dallas, Jacksonville, Nashville, Phoenix, Seattle), 2018–2024:
- By 2024, institutional investors owned 1–3% of ALL single-family homes in these metros (below 1% in Cincinnati and Seattle), but 4% (Seattle) to 22% (Jacksonville) of single-family RENTAL homes.
- Portfolio growth peaked in 2021–2023 and declined in 2024 — i.e., the cooling began before the ROAD Act, on high rates and softening rents, not regulation.
- Acquisition mix (Nashville example): 35% of institutional-owned homes were acquired from owner-occupants; 15% were newly constructed.
- Dispositions: sales to owner-occupants never exceeded 8% of holdings in any year in any of the six metros — the "investors will sell back to families" rebalancing channel is empirically small so far.
Verbatim from the report: "While these investors owned from 4 percent (Seattle) to 22 percent (Jacksonville) of single-family rental homes, they owned from less than 1 percent (Cincinnati and Seattle) to 3 percent of all single-family homes.12 The largest year-to-year increases generally occurred during 2021–2023 and declined in 2024."
Purchase-flow data (the other half of the picture):
- Cotality (Q2 2026 investor report, Sept 3; re-reported in its September "10 things" release and Realtor.com's Oct 2 writeup): mega investors (1,000+ homes) were 2.2% of home sales in August 2026, up from a 1.4% February low — the post-ROAD Act rebound. Cotality's Thom Malone designated Q3 data as the real test of whether the chill was "a permanent retreat or just a pause."
- Realtor.com's annual investor report (June 2026): all investors bought 534,000 homes in 2025; investor share of sales ticked up to 11.3% (+0.3pp YoY); median investor purchase amount rose 5.6%. Mom-and-pop buyers dominate that universe; large investors are concentrated in specific large metros.
Definitional reconciliation (why the headline numbers don't actually conflict): GAO's 1–3% measures the stock of all single-family homes owned by the very largest investors (5,000+ homes) in six metros; AELP's ~13–14% of Phoenix SFR stock measures the rental stock share in a high-concentration metro; Realtor.com's 11.3% measures the flow of transactions by investors of all sizes. The public debate routinely mixes these three — any claim that "investors own X% of homes" should be forced to say which one it means.
What this means for the price/rent adjudication: the neutral read holds — national stock share is small (1–3% even in GAO's investor-heavy metros), local rental-stock concentration is real (up to 22% in Jacksonville), and mega-investor purchase share is small but rising again post-ROAD Act. See Academic Evidence on the Causal Impact of Institutional Landlords on House Prices and Rents for what the causal literature does with this concentration gradient, and Builder-Affiliated SFR Growth: Lennar's Spinoff Millrose Becomes an Institutional Landlord Through the Build-to-Rent Exemption for a new purchase channel (builder-affiliated BTR) that GAO's existing-home stock data does not capture.
-
An instance of Tiny national housing footprints mask the severe local squeeze corporate capital exerts on starter homes. — The neutral GAO benchmark reproduces this theme's reconciliation — a negligible total-stock share coexisting with heavy concentration in the rental stock where the squeeze lands. ↩︎
-
An instance of The share of homes investors own is three different statistics wearing one number. — The GAO audit supplies the authoritative property-level benchmark for the theme's point that stock, rental-stock, and flow shares are divergent statistics an order of magnitude apart, quoted selectively by each side. ↩︎