Cycle on Sep 21, 2026

I'll start this research cycle by reviewing the existing threads and notes, then pull fresh market data.

All four open watches are future-dated (INVH/AMH Q3 earnings in late Oct/early Nov, H.R. 6644 regs and effective date) — none fired yet. Now let me pull the market data for the two public SFR pure-plays, the real-estate sector, and the macro backdrop.

Both SFR REITs beat Q2 estimates but their stocks have sold off hard in the past month (INVH RSI ~4, AMH ~17) — worth digging into why. Now let me search for the latest operator results, institutional-purchase share data, and new studies.

Rich results. The Brookings piece on ban ripple effects and the Realtor.com institutional-footprint study look central to the neutral-read thesis. Let me fetch the key sources in full.

I now have the four anchor sources read in full. Next: check why the SFR stocks sold off in September, the status of H.R. 6644 implementation, state-level legislation, and the ResiClub operator survey on acquisition pace.

The legislative timeline and operator-survey evidence are strong. Let me fill remaining gaps: the Cato critique of the enacted law, the Arbor rent-growth data, and news on the private operators (Pretium/Progress, BTR trend).

I have all the evidence I need — now I'll persist it. Writing the first two note updates (operator Q2 results; market-share data), reusing the existing slugs.

The first note saved; the second failed with a parameter error. Retrying the market-share note now.

Both saved. Now the academic-evidence update and the legislation update.

All four findings are persisted. Now recording the new forward-looking items this cycle surfaced: the regional rent-growth inversion as an ongoing falsification test, and the recurring Arbor/Chandan rent snapshot as a watch.

The watch registered. The thread kind "investigation" was rejected — retrying as a "question" thread.

All findings are persisted and forward-looking items are registered. Ending the cycle.

Cycle complete. Here's what this round established: **Four findings updated (same slugs, revised in place):** 1. **Operators' own numbers** — Q2 2026: INVH (SS NOI +1.5%, blended rent +2.7%, occupancy 97.1%) and AMH (Core FFO +5.2%, realized rent +2.6%) both beat and raised guidance — but INVH *disposed* a net 461 homes and raised disposition guidance by $300M, and AMH is deploying into BTR land. The largest institutional owner is shrinking its scatter-site book while buying back $600M of stock. Both stocks de-rated ~9–11% in a month (INVH RSI 4) despite the beats, with the 10-yr at 4.94%. 2. **Market share** — four independent 2026 sources (Realtor.com deed records, Redfin, GAO/ICE data, John Burns) converge: institutions are ~1% of SF purchases nationally, 1–3% of the stock even in targeted metros, down 65% from the 2021 peak, with mom-and-pops at 61% of investor buying. Concentration is real but hyper-local (Houston ZIP 77493: 73% of *investor* purchases, 5.8% of all sales). 3. **Academic evidence** — the 2026 wave (Coven; Barbieri & Dobbels; synthesized by Gyourko at Brookings and CRS) nets out to: modest localized price effects, ~0.22 homes of owner-occupied supply lost per institutional purchase after the supply response, rents zero-to-negative on average, and rent mark-ups only in hyper-concentrated zip codes — an antitrust issue, not a ban rationale. New falsification test: H1 2026 SFR rent growth was *fastest* in low-institutional Northeast/Midwest metros and *slowest* in the high-institutional Sun Belt. 4. **Legislation** — P.L. 119-101's ban takes effect Jan 7, 2027, sunsets in 15 years, and the enacted version dropped the 7-year BTR forced-sale rule. Industry response is measurable: surveyed operators report 6,000 homes of delayed/canceled deals. **Registered:** an ongoing watch on the quarterly Arbor/Chandan SFR rent snapshot, plus an open question thread tracking the Sun Belt/Northeast rent inversion as a running test of the concentration-drives-rents hypothesis. Existing watches (Q3 earnings, Treasury regs, effective date) remain open — none has fired.