← All reports

Is the Money Following the People? Six Markets Where Investors and Movers Disagree

By David, Next Generation Data Analytics

Every investor underwriting a market looks at the same handful of signals — price, rent, days on market, maybe population growth if they’re thorough. Almost nobody looks at the two I want to put in front of you today, together: where the people are actually moving (IRS county-to-county migration, built from real tax returns) and where the investor money is actually going (CFPB HMDA mortgage data, which flags whether each purchase loan was for a non-owner-occupied property).

Cross them, and one question falls out:

Is the money following the people, or betting against them?

Most of the time, the two agree. The interesting markets are the ones where they don’t. So I pulled every U.S. county where investor mortgage activity and resident migration point in opposite directions, and split them into two lists: places people are flooding into that investors haven’t chased yet, and places investors are piling into while residents head for the exits.

How to read this

Two numbers anchor every market below.

Net migration is IRS data: household tax returns that moved in, minus returns that moved out, for the most recent year the IRS publishes at the county level — 2022. (It lags about two years. That’s the single biggest caveat here, and I’ll come back to it.)

Investor share is the percentage of mortgaged home-purchase loans in that county that went to non-owner-occupied properties (HMDA, same year). Higher = more of the financed buying is investors rather than residents.

“Divergence” just means those two arrows point opposite ways.

List A — Overlooked: people are arriving, investors aren’t (yet)

High-inflow counties where the investor share of financed purchases is still in the single digits. The story is simple — households arrived faster than the investor flywheel could chase them. The whole Florida I-4 corridor (Polk, Pasco, Lee) and the Dallas and Phoenix suburbs dominated this list. Here are the three sharpest.

Pasco County, FL (Tampa metro) — investor share 5.5%, net inflow +9,800 households The cleanest divergence on the board: nearly ten thousand net new households, and only about one in eighteen financed purchases was an investor. - The opening: genuine end-user demand with almost no investor competition for financed deals. - The trap: Tampa-area insurance costs and the 2024–25 Florida price softening. Inflow doesn’t guarantee that rents or values hold.

Collin County, TX (Dallas metro) — investor share 7.4%, net inflow +8,600 households, ~21,000 purchase loans High inflow and deep liquidity — 21,000 financed purchases means you can actually transact and exit here. - The opening: one of the most liquid low-investor-share growth counties in the country. - The trap: Texas property taxes plus a lot of new supply can cap appreciation even with strong demand.

Maricopa County, AZ (Phoenix metro) — investor share 7.8%, net inflow +8,200 households The surprise pick. Phoenix was investor-central in 2021, so a 7.8% share looks too low to believe. - The opening: if it’s real, it’s a major metro people are still pouring into without heavy financed-investor competition. - The trap — and this is the important one: HMDA only sees mortgaged purchases. Phoenix’s investor wave was disproportionately cash (iBuyers, institutional buyers), which this data can’t see. Maricopa is the market that best shows why you read the methodology before you act on the number.

List B — Overheating: investors are piling in, people are leaving

Net-outflow counties with high investor shares. This list needs more care than List A — most candidates were small markets where a swing of a few hundred households is noise, not signal. I kept the three where the pattern is real, and named what each one actually is, because the headline number lies in two of them.

Bronx County, NY — investor share 24.2%, net outflow −22,700 households The one unambiguous big-market divergence: the largest resident outflow of any county on either list, with roughly a quarter of financed purchases going to investors. - The read: a classic outflow-plus-consolidation pattern — investors buying into rental cash flow as residents leave. The most textbook version of the thesis on this page.

Sevier County, TN — investor share 31.3% (highest on the board), net outflow −480 households The highest investor share anywhere in the data — and almost entirely not what it looks like. This is Pigeon Forge and Gatlinburg: a short-term-rental cluster. That 31% is structural vacation-rental demand, not investors quietly displacing residents. - The read: the highest number on the page is the least alarming one. Context beats the leaderboard.

Dougherty County, GA (Albany) — investor share 28.6%, net outflow −280, loan-denial rate 16.5% A small, shrinking Southern city where investors are consolidating cheap stock — and note the 16.5% denial rate, among the highest here. Tight credit plus investor accumulation in a declining market is a recognizable distressed-consolidation pattern. - The read: small in absolute terms, but it stands in for a whole cluster (Albany GA, Danville VA, the north-Louisiana parishes) doing the same thing.

One oddball worth a line: Lubbock County, TX showed a net household outflow but a net income inflow — fewer bodies, higher earners, and a 26% investor share. When people and money move in opposite directions inside the same county, that’s its own signal.

The honest part

This is built from public data with real limits, and the data-literate among you will (rightly) poke at it. So here’s where it’s soft:

  • The migration data lags ~2 years. 2022 is the most recent county-level IRS release. A market’s trajectory can shift in two years — treat these as structural tendencies, not this-quarter calls.
  • Investor share is mortgaged purchases only. Cash and institutional buying is invisible here (see Maricopa). High-cash markets understate their true investor presence.
  • County-year grain is coarse. A county isn’t a neighborhood; the right ZIP inside a “boring” county can tell a completely different story.
  • Investor share is not your competition. It’s a market-temperature reading, not a deal-by-deal forecast.

None of this kills the signal. It just means each of these is a place to start a thesis, not finish one.

Where this came from — and your turn

I built a free tool that scores any U.S. market on a composite of this same kind of cross-source data — housing, jobs, migration, and risk in one view: zonevista.io/composite-market-score. No signup.

Now I want the argument: of these six, which one do you think is a trap, and which is the real opportunity? I’ve got a hunch about Maricopa. Tell me where I’m wrong — that’s half the reason I publish these.

— David