Which States Are Losing Population — and Why Home Prices There Rarely Fall

San Francisco County lost residents last year. Home prices there rose 11.6% anyway — on a median home already worth about $1.39 million. That is not a glitch. It is the single most misunderstood thing about a shrinking place: losing people almost never makes the housing cheap.

The intuition feels airtight. Fewer buyers, softer demand, falling prices. It is also wrong most of the time. Across the counties we track, the ones bleeding population are, with striking regularity, the ones where the sticker price kept climbing.

First, the question you actually came here for: who's shrinking?

At the state level, outright population decline is still rare. Weighting every county by its population, only three states are net-negative: New York (−0.09%), Louisiana (−0.06%), and Hawaii (−0.03%). Behind them sits a cluster running on empty — Illinois, West Virginia, California, Michigan, Pennsylvania, and Ohio are all growing by a rounding error, a few hundredths of a percent to three-tenths.

Zoom in and the picture sharpens. Of the counties in our dataset with complete figures, 228 — nearly one in four — are flat or shrinking. That is the real story the state averages hide: population loss in America is a county-level phenomenon stitched together underneath states that still look stable on paper. A shrinking county in upstate New York and a booming one outside Dallas can average out to "no change" for the whole state.

The national weather behind all of this changed in 2026. The Census Bureau's Vintage 2025 estimates showed net international migration falling from roughly 2.7 million in 2024 to 1.3 million in 2025 — more than a 50% drop, and the reason the Bureau explicitly cited for the national slowdown. That decline touched 90% of all counties. At the same time, the country's 50 largest counties shed a net 637,634 domestic movers to smaller places. Immigration had been quietly filling the gap in a lot of counties. When it thinned out, the decline stopped hiding.

A quick honesty note on the numbers: the national figures above are Census Vintage 2025 estimates (measured mid-2024 to mid-2025). The county-level population and price figures in this piece come from our own dataset, built on American Community Survey data and Zillow's home-value index over a trailing 12-month window. They are adjacent measurements, not the same ruler — so treat them as two views of the same trend, not decimal-precise matches.

Now the part nobody expects: the prices didn't follow

Here is the finding that should reset your instincts. Of those 228 flat-or-shrinking counties, 196 — 86% of them — still posted a positive home-price change over the past year. Not stabilized. Rose. All three of the net-negative states did too: New York up 4.75%, Louisiana up 1.66%, Hawaii up 0.91%, every one of them losing people while prices climbed.

The examples are not exotic outliers. Winnebago County, Illinois — the Rockford metro — held flat on population and gained 9.2% on home values. Scioto County, Ohio, lost residents and rose 8.0%. Darlington County, South Carolina, shrank and climbed 12.0%. And San Francisco, losing people on a seven-figure median, tacked on another 11.6%. These are Rust Belt, rural South, and superstar-coast counties alike. The one thing they share is a population line pointing down and a price line pointing up.

Why fewer people doesn't unlock cheaper homes

The mistake in the intuition is treating a housing market like a concert with empty seats — fewer fans, cheaper tickets. Housing does not clear that way, for two reasons.

The supply side stopped building too. A shrinking place is not adding inventory. In our data, the flat-and-shrinking counties permit new housing at roughly a third the per-capita rate of growing ones. Scioto County, Ohio, is pulling permits for a literal handful of homes per thousand residents. When both demand and supply shrink together, the market doesn't loosen — it just gets quieter at the same price, or tighter. (Low permitting is partly an effect of weak demand, not only a cause of tight supply; the two feed each other.)

The people leaving aren't freeing up the homes buyers want. This is well-documented housing economics rather than something our county data proves directly: out-migration skews toward renters, younger adults, and lower-income households — people who were not occupying the owner-occupied, move-in-ready stock an inbound buyer is shopping for. A county can lose thousands of residents without a single additional desirable house hitting the market. The population leaves through the rental and starter-home door; the for-sale inventory an outside buyer wants barely budges.

The catch worth saying out loud

Two honest caveats keep this from being a fairy tale. First, these are nominal prices — the sticker number on Zillow. With inflation running a couple of points, a county up 1.5% in nominal terms is roughly flat, or slightly down, in real dollars. What we can say cleanly is narrower and more useful to a buyer: the number on the listing is not falling.

Second, 14% of shrinking counties did see prices drop — and where they dropped tells you something. Pike County, Kentucky, deep in coal country, fell 11.6%. Maui slid after the wildfire. And a knot of expensive California counties — Napa, Alameda — gave back a few percent, which is less "decline spiral" than the pandemic-boom hangover working itself out. Genuine economic collapse and a cyclical correction both show up as a minus sign; they are not the same animal, and neither describes the 86% where prices rose.

What this changes for anyone deciding where to move

The population headline is close to useless as an affordability signal. "This place is losing people" tells you nothing about whether homes there are cheap or getting cheaper — the San Franciscos and the Rockfords both lose residents, and both keep getting pricier. If you are hunting for value, skip the demographic trend line entirely and look at two things instead: the actual price-to-income math on the ground, and whether the local market is priced below what its fundamentals justify. A shrinking population is not a discount code.

It is the same trap in reverse from the one on the growth side, where the counties pulling in the most newcomers often turn out to be college towns that barely grow, and the places absorbing the most movers rank worst for home-price outlook. Population flow — in or out — is a weak predictor of what happens to prices. Supply, income, and what the housing is actually worth do the real work. If you want to see how we separate a cheap market from an undervalued one, that distinction is the whole point of valuing a market like a stock.

So the next time a headline tells you a state or a county is emptying out, resist the reflex to picture a bargain. Look at what the homes actually cost, and what anyone is building. The people can leave and the prices can climb — and across most of shrinking America right now, that is exactly what's happening.