America Is Getting Hotter. People Aren’t Leaving Because of It

The assumption has a certain intuitive force to it: temperatures rise, places become unbearable, people pack up and leave. A slow-moving version of the disaster movie, basically, where the climate drives the plot. It’s not an unreasonable story to tell. Except that, according to a new analysis of migration data across more than 3,000 US counties, it mostly isn’t happening. Not yet, anyway. What’s occurring is something rather more subtle, and in some ways more unsettling.

Researchers at Florida Atlantic University have been mapping the relationship between temperature anomalies and where Americans actually move, drawing on IRS tax-filing records from 2020 to 2022 alongside climate data from NOAA and the CDC. The headline finding cuts against a lot of climate journalism: hotter counties are not, on average, losing residents faster than cooler ones. If anything, the relationship runs the other way.

The study, published in Sustainability, found that a one-degree Fahrenheit increase in the local temperature anomaly was associated with roughly 0.34 fewer departures per 1,000 residents. That’s not a typo. Rising heat appeared to slightly suppress out-migration, rather than accelerate it. The most plausible explanation isn’t that people are content to swelter; it’s that heat traps them. Worsening conditions raise the costs of staying, but they also tend to coincide with economic hardship and tight housing markets that make leaving extremely difficult, leaving people stuck in place through something closer to constraint than choice.

The sharper effect showed up on the other side of the ledger. Extreme temperature anomalies, those reaching 2.6 degrees Fahrenheit or more above historical norms, were associated with roughly 1.9 fewer in-migrants per 1,000 residents. Places aren’t emptying; they’re becoming less attractive to newcomers. That’s a meaningful distinction. It changes how you think about the policy problem entirely.

“As extreme temperature anomalies increase, we don’t see more people leaving,” said Yanmei Li, senior author and an associate professor of urban and regional planning at FAU. “Instead, fewer people are moving in – especially to unusually hot areas – slowing population growth. It’s less about people being pushed out and more about places becoming less attractive. At the same time, consistently warm climates still draw people, highlighting a contrast between steady warmth and extreme heat.”

The Sun Belt Paradox

That contrast is, perhaps, the study’s most clarifying insight. The Sun Belt keeps growing. Florida, Texas, Arizona, Nevada: all of them show up in the net migration data as consistent gainers, despite registering some of the highest temperature anomalies in the country. People moving to Phoenix or the Tampa-Orlando corridor are apparently not much deterred by rising heat, at least when the job market is good and the housing is (comparatively) cheap. The Southwest and parts of the urbanized Northeast show the strongest warming, with roughly 394 counties now sitting more than 2.6°F above their 1901–2000 baseline and about 121 counties exceeding 3°F. The four counties with the largest anomalies, all around 3.7°F, are in Colorado, which doesn’t map neatly onto any simple narrative about warm-weather flight.

The statistical analysis deployed spline regression models to hunt for a temperature “tipping point,” a threshold beyond which migration patterns shift significantly. The models identified something around 2.6 to 2.7°F as the level at which the migration-temperature relationship begins to change slope. Below that, patterns are mostly flat. Above it, there’s a slight steepening of the curve; not dramatic, not yet statistically unambiguous, but present in the data. Whether that constitutes a genuine behavioral threshold or a quirk of the county-level averages is genuinely hard to say.

Who Gets Trapped

Income turns out to be the crucial variable threading through everything. In high-poverty counties, the relationship between temperature anomalies and out-migration flips: the interaction term is positive and strongly significant, meaning that poorer communities exposed to more heat really do show higher departure rates. It’s wealthier counties, with their adaptive buffers, air conditioning, insurance markets, and greater residential stability, where heat seems to suppress rather than accelerate movement. Medium-income households are likely the most mobile; they face enough environmental pressure to consider leaving and just about enough resources to act on it. This isn’t a uniform national story. It’s several different stories sorted by zip code.

The findings sit uneasily alongside some of the climate-migration literature, which has documented robust out-migration responses to heat in agricultural economies and lower-income settings, particularly in the Global South. The US context is different, buffered by infrastructure and economic complexity, which may be muffling signals that would otherwise come through more clearly. Li and her co-author Diana Mitsova are careful not to interpret the muted aggregate effects as meaning climate simply doesn’t matter for migration. The data are from 2021, a single cross-sectional year, and a period possibly too early for the full weight of cumulative exposure to register.

“The absence of strong effects today does not mean climate will remain a minor factor,” said Mitsova, who chairs FAU’s Department of Urban and Regional Planning. “Our findings suggest that stronger migration responses could emerge in the future, particularly as rising temperatures interact with extreme events, long-term exposure, or constraints such as housing availability and insurance markets. Potential ‘tipping points’ may still lie ahead.”

What the research does fairly robustly establish is that housing and economic structure shape climate responses at least as much as temperature itself. Counties with high vacancy rates show lower overall mobility. Counties with high housing costs show more movement in both directions. Homeownership rates are strongly associated with in-migration. None of that is surprising on its own, but it suggests something important: the effects of climate stress arrive filtered through systems, insurance, mortgages, labor markets, that can amplify, delay, or entirely suppress the underlying signal. By the time heat shows up as a migration statistic, it may have traveled through a housing market crash, an insurance withdrawal, or a sequence of drought-driven job losses first.

The image that stays with you from the research isn’t the Sun Belt boomtown or the slowly emptying rural county. It’s what the authors call the “trapped population”: households in high-risk, high-heat, low-income areas who cannot move not because they don’t want to, but because nothing in the economic landscape makes movement possible. Immobility, in this framing, is not resilience. It is the shape that vulnerability takes when it runs out of options. Whether that changes as anomalies climb toward and past the three-degree mark may be the most important demographic question of the next few decades.

DOI: 10.3390/su18042040

Frequently Asked Questions

If heat is making places unliveable, why aren’t people just moving away?

Moving is expensive and disruptive, and the economic conditions that tend to accompany heat-stressed areas, tight budgets, limited job markets, restricted housing, often make relocation less feasible rather than more. Researchers call these immobile households “trapped populations”: people who face worsening conditions but lack the resources to leave. The study found that in wealthier counties, rising heat actually appeared to slightly suppress out-migration, likely because residents could adapt in place rather than relocate.

Why do places like Phoenix and Florida keep growing if they’re getting so much hotter?

The study found a key distinction between baseline warmth and anomalous heat. Historically warm climates continue to attract migrants because of jobs, lower costs, and amenities; what slows growth is extreme departures from what people already expect. Phoenix residents know it’s hot; what matters is whether conditions cross into genuinely unusual territory. The research suggests that threshold may be around 2.6 to 2.7 degrees Fahrenheit above historical norms, after which destination attractiveness begins to decline.

Is there a point at which heat really will start driving mass migration in the US?

The researchers found tentative statistical evidence of a tipping point around 2.6 to 2.7°F above baseline, where the migration-temperature relationship begins to shift, though the effect remains small. About 394 US counties have already crossed that level. The authors are clear that the absence of strong effects in 2021 data doesn’t mean climate will stay a minor factor; as cumulative exposure grows and extreme events compound, the response could become considerably larger.

Who is most at risk from climate-driven population shifts?

Low-income communities face a double burden. In high-poverty counties, rising temperatures are associated with higher out-migration, suggesting that vulnerable populations may be pushed out when conditions deteriorate. At the same time, poorer households that can’t leave face growing exposure without the resources to adapt. The study found that the effects of temperature stress are shaped as much by socioeconomic context as by temperature itself, making inequality a central variable in any serious climate adaptation policy.