With Less Immigration, Urban Growth Slowed in 2025

Plus, revisions to 2022-2024 local population estimates were unusually large.

Urban counties grew more slowly than suburban counties, smaller metropolitan areas, and rural areas in 2025. All types of counties grew more slowly in 2025 than in 2023 and 2024 because immigration fell. However, the slowdown was especially sharp for urban counties. The only other years this century when urban counties lagged all other types of places were in 2020-2022, when people moved out of cities during the pandemic, and in 2003-2006, when the housing bubble boosted housing construction in outlying suburban areas and drew new residents from cities. (This is based on vintage 2025 Census population estimates, released this morning. See how I classify counties here.)

Urban counties rely more on immigration for growth than other places do. In 2025, even with less immigration nationally, international migration was the main source of urban population growth, and international migration contributed more to population growth in urban counties than in any other places. At the same time, urban counties lose population due to domestic migration, whereas most other types of places gain. Without immigration, urban counties would shrink.

Recent population trends in urban counties have tracked national immigration patterns. Looking just at these counties, immigration contributed twice as much to population growth in 2023 and 2024 as was typical in the 2010s, then added much less population in 2025: this reflects the timing of the national immigration surge. But domestic migration follows a different pattern: domestic out-migration from urban counties has been increasing steadily since the early 2010s, accelerating dramatically during the pandemic as people moved from cities but then returning to the pre-pandemic trend by 2024 and 2025.

Immigration doesn’t appear to compete with domestic migration: that is, domestic migration into urban counties didn’t fall during the immigration surge of 2022-2024 and then rise when immigration slowed in 2025. Rather, domestic migration is driven more by affordability, with people leaving expensive urban counties for more affordable suburban areas and smaller communities.

While domestic out-migration from urban counties has recovered from the pandemic, out-migration has accelerated in a few metros. The rate of people leaving metro Miami to other places in the U.S. was higher than that of any other large metro in 2025. This is a reversal: during the pandemic, the rate of out-migration from Miami was relatively modest compared with metro New York and San Francisco, which experienced outflows (including to Miami!). But domestic out-migration has moderated in New York and San Francisco since the pandemic, whereas Miami is the new San Francisco – at least in the sense that housing affordability is pushing people out.

Under the hood: 2025 revisions changed 2022-2024 local immigration estimates

Trends in local population growth hinge on immigration patterns. Today’s release included an important methodological change that affected earlier immigration estimates this decade, revising some local populations upward and others downward, in many cases by a lot. As the Census Bureau explained when it released the national and state population and immigration estimates in January, they revised how “humanitarian migrants” from 2022 to 2024 were distributed geographically to align more with immigration court filing data rather than with survey response data that missed many of these migrants. The state data showed upward revisions to immigration estimates in New York and downward revisions in Florida, for instance. But large states hide important local variation, and the revisions are starker at the metro level, as today’s data reveal.

Immigration estimates were revised upward in places where the 2022-2024 surge was large relative to immigration through traditional pathways, such as in border areas like Laredo TX and widely reported immigrant destinations like Denver and Springfield, OH. Estimates were revised downward in places far from land border crossings, like Honolulu and Miami, and in tech hubs and college towns where immigrants tend to enter through traditional pathways. In places with large revisions where immigration contributes meaningfully to population growth, this somewhat obscure methodological revision added or subtracted as much as 1% to the total local population – which is bigger than the magnitude of annual population change in most metros. These local population estimates inform local planning as well as federal funding allocations; big data revisions have real effects.