Moving Trends Forecast for 2027: Six Scenarios to Watch

2027 is more likely to produce several migration maps than one

The most defensible base case is continued selective redistribution: slower national population growth, more movement away from the largest counties, and continued gains in midsized cities and outer counties that combine housing supply with access to jobs or retirement destinations. That does not mean every Southern market will grow or every large city will shrink.

The latest Census estimates illustrate the split. From July 2024 to July 2025, counties with at least 1 million residents collectively lost 637,634 people through domestic migration, while counties with 50,000 to 999,999 residents gained 533,766. At the same time, lower international migration slowed many large metros, so a city can lose population even when its housing or labor market follows a different path.

Scenario 1: Southern growth continues, but the strongest gains move outward

What would drive it: employment near major job centers, new housing on the metro edge, warm-weather retirement demand and lower costs than many coastal origin markets. The Census Bureau found that 45 of the 50 fastest-growing counties from July 2024 to July 2025 were in the South. It also found that the fastest-growing counties around large metros were often on their outer edges.

Where it would show up: the Charlotte region, Raleigh-Cary and Wilmington in North Carolina; Myrtle Beach, Spartanburg and Jasper County in South Carolina; northern Atlanta counties such as Jackson and Dawson; the Dallas-Fort Worth and Houston outskirts in Texas; and Ocala, Lakeland-Winter Haven and Punta Gorda in Florida. Charlotte added 20,731 residents in the latest city estimates, yet nearby Fort Mill, South Carolina, grew 6.8%. Around Dallas, Celina, Princeton, Melissa and Anna were among the country’s fastest-growing cities. That is the pattern to expect if outward growth continues: the employment center remains important, but a larger share of growth lands beyond the core.

How the scenario could change: the trend would weaken where housing construction fails to keep up, commuting costs erase the price advantage, job growth slows, or insurance costs make ownership less affordable. Florida’s fastest-growing metros may therefore behave differently from the Carolinas or Texas even when all are grouped under a broad “Sun Belt” label.

Scenario 2: The Midwest shifts from population loss to retention and modest gains

What would drive it: a smaller cost gap between renting and owning, established employment and health-care centers, and fewer residents leaving. The Midwest recorded positive net domestic migration of about 16,000 from July 2024 to July 2025, its first positive result of the decade. Ohio gained 11,926 residents through domestic migration and Michigan gained 1,796, a sharp change from their 2021 losses.

Where it would show up: Columbus, Cincinnati and Cleveland in Ohio; Detroit and Grand Rapids in Michigan; Milwaukee in Wisconsin; Indianapolis in Indiana; and parts of the Chicago region. Columbus added 7,696 residents in the latest city estimates, and its unemployment rate had one of the largest year-over-year declines among major metros in June 2026. A durable version of this scenario would look more like stabilization and selective growth than a sudden nationwide rush to the Midwest.

How the scenario could change: Midwestern housing is not standing still. FHFA reported 4.4% annual home-price appreciation in the East North Central division through the first quarter of 2026, the strongest of the nine census divisions, and Illinois rose 7.3%. Continued price growth without stronger incomes or employment could narrow the region’s affordability advantage. Weak manufacturing or office employment could also restore domestic outflow.

Scenario 3: High mortgage rates keep moves suppressed, or a rate decline releases delayed demand

What would drive it: the cost of replacing an existing mortgage, the number of homes listed for sale and whether local prices are rising or falling. In June 2026, the Mortgage Bankers Association said it expected mortgage rates to average about 6.5% over its forecast horizon. If rates remain near that level, many owners with older, lower-rate loans will continue to have a financial reason not to sell. That reduces both local and interstate homeowner mobility.

Where it would show up: markets where prices have already softened or stopped rising quickly, including Austin, San Antonio, Phoenix, Raleigh, parts of Colorado’s Front Range, and some Florida Gulf Coast metros. FHFA reported that national house prices rose 1.7% over the year through the first quarter of 2026, but prices fell in eight states and the District of Columbia. Austin had the largest decline among the 100 largest metros, down 6.9%, while San Antonio was down 3.7% and Raleigh was down 1.0%.

Two possible outcomes: if borrowing costs stay high, job, family and renter moves are likely to make up more of the market while discretionary homeowner moves remain limited. If rates fall enough to change monthly payments materially, listings and purchases could rise first in places where inventory has rebuilt and prices have adjusted. A rate decline would not automatically restore the pandemic-era destination map; local jobs, insurance and housing supply would still determine which cities benefit.

Scenario 4: Jobs and generative AI matter more than broad regional labels

What would drive it: a highly uneven labor market. In June 2026, payroll employment rose over the year in only 15 of 387 metro areas, fell in three and was statistically unchanged in the other 369. Dallas-Fort Worth added 54,600 jobs, Las Vegas added 33,500 and Phoenix added 33,200. Among million-plus metros, Raleigh had one of the strongest percentage gains at 2.6%.

Where it would show up: continued employment-led demand in Dallas-Fort Worth, Las Vegas, Phoenix, Raleigh and Salt Lake City, and weaker relocation demand around Washington, D.C., and Portland if their job losses persist. Washington-Arlington-Alexandria was down 83,500 jobs, or 2.4%, from June 2025; Portland-Vancouver-Hillsboro was down 25,700, or 2.1%. These are preliminary monthly estimates, but the gap is large enough to make employment a central 2027 variable.

Why office policy will not produce a simple reversal: remote work has declined from its pandemic peak but remains embedded in the labor market. BLS reported that 22.6% of people at work teleworked at least some hours in March 2026, within a 21.5% to 23.0% range over the prior year. More office attendance could strengthen demand near major employment centers, while persistent hybrid work could preserve demand in outer suburbs and smaller cities within occasional commuting distance. The result depends on each metro’s industry mix, not a single national return-to-office rule.

Generative AI creates two opposite migration paths

The immediate 2027 question is not whether generative AI will eliminate knowledge work everywhere. It is whether employers use it mainly to automate tasks and reduce hiring or to augment workers and expand output. The evidence available so far points to both. A June 2026 Stanford Digital Economy Lab analysis found only modest employment differences between more- and less-AI-exposed occupations across workers of all ages. Among workers ages 22 to 25, however, employment in AI-exposed occupations was contracting at an annual rate of 3.8%, while the least-exposed occupations were growing at 2.0%. The researchers describe this as an early signal from a selected sample, not proof of economy-wide displacement.

BLS projections show why the outcome is unlikely to be a general collapse in high-paid work. From 2024 to 2034, BLS expects employment to grow 33.5% for data scientists, 28.5% for information security analysts and 15.8% for software developers. It expects declining employment for several office and administrative occupations, including customer service representatives, legal secretaries and claims adjusters. The likely split is therefore between work that AI can perform with limited human judgment and work needed to build, secure, integrate, supervise or apply AI in a specialized field.

If demand for high earners falls in expensive cities: the first effect would probably be fewer openings and smaller hiring classes in software, finance, consulting, media, design and other information-heavy functions, not the immediate departure of every established worker. San Francisco-San Jose, Seattle, New York, Boston, Washington, Los Angeles and Austin would be especially exposed because high housing costs overlap with large concentrations of knowledge work. Upper-end rental and condominium demand near employment centers could soften first. If weaker hiring becomes sustained job loss, restaurants, retail, office districts and local tax collections would feel a second-round effect. Some households would move to lower-cost metros or nearby outer counties, although low mortgage rates on existing homes could delay that response.

If AI creates more valuable work than it removes: many of the same expensive cities could strengthen. Stanford’s 2026 AI Index counted 17.2% of U.S. AI job postings in California, 8.1% in Texas and 6.6% in New York in 2025, roughly one-third of the national total across those three states. Capital, research institutions and experienced technical workers may keep new AI companies and their highest-paid jobs concentrated in the Bay Area, New York, Seattle, Boston and selected Texas metros. In that version of 2027, AI does not disperse high earners; it increases competition for a smaller group of specialized workers.

Moveline’s base prediction for 2027: generative AI will change who gets hired before it changes where millions of people live. The clearest early signs are likely to be weaker entry-level hiring in automatable knowledge work, continued growth in AI-complementary occupations, and wider differences inside the same metro. An expensive city could lose routine or junior positions while adding highly paid AI, data and security roles. That would make its labor and housing markets more polarized without necessarily causing a broad population decline.

Scenario 5: Lower international migration changes the apparent winners and losers

What would drive it: whether the steep decline in net international migration continues. Census estimates show that net international migration fell from 2.7 million in 2023-2024 to 1.3 million in 2024-2025. If the then-current trend continued, the agency projected approximately 321,000 by July 2026. International migration is not an interstate move, but it changes total population growth, rental demand, labor supply and the headline rankings often described as “moving trends.”

Where it would show up: New York, Los Angeles, Miami, Houston and Chicago, along with border metros such as Laredo, Yuma and El Centro. Florida, Texas, California and New York received the largest numeric international-migration gains among states in 2024-2025, so they also have the greatest exposure to another decline. New York City lost 12,196 residents in the latest city estimates after gaining population in the prior vintage; Census linked the broader large-city slowdown partly to lower international migration.

Two possible outcomes: if international inflows remain low, domestic outmigration will have a larger effect on large gateway metros and their growth may stay weak even when their job markets improve. If international migration rebounds, those cities could return to growth without any reversal in domestic migration. That distinction is essential: a gateway city’s total population can rise while it still loses residents to other U.S. counties.

Scenario 6: Insurance costs slow or redirect demand in high-risk markets

What would drive it: rising premiums, policy nonrenewals and the availability of coverage. A U.S. Treasury analysis of more than 246 million policy records found that homeowners in the 20% of ZIP codes with the highest expected climate-related losses paid average premiums 82% higher than residents of the lowest-risk ZIP codes. Their average nonrenewal rates were about 80% higher. The dataset covers 2018 through 2022 and excludes flood insurance, so it establishes a housing-cost pressure rather than proving a 2027 migration reversal.

Where it would show up: South Florida, Tampa Bay, Cape Coral and Punta Gorda; Houston and other Texas Gulf Coast communities; New Orleans and coastal Louisiana; and wildfire-exposed parts of California, including the edges of the Los Angeles region. Coastal growth markets such as Myrtle Beach and Wilmington also face this pressure even though their recent population numbers remain strong.

How the scenario could unfold: higher insurance costs may first redirect demand within a state or metro, toward lower-risk neighborhoods and inland counties, rather than cause a wholesale exit. Retirees and buyers on fixed budgets are likely to be more sensitive because insurance directly changes the monthly cost of ownership. Strong job growth or abundant new housing could offset that pressure in one metro while insurance becomes decisive in another.

The clearest 2027 forecast is greater divergence

The evidence available in August 2026 does not support one national migration winner. It supports a more uneven year: outward growth around selected Southern job centers, possible Midwestern stabilization, limited homeowner mobility unless rates fall, stronger differences between job markets as generative AI changes their occupational mix, weaker growth in gateway cities if international migration stays low, and increasing insurance pressure in high-risk housing markets.

The most important uncertainty is how these forces interact. Dallas-Fort Worth could keep gaining because jobs and housing offset higher costs. A Florida metro could keep adding residents while its growth rate slows because insurance reduces demand at the margin. New York could gain population through renewed international migration while continuing to lose residents domestically. Those are not contradictions; they are different components of population change moving at the same time.

How this forecast was built

The evidence cutoff is August 3, 2026. Population findings use Census Vintage 2025 estimates covering July 1, 2024, to July 1, 2025. Housing-price findings use FHFA data through the first quarter of 2026. Employment findings use preliminary BLS estimates for June 2026, and the telework measure is for March 2026. The AI discussion uses BLS 2024-2034 occupational projections, Stanford payroll indicators through April 2026 and 2025 AI-job-posting data. The insurance evidence covers 2018 through 2022 and was released by Treasury in January 2025.