Metro unemployment rates show uneven labor stress across 387 areas

The latest Bureau of Labor Statistics metro-area release points to a labor market that is cooling unevenly, not uniformly. In August 2026, unemployment rates were lower than a year earlier in 234 of 387 metropolitan areas, yet payroll employment rose year over year in only 9 areas and fell in 5, with 373 essentially unchanged, according to the BLS metropolitan-area release published September 30, 2026.

Those measures describe different things. The unemployment estimates count people by residence, while payroll employment counts jobs where establishments are located. Both series are not seasonally adjusted in this comparison, so local seasonal patterns matter. Still, the combination says the national headline can conceal pockets of weakness and pockets of expansion.

Related reading: August Payrolls Rebounded to 162,000, but June and July Added Just 52,000 Jobs.

Washington's job base remains the clearest large-metro pressure point in the latest print. The Washington-Arlington-Alexandria area had 66,000 fewer payroll jobs than in August 2025, a 2.0% decline, BLS reported on September 30 in its metro employment release. Portland-Vancouver-Hillsboro was down 18,800 jobs, or 1.5%, while Atlantic City-Hammonton lost 5,800, or 3.1%. Bloomington, Indiana, recorded the largest percentage decline among metros at 5.9%.

Houston-Pasadena-The Woodlands added 44,700 jobs over the year through August, the largest numerical gain in the BLS release. Minneapolis-St. Paul-Bloomington added 27,800 and Charlotte-Concord-Gastonia gained 21,400. Baton Rouge and Myrtle Beach-Conway-North Myrtle Beach led percentage gains at 3.2% each.

Related reading: State Unemployment in August Hid a 1.8 Point Spread Behind the 4.1 Percent Average.

A few metro areas are expanding quickly, but that does not describe the typical area: BLS classified 373 of 387 as essentially unchanged in payroll employment.

Unemployment rates add another layer. Cleveland's August rate was 1.8%, the lowest among metro areas, while El Centro, California, reached 21.9% and Yuma, Arizona, 19.0%, BLS said September 30 in the same release. These extremes often reflect seasonal industries and local labor-force composition, so they should not be read as direct rankings of job creation. Across all areas, 234 rates were lower than a year earlier, 126 were higher and 27 were unchanged.

Related reading: The Fed Hiked Into a Labor Market That Is Quietly Firming.

The county and city rows are not a substitute for the national series, but they help distinguish a broad slowdown from local adjustment. BLS reports the metro unemployment data from its Local Area Unemployment Statistics program and payroll data from Current Employment Statistics. The former follows residents, including people who may commute elsewhere, while the latter follows jobs at worksites. A metro can therefore show a change in payrolls that does not line up neatly with its resident unemployment rate.

For the Fed and national labor watchers, metro data are a distribution check, not a replacement for the monthly Employment Situation. The September national report released October 2 showed payrolls up 29,000, according to the BLS Employment Situation release. That weak national gain, alongside metro dispersion, argues against treating any one city or national average as a complete measure of hiring conditions. The metro series is also unadjusted, so annual comparisons are more useful than month-to-month swings.

Related reading: The Unemployment Rate Fell to 4.1 Percent for the Worst Reason: Workers Vanished.

The counterpoint is that the metro release does not yet show a broad wave of local job losses. Payroll employment declined in just 5 areas and increased in 9, while most were effectively flat year over year. Unemployment rates also declined in more areas than they rose. One cautious reading is a low-growth market with concentrated stress, rather than synchronized contraction.

Another month of national payroll gains and revisions will be needed to tell whether the latest softness spreads geographically.

There is also a measurement reason to avoid turning the most dramatic local figures into a national claim. The BLS release compares August 2026 with August 2025 and labels the metro estimates not seasonally adjusted. In a tourism-heavy place such as El Centro or Yuma, the level of unemployment can reflect recurring seasonal work as well as underlying demand. The annual comparison helps, but it does not remove all local composition effects.

For households, the divergence matters because the national unemployment rate does not describe every job search. A worker in a contracting federal-adjacent market faces a different set of openings from a worker in a metro adding payrolls. The August data do not identify the causes of each local gain or loss, so attributing Washington's decline to any one employer or policy would go beyond this release. BLS provides the geography and totals, not a causal breakdown.

The next dated BLS checkpoints are the September state employment and unemployment release on October 20, 2026, at 10:00 a.m. Eastern, and the September metro release on October 28 at 10:00 a.m. Eastern, per the BLS release calendar retrieved October 6, 2026. Until then, the main signal is dispersion: weak payroll centers coexist with growing metros, and area unemployment changes do not all point in the same direction.

This content is for informational purposes only and does not constitute financial advice. Past performance is not indicative of future results. Consult a qualified financial advisor before making investment decisions.