Unemployment by county · BLS LAUS · August 2026 preliminary
What is the unemployment rate where you live?
Pick a state, then a county. You get the share of its labor force that was unemployed in August 2026, the same month a year earlier so the season is taken out, and the state and the nation on the same basis. Every figure is a published Bureau of Labor Statistics number, with the counts behind it so you can check the division.
A real page, in one line
Harris County, TX, August 2026:
126,999 unemployed ÷ 2,548,851 in the labor force × 100 = 4.98%, published as 5.0%; a year earlier it was 5.0%.
The Harris County page shows the last 13 months, the state and the United States beside it, and what the figures leave out.
Informational only, not economic, financial or career advice. The pages present published statistics; they do not forecast or advise.
Counties by state
The United States and every state on the same month and the same not-seasonally-adjusted basis as the county pages. States are alphabetical, not ranked.
State figures are BLS's own statewide LAUS estimates, not seasonally adjusted; the national figure is the Current Population Survey's, on the same basis. Puerto Rico's municipios are included because BLS publishes them in the same county file on the same basis.
Methodology
Source and formula. Every county, state and national figure is a published estimate from the U.S. Bureau of Labor Statistics: Local Area Unemployment Statistics (LAUS) for counties and states, and the Current Population Survey for the United States, all not seasonally adjusted. Unemployment rate = unemployed ÷ civilian labor force × 100. The pages recompute it from the published counts and print the published rate beside it; nothing is modelled or estimated here. Snapshot: August 2025 to August 2026, read from the BLS flat files on 2026-10-08; the site refreshes it monthly.
Why the year-earlier month, not last month. BLS seasonally adjusts state and national series but not county series. A county's rate moves with the calendar every year, so a change from one month to the next is mostly season. The same month a year earlier removes that, which is why it sits beside the latest month on every page and why there is no month-over-month figure.
Preliminary month. August 2026 carries BLS's preliminary flag. The newest month BLS publishes is preliminary and is revised when the following month is released; the revision is usually a tenth or two. Each March, BLS also revises the prior years for every area.
County figures are model-based. BLS does not survey each county. State estimates come from a statistical model that combines the Current Population Survey (a household survey), the Current Employment Statistics survey of employers, and counts of unemployment insurance claims; county estimates are built from the same inputs and constrained to sum to the state, and the states to the nation. They are the official figures, and they carry sampling and model error that BLS does not publish at the county level.
What these pages cannot tell you
- "Unemployed" means actively looking. A person counts as unemployed only if they had no job, were available for work and had looked for work in the previous four weeks. People who have stopped looking are not in the labor force and are in neither figure.
- Small counties move a lot. In a labor force of a few thousand, a hundred people changes the rate by several points; a single plant closing or a seasonal workforce arriving shows up as a swing. The counts on the page are there so the size of the base is visible.
- Where people live, not where they work. LAUS counts residents. A county whose residents commute elsewhere shows their employment, not its own employers'.
- No ranking. Model error and seasonal patterns differ by county, so an ordered list of counties would compare noise. Each page carries its state and the nation for scale and nothing else.
- October 2025 has no figure. Household survey data were not collected during the 2025 lapse in federal appropriations, and BLS has not estimated the month; 2025 annual averages are 11-month averages for the same reason.
Informational only, not economic, financial or career advice. Last reviewed: October 2026. Data: BLS LAUS and CPS, August 2026 (preliminary), pulled 2026-10-08.
Frequently asked questions
Why compare with a year ago instead of last month?
Because county figures are not seasonally adjusted. BLS removes the regular calendar pattern from state and national series, but not from counties, so a county's rate rises and falls every year with harvests, tourism, school calendars and holiday hiring. A move from one month to the next is mostly that pattern. The same month a year earlier has the same season in it, so the difference between the two is the part that actually changed.
What does "preliminary" mean on the latest month?
BLS publishes each new month as preliminary and revises it when the following month is released, usually by a tenth or two of a point. The pages mark the preliminary month and say so. Each spring BLS also revises the prior years for every area, so the year-earlier figure on a page can change at the next refresh.
Where do county figures come from if nobody surveys each county?
From a model. BLS estimates each state from the Current Population Survey of households, the Current Employment Statistics survey of employers and unemployment insurance claims, then builds county figures from the same inputs so they add up to the state. They are the official statistics, but they are estimates with error BLS does not publish at the county level, and small counties swing more than large ones.
Why is there no list of the highest and lowest unemployment counties?
Because an ordered list would mostly compare noise. A county of 4,000 workers moves a full point when 40 people change status, and the model error differs from county to county. Each page shows its own county next to its state and the United States for scale, with the counts behind the rate, and leaves ranking out.
Related tools
See which jobs employ the most people in a city, or which cities have the most jobs of one kind — official Bureau of Labor Statistics job counts, not estimates scraped from job postings.
Open tool →Cost-of-Living Salary CalculatorSee what a salary in one state is worth in another once local prices are counted — using the federal government's own price index (BEA Regional Price Parities) and one division you can check by hand.
Open tool →Metro ComparisonSame job, different city: see how pay for your occupation changes across metro areas before you relocate or take a remote offer.
Open tool →