US State Unemployment July 2026: Jobs Data Explained

Ten states had lower unemployment rates, but broad payroll growth was absent

The U.S. Bureau of Labor Statistics reported that unemployment rates fell significantly in 10 states in July 2026 and were statistically stable in 40 states and the District of Columbia. The national unemployment rate was 4.1%, little changed over both the month and the year.

The payroll survey delivered a narrower result. Nonfarm payroll employment increased significantly in one state, decreased in one and was essentially unchanged in 48 states and the District. Maryland added 11,700 jobs, or 0.4%, while New Jersey lost 25,600 jobs, or 0.6%.

This combination is not contradictory. The unemployment rate and payroll employment come from different statistical programs, measure different things and can move differently in the same month.

The report combines two surveys with different definitions

State unemployment rates are modeled largely from the household survey and describe people according to where they live. A person is unemployed only when they do not have a job, are available for work and have actively looked for work under the survey definition.

State payroll estimates come from the establishment survey and count jobs according to where employers are located. One person with two payroll jobs can be counted twice in the payroll measure, while self-employed workers are generally outside the nonfarm payroll total.

Because the sources, samples and concepts differ, a state's unemployment rate may fall even when payroll employment is statistically unchanged. Reading the two measures together gives a fuller picture than treating either as a complete scorecard.

A falling unemployment rate can have more than one explanation

An unemployment rate can decline because unemployed people find work. It can also fall when some people stop looking and leave the labor force, because people outside the labor force are not counted as unemployed. Population changes and statistical revisions can affect the rate as well.

That is why a lower unemployment rate should be checked against employment levels, labor-force participation and payroll data. The July state release identifies statistically significant changes, but it does not claim that every numerical movement represents a real change rather than normal sampling variation.

The 10 significant monthly declines were in Delaware, Florida, Illinois, New York, North Dakota, Ohio, Pennsylvania, Rhode Island, South Carolina and Washington. Seven fell by 0.2 percentage point and three fell by 0.1 point.

State unemployment ranged from 2.0% to 5.9%

South Dakota had the lowest seasonally adjusted state unemployment rate in July at 2.0%. North Dakota followed at 2.2%, while Vermont was at 2.6%. At the other end, the District of Columbia was highest at 5.9%; Connecticut and Oregon were at 5.2%, and California was at 5.1%.

Sixteen states had rates significantly below the national 4.1% rate. Nine states and the District had rates significantly above it, while 25 states were not appreciably different from the national figure.

These rankings describe labor-market conditions, not economic quality in isolation. Industry composition, migration, demographics, participation, housing costs and the speed at which people enter or leave the labor force differ substantially across states.

Maryland and New Jersey were the only significant monthly payroll moves

Maryland's estimated nonfarm payroll total rose by 11,700 in July, making it the only state with a statistically significant monthly increase. New Jersey's total fell by 25,600, the only significant decrease. The other 48 states and the District were classified as essentially unchanged.

Essentially unchanged does not mean that every employer kept the same headcount or that the published estimate moved by exactly zero. It means the measured change was not large enough, relative to the estimate's uncertainty, for BLS to identify it as statistically significant.

Monthly state estimates can be volatile and are preliminary. A single release should therefore be treated as a snapshot that later revisions and additional months may strengthen, weaken or reverse.

Annual job growth was concentrated in a small group of states

Compared with July 2025, payroll employment increased significantly in six states, decreased in Virginia and the District of Columbia, and was essentially unchanged in 43 states. Texas recorded the largest numerical gain at 165,600 jobs, followed by California at 112,700 and North Carolina at 51,600.

Minnesota had the largest percentage increase, 1.4%, followed by Louisiana and South Carolina at 1.3% each. Virginia lost 47,900 jobs, or 1.1%, while the District lost 31,500, or 4.2%.

Levels and percentages answer different questions. Large states can add more jobs in absolute terms because their employment bases are larger, while smaller states can post faster percentage growth from much lower starting totals.

Year-over-year unemployment changes pointed in both directions

Twelve states had significantly higher unemployment rates than a year earlier. Connecticut recorded the largest increase, up 1.3 percentage points, and Oklahoma followed with a 1.0-point increase.

Ten states and the District had lower rates than in July 2025. New Jersey and Ohio recorded the largest declines, down 1.1 percentage points each. The contrast in New Jersey is instructive: its unemployment rate improved over the year even though the July monthly payroll estimate fell significantly.

Different comparison periods can tell different stories. A one-month payroll decline can occur inside a longer improvement, and a lower unemployment rate can coexist with weak recent hiring.

What the state report can—and cannot—tell financial markets

State data help reveal whether national labor conditions are geographically broad or concentrated. That matters for regional banks, consumer demand, housing, tax receipts and businesses with exposure to particular states.

The release is not a direct trading signal. It arrives after the national employment report, contains preliminary estimates and can be revised. Interest rates and asset prices respond to a larger set of information, including national hiring, wages, inflation, productivity and central-bank communication.

A careful reading therefore treats the state release as a distribution map: it shows where conditions differ, not what every stock, bond or currency must do next.

A practical checklist for reading the next state jobs report

Start with the release date and confirm whether the estimates are preliminary. Separate the unemployment-rate section from the payroll section, then distinguish monthly changes from annual changes. Look for BLS language such as significantly different and essentially unchanged instead of ranking every small numerical move.

Compare job levels with percentage growth and read labor-force participation when available. Examine several months, because one observation can be noisy. Finally, check the technical note and subsequent revisions before building a strong economic conclusion.

BLS scheduled the August 2026 state report for September 18, 2026. The July estimates may be revised when that report is published.

Editorial boundary

This article reports and explains public labor-market statistics. It separates the official observations from interpretation and does not forecast a recession, interest-rate decision or market return.

The information is for education only. It is not a recommendation to buy, sell or hold any security, fund, currency, commodity or other financial product.