In August, the United States added 162,000 non-farm jobs, which on the surface appears to be a solid national performance report. However, state-level data released by the U.S. Bureau of Labor Statistics on September 18th showed that the growth was not widespread: only California, Wisconsin, South Carolina, and New Mexico experienced statistically significant increases in non-farm employment, while the remaining 46 states and the District of Columbia saw virtually no significant changes.
Monthly changes in unemployment rates are also concentrated. Unemployment rates have significantly decreased in 8 states and the District of Columbia, while they remain relatively stable in 42 states, with the national unemployment rate staying at 4.1%. South Dakota has the lowest rate at 2.0%, followed by North Dakota at 2.2%; the District of Columbia has the highest rate at 5.7%. Ohio’s unemployment rate of 3.3% has reached a new low since 1976 when records began to be kept for that state. Although the national overall figure appears calm, the supply and demand for labor across the country are at different stages.
In August, California saw the largest increase in job positions, with 39,400 new jobs; Wisconsin added 11,800, South Carolina 10,800, and New Mexico 5,500. In terms of percentage increase, New Mexico had the fastest growth at 0.6%, followed by Wisconsin and South Carolina both at 0.4%, and California at 0.2%. These are changes that meet statistical significance criteria, but it does not mean that other states did not experience any job growth at all; it's just that the estimated changes in those states are not significant enough to be confirmed due to sampling errors.
Behind the stability across the country are both regional trends and statistical methods at play.
On a year-on-year basis, the unemployment rate decreased in 14 states and the District of Columbia, increased in 10 states, and remained relatively unchanged in 26 states. New Jersey saw a decrease of 1.2 percentage points, Ohio by 1.1 percentage points; Connecticut saw an increase of 1.0 percentage points, and Oklahoma by 0.9 percentage points. The unemployment rates in California, Connecticut, and Oregon were all at 5.1%, which is significantly higher than the national average; whereas states such as Georgia, Indiana, and Iowa had rates significantly lower than the national average.
Employment and unemployment rates come from two different sets of surveys. State-level labor force, employed population, and unemployed population data are primarily based on household surveys and estimated through models, and are categorized according to the residents' place of residence; non-agricultural employment data, on the other hand, comes from institutional surveys and is counted based on job positions at the location of the enterprises. If a person lives in one state but works in another state, they may be recorded in different locations in both sets of data. When part-time workers have two jobs, the household survey counts them as one employed person, while the institutional survey may count them as two job positions.
Therefore, even though the number of job positions in a state may increase, the unemployment rate does not decrease accordingly. An increase in labor force participation brings more job seekers, and population migration also affects the denominator. Conversely, a decline in the unemployment rate sometimes results from workers ceasing to look for jobs, rather than necessarily from increased hiring. To understand local labor markets, one must consider employment opportunities, the unemployed population, labor force participation rates, and wages simultaneously, rather than relying solely on the unemployment rate as a single indicator.
Only 8 states saw a significant increase in non-farm employment year-on-year, while the District of Columbia saw a decline, with the remaining 42 states remaining relatively stable. The states with the largest absolute increases in jobs were Texas with 159,400, California with 138,500, and North Carolina with 65,600; the states with the fastest growth rates were Louisiana, New Mexico, and South Carolina, all at 1.6%. The District of Columbia experienced a decrease of 27,000 jobs, a decline of 3.6%, making it the only region to show a significant year-on-year decrease.
Absolute numbers and percentages tell different stories. California and Texas have large populations and employment bases, so even if their growth rates are average, they can still contribute a significant number of jobs; New Mexico and South Carolina, with smaller bases, can achieve higher percentages with less new employment. When comparing the performance of states, one should first determine whether the question is “who contributes more jobs to the nation” or “who expands relatively faster compared to their own base,” and not confuse the two lists.
For interest rates and corporate decision-making, "no significant changes" is in itself an important signal.
There were no significant changes in monthly non-farm employment across 46 states, indicating that the growth in hiring in August was concentrated in a few regions rather than a nationwide simultaneous acceleration. This does not equate to an immediate economic recession. Mature labor markets are naturally capable of slowly adding jobs even with low unemployment rates, and state-level samples are also more susceptible to errors compared to national samples. However, when companies face uncertainties in financing, energy, wages, and demand, it is common for them to pause expansion rather than lay off employees immediately as a way to cool down the situation.
For the Federal Reserve, a national unemployment rate of 4.1% and 162,000 new jobs still indicate that the labor market is not broken, while regional disparities suggest that demand is not evenly distributed. If a small portion of the population and major industrial states bear most of the new job creation, the impact of policies on different regions will vary. Interest rate cuts may alleviate the financing pressures on real estate and small businesses, but they will not automatically erase the regional differences in employment in technology, manufacturing, tourism, and government sectors.
When selecting a location for a business, one cannot solely rely on the lowest unemployment rate. An unemployment rate of 2% may indicate a tight labor supply and high recruitment costs; a rate above 5% might suggest an abundance of talent, but it could also reflect weak local demand. State-level data needs to be considered in conjunction with industry composition, migration patterns, wages, and productivity. California, for example, has seen the largest increase in job creation, yet its unemployment rate remains high, which is an example of the simultaneous effects of scale, industrial restructuring, and labor supply dynamics.
The industry structure will further amplify these differences. Energy-producing states are affected by oil and gas investments and prices, while tourist states are more susceptible to seasonal fluctuations in visitor flow. Regions with large state governments, universities, or military facilities may be dominated by the public sector. When a national interest rate change affects local economies, it can lead to different outcomes in housing, manufacturing, technology financing, and fiscal budgets. Therefore, state-level data is an indispensable supplement to national reports.
This month's state-level estimates are still preliminary and will be revised later on. The Bureau of Labor Statistics also reminds that although some figures may seem to show significant changes, they may not pass the significance test. The most cautious conclusion is not that "employment in the United States is booming across the board" or that "most states have come to a standstill," but rather that the national market continues to grow, although the new momentum is highly concentrated. Only if the number of jobs created continues to decrease in states with significant increases and the number of regions showing year-on-year declines expands will that constitute clearer evidence of weakening.
