Job openings are one of the most quoted labor-market numbers, and one of the most easily misread. In July, U.S. employers reported 7.3 million openings, according to the U.S. Bureau of Labor Statistics. In the same month, 5.1 million people were hired. Both figures were little changed from June. Put side by side, they invite a tempting conclusion: that there are 2.2 million unfilled jobs waiting for workers. The data do not support that reading.
Openings measure advertised employer demand at a point in time. They are useful, but on their own they do not tell workers how easy it is to get hired or tell employers how hard it will be to fill a role. A clearer picture comes from reading openings as one signal among several.
A stock and a flow are not the same kind of number
The Job Openings and Labor Turnover Survey (JOLTS) counts an opening if a position is open on the last business day of the month. That makes openings a stock, a snapshot on one day. Hires are a flow: all additions to payrolls during the entire month. A position can be posted and filled within the month and never appear as an opening at month’s end. Another can stay open for months, counted in each snapshot, while producing no hire at all.
Subtracting hires from openings therefore does not measure a shortage of workers. Nor can aggregate openings be converted into any individual applicant’s odds of getting a job. The two series move for different reasons and should be read in rates and over time.
On that basis, July’s numbers look like this. The openings rate, openings as a share of employment plus openings, was 4.4%, compared with 4.3% in July 2025. The level was about 7.27 million, down from about 7.59 million in April. The hires rate was 3.2% (preliminary), down from 3.4% in June and 3.3% a year earlier. In other words, advertised demand is roughly where it was a year ago, while the pace at which people actually move into jobs has eased.
The signals worth reading together
No single series answers the question employers and workers care about most: are advertised opportunities turning into jobs? A more useful stack of signals includes:
- Hires and the hires rate. The most direct measure of completed hiring. July: 5.1 million, 3.2%.
- Quits. Voluntary departures are often read as a gauge of workers’ confidence about finding something better. July: 3.1 million, a rate of 1.9%, versus 2.0% a year earlier.
- Layoffs and discharges. The clearest sign of employer-initiated job loss. July: 1.7 million, a rate of 1.0%, versus 1.1% a year earlier.
- Net payroll change. The Employment Situation report showed payrolls up 162,000 in August, following revised gains of 31,000 in June and 21,000 in July.
- Pay. Average hourly earnings rose 3.1% over the year to August. The Federal Reserve Bank of Atlanta’s Wage Growth Tracker, which measures the median wage growth of individual workers observed 12 months apart, was 4.1% in August, up from 3.8% in July. These are different measures of different things: a mean for all private payroll employees versus a median for people employed in both periods. Neither is adjusted for inflation.
- Sector composition. Aggregates hide divergence. In July, openings rose by 76,000 in durable goods manufacturing while hires fell by 188,000 in professional and business services.
Read together, these signals are more informative than openings alone. Openings near last year’s rate, combined with lower hires and quits and low layoffs, point to a market where demand has not collapsed but movement between jobs has slowed. That is a description of flows, not a forecast.
National signals are not local signals
JOLTS headline figures are national. They provide context for a city or region, not a direct measurement of it. Local labor markets have their own data, with their own definitions and publication schedules.
For metro Atlanta, the latest BLS metropolitan data cover July. The Atlanta-Sandy Springs-Roswell unemployment rate was 3.2%, not seasonally adjusted and preliminary, compared with 3.4% in July 2025. Nonfarm payroll employment in the metro was 3,107,300 in July, down 3,300, or 0.1%, from a year earlier. Those two series, a lower unemployment rate and essentially flat payroll employment, do not tell the same story, and neither can be read off the national openings figure. For comparison on the same unadjusted basis, the national unemployment rate in July was 4.4%.
The practical point is simple: a national openings number should not be used as a proxy for recruiting conditions in Atlanta, or anywhere else, without local data alongside it.
What it means, and what to watch next
For employers, an opening is an intention. What matters for planning is how long comparable roles take to fill, how many offers are accepted and whether turnover is rising or falling. The national data suggest lower quit rates than a year ago, which can ease backfill pressure. They say nothing about any particular occupation’s candidate pool.
For workers, a high count of openings does not mean hiring is fast. The median duration of unemployment was 11.4 weeks in August, compared with 9.9 weeks a year earlier. Postings are a starting point; the hires rate is the better guide to how much movement is actually happening.
Openings also have limits as a measure. The JOLTS figures are survey estimates, the latest month is preliminary, and BLS revises data each year when it incorporates annual updates. BLS described July’s hires, which fell by about 278,000 from June, as little changed.
JOLTS data for August are scheduled for September 29. The key question is whether the hires rate recovers from July’s 3.2% or stays there while openings hold near 4.4%. The September Employment Situation report follows on October 2. For Atlanta, August metropolitan data are due October 28. Watching all of these together, rather than the openings headline alone, gives a better answer to whether advertised demand is becoming jobs.
Sources
- U.S. Bureau of Labor Statistics, “Job Openings and Labor Turnover — July 2026,” 2026-09-01. https://www.bls.gov/news.release/jolts.nr0.htm
- U.S. Bureau of Labor Statistics, “JOLTS, Table 1: Job openings (July 2026),” 2026-09-01. https://www.bls.gov/news.release/jolts.t01.htm
- U.S. Bureau of Labor Statistics, “JOLTS, Table 2: Hires (July 2026),” 2026-09-01. https://www.bls.gov/news.release/jolts.t02.htm
- U.S. Bureau of Labor Statistics, “JOLTS, Table 4: Quits (July 2026),” 2026-09-01. https://www.bls.gov/news.release/jolts.t04.htm
- U.S. Bureau of Labor Statistics, “JOLTS, Table 5: Layoffs and discharges (July 2026),” 2026-09-01. https://www.bls.gov/news.release/jolts.t05.htm
- U.S. Bureau of Labor Statistics, “The Employment Situation — August 2026,” 2026-09-04. https://www.bls.gov/news.release/empsit.nr0.htm
- U.S. Bureau of Labor Statistics, “Employment Situation, Table A-12 (August 2026),” 2026-09-04. https://www.bls.gov/news.release/empsit.t12.htm
- Federal Reserve Bank of Atlanta, “Wage Growth Tracker,” 2026-09-10. https://www.atlantafed.org/research-and-data/data/wage-growth-tracker
- U.S. Bureau of Labor Statistics, “Metropolitan Area Employment and Unemployment — July 2026,” 2026-09-02. https://www.bls.gov/news.release/metro.nr0.htm
- U.S. Bureau of Labor Statistics, “Metropolitan Area Employment, Table 1 (July 2026),” 2026-09-02. https://www.bls.gov/news.release/metro.t01.htm
- U.S. Bureau of Labor Statistics, “Metropolitan Area Employment, Table 3 (July 2026),” 2026-09-02. https://www.bls.gov/news.release/metro.t03.htm
- U.S. Bureau of Labor Statistics, “Release calendar, October 2026,” 2026-09-24. https://www.bls.gov/schedule/2026/10_sched_list.htm
Featured photo: A "Now Hiring" sign in Pittsboro, North Carolina. A job opening signals demand; it is not a completed hire. Photo: DiscoA340 (photo; sign logo by Chatham County Government), via Wikimedia Commons (CC0). Resized from the original.
