The Labor Department's unemployment rate has been edging lower in recent months, reversing last year's uptick—but an alternative gauge paints a very different picture of the labor market.

In July, the official unemployment rate dipped to 4.1%, down from 4.2% in June and 4.5% in November. The August jobs report is due out this Friday, and Wall Street expects the rate to hold steady at 4.1% while payrolls expand by 50,000, rebounding from a surprise loss of 23,000 in July.

Even though job gains have been weak lately, the Labor Department's measure of joblessness has kept falling as retiring baby boomers and President Donald Trump's immigration crackdown shrink the overall labor force. That is because the official rate counts people as unemployed only if they are actively looking for work—so a shrinking labor pool can push the rate down even when hiring is soft.

The breakeven rate of employment growth—the number of net new jobs needed each month to keep the unemployment rate steady—actually turned slightly negative during the summer and fall of 2025. Economists expect that to happen again in 2028, meaning the economy would have to shed workers just to keep unemployment steady.

At the same time, jobless claims have remained low, sustaining a low-hire, low-fire labor market as businesses stay cautious amid Trump's tariffs and the war on Iran.

The official unemployment rate is so low that Federal Reserve policymakers view it as a signal that the economy is at or near full employment. Fed Chairman Kevin Warsh said as much during his speech in Jackson Hole, Wyo., on Friday. As a result, the Fed's attention is now fixed on fighting inflation rather than on the other half of its dual mandate: supporting the labor market. The Fed's statutory mandate, set by Congress, requires it to pursue both maximum employment and price stability—so which labor-market signals officials rely on directly shapes which side of that mandate gets priority.

The Ludwig Institute for Shared Economic Prosperity (LISEP), however, takes a far less rosy view of the workforce. Founded by Gene Ludwig, a former U.S. Comptroller of the Currency, the think tank focuses on the economic well-being of middle- and lower-income households. Its True Rate of Unemployment tracks the "functionally unemployed"—a group that includes the jobless, those involuntarily working part-time, and those earning a poverty wage.

That measure posted its fourth consecutive increase in July, in contrast to the official jobless rate, which has been steadily declining this year. The share of the labor market that is functionally unemployed now stands at 24.9%, up 1.3 percentage points since March.

Similarly, LISEP's measure of the percentage of the working-age population that is not functionally employed—including those who dropped out of the labor force—reached 53.8%, up 0.8 percentage points since the start of the year.

"Functional unemployment is moving higher while workforce participation is moving lower. If this continues, it would suggest the labor market is losing strength despite what we may see in the headline unemployment numbers," LISEP Chairman Gene Ludwig said in a release on Aug. 20.

The functional unemployment rate for Black workers was flat at 27.3% last month. It rose 0.6 percentage points to 23.8% for white workers, while it fell 1.5 percentage points to 26.7% for Hispanics.

Among men, the rate dropped 0.9 percentage points to 19.5%. For women, however, it jumped 1.6 percentage points to 31%—the highest level since March 2021, when the economy was still recovering from the COVID shock.

Some of these demographic divergences may reflect crosscurrents in the economy. The AI boom has generated massive demand for construction workers and skilled trades—traditionally male-dominated jobs. Meanwhile, a crisis in family-care services has forced many women to pull back from their careers.

"In a strong labor market, good jobs and rising wages should bring more people into the workforce, not fewer," Ludwig said. "We need to pay attention when that starts moving in the other direction. It could be a sign that people aren't finding the opportunities they want or need, which matters for the broader economy."

The widening gap between the two measures is part of a longer-running debate among economists and policymakers over whether headline statistics capture the quality, not just the quantity, of jobs in the economy. What to watch next is whether the divergence persists in coming months' LISEP readings and whether it influences how Fed officials characterize labor-market conditions in future policy discussions.

This story was originally featured on Fortune.com