Silos, Stagnation, and Soccer. Also Known As: The May 2026 Jobs Report

By Jillian O'Malior, Founder & CEO, New World Labs

PUBLISHED: June 8, 2026

By this point, we’ve all likely seen the headlines regarding the Jobs Report for May. Namely, that unemployment has remained stable in the 4.3-4.5% since July 2025, that payroll grew 3.4% year-over-year with a 12 cent increase on average hourly wages, and most notably, an addition of 172,000 jobs, far surpassing consensus estimates of 80,000. On the surface, this is a labor market in equilibrium, and moving towards a surprisingly strong one at that.

The truth is not quite as sunny as it would seem.

My goal here is not to look for the negative in every positive. My goal is to keep us firmly focused on the reality of these numbers, what they mean for now and what they mean for the future of the job market. And to ensure that we all see the difference between the facts and the story.

Sign Up for The Dispatch Newsletter

172,000 Jobs Added

Here is the number that explains the entire beat: leisure and hospitality added 70,000 jobs in May. As we mentioned, the consensus estimate for the whole economy was 80,000, which on paper looks like either a massive win or a massive miscalculation.

The truth is: the World Cup hiring surge alone (June 11 to July 19) nearly accounts for everything economists thought the entire labor market would produce. Strip it out, and the underlying economy added roughly 102,000 jobs – barely above consensus, well within the margin of statistical error, and consistent with a labor market that is neither growing nor collapsing.

But as I’ve discussed every month, beneath that is a job economy defined by quality deterioration, a growing long-term unemployment crisis, and a one-time world event distortion that will make the next three months of data functionally unreadable. The number of people stuck in long-term unemployment (27 weeks or more) hit its highest share in over a decade. The jobs being created increasingly don’t come with benefits, full-time steady hours, or wages that lift families above the poverty line. And the biggest driver of May’s numbers, leisure and hospitality, has almost nothing to do with the underlying economy.

The World Cup Halo is Real, and It Will Likely Mislead Us All Summer

Leisure and hospitality added 70,000 jobs in May. Food services and drinking establishments alone (restaurants, bars, catering operations) accounted for 48,000 of those positions. That sounds good until you compare it to the sector’s 12-month average monthly gain of 14,000 jobs. May’s number being 5 times the trend should be alerting economists and labor leaders to the fact that this is an anomaly.

As we mentioned above, the consensus estimate for the entire economy sat at 80,000, which is an interesting consideration. Wall Street economists knew the World Cup was coming, and they likely modeled for it. And yet they still didn’t fully account for the scale of pre-tournament hiring, evidenced by the 70,000 in L&H almost matching the entire economy estimation. But this is what a one-time demand event looks like in payroll data; it’s enormous, it’s visible, but it’s temporary.

This would not have been factored into the season adjustments that are made every month, but it is still as categorically different from structural job growth as any other seasonal/holiday hiring. Seasonal adjustment models are built for predictable cycles – summer hiring, back-to-school, winter holidays. They’re understandable and ebb and flow in many of the same ways every year, which is why they are left out of the larger number set so as not to skew the data. But a global mega-event that will see a high level of travel, accommodations, and celebration across major cities in the U.S. throughout a six-week period does not factor into this predictable cycle.

Even the adjusted April numbers are showing an interesting point to this: L&H was adjusted up to 30,000 jobs added. This doesn’t mean a sudden steady surge of hospitality roles opening; it’s early hiring in preparation for the games. But what this does mean is that four months of employment data (April through July at minimum, with an August correction) will likely be distorted by this single situational factor. Not only that, but once the games are done and the demand for higher staffing is no longer necessary, we may see a massive pendulum swing back in the other direction with job loss. It’s something to keep in mind over the coming months, that these will not be standard numbers that are predictive of overall stability or growth.

The Jobs Being Lost & The Jobs Being Created Are Not Created Equal

Looking at the actual breakdown of where jobs are expanding and contracting tells another side of this story:

Leisure and Hospitality (+70k): Average hourly earnings of $23.58, average weekly earnings of $598.93. Less than half the private sector average of $1,287. These workers average approximately 25 hours a week, well below the 30-hour threshold for employer-sponsored health insurance under the ACA. Most are ineligible for retirement contributions, paid leave, or any employer-provided benefits. At $31,100 in annual income, a full-time equivalent L&H worker earns below the federal poverty line for a family of four ($32,200).

Local Government (+55k): The second-biggest contributor to May’s gains was local government, predominately local government excluding education (+44k). More stable than L&H, but not an indicator of private sector vitality. 

Health Care (+35k): The most durable growth sector in this current labor market, consistently adding 30,000-40,000 jobs per month regardless of the broader economic environment. This reflects demographic inevitability, a large aging population generating structural demand for care, not cyclical strength. And healthcare is not a monolith: home health aides, who account for a significant share of the gains, earn around $17-$18 per hour. Meaningful below both the private sector average of $37.53. 

Meanwhile, the sectors that on average pay the most are the ones continually contracting. Financial activities shed 22,000 jobs in May and is now down 107,000 from its May 2025 peak, with concentrated losses in insurance carriers and commercial banking. Information sector employment has been in decline since November 2022 – down roughly 342,000 jobs, or more than 10% of its peak workforce. These are sectors where average hourly earnings run $49/hour and $55/hour respectively. We are adding $23/hour jobs and subtracting $50/hour jobs. The aggregate wage number is currently holding up because of the weight of existing higher-wage workers, but the marginal direction of the labor market is running downhill.

Long-Term Unemployment is the Story Everyone is Missing

In my opinion, the most significant number in last week’s report is not 172,000. It’s 27.5%.

That is the share of all unemployed people who have no been out of work for 27 weeks or more, the Bureau of Labor Statistics definition of “long-term unemployment.” In absolute terms, roughly two million Americans fall into this category, up 524,000 over the past year alone. One year ago, that share stood at 22%. At 27.5% today, it’s now reached the highest number in this current economic cycle.

To understand why this matters, we need to consider what is happening structurally. The headline U-3 rate of 4.3% has barely moved in months. That stability is often read on the surface as a sign of resilience; and on the surface, it partially is. We have seen mass layoffs slow down. New displaced workers are not flooding the unemployment rolls in large numbers. But the people who fall into unemployment are increasingly not finding their way back out. The short-term unemployed pool (people jobless for less than five weeks) actually declined by 286,000 in May, largely reversing a one-month spike. The flow into unemployment is not our current problem; the flow out of it is.

What is really happening is a slow migration from short-term to medium-term to long-term unemployment, particularly for workers displaced from higher-wage sectors. Someone who loses a job in financial services or technology in this environment faces a labor market where the openings are concentrated in healthcare and hospitality – sectors with different skill requirements, lower wages, and often part-time structures. The longer that mismatch persists without resolution, the more workers line into long-term status. At 27.5% and consistently rising, we’re seeing that dynamic play out at scale.

For historical reference: before the pandemic, long-term unemployment accounted for roughly 18-21% of total unemployment in a healthy labor market. In the depths of the 2008 financial crisis and its recovery, that share peaked at around 45%. We are not at crisis levels right now, but we are meaningfully above the healthy baseline and moving in the wrong direction.

The broader unemployment measures confirm the same story. The U-6 rate (which captures official unemployment, discouraged workers, and people working part-time who want full-time work) sits at 8.1% for May, but from 7.8% a year ago. Not only that, the gap between U-3 and U-6 has widened from 3.5 percentage points to 3.8 percentage points over that period. The folks at the margins of the labor market are accumulating.

Part-Time Work and the Benefits Gap

One of the most underreported dimensions of this labor market is the relationship between job quality and benefits access. The 70,000 jobs added in leisure and hospitality this month are not just low wage – they are overwhelmingly part-time.

Food service and drinking establishment workers, who account for the bulk of May’s L&H gains, average approximately 25 hours of work per week based on the sector’s reported earnings data, which accounts for a benefits access issue. Under current law, employers are not required to provide health insurance to employees working fewer than 30 hours per week. Most employer retirement plans have similar thresholds. Paid leave, when it exists in low-wage service work, is often accrual-based with very low caps.

This means that a large share of the job added in May do not provide the basic employment package that most workers would define as “a job” in the full sense. They provide income (often insufficient) without the accompanying stability or benefits that enable workers to plan for the future, absorb health shocks, or retire with any security.

Simultaneously, 4.8 million Americans remain employed part-time for economic reasons; meaning they want full-time work but cannot find it or have had their hours reduced. That number was little changed in May from the month before, but in a job market where the marginal work being created is itself part-time, it’s not surprising that this pool is not shrinking.

The Multiple Jobholder Picture is More Troubling Than it Looks

The overall count of Americans working multiple jobs fell year-over-year; from 9 million in February 2025 to 8.5 million in February 2026, or from 5.6% to 5.2% of the employed population. On the surface, that reads as a sign of reduced economic pressure. However, when we look at where the change actually comes from, it tells a different story.

The decline was driven almost entirely by the “primary full-time, secondary part-time” category (what would be considered classic “moonlighting”) which fell by 639,000 year-over-year. These are workers who have a stable full-time job and picked up a second gig for supplemental income. And these are the numbers that significantly have gone down.

At the same time, the category that actually increased was people holding multiple part-time jobs with no full-time position at all – up 55,000 year-over-year to 2.2 million. People we variable or irregular hours across their multiple jobs also rose by 62,000. The populations with a full-time anchor fell sharply. The populations without one grew.

The is one of the compositions that the headline numbers obscure. The labor market is not producing fewer struggling multiple job holders, it’s producing fewer economically stable ones. The worker with a full-time job moonlighting for extra cash is fundamentally different from the person stitching together two or three part-time positions with no benefits, no schedule predictability, and no single employer responsible for their economic welfare. 

Also to consider: these workers are counted as “employed” in the household survey, which counts each person once regardless of how many jobs they hold. Their economic precarity is invisible in our broad reporting, and they don’t show up in the unemployment rate. They are not counted among the 4.8 million employed “part-time for economic reasons,” because they chose (or were effectively forced) to assemble income from multiple sources rather than waiting on a single employer to offer full-time hours. And in a labor market adding 70,000 jobs a month in a sector that is heavily dominated by part-time work, there will likely be more of them by the end of the summer.

Reading the Revisions

One more positive piece of news in last week’s report: prior months were revised upwards substantially. March was revised from +185,000 to +214,000 (+29,000), and April was revised from +115,000 to +179,000 (+64,000). Combined, the two months were 93,000 higher than initially reported.

The April revision is particularly notable, and the finalized sector breakdown tells a clear story. The top contributors to April’s revised numbers were healthcare and social assistance (+58,000), private education and health services (+54,000), transportation and warehousing (+39,300), professional and business services (+22,000), and leisure and hospitality (+30,000).

That last number deserves attention. The original April report described L&H as showing “little change” and wasn’t event flagged in the BLS summary. With the revision to 30,000, we’re seeing an addition of more than double the 12-month average of 14,000. Not as dramatic as May’s 70,000, but it was already running above trend, and it hasn’t been called out.

The genuinely durable strength in April is in healthcare and social assistance, which reflects the structural demographic demand we see of an aging population. Professional and business services adding 22,000 is a more mixed signal: the sector has been weak for months, so stabilization is welcome. But temporary help services remain elevated within it as a share of the total, which is unfortunately not a sign of employers making confident permanent hiring decisions.

What the revisions do confirm is that the labor market in early 2026 was somewhat stronger than the initial reporting suggested. The narrative of a labor market in free fall following February’s -156,000 headline was overstated. But the strength that emerged in March, April, and now May is more concentrated, more event-driven, and less broadly based than the overall numbers may imply.

The Number That Matters Going Forward 

Set aside the World Cup noise, the sector mix, the revision dynamics. And once again the single most important number to track over the coming months is the long-term employment share: currently 27.5% and rising.

If that number continues to increase, it means the labor market is stratifying: workers with the right credentials and connections are finding employment, and everyone else is waiting longer and longer. That is not necessarily a recession indicator in the traditional sense. But the headline rate can stay at 4.3% indefinitely while the long-term pool continues to grow underneath. And that is a structural deterioration that will eventually show up in consumer spending, in debt reporting, in bankruptcy filings, and in the social fabric of communities where workers have been out of the labor force for a year or more.

The labor market is not in active collapse. But it is not delivering for more workers either. There are jobs there; the question is whether they are the right jobs, at the right wages, with the right hours and benefits to constitute genuine economic participation and security. And right now, for a growing share of Americans, the answer is no.

Key Data At-A-Glance

Indicator	May 2026	May 2025	Direction
Nonfarm payrolls (monthly)	+172k (consensus: 80k)	+139k	→
Unemployment rate (U-3)	4.3%	4.2%	→
Broad unemployment (U-6)	8.1%	7.8%	↑
Long-term unemployed	2.0M	1.5M	↑↑
LT unemployed (% of total)	27.5%	22%	↑↑
Labor force participation	61.8%	62.6%	↓
Part-time for economic reasons	4.8M	4.7M	→
Multiple jobholders (Feb comp.)	8.5M/5.2%	9M/5.6%	↓
Avg hourly earnings (all private)	$37.53	$36.28	+3.4% YoY

Sources: BLS Employment Situation, May 2026 (USDL-26-0786); April 2026 (USDL-26-0687), Table A-15, A-16, B-1, B-3.

Sign Up for The Dispatch Newsletter