Does the World Cup Create Jobs in Host Cities?
Our analysis says no. In our second piece on the World Cup, we show that the event has no meaningful effect on local labor markets — even when just considering hospitality.
Photo credit: Vienna Reyes
In our previous piece, we looked at whether the 2026 World Cup sparked a hiring boom in U.S. hospitality. The national data gave a clear answer: no. Employment in the sector fell in June, and economic research suggests that a muted response is exactly what we should expect from such mega-events.
But national figures can hide a lot. Even if the World Cup barely registers in a $32 trillion economy, its effects should be concentrated where the games are actually played — in the 11 U.S. host cities. If there is a hiring bump to be found, that’s where it would show up. That is why we zoom in and look at the local labor markets of the host cities. Spoiler alert: we still can’t find any meaningful effect.
Local trends: Comparing hosts with non-host cities
Weak employment growth in hospitality at the national level is by itself not definitive proof that the World Cup is not creating additional jobs: the sector could have contracted even more without the event taking place. To dig deeper, we therefore need to look at regional data. In this piece, we compare the 11 host cities with a group of similar non-host cities, our control group. While the U.S. cities where the games are played are somewhat larger than the non-host cities, this is driven by New York and Los Angeles as outliers.
Our analysis reveals that both groups display similar labor market trends in the run-up to the event — the most important assumption of this kind of event study. Rest assured that we also experimented with swapping metropolitan statistical areas (MSAs) in the control group, but that doesn’t seem to affect our results.
What the data says
Unfortunately, employment data at the MSA level is only available through May, just before the World Cup kicked off — pun intended. However, one would expect businesses to hire staff in anticipation of higher demand ahead of the event, not just during it. But our chart shows no meaningful difference in total employment between host and control cities. Even more disappointing, that is also true when just considering employment in leisure and hospitality.
Indeed job postings allow us to look at data all the way through June 26. There is no significant uptick in job advertisements before or during the event. And again, we fail to detect any difference between host and comparison cities.
With data from Lightcast, a labor market analytics firm, we can narrow the focus to job postings within hospitality. We identify four subsectors within accommodation and five within food and drinks that should have the highest exposure to the World Cup.
Alas, we need to disappoint you again. Even within these core hospitality postings, there is no divergence between the two city groups.
Additional checks: Difference-in-differences estimation
Feel free to skip this section as our statistical analysis shows the same non-result!
If you were worried that our previous analysis wasn’t quite rigorous enough, we also tested our results more formally with a regression model. Using a difference-in-differences approach, we estimate whether the gap between the two groups changed significantly in the weeks just before and after the tournament began.
For the BLS employment data, we use the months up to March 2026 as the pre-period, while April and May are included in our analysis as the treated period — although the World Cup only started in June, one might expect businesses in the host cities to increase headcount in anticipation a few weeks before kickoff. With the Lightcast data, we change the treated period to May and June. Our analysis includes time and metro fixed effects (FE) — these are statistical controls — don’t ask.
The two graphs below show that there is no statistically significant difference between the host and comparison cities. Every dot sits close to zero before and throughout the treated period, and the error bars always include the null effect. In plain terms: host cities were hiring staff at restaurants, bars, and hotels at essentially the same pace as everywhere else. Whatever the World Cup was about to do to host metros, a statistically significant surge in hospitality jobs isn’t part of the story — at least not yet.
Caveats to our analysis
The main issue is that the metropolitan employment data only goes through May, while the job postings data includes June but not July. Neither dataset covers the full length of the World Cup. We do not believe this to be a big problem since employers typically don’t hire at the last minute. Any meaningful effect should have already shown up in the run-up to the event and in the June job postings. Nevertheless, we will follow up with an extended analysis once we have all available data — in early September.
What does this mean for recruiters?
Zooming in on the host cities doesn't change the conclusion. When we compare the 11 U.S. host metros with a control group, we find no significant divergence in employment or job postings — neither in total metro employment, nor in the hospitality industries most exposed to the event. Our statistical analysis points to the same non-result: the cities with games were hiring at essentially the same pace as everywhere else.
For recruiters, this means the local labor market in host cities likely stayed closer to business as usual than the pre-tournament hype suggested. Competition for hospitality talent in New York, Los Angeles, and the nine other hosts was probably no fiercer than in comparable cities without the World Cup on their calendar. That doesn’t rule out smaller, shorter-lived effects. Some businesses near the stadiums certainly stretched their current staff with extra shifts, and a few may have added headcount for the tournament. But those adjustments are too localized and too brief to move the metropolitan numbers.
One caveat remains: the regional data only runs through the early part of the event. We’ll revisit these host cities once we have data covering the full tournament — so stay tuned!
Last but not least, do not forget to sign up for our Substack live event on August 7 in which we break down the U.S. July jobs report. Here is the link to the event.