[Editor’s Note] After the release of the nonfarm payroll data and the huge volatility in the market, many investors have raised similar questions, mainly focusing on two points. First, whether recent employment is related to the sporting event; second, why the employment data have significantly exceeded expectations. To answer this, we compared the initial jobless claims data in recent host states versus non-host states, and looked back at employment changes after large-scale sporting events historically held in the US.The conclusion is that current evidence is insufficient to prove that sporting events are the main reason behind this month’s surge in leisure and hospitality employment.We believe that,the signs of US employment recovery have become clearer since the beginning of the year, driven by the lagged effect of last year’s precautionary rate cuts and the recent revival of the global manufacturing cycle. Simply attributing recent strong job numbers to a single sporting event may overlook more important cyclical signals worth focusing on.
1. Question 1: Is the strength in this month’s nonfarm payrolls driven by the sporting event?
In this month’s nonfarm employment, leisure and hospitality jobs grew by 70,000, significantly outpacing March and April. Some opinions argue that the robust hiring in leisure and hospitality reflects advanced hiring practices by employers. We take a cautious view:The current evidence is insufficient to confirm it as the primary driver for the surge in leisure and hospitality employment this month. There are three main reasons for this:
Reason 1: According to initial jobless claims, there’s no significant divergence between host states and non-host states
Although May’s regional employment data haven’t been released yet, the initial jobless claims data are available up to the week of May 23, covering the reference period of May’s nonfarm payrolls. Thus, by comparing initial claims, we can assess the impact of the event on regional employment conditions.
We grouped the 10 states (California, Texas, Georgia, Florida, New York, New Jersey, Pennsylvania, Massachusetts, Missouri, Washington) with 11 host cities as “host states”, while all others are “non-host states”. Figure 1 compares initial jobless claims (unadjusted) between the two groups from April to May. Since the beginning of the year,initial claims in both groups have fallen year-over-year, with similar trends. There is no indication—either in absolute numbers or in year-over-year growth rates—of a tighter labor market in the host states.In fact, the year-over-year drop in initial claims in non-host states since April is even slightly greater.
Figure 1: Comparison of Initial Jobless Claims and Year-over-Year Changes in Host States vs Non-Host States ‘ initial jobless claims and annual changes
Data Source: Fred, GMF Research
Reason 2: Historically, US major sporting events have not caused significant disruption in employment.
The US has hosted many major sporting events in history, such as the 1984 Los Angeles Olympics, the 1994 World Cup, the 1996 Atlanta Olympics, and the 2002 Salt Lake City Winter Olympics.
Using the holding month as the baseline (t=0), we examined monthly additions in leisure and hospitality versus total employment in nonfarm payrolls, six months before and after the event (t-6 to t+6).Figure 2 shows that, nationwide, there was no clear employment surge before or after the event. Nor was there any “leisure and hospitality hiring surge, followed by retreat” pattern.The increase in leisure and hospitality jobs during the event month was mostly in line with (or slightly below) pre-event baselines on all four occasions. Total nonfarm payroll did not systematically jump either. Logically, even if there was a leap at the host-state level,its share of total national employment is relatively small and is often outweighed by fluctuations in the national total.
Figure 2: US Nonfarm Payrolls and Leisure & Hospitality Employment Before and After Four Major Historical Sporting Events
Data Source: BLS, GMF Research
Reason 3: Neither the Beige Book nor enterprise comments in the Services PMI mention event-related impacts.
In qualitative evidence, the latest US Fed Beige Book (including the Atlanta Fed district with two host cities Atlanta and Miami) clearly stated that,local tourism demand weakened due to higher energy and travel costs. The main narrative revolves around rising oil prices due to Middle East tensions, tariffs, and AI/automation boosting productivity—there is no mention at all of sporting events.ISM Services PMI survey comments only mentioned “spring weather led to more activities,” and “low consumer confidence and high mortgage rates create pressure,again, no mention of events.”
Question 2: Why did employment numbers far exceed expectations?
In fact, signs of recovery in the US job market have gradually become clearer since the start of the year. One could say that May’s nonfarm payrolls simply confirmed this trend.Figure 3 shows the three-month moving average of net job gains for total nonfarm payrolls, private sector, and industrial sector employment (left axis). Even excluding May’s data, the trend has already reversed its negative trajectory from the second half of 2025 and started to rebound since the start of the year. Weekly ADP small business job data also has rebounded since the start of the year and does not depend on May’s single reading.
Figure 3:New Jobs ,New Private Sector Jobs and Industrial Sector New Jobs (3mma)
Data Source: Haver, GMF Research
Figure 4: Weekly ADP Small Business Net Job Increases (Weekly Value vs Monthly Value/4)
Data Source: Haver, GMF Research
Since the beginning of the year, the marginal improvement in the labor market aligns with the historical pattern following soft-landing rate cuts.Historically, after a soft-landing rate cut, the unemployment rate usually shifts from “continuous decline” to “stable sideways” shortly after cuts begin, with further declines three quarters later. In the current cycle (since September 2025), the unemployment rate has behaved consistently with this pattern (spiking, then falling, then stabilizing). Based on this, the jobless rate is expected to continue falling toward mid-year and approach 4% by year-end.
Notably, the rebound in industrial sector employment deserves more attention than service sector employment—and is more informative. In the past three months, the US industrial sector (goods-producing industries) has averaged 75,000 new jobs per month, compared to -77,000 in mid-2023. Historically, manufacturing employment is more cyclical and more highly correlated with the PMI cycle (see Figure 6). Since early this year, global PMIs have turned upward for multiple reasons, US ISM PMI has risen to 54, driving a revival in industrial sector job data.
Figure 5: US Unemployment Rate 2 Years Before and After Historical Soft-Landing Rate Cuts
(Purple: this cycle; Red: historical median)
Data Source: GMF Research
Figure 6: High Correlation Between Industrial Sector Employment and US ISM PMI
Data Source: Haver, GMF Research
In summary, the recent improvement in US employment data is a natural outcome of prior monetary policy easing and the global manufacturing recovery cycle.We believe that to attribute this entirely to a temporary factor such as a sporting event is to “miss the forest for the trees,” potentially underestimating the persistence of the labor market recovery. In short, tying this month’s strong employment purely to a single event is not supported by data and could obscure more important cyclical signals.
Of course, the equity market’s reaction to Friday’s jobs data remains “excessive.”Theoretically, a 10bp rise in the 10y US Treasury real yield corresponds to roughly a 0.7% drop in the S&P 500. On Friday, the 10y yield rose only 8bp, while the S&P 500 fell more than 2.6%, far beyond historical norms.This shows that the fragility of positions after the previous sharp rally is the real reason behind the market’s panic and excess move.
Source: Original Article
