The post-pandemic labor market recovery has been surprisingly strong, characterized by low unemployment and booming job creation, but a possible warning sign lies within the data—a decline in median hours worked and rise in part-time jobs, according to an analysis conducted by the ADP Research Institute.
In December 2019, part-time jobs accounted for 43 percent of all hourly jobs, but by December 2023, they accounted for 47 percent, ADP found. During the same period, the median number of hours worked by hourly workers fell from 38.4 to 37.7 a week, a decline of almost two percent.
The ADP Research Institute examined the payroll records of private-sector, hourly workers, tracking about 13 million individual jobs each month.
“We measured average weekly hours worked within a given month for each job, then calculated the median of those observations,” said Liv Wang, the lead data scientist with ADP who conducted the analysis. “The pandemic affected the labor market in ways small and large, including hours worked,” she said.
Julia Pollak, the chief economist at ZipRecruiter, said the relationship between hours worked and the health of the labor market is complicated, but the data could be signaling that the labor market is not as hot as other measures indicate and that layoffs and job loss may be around the corner.
“In the U.S., working hours tend to plummet during recessions, and tend to spike when demand for labor is very high,” Pollak said.
Between 2015 and 2019, the workweek for all private-sector nonfarm workers ranged from 34.3 to 34.6 hours, according to the Bureau of Labor Statistics (BLS). In February, the average workweek for all employees on private nonfarm payrolls edged up to 34.3 hours, at the bottom end of what’s considered healthy.
“Below that tends to signal layoffs,” Pollak said. “Above that is not great either, like during the pandemic reopening period when companies that were short-staffed were excessively relying on overtime.”
In this case, it’s difficult to figure out whether working hours are declining because of a rise in involuntary part-time work, or whether people are voluntarily working less because they want to, she added.
Pollak also noted that the labor market has cooled over the past year. “Employers are finding it easier to hire people and are not as reliant on squeezing as many hours as possible from a limited supply of workers,” she said.
Wang said possible reasons for the downward trend could be coming from both the employee and employer sides. “Employees may be asking for more flexibility, while rising wages have also enabled them to work less without reducing income,” she said. “Child care responsibilities may be impacting the workweek for female employees. The gig economy has also offered opportunities to work multiple jobs, which may mean less time on a single job.”
She added that “employers may want to avoid overtime labor costs and could also be reacting to economic fluctuations, which could lead to hiring more employees but giving them less hours instead of reducing and scaling headcount.”
Pollak offered additional suggestions, including the rise in labor force participation among women and young people, two groups who have traditionally worked fewer hours.
“In such a hot market, employers are casting a wider net, recruiting more women and young people,” she said. “Whenever unemployment goes under 4 percent, labor force participation rises for these groups, who have higher percentages of people who are indifferent between not working and working part time. Some in these groups don’t want to work full time because they have family and caregiving responsibilities, or they are students.”
Adults 35 and younger are working an hour less than they did four years ago, while hours worked by older age groups held steady, according to the ADP Research Institute.
“While women accounted for 47 percent of all hourly paid workers in December 2023, they made up 56 percent of all part-time hourly workers,” Wang said. “Now women who are paid by the hour are putting in an hour less per week than they did in 2019. Time worked by male hourly workers, in contrast, is little changed at about 40 hours a week. Over the past four years, the hours-worked gender gap has widened from 4.4 to 5.4 hours a week.”
The Wealth Effect
There has also been an increase in wealth among every income level since the pandemic, Pollak said. And younger workers in front-line, hourly roles have received the largest wage increases over the last few years.
ADP payroll data shows that workers who stayed in their jobs were able to command more than five percent annual pay increases in 2022 and 2023. “For people who changed jobs, the numbers were even higher,” Wang said. “In June 2022, job-switchers recorded a 16.4 percent pay gain. This historic jump helped offset the effects of the reduced hours that followed.”
Wang said that among people who worked fewer hours than a year ago, a majority have a higher hourly wage. “These hourly wage increases offset the effect of reduced hours on workers’ annual gross pay,” she said. “For example, in October 2022, 52 percent of workers whose hours were reduced saw an increase in their annual gross pay because their hourly wage had grown so much. By December 2023, that number had dropped to 43.7 percent—still more than 4 in 10 workers.”
Pollak pointed out that the declines in median and average hours worked do not necessarily imply that the same individuals who were previously working 40 hours a week are now working 35 hours, for example. “It could be that they are all still working 40 hours—or even working more hours than before—and that the workforce has simultaneously absorbed a whole lot of new marginal workers who are only prepared to work 20 hours a week, and only when jobs are sufficiently attractive.”
Have questions or feedback? Get in touch with Roy at roy.maurer@shrm.org.
An organization run by AI is not a futuristic concept. Such technology is already a part of many workplaces and will continue to shape the labor market and HR. Here's how employers and employees can successfully manage generative AI and other AI-powered systems.