The views of the authors, including guest contributors Shelby Cook (Data Analyst, Michigan Unemployment Insurance Agency) and Leonidas Murembya (State Administrative Manager, Data Analysis and Reporting, Michigan Unemployment Insurance Agency), do not necessarily reflect the views of the Federal Reserve Bank of Minneapolis, the Federal Reserve System, or the Michigan Unemployment Insurance Agency.
Shared work, also known as short-time compensation, is a provision of the unemployment insurance (UI) system that provides benefits to workers who have their hours temporarily reduced. Rather than laying off workers during brief downturns, employers can reduce workers’ hours, and workers can receive unemployment benefits to partially offset their lost income.
Who’s participating in shared work programs? New data from the State of Michigan, presented during a recent Federal Reserve Bank of Minneapolis virtual event on shared work, provide some insights. First, participating employers in this Ninth Federal Reserve District state are disproportionately, though not exclusively, from the manufacturing sector. Second, they have roughly the same number of employees as employers that employ regular UI claimants—a surprising finding given concerns about the administrative requirements of shared work. And finally, shared work participants tend to earn more (before entering the program) than regular UI claimants.
Decision-makers in other states that are considering developing or expanding shared work programs can learn from Michigan’s experience. In particular, analysis of who currently participates in shared work is useful for identifying new opportunities to use the program, including expanded usage by low- and moderate-income workers.
How shared work works
To participate in shared work, employers must submit plans to their state UI agencies. These plans must list the timing of the temporary hour reductions, the percentage that hours will be reduced, and the affected employees. Once a plan is accepted and implemented, participating workers will receive partial benefits from the UI agency to offset their lost earnings. These continue for as long as the plan specifies, unless the employer changes the end date of a plan.
Specific rules about the plans vary by state within guidelines established in federal law.* The reduction in employee hours must be between 10 and 60 percent. For those reduced hours, workers receive prorated payments equivalent to what they would have received in the regular UI program had they been laid off. Importantly, eligibility at the worker level is largely similar to eligibility for regular UI, including the requirement that workers lose their job through no fault of their own, with the exception that workers participating in a shared work program are not required to search for new employment. States set the maximum duration of a plan, with that cap typically set at six or 12 months.
With 30 states having shared work programs, shared work participants make up only about 1 percent of all UI claimants nationally.
Who uses shared work
According to the Michigan data, which cover 2022 through 2024, nearly half (49 percent) of all employers participating in the state’s shared work program were from the manufacturing industry, in contrast to just 11 percent of employers participating in the regular UI program (and who had claims during the period). Individual manufacturing employers using shared work programs tended to have more workers participating in the program, such that over 80 percent of all claimants were from the manufacturing industry. Figure 1 shows these differences. The overrepresentation of the manufacturing industry in Michigan’s shared work program is consistent with recent data from California and Illinois and older survey data from other states.
The shared work program may be well suited for industries like manufacturing in Michigan and elsewhere, where employers want to retain specialized workers but often experience fluctuations in demand that temporarily reduce their labor force needs. Other sectors that also experience fluctuating staffing needs, like construction or retail trade, are less likely to use shared work in Michigan.
As shown in Figure 2, the distributions of employer size for employers with claims in the regular UI program and employers participating in the shared work UI program are similar in Michigan, with about three-quarters of both regular and shared work employers having 50 or fewer employees. Nearly all of the other employers were of medium size with 51 to 999 employees. Of all regular and shared work employers, just about 1 percent were large employers with 1,000 or more employees.
The absence of meaningful differences between shared work employers and regular UI employers is surprising, standing in contrast to earlier survey analysis that found an overrepresentation of larger employers among those participating in shared work programs. This new finding from Michigan suggests that many smaller employers overcame information hurdles and administrative requirements that earlier research hypothesized had limited their usage of shared work programs.
The Michigan data also show that—over time—participation in shared work is not concentrated among the same set of employers. Three-quarters of employers who used shared work participated only in one year of the program’s 10-year history. On average, employers participated for about 1.4 years. These statistics suggest that the program has reached a variety of employers as they temporarily reduced hours, including during the COVID-19 pandemic.
Demographics of shared work claimants
The new Michigan data provide details about claimant characteristics such as wages, race, and gender that show substantial differences between individuals claiming shared work benefits and those claiming regular UI benefits. As shown in Figure 3, shared work claimants generally earned more than regular UI claimants before their hours were reduced: 73 percent of shared work claimants earned more than $35,000 annually, compared to 46 percent of regular UI claimants. Conversely, only 6 percent of Michigan’s shared work claimants had annual wages below $20,000, compared to 33 percent of regular UI claimants. The higher wages of shared work claimants could reflect their specialized skills, the types of employers participating in shared work programs, or other factors.
Of shared work claimants, the share who are Black was 18 percentage points lower than the share of regular UI claimants who are Black, which can be seen in Figure 4 by taking the difference between the purple bars under the “Race and ethnicity” demographic. In addition, the fraction of shared work claimants who are men was 6 percentage points higher. These differences do not take into account the unspecified population of claimants, which could make the gaps larger. Gender differences likely reflect the current manufacturing-heavy industry composition of shared work in Michigan, but it is less immediately clear what accounts for racial and ethnic differences.
Applying data insights
The data on Michigan’s shared work program suggest avenues for future research, which could examine whether the patterns presented here are due to the prevalence of specialized skills in certain segments of the workforce, the administrative requirements of shared work programs, or uneven awareness of and familiarity with shared work. This research is needed because much of the pioneering analysis of shared work is based on overall counts of claimants, describes few details of claimant characteristics, or is drawn from European contexts that may not fully apply in the United States. Going deeper with new research—such as California Policy Lab’s recent exploration of data from that state—could help decision-makers consider whether and how to expand shared work programs. In turn, this helps advance the Federal Reserve’s work to support an economy that works for everyone.
Endnote
* See Whittaker (2016) for a detailed description of shared work programs and the applicable federal guidelines, and the U.S. Department of Labor’s Short-Time Compensation fact sheet for links to states’ shared work websites.
More On
Ayushi Narayan conducts research on labor market institutions to help the Community Development and Engagement team understand how employment-related policies and trends affect low- and moderate-income communities. Prior to joining the Bank, her work included roles at the Council of Economic Advisers, Nike, and Amazon.