Last week in a confirmation hearing, Russel Vought, President Trump’s nominee to run the Office of Management and Budget, said he would support work requirements for Medicaid, the government health insurance program for low-income people. His position—which has also shown up in Republican proposals for the House reconciliation package—was couched in language to “encourage people to get back into the work force, increase labor force participation and give people again the dignity of work.”
In reality, work requirements have nothing to do with getting people into the workforce. While increasing labor force participation and helping people obtain the dignity of work are important goals, people don’t actually need encouragement to do this. The incentive to work is already there: It gives people sufficient income to not live in grinding poverty. People with income low enough to qualify for social safety net benefits need support from policymakers to access programs like Medicaid and SNAP, not new rounds of bureaucratic paper pushing, which is what work requirements mainly achieve.
If Mr. Vought was more serious about improving access to work, he would be clear-eyed about the core barriers to work that low-income workers have traditionally faced: weak macroeconomic conditions, the volatile nature of low-wage work, and other barriers to work like caregiving responsibilities.
With respect to macroeconomic conditions, while today’s labor market is extremely strong, this has not been the norm nor is it something we can assume will persist in the future. The United States has spent far too much time with excess unemployment rates in recent decades. This macroeconomic failure is the responsibility of policymakers—individual workers have little control over the macroeconomic situation, yet it determines whether they are able to find regular work at sustaining wages. Employment rates for low-income adults are highly cyclical, rising when the macroeconomic environment is more favorable and overall unemployment rates fall, and falling when overall unemployment rises due to slack job markets. This is a key signal that these workers mostly do not need “encouragement” or “incentives” to work—they need opportunities. When opportunities arise in the form of strong labor markets, these workers flock to them.
In my analysis, I explored the association between number of hours worked for low-income adults and the unemployment rate between 1979 and 2019 to see how excess unemployment was related to work time. Figure A shows that as unemployment increases, the number of available jobs in a given local labor market becomes scarce and workers work fewer hours, suggesting that the jobs low-income adults take are much more tied to aggregate labor market health than to work requirements.
Higher unemployment is associated with fewer work hours for low-income adults: Total hours worked and unemployment rate for the bottom 30% of households, 1979–2019
| year | Unemployment rate (%) | Hours worked |
|---|---|---|
| 1979 | 4.95 | 1235.1 |
| 1980 | 6.29 | 1213.1 |
| 1981 | 6.84 | 1193.7 |
| 1982 | 8.98 | 1150.7 |
| 1983 | 8.92 | 1167.3 |
| 1984 | 6.86 | 1229.6 |
| 1985 | 6.61 | 1266.9 |
| 1986 | 6.45 | 1285.2 |
| 1987 | 5.69 | 1308.5 |
| 1988 | 5.03 | 1320.2 |
| 1989 | 4.88 | 1365.5 |
| 1990 | 5.22 | 1361.2 |
| 1991 | 6.46 | 1334.1 |
| 1992 | 7.20 | 1328.4 |
| 1993 | 6.65 | 1352.2 |
| 1994 | 5.89 | 1361.0 |
| 1995 | 5.43 | 1374.9 |
| 1996 | 5.27 | 1375.0 |
| 1997 | 4.82 | 1382.2 |
| 1998 | 4.46 | 1400.1 |
| 1999 | 4.20 | 1409.0 |
| 2000 | 4.04 | 1400.6 |
| 2001 | 4.78 | 1356.1 |
| 2002 | 5.93 | 1322.7 |
| 2003 | 6.17 | 1307.4 |
| 2004 | 5.71 | 1311.8 |
| 2005 | 5.35 | 1317.9 |
| 2006 | 4.90 | 1349.3 |
| 2007 | 4.94 | 1339.4 |
| 2008 | 6.18 | 1277.9 |
| 2009 | 9.72 | 1177.1 |
| 2010 | 10.07 | 1144.2 |
| 2011 | 9.41 | 1147.4 |
| 2012 | 8.65 | 1173.2 |
| 2013 | 7.99 | 1199.0 |
| 2014 | 6.84 | 1215.9 |
| 2015 | 5.98 | 1255.2 |
| 2016 | 5.62 | 1279.4 |
| 2017 | 5.16 | 1289.9 |
| 2018 | 4.71 | 1319.2 |
| 2019 | 4.55 | 1361.0 |
Note: The relationship between unemployment rate and hours worked displayed in the figure are adjusted for a time trend. We define the bottom 30% of households based on their total income from wages and salaries.
Source: Author’s analysis of Current Population Survey Annual Social and Economic Supplement (CPS ASEC) (Flood et al. 2024).
Further, jobs available to low-income adults often pay low wages and have scheduling practices (such as little advance notice or time-varying schedules), which decrease the regularity and predictability of work time and can make it hard for workers to maintain consistent work hours needed to satisfy the requirements (by either working 80 hours per month or 20 hours per week). A 2014 study showed that disproportionately large share of workers in low-wage jobs (66% of janitors and housekeepers, 90% of food service workers, 87% of retail workers, and 71% of home care workers) reported their hours varied within the last month, highlighting the pervasiveness of such practices.
Finally, given that many low-income workers on programs like SNAP and Medicaid have caregiving duties, policies that improve access to care would do much more to increase employment than simply mandating workers to work more. Studies show that when barriers to care are reduced or policies like paid sick leave are passed, women experience economically meaningful increases in their employment. This suggests that if the Trump administration wants to get serious about improving labor market outcomes for low-income adults, policies to support caregiving would be more effective than work requirements at achieving this goal.
In the end, work requirements function as reporting requirements for all recipients, making the process more onerous and burdensome. All recipients, including people with documented disabilities getting Medicaid would have to jump through additional bureaucratic hoops to prove that they’re exempted from work requirements, further risking lapsing or losing their coverage. These burdensome practices and paperwork ultimately lead people to withdraw from programs (see examples for SNAP and Medicaid).
Mr. Vought’s view that work requirements would increase labor force participation and employment is flawed and reflects inaccurate beliefs (or just lack of concern) about the barriers to work for low-income adults. My analysis shows that there are still plenty of barriers that keep low-income adults out of the workforce, but insufficient incentives are not one of them. When labor market conditions are right, low-income workers do work and earn more than they do when unemployment is high, suggesting that macroeconomic policy has more to do with low-income adults’ ability to work than any work requirement-imposed threat to take away their health care or nutrition assistance. If policymakers were serious about creating opportunities to work, they would pass policies like secure scheduling laws and affordable care policies that would meaningfully reduce barriers low-income adults face in gaining employment.
Sign up for EPI’s newsletter so you never miss our research and insights on ways to make the economy work better for everyone.
