There’s been a lot of talk recently about adding work requirements for in-kind welfare programs such as the Supplemental Nutrition Assistance Program, which provides food assistance, and Medicaid, the health care program for the poor.
It’s perhaps understandable that policymakers want to limit the cost of these programs. Annual spending on SNAP has grown tremendously since the 2008-2009 financial crisis – from roughly $33 billion in 2007 to about $70 billion today. Medicaid is an even bigger expense – at more than $500 billion a year, it’s a major contributor to the U.S.’ widening budget deficit.
But as my Harvard colleague Jason Furman, who headed former President Barack Obama’s Council of Economic Advisors, details in a Wall Street Journal op-ed, the working poor face substantial employment uncertainty and volatility. As most work requirements entail continuous employment, the poor have trouble meeting them because their jobs often tend to be intermittent.
There’s also a conceptual problem with work requirements for basic services: A lot of the social value of in-kind transfers comes from getting essential aid to people who can’t find jobs, or who can’t work at all.
Some types of barriers to employment – such as old age or physical disability – are relatively easy to identify and verify. But others – such as homelessness or the need to take care of family members – are not. Most work-requirement policies make exceptions for people in the former categories. But they often exclude people in the latter ones.
This is a major defect: Many poor people who can’t work are in some sense most in need of help. So, the burden of a work requirement can impose the greatest hardship on exactly those people who have the highest needs for social services. If you’re unemployed because of health problems, for example, cutting your access to Medicaid further ensures your joblessness.
Many poor people who can’t work are in some sense most in need of help.
Of course, there’s an incentive problem: If government provides services to people who aren’t working, then some people might choose to remain unemployed. Nevertheless, research by Hilary Hoynes of the University of California-Berkeley and Diane Whitmore Schanzenbach of Northwestern University suggests that this effect is likely to be modest, at least for SNAP. And that’s not too surprising – programs such as SNAP and Medicaid aren’t enough for most people to live off of, so there’s still a lot of value to having a job.
Besides, some incentive problem is inevitable. Once we are committed to helping people who aren’t able to work, we have to accept the possibility that we’ll also help some people who simply choose not to work, since we can’t always tell the difference.
And even if we do want to incentivize labor directly, we have better ways to do that – particularly, as my Harvard colleague Nathan Hendren has pointed out, through programs such as the earned income tax credit, which provides income tax refunds to low-income individuals who hold jobs. At the bottom of the income distribution, at least, the value of the EITC ratchets up with how much you work. And unlike in-kind transfers, there’s no concern that the EITC may be more valuable to those who aren’t in the labor market. So if policymakers really care about encouraging the poor to work, they could think about shifting money into the EITC, or simply expand it! Unfortunately, proposals to do that don’t seem to be on the table.
All told, the work requirements currently under consideration are unlikely to work out well. They’ll cut costs – at least in the short run – but will do so at the expense of the vulnerable people most in need of social services. And at the same time, if incentivizing labor is the goal, then work requirements are a blunt instrument – and policymakers have better options.
Scott Duke Kominers is the MBA Class of 1960 associate professor of business administration at Harvard Business School. Distributed by Bloomberg Opinion.