The long-standing debate over work requirements for social programs in the United States heated up this spring during the debt ceiling negotiations. A provision in the debt ceiling bill expanded work requirements for some recipients of the Supplemental Nutrition Assistance Program (or SNAP, commonly known as food stamps). But politicians and editorial pages have mostly obfuscated the issue in clouds of moral indignation: on the left, “How could you be so heartless as to force unfortunate people to work?” On the right, “You’re bankrupting the country and undermining the culture of self-sufficiency.”

Economics, as usual, offers a straightforward, value-free way to think about the issue: incentives. When you put all of the country’s social programs together, low-income Americans face roughly 100 percent marginal tax rates. Earn an extra dollar; lose a dollar of benefits. There are also cliffs of infinite tax rates, where people earn an extra cent and lose a program entirely. The disincentive also depends on how many and which programs people sign up for. But the order of magnitude is right.

The incentive effect is clear: don’t work, or at least don’t work legally; when there are asset tests, don’t accumulate any wealth. As former senator Phil Gramm (Republican of Texas) and Mike Solon of lobbying firm US Policy Strategies write in the Wall Street Journal, “Since 1967, average inflation-adjusted transfer payments to low-income households—the bottom 20%—have grown from $9,677 to $45,389. During that same period, the percentage of prime working-age adults in the bottom 20% of income earners who actually worked collapsed from 68% to 36%.”

Thirty-six percent.

Incentives are a budget constraint to government policy, hard and immutable. Your feelings about people one way or another do not move the incentives at all. A gift of money with an income phaseout leads people to work less, and to require more gifts of money. That’s just a fact.

What to do?

We could just remove the income phaseouts. Give everyone food stamps, Medicaid (the US healthcare program for low-income individuals), housing subsidies, the Earned Income Tax Credit, college tuition subsidies, and so forth. Don’t reduce benefits with income. Then the disincentive to work is much diminished.

Rather obviously, that’s impractical. Even the US would run out of money quickly. Just $20,000 x 331 million people = $6.6 trillion, essentially the entire federal budget right there, and $20,000 of total support is a lot less than people with no income get right now.

There is always a trade-off between help and disincentives.

And even then, the disincentive is not eliminated. To eliminate the work disincentive in social programs by eliminating phaseouts, we would have to jack up marginal tax rates for everyone to such stratospheric levels that nobody would work. You can’t escape budget constraints.

Support for the unfortunate must be limited somehow. That’s why we restrict help to people below a certain income level. But even if each individual program maintains a reasonable marginal phaseout, together they add up across programs, and the next thing you know, we’re back to a 100 percent marginal tax rate.

Posit that work is still desirable—to earn some money, to contribute to your fellow citizens, to reduce the need for income assistance, and to build human capital. (And let’s not dismiss the more ephemeral benefits widely ascribed to work: self-reliance, life meaning, self-respect, participation in society, and so forth. Few people think that a large portion of a citizenry living on government checks for a lifetime makes for a desirable society, no matter whom they vote for.) If so, if we still value work, and if the social safety net creates a 100 percent marginal tax rate on it, and if abandoning income phaseouts will bankrupt the state, we have a problem.

Work requirements are an imperfect method to try to replace the incentive to work that social programs eliminate. But imperfect is better than nothing.

Work requirements are inefficient, as you can tell from the brouhaha. It’s much more efficient to get people to work by saying, “If you earn a dollar, you can keep it,” rather than, “If you earn a dollar, we’ll take it away from you, but we’re going to force you to work.” The rules around work requirements are complex, and people and governments game them.

Just who should or shouldn’t have to work, and under what terms? Progressives will quickly find a sick single mother taking care of elderly parents and commuting to some horrible fast-food job who falls through the cracks, and they are right. Rules and bureaucracies are rough substitutes for market incentives. More importantly, if you’re working for money that you get to keep, you find the best job you can, you work hard, and you look for better opportunities. If you’re working to satisfy a bureaucratic requirement in the face of a 100 percent tax rate, you find the easiest job you can, you don’t care about the money and thereby the social productivity of the work, and you do as little as possible.

I do not, therefore, defend work requirements as a perfect offset to a 100 percent marginal tax rate. But they are there for a reason, as a rough offset to some of the disincentives that means-tested programs pose. The point is that we should start to understand and debate work requirements in this framework. If we’re going to remove market incentives, we need some replacement incentives.

Recommended Reading

Is there a better way? I can think of at least three other changes to the system that would help fight disincentives:

Limit assistance by time rather than by income. This is how unemployment insurance works. We understand that replacing people’s paychecks forever if they lose their job has bad incentive effects. Unemployment is understood as a temporary misfortune, and you get unemployment checks for a limited amount of time. Could not many other programs aimed at misfortune also be limited by time—but then allow you to keep each extra dollar of earnings? Perhaps even unemployment should be a fixed amount of time, and you can keep receiving it for the full (normally) 26 weeks even if you get a job.

The trouble with this solution, of course, is that some people will not solve their issues in the required time, and then you have to be heartless. But is it not just as heartless to say to a person who had been benefiting from food stamps, the Earned Income Tax Credit, social-security disability payments, and a housing voucher, “Well, congrats on getting a job, and a good one. Now we’re taking away all your benefits?”

Also, the safety net does include a detailed bureaucracy to determine who is needy. Disability, unemployment, and so forth look hard at these issues. Replicating that with a different set of rules for each program seems wasteful.

Pay more attention to consumption. Good economists all understand that consumption, not income, is the right measure of well-being. That’s why consumption taxes are a good idea, and why we should measure consumption diversity rather than income diversity. (I don’t use the word inequality anymore as it prejudices the right answer.) One advantage of a consumption tax is that it would be easier to condition benefits on consumption rather than income. If you work and save the results, you can keep your benefits.

Take a more comprehensive view of income. Income phaseouts in social programs take away benefits based on market income, but not on social program income. If you have food stamps and earn an extra $10,000 of income, you can lose your food stamps. If you get housing worth $10,000, you don’t lose anything. This is a large distortion toward putting effort into obtaining more social program benefits rather than working.

Disincentives come from the combination of social programs and taxation, and any hope of eliminating disincentives must take a similar integrated-system approach.

There is also an argument over how much money the programs cost. This leads back to “how could you be so heartless” versus “but the country will go broke.” A focus on incentives offers the way out. Fix the incentives, and we end up better helping people who need it, we get a lot fewer people who need help, and we spend a lot less money. Win-win-win.

Even so, there is no clean answer. A main lesson of economics is that there is always a trade-off between help and disincentives. We can’t totally eliminate this trade-off, but we can make it a lot more efficient than it currently is. And we can have a much more productive discussion about how to manage social programs if we talk about the constraint posed by incentives, rather than engaging in the usual moral mudslinging.

John H. Cochrane is a senior fellow of the Hoover Institution at Stanford University and was previously a professor of finance at Chicago Booth. This essay is adapted from a post on his blog, The Grumpy Economist.

More from Chicago Booth Review

More from Chicago Booth

Your Privacy
We want to demonstrate our commitment to your privacy. Please review Chicago Booth's privacy notice, which provides information explaining how and why we collect particular information when you visit our website.