Work Requirements: Moral Judgements vs. Incentives.

The work requirements for Medicaid are in the news. All the news coverage I have seen describes this as a morality play. On the right, we shouldn’t give federally paid free health care to “able-bodied” young men who choose not to work. On the left, “people will die!” NPR set me off, of course, with a long bleating story that the paperwork alone to document 20 hours of week of work or training or school or volunteering or whatever is such a burden that, well, people would choose to die instead. Even a $35 copay is heartless.

Economists, at least those mindful of the limits of our expertise, really have nothing special to say about this sort of moral judgment. Except that it is entirely the wrong question to ask. Debating whether one should support the spending of other people’s money, based on the anecdotal moral worthiness of the recipients is not a great way to produce effective public policy.

Economics knows about incentives, not transfers. It knows about cause and effect, not moral worthiness of forced philanthropy. Incentives are the key case for work requirements.

Medicaid is free health insurance so long as you earn less than un upper limit. If you earn more than this limit, you lose Medicaid. Consulting Google,

For Medi-Cal in California, the income limit is generally up to 138% of the Federal Poverty Level (FPL) for adults, and up to 266% FPL for children. For example, in 2025, a single adult can earn up to $21,597 annually to qualify, while a family of four can earn up to $44,367. For children, the income limit is higher, with a family of four potentially qualifying with an income up to $85,519.

$21,598 and you lose Medi-Cal.

So, Houston, we have a problem: People whose incomes are near this limit have a big disincentive not to work more, harder, study, take a better job, move, etc., if that would cause them to exceed the income limit.

You may dismiss losing Med-Cal as a significant disincentive, but work disincentives come from all social programs working together, not each acting in isolation. (This is a frequently forgotten and important lesson!) A Medicaid recipient might also have food stamps (SNAP). SNAP also has an income limit, also about 130% of the poverty line. (SNAP has a work requirement after three months.) So, if you rise from 120% to 140% of the “poverty line,” you lose both SNAP and Medicaid. Add taxes, phaseout of earned income and child tax credits, “affordable” or section 8 housing income limits, income limits for education subsidies, all the say down to the low-income bus pass and free tolls, and the marginal tax rate on low income Americans is as high as 100%, with many cliffs. Perversely, the income tests do not include non-cash benefits, so that signing up for another government program doesn’t trigger loss of one but working does. Indeed programs increasingly cross-qualify so that signing up for one makes it easier to sign up for another.

Why not just remove the income limits, you ask? Well, Medicaid spends roughly $10,000 per recipient. If everyone gets it, eliminating this part of the disincentive effect, then it would cost 340 million people x $10,000 = $3.4 trillion. That’s half the federal budget right there. Government transfers must be limited in some sense to people who “need” them. But income as an indicator of need has a problem, in that (this is news to the Left) income is not an immutable characteristic. I hate the word “low-income person.” There is no such thing as a “low-income person.” There is a person who at the moment is not earning much, for a variety of reasons including choices and mutable circumstances as well as a much smaller component of innate and immutable characteristics.

And even were we to remove the phaseout disincentive, there is a second disincentive. Whether or not the program phases out with more income, why bother working if you can get free (or highly subsidized) food, housing, health care, and often some cash too? Work is hard, and not much fun (except if you’re a senior fellow at a great think tank, but sadly not everyone can have my job). Indeed, in the Obamacare debates the mainstream news was full of stories about the awful situation of people who worked, though they didn’t really need the money, just to get health insurance. That’s a darn good reason to work!

Economists call these the “substitution” effect (working less if the government takes away benefits when you earn more) and the “income” effect (working less if you have money from other sources). Trust fund kids generally don’t work as hard as normal people, even when they face the same (or better) opportunities to turn time and effort into after-tax consumption. These are common sense effects which your grandmother could tell you about. Economics helps to formalize and quantify them (and occasionally to ridiculously deny them under a cloud of fashionable econometrics.)

The tradeoff of help for disincentive has no easy answer.

Enter work requirements. Well, if work does not pay, literally, one can try to offset some of the disincentive by forcing people to work in order to receive benefits. The “incentive” of the work requirement formula does something to offset the disincentive of the (necessarily) income-capped benefits.

It is an imperfect fix for sure. Work that you do though you don’t want to in order to persuade a bureaucrat to sign your check is going to be much different than work you do voluntarily in order to get money. Would you want such an employee? The right measure of work is, will someone pay you to do it, not how many hours a week you spend at it. There is a Swiss Cheese of exemptions, naturally. It will be interesting to see just how quickly the (say) State of California bureaucracy that certifies if your “work” counts as “work” and program beneficiaries game the system so that everyone qualifies. (I several acquaintances on Medi-Cal. One is an “artist,” mostly involved in left-wing politics, and mostly living off his girlfriend. He’s probably exactly what the right does not want to support. I can’t wait to see how Medi-Cal treats his “work.” Does 20 hours a week as a “self-employed” political artist count? However, when he recently had a health issue I learned just how awful Medi-Cal is. The NHS looks like the Mayo Clinic in comparison. ) But some exemptions make sense. The program was originally intended for people who really could not work, not as a NHS for anyone with low income. That too is more a moral question than an economic one.

But if you have a welfare system with, necessarily, huge incentives (and moral hazard), you have to do something to offset the disincentives. The Nordic countries, which our leftists last visited in the late 1980s before they ditched socialism, are pretty hard nosed about controlling disincentives and requiring people to work in order to receive benefits.

If we could get our public discourse to frame things this way, we could, I think produce much better programs. Time, for example, is a better limit than income. I think we expect “able bodied” people who have suffered misfortune to get back to a normal life.

There are 100 other things wrong with Medicaid, and I simplified a lot to make the basic point in this post. If you earn out of Medicaid, you typically then go on to subsidized Obamacare so it’s not really a cliff. Those subsidies phase out with income as well, however, so it’s still a greased flagpole. On the other hand, it’s not clear to me what happens if you are dropped from Medicaid by failing the work requirement. Obamacare has a minimum income.

The media also paint the issue as Federal Government or die. States can raise taxes and pay for insurance too. Many people on Medicaid have alternatives. Economics teaches us there are a lot of substitutes for anything.