One of the clearest patterns in public health research is the correlation between income and health. The richer you are, the more likely you are to have good health, and live a long life. The poorer you are, the more likely you are to be sicker, and die younger.
That data could be an argument for just giving poor people money to improve their health. But in general, the way states and the federal government try to make poor people healthier is by giving them health insurance instead.
CreditJulio Cortez/Associated Press
Now the Trump administration wants to make a major change to Medicaid, the main program used to provide health care for the poor. Instead of simply providing Medicaid to poor Americans whose incomes are low enough, it is encouraging states to require childless adults without disabilities to work or take part in “community engagement activities” in order to obtain Medicaid.
The idea behind the change is that requiring work will help move more Medicaid beneficiaries into the work force and out of poverty. And, legally, it rests on the contention that the requirements themselves will make those people healthier, since improving health is the stated purpose of the Medicaid statute.
There is not strong evidence for that contention.
“Higher earnings are positively correlated with longer lifespan,” says a guidance document published by the Centers for Medicare and Medicaid Services on Thursday.
One comprehensive review of existing studies found strong evidence that unemployment is generally harmful to health, including higher mortality; poorer general health; poorer mental health; and higher medical consultation and hospital admission rates. Another academic analysis found strong evidence for a protective effect of employment on depression and general mental health. A 2013 Gallup poll found that unemployed Americans are more than twice as likely as those with full-time jobs to say they currently have or are being treated for depression. Other community engagement activities such as volunteering are also associated with improved health outcomes, and it can lead to paid employment.
(One citation in that passage is to a health insurer’s survey that asked people who volunteer how it made them feel.)
But it is not at all clear how much work or income alone improve health. In fact, there’s quite a lot of evidence that causality can move in the opposite direction, since health problems can make it difficult to work or go to school. It might be, as the guidance document suggests, that people who don’t work become depressed. Or it could be that some people who are depressed have a harder time getting or keeping a job. Treating their depression might ease their symptoms and improve their employment prospects. “Having the medical coverage helps people to get a job,” said LaDonna Pavetti, a vice president at the liberal Center on Budget and Policy Priorities, who has studied work requirements extensively.
Some research does suggest that a boost to income can improve people’s health. Yet a study that examined the earned-income tax credit, a program established specifically to raise the incomes of low-wage workers, wasn’t able to find any clear health benefit.
More crucially, a work requirement does not guarantee that Medicaid beneficiaries will have an increased income or even find a job. Some people may be nudged by a work requirement into work they wouldn’t have done otherwise. But others may simply be kicked out of the health insurance program because they aren’t able to meet the requirement. States will be allowed to cut off Medicaid benefits to people who can’t demonstrate that they’ve completed the required number of work or activity hours or who don’t qualify for an exception related to a health problem or having a young child. Others might meet the work requirement rule and keep their Medicaid, but still remain poor, because they are able to get only a volunteer or part-time job that keeps their income low.
Studies of the Medicaid population suggest that most of them work already or would qualify for exceptions. The guidance says that states will need to carve out exceptions for people who are medically frail and for people with opioid addictions, so they won’t lose their benefits. But anyone who fails to complete or document their required work and loses benefits won’t get the health benefits of Medicaid coverage or any of the described health benefits that the administration says come from work. States that have asked for work requirements — 10 have submitted applications already — have estimated reductions in Medicaid enrollment, and not all because of people finding other kinds of health coverage.
Work requirements have been used in other social welfare programs, and they have been studied extensively in cash assistance programs, both as pilot programs in the old welfare system and as a standard feature of the Temporary Assistance for Needy Families program initiated by welfare reform in the 1990s. Work requirements in cash programs share features with the Medicaid requirements described in the document. (Indeed, the document suggests that states might wish to line up the rules and support services between the programs, since some people will be enrolled in both.)
The general conclusion is that the work requirement increased the number of welfare beneficiaries who worked in the short term but had little impact on their long-term employment prospects or their long-term earnings. While work requirements appear to have nudged some beneficiaries who were close to the poverty line above it, they do not appear to have meaningfully improved the long-term financial circumstances of eligible people, who were predominantly parents of young children with low levels of education.
“The work requirement itself doesn’t necessarily lead to employment, and especially not into work that will lead them out of poverty,” said Heather Hahn, a senior fellow at the Urban Institute, who was recently a co-author of a summary of the evidence about work requirements.
Robert Rector, a senior research fellow at the Heritage Foundation, who helped write the welfare work requirements and who is a strong advocate for them in many more social welfare programs, said he does not think they are a good match for a health insurance program. Mr. Rector said he’s not surprised that work requirements failed to push people into middle-class jobs. “It is an irrational expectation, and it’s not an expectation I have,” he said.
He said the goal of work requirements in programs like welfare is not to reduce poverty, per se, but to make sure that people who are using public programs are taking steps to contribute to society and improve their independence. He described the rationale as a moral one.
The Trump administration is making a different argument, in part because it is trying to make the program change through state waivers, and has to work within the confines of Medicaid law.
Medicaid has a lot of rules about whom it must cover and what services must be included, but it includes a very broad exemption. States that want to undertake demonstration projects that may further the goals of the program can waive many of the normal rules. But the Medicaid statute doesn’t include promoting work as one of its explicit purposes, which is why the document is trying to make a connection between a requirement and health.
There are likely to be legal challenges to any approved requirements, and the court cases are likely to hinge on the question of whether a work requirement can promote health. Eliot Fishman, who ran Medicaid’s state demonstrations program until May, and is now a senior director at the consumer advocacy group Families USA, said the Obama administration examined the evidence when states asked for work requirements before, and found them inconsistent with the program’s purpose.
“I don’t think this is an empiricist’s approach,” Mr. Fishman said.