How Medicaid for Children Recoups Much of Its Cost in the Long Run
Medicaid payments
By: Margot Sanger-Katz (NY Times) January, 2015
When advocates talk about the advantages of government health care, they often talk about a moral obligation to ensure equal access. Or they describe the immediate health and economic rewards of giving people a way to pay for their care.
Now a novel study presents another argument for the medical safety net, at least for children: Giving them health coverage may boost their future earnings for decades. And the taxes they pay on those higher incomes may help pay the government back for some of its investment.
The study used newly available tax records measured over decades to examine the effects of providing Medicaid insurance to children. Instead of looking at the program’s immediate impact on those children and their families, it followed them once they became adults and began paying federal taxes.
People who had been eligible for Medicaid as children, as a group, earned higher wages and paid higher federal taxes than their peers who were not eligible for the federal-state health insurance program. And the more years they were eligible for the program, the larger the difference in earnings.
“If we examine kids that were eligible for different amounts of Medicaid over the course of their childhood, we see that the ones that were eligible for more Medicaid ended up paying more taxes through income and payroll taxes later in life,” said Amanda Kowalski, an assistant professor of economics at Yale and one of the study’s authors.
The results mean that the government’s investment in the children’s health care may not have cost as much as budget analysts expected. The study, by a team that included economists from the Treasury Department, was able to calculate a return on investment in the form of tax revenue.
The return wasn’t high enough to pay the government back for its investment in health insurance by the time the children reached age 28, when the researchers stopped tracking the subjects. By that age, the Treasury had earned back about 14 cents for every dollar that the federal and state governments had spent on insurance. But it did suggest that, if the subjects’ wages continued to follow typical trajectories as they aged, the federal government would earn back about what it spent on its half of the program by the time the children reached 60 — about 56 cents on the dollar, calculated using a formula that took into account the time value of money.
The split in spending between the federal and state governments for Medicaid varies by state, but, on average, federal taxpayers pay 57 cents of each dollar. There may also be some return on investment for states that collect income taxes, but the researchers didn’t measure that.
Here’s what that means in real numbers: The average person in the study with no Medicaid earned about $149,000 by age 28. For each year a person was eligible for Medicaid, that income went up by $250, and the taxes the person paid went up accordingly.
“What’s exciting about this is how good the outcome variables that they can look at,” said Janet Currie, a professor of economics and public affairs at Princeton. A few studies have tracked the health outcomes of children who were eligible for Medicaid over time, including one Ms. Currie wrote, but the study’s measures of economic outcomes are new.
The new paper was made possible by a series of policy changes throughout the 1980s and 1990s that slowly expanded Medicaid to cover more and more American children. The changes essentially happened in two phases: First, the federal government allowed the program to include older children, and then individual states approved expansion to those groups. The slow, state-by-state spread of the policy enabled the researchers to compare children who were eligible for Medicaid with a control group of similar children of the same age and family income level who were not eligible for the program. The study looked at children who were eligible for Medicaid, even though not every eligible child actually signed up.
Expanded eligibility had two other important effects closely related to the earnings statistics: Children who were eligible for coverage were less likely to die before reaching 28, and they were more likely to attend college. Those are outcomes that, Ms. Kowalski points out, the government may value even if the program doesn’t return any money to the Treasury.
The study can’t entirely explain how access to childhood health insurance helped low-income children earn more later in life. But Ms. Kowalski has a few theories. One is that it may have helped the girls, in particular, by offering them a way to get contraception (which Medicaid covers to varying degrees in all states) and avoid unplanned pregnancies. The earnings effect was much more pronounced for girls than it was for boys.
The difference may also come from the way that public health insurance changed the budgets of the children’s families, she said. By taking care of health care bills, Medicaid may have freed the parents to make other investments in their children’s development that paid off.
Ms. Currie said that earlier studies of children’s health outcomes also suggest that children with serious illnesses often go on to be sick as adults as well — meaning they are more likely to miss work or have limited career options. Medicaid supports and funds a lot of important preventive health care for very young children. She said the lesson could be that “an ounce of prevention is worth a pound of cure.”
Now that the earlier expansions have had a chance to spread, nearly every low-income child in the country is eligible for Medicaid, and more than a third of all American children are currently enrolled in either Medicaid or a closely related federal-state program, called the Children’s Health Insurance Program.
“If this is right, then we’re going to be seeing a lot more impact for the kids that are born now and in the future,” said Judy Solomon, a vice president for health policy at the left-leaning Center on Budget and Policy Priorities.
