Which Voters Want to Expand Medicaid? Maybe Not the Ones You Think

Today, however, voters in some of those states have overruled their Republican lawmakers and approved Medicaid expansion. In 2017, Maine became the first state where voters approved a binding expansion resolution, despite strong opposition by the GOP governor. In 2018, three more reliably Republican states—Idaho, Nebraska, and Utah—also flouted their leaders’ positions and passed Medicaid expansion referendums.

Clearly, then, Republican voters are not reliably opposed to Medicaid expansion. So, if political party doesn’t predict voter behavior on the issue, what does?

A recent line of research suggests that one important factor is whether voters possess a college diploma.

David Matsa, a professor of finance at the Kellogg School, worked with the University of Virginia’s Amalia Miller to dig into Maine’s referendum results. They discovered a striking correlation between voters who had a four-year degree and those who supported Medicaid expansion.

“It seems counter-intuitive at first, because a college degree is closely associated with earning more money, and to qualify for Medicaid your income has to be close to the poverty line,” says Matsa.

Nonetheless, he and Miller found that a four-year degree holder was far more likely to vote for expansion than someone with less schooling. What’s more, they discovered that a college graduate’s field of study mattered. Maine areas populated more heavily by college graduates with humanities degrees were the most supportive of Medicaid expansion. Places with more associate’s degrees, meanwhile, tended to vote against expansion.

This held even when the researchers controlled for party affiliation, income, and racial and ethnic backgrounds.

“What’s not clear is to what extent the education itself helps form people’s views about social safety nets, or whether people who are inherently more altruistic are more likely to study humanities,” Matsa says. “Does reading literature and thinking about what it would be like to be someone else increase empathy and openness to government programs for the poor? It’s a tantalizing question.”

The Affordable Care Act and Medicaid Expansion

Medicaid has provided health insurance for poor families and children, as well as pregnant women, the elderly, and those with disabilities, since its 1965 inception. The Affordable Care Act sought to reduce the number of uninsured Americans by extending Medicaid coverage to low-income working-age adults without dependent children.

When the ACA first passed, it required states to expand Medicaid to all adults with income up to 138 percent of the poverty level, or risk losing federal funds for the program. Several states sued, and in 2012 the Supreme Court held that states must be allowed to reach their own decisions about whether or not to increase Medicaid coverage.

By early 2017, 31 primarily blue states had passed Medicaid expansion, while 19 mostly red states had not. Changes in the party in power also produced the expected effects. Democratic gains in Virginia’s state legislature in November 2017, for example, prompted a vote for Medicaid expansion there the following summer.

But that clear-cut partisan divide has blurred as citizens increasingly add their own initiatives to state ballots in the 24 states that allow such direct action. (In 18 of those states, those referendums are binding.)

In Maine, Republican Governor Paul LePage vetoed five different bills passed by the state’s legislature to expand Medicaid. Citizens then successfully petitioned to pose the question directly to voters, who passed the measure.

Who Supports Safety Nets?

As the first of its kind, the Maine Medicaid referendum intrigued Matsa. Although Medicaid expansion had been a hot topic for years, no one really knew how the issue aligned with individuals’ specific social and economic characteristics.

“We’ve always known that voters don’t necessarily agree on every issue with the politicians they support, but it’s difficult to delineate those differences,” Matsa says. Opinion polls are sometimes unreliable, he explains, because survey answers may not be thought-through and carry no real repercussions. One’s vote on a binding referendum, by contrast, follows an election campaign and involves real financial consequences and therefore better reflects voters’ true will.

Matsa and Miller examined official election results from the November 2017 referendum across 495 districts and compared the results to Census data in order to draw conclusions about Medicaid voters.

First, the researchers demonstrated that it was specifically a four-year degree that drove Medicaid support, rather than a two-year degree. In fact, they found that places with high rates of associate’s degrees were more likely to oppose expansion.

While previous research indicated that Medicaid expansion was greater in states with more affluent residents, Matsa and Miller found that high incomes alone did not correlate to expansion support. In fact, they uncovered a major split in areas populated by college-educated high earners (defined as making more than $100,000), which overwhelmingly supported expansion, and areas with high earners who had less education, where it was opposed.

The researchers also considered different aspects of Maine’s relative racial and political makeup.

While previous research, based on opinion polls, found that those who identify as African-American and Latino were among the most likely to support Medicaid expansion, Matsa and Miller found race had virtually no effect on the impact of a college education in determining one’s Medicaid viewpoint.

Similarly, the researchers found that while party affiliation slightly dampens the relationship between a degree and support for Medicaid expansion, Republican college graduates still favor it more than other Republicans.

Why Does College Make Such a Difference?

You might expect that earning more money lowers support for safety-net programs, because such programs raise wealthier people’s taxes while offering limited personal benefit.

A four-year degree strongly mitigates that self-interest effect—but why?

Matsa is quick to point out that he is not sure. After all, showing there’s a relationship between education level and political support does not illuminate which causes the other. And yet, the current research offers a few clues.

The findings suggest that economic inequality in neighborhoods may impact whether someone supports Medicaid expansion.

Based on Census data, Matsa and Miller noted that Maine’s college grads tend to live in more densely populated urban areas, where residents are more likely to have lower-income neighbors, than in suburban and rural areas, where high earners with less education tend to live and which are generally more economically stratified. Nevertheless, the correlation with education still holds when controlling for both population density and local income inequality.

Stark differences in support for Medicaid expansion depending on college major also suggest possible answers.

The Census tracks residents’ area of study, so the researchers were able to distinguish between the percentage of arts and humanities majors, education majors, science and engineering majors, and business majors across Maine’s 495 voting districts.

Areas with more humanities majors voted most strongly in favor of expansion, while areas with more business majors had a negative correlation almost as significant as that of areas with lots of two-year college grads.

People who earned master’s and doctoral degrees were even more supportive of expansion, while those who earned professional degrees—including medical degrees—were less.

Taken together, all these findings raise “the big question of whether a college education—and, specifically, a college education that isn’t vocationally focused—actually increases support for government programs that help the disadvantaged,” Matsa says.