Another study, from the American Political Science Review, zoomed in more narrowly, looking just at similar counties on the borders between states that expanded and those that didn’t. Both papers showed increases in voter registration and voting. In the neighboring-counties research, people in the expansion counties were more likely to register and to vote, by three to four percentage points.
A third study, released as a working paper this week, looked at an earlier Medicaid expansion, when Oregon opened up its program to some childless adults in 2008. The state set up a lottery to determine who got coverage, which led to an important randomized experiment of how Medicaid coverage affects people’s lives. Unlike the 2014 research, the Oregon study was able to track voting by individual, not just region. It found that the Medicaid expansion increased voter turnout by 7 percent among the newly covered group in Oregon.
Voting is private, so none of the research could say whether the new Medicaid voters supported Democrats or Republicans. Mr. Haselswerdt, who did the statewide study, said it might even be possible that some of the voting increase by Medicaid beneficiaries was matched by increased voting by expansion opposers.
But the people eligible for Medicaid expansion tend to be poor, single adults, a demographic more likely to be Democratic-leaning. And the Oregon study showed bigger voting effects in more heavily Democratic parts of the state.
Even though all the studies detected an effect of Medicaid on voting, none of them showed one that lasted. Political science research suggests that voting can be a habit-forming behavior, but that may not be the case for this particular group. In each study, the voting boost after expansion had faded by the next election. That means that the coming expansions may make an electoral difference only once.
“Medicaid coverage policy is still front and center in the political debate,” said Katherine Baicker, a co-author on the Oregon work and the dean of the Harris School of Public Policy at the University of Chicago. “Having this kind of information about how Medicaid coverage affects not only health and health care use and financial security, but also civic engagement, seems like a really important relationship to have a little more information about.”