Political Underrepresentation Among Public Benefits Recipients
Evidence from Linked Administrative Data
Seth Chizeck (Carnegie Mellon University), Kelley Fong (University of California, Irvine), Rebecca Goldstein (University of California, Berkeley), and Ariel White (MIT)
In November 2020, all eyes were on Pennsylvania in the lead-up to the hotly-contested presidential election. Four years prior, Donald Trump had carried the state by under 45,000 votes, out of more than six million ballots cast. Given its pivotal position as a presidential swing state, campaigns and grassroots groups blanketed the state to register and then turn out people to vote. Turnout in that election broke modern records.
But one key group of eligible voters was underrepresented among the record-high electorate: people receiving means-tested public benefits. Studying voter registration and voting in a large county in Pennsylvania, we found that people enrolled in means-tested public benefits programs register to vote and vote much less often than non-recipients.
Measuring voter turnout among public benefits recipients has been difficult due to data limitations. But by linking the county’s administrative data on public benefits receipt to state voter records, we were able to fairly accurately compare turnout rates of eligible voters who do and do not receive means-tested public benefits, including Temporary Assistance to Needy Families (TANF), the Supplemental Nutrition Assistance Program (SNAP), Medicaid, Supplemental Security Income (SSI), childcare assistance, and housing assistance.
In the 2020 general election, we found that 45% of program recipients voted, compared with 84% of non-recipients. Public benefits recipients were thus over 20% of the voting-eligible population but only 12% of voters.
Put another way, there was a nearly 40-percentage-point gap in turnout between benefits recipients and other residents in the 2020 general election. If people enrolled in the public benefits programs we study had registered and turned out to vote at the same rate as the rest of the county, the county would have seen more than 75,000 additional ballots cast in November 2020, roughly 10% of its actual observed vote total. (Despite the Covid-19 pandemic creating unusual voting circumstances and public benefit needs in 2020, the voter turnout rates of public benefit recipients – and the relative turnout rates among recipients of different public programs – were similar in 2016 and 2018.) This pattern is consistent with what we know about how groups overrepresented among public benefits recipients, such as lower-income people, Black people, and younger people, face greater barriers to voting.
This turnout gap matters because the underrepresentation of benefits recipients among voters likely yields an electorate with different political knowledge and views than we would see under participatory parity. National survey data finds means-tested benefits recipients are more likely to see the government as responsible for assisting those in need. And new research finds that public benefits recipients are less supportive of governments imposing “administrative burdens” that make it more difficult to access benefits, even accounting for demographic characteristics like age, gender, race, education, and income.
Political science research documents persistent “turnout bias,” wherein policy tends towards the preferences of those who turn out to vote. Thus, to the extent that benefits recipients are more supportive of generous welfare policy than non-recipients, this underrepresentation may obscure popular preferences for social welfare provision. The relative absence of program recipients in the electorate may also mean that politicians are less likely to see administrative burdens as issues of concern for their constituents, potentially making them more attentive to other concerns, such as concerns about fraud.
Public benefits recipients have personal experience with and intimate knowledge of the details of government policy implementation. Yet we find that this knowledge is not being incorporated into the formal political process. Moving towards a more representative democracy requires broadening political participation to include people from all walks of life.
Seth Chizeck is a PhD student at Carnegie Mellon University. His research explores the economic effects of the social safety net and the effectiveness of interventions that are meant to help lift people out of poverty.
Kelley Fong is an assistant professor of sociology at the University of California, Irvine, where she studies social inequality and family life. In particular, her research examines state interventions into families and families’ experiences with state and social service systems.
Rebecca Goldstein is an assistant professor of law and political science (by courtesy) at the University of California, Berkeley. She studies how politics shape punitive state institutions and how punitive state institutions shape politics.
Ariel R. White is an associate professor of political science at MIT. She studies people’s everyday experiences with government and how they shape our political life, with a particular focus on punitive interactions.