Poor People's Vote: Do Welfare Sanctions Impact Civic Participation?

Background/Purpose:People living in poverty in the U.S. are consistently less likely to turn out for U.S. elections than more wealthy citizens. Yet poor people also interface directly with government entities through participation in public assistance programs. Past studies have indicated that experience with some public programs, such as the GI Bill, may foster civic engagement (Mettler 2005). Conversely, Soss (2000) demonstrated that experiences with more “authoritarian” programs, where participation has stringent requirements and failures are sanctioned, leads to decreased civic participation. Bruch, Ferree, and Soss (2010) modeled voting and participation in voluntary organizations as a function of paternalism in a state’s welfare program (e.g., time limits, family caps, etc.) and found state-level paternalism significantly decreases political participation, even controlling for selection effects. But does this relationship hold when considering individual’s varying experience of paternalism within one state welfare program? Building on Lee, Slack, and Lewis’ finding (2004) that welfare sanctions negatively impact economic well-being and employment outcomes, we examine the relationship of sanctions with civic participation in Illinois.

Methods: We use data (n=1,005) from the Illinois Families Study (IFS), a longitudinal study of families who were receiving Temporary Assistance for Needy Families (TANF) in 1998 that used stratified random sampling to generate representative results from a survey of urban and rural county residents in Illinois, supplemented with individual level administrative data. Our key outcome of civic participation is measured by self-report of voting behavior in the 2000 election and participation in civic or voluntary organizations. Our key predictor is the imposition or threat of TANF sanctions as indicated in the administrative data. Covariates include county, demographics (such as race, age, gender, and marital status), whether respondent had a car or driver’s license, and past use of welfare. We model an interaction between threat of or imposed sanctions and education, as respondents without a high school degree (38% of sample) are both more likely to receive a sanction and less likely to vote.  We examine relationships with logistic regression.

Results: Illinois has a mixed record on TANF paternalism, with relatively generous provisions for childcare and income disregards but stringent rules including family caps and full-grant sanctions for noncompliance. In our sample, 25% of respondents had received a threat or sanction in the recent past. Of respondents who were not sanctioned, 62% report voting in the 2000 election, compared to 54% of those sanctioned (p<0.05). Logistic regressions indicate that, holding all else equal, a sanction threat or imposed sanction decreases the odds of voting by 54% (p<0.01). Similar patterns hold for participation in voluntary organizations.

Conclusions/Implications: This study provides new evidence that individuals’ personal experience of authoritarianism in public benefit programs, measured by TANF sanctions, depresses voter turnout.  Given that low-income voters are already less likely to vote, this raises serious concerns about how social policies foster or dampen civic engagement and the enfranchisement of marginalized communities.