Elizabeth Linos, the Emma Bloomberg Associate Professor of Public Policy and Management and faculty director of The People Lab, and Jessica Lasky-Fink, a PhD student at the University of California, Berkeley and research director at the People Lab, tested the role of stigma as a barrier to participation in social safety programs.

In a recent HKS faculty working paper, they find that “subtle changes to the framing of rental assistance—one highly stigmatized benefit—increased interest in the program by 36% compared to providing information only and increased completed program applications by about 11%, with potentially larger effects for renters of color.”

Stigma as a barrier

Although government programs can alleviate poverty for Americans, studies have shown that 20 to over 50% of households that are eligible for such programs do not use them. There are many factors that contribute to this ”take-up gap,” including logistical and informational hurdles. One potential barrier, however, may be the stigma that is associated with using government assistance. There are widespread—and racialized and gendered—stereotypes that people living in poverty are lazy or even morally inferior. At the same time as Linos and Lasky-Fink write, “poor people who receive government assistance are more likely to be seen as lazy and undeserving of help than poor people who do not participate in benefit programs.”

While there is much research on the stigma associated with poverty in the United States, there are fewer studies on how stigma affects people’s choices to enroll in benefit programs. And the existing studies in this space show mixed results. In one study, a letter that aimed to reduce the perceived stigma associated with the Earned Income Tax Credit (EITC) did not increase take-up. Meanwhile, another study found that reframing Supplemental Nutritional Assistance Program (SNAP) benefits (i.e., “food stamps”)—which, unlike the EITC, is “traditionally one of the most highly stigmatized government programs”—did increase interest in the program.

De-stigmatizing emails received 36% more engagement than information-only messages

Reframing rental assistance

Linos and Lasky-Fink focused on rental assistance in their research. “Housing assistance is a relevant test case for this research because it is both a central component of the social safety net and more stigmatized than many other means-tested programs,” they write. The COVID-19 pandemic also presented a unique opportunity for this research since states and counties received a large influx of relief funds for temporary and emergency rental assistance programs, but struggled to disburse these funds to renters.

In a series of four studies, the researchers tested whether the framing of government rental assistance programs could reduce stigma and increase participation. They looked at both “anticipated stigma” (a person’s expectation of being discriminated against by others for using public benefits) and “internalized stigma” (the ways in which negative stereotypes make people feel disempowered or shamed).

In a randomized study conducted in Austin, Texas, they found that subtle changes to the language used to describe the city’s emergency rental assistance program in a mass outreach campaign significantly increased engagement with the message. They tested an “information only” message against a “de-stigmatizing” message (both delivered via email) that targeted potential sources of both anticipated and internalized stigma. The de-stigmatizing email received 36% more engagement than the one that only shared information.

Linos and Lasky-Fink replicated this study in Denver, Colorado, with mail-based (rather than email) outreach that aimed to connect residents with the county’s temporary rental assistance program. They found that “de-stigmatizing” outreach delivered via mail increased application requests by 18% and submitted applications by 11% compared to providing information alone, although these differences were not quite statistically significant. Moreover, they found suggestive evidence of larger effects for renters of color.

The authors followed their experiments in Austin and Denver with two online studies designed to tease out whether the “de-stigmatizing” message was, in fact, reducing stigma or if there was another explanation for the different responses to the messages—for instance, if recipients of the “de-stigmatizing” messages believed that it would be easier to apply for the programs or that there was a greater chance of receiving assistance should they apply. The online studies suggest that the “de-stigmatizing” message did reduce internalized stigma and, to a lesser extent, anticipated stigma relative to providing information alone.

Taken together, the results of their four studies, the authors write, suggest that stigma may indeed be a consequential barrier to participation in some safety net programs.

“Expanding on an expansive literature documenting the existence of societal stigma against low-income households who use government assistance,” the authors write, “these studies offer causal evidence that stigma may be a meaningful barrier to take-up of benefit programs, and demonstrate that internalized stigma can be shifted even in the presence of pervasive societal stigma.”

Photo by Frederic J. Brown / AFP via Getty Images