Elizabeth Linos on Reducing Stigma To Increase Participation in Safety Net Programs

Judith Siers-Poisson [00:00:06] Hello and thanks for joining us for the Poverty Research and Policy podcast from the Institute for Research on Poverty at the University of Wisconsin–Madison. I’m Judith Siers-Poisson. For this episode, we’re going to be talking with Dr. Elizabeth Linos about the paper that she coauthored with Jessica Lasky-Fink, titled “It’s Not Your Fault: Reducing Stigma Increases Take Up of Government Programs.” It was published as part of the Harvard Kennedy School Faculty Research Working Paper series. Dr. Linos is the Emma Bloomberg Associate Professor for Public Policy and Management and faculty director of the People Lab at the Harvard Kennedy School of Government. The majority of her research focuses on how to improve government by focusing on its people and the services they deliver. Elizabeth, thanks for joining us today.

Elizabeth Linos [00:00:51] It’s a pleasure to be here.

Siers-Poisson [00:00:53] So what do we know about how many households that are eligible for social safety net programs don’t use them?

Linos [00:01:00] So it turns out that this is a difficult question to answer correctly. One of the challenges is figuring out who is eligible but hasn’t already interacted with a government service so that we can actually count them or know more about their household and their income. But depending on kind of the estimate that we look at, somewhere between 20 to 50% of households are missing out on benefits that we think they’re eligible for, but they’re not, in fact, accessing them right now. And that really kind of ranges based on the benefits.

Siers-Poisson [00:01:31] So what are some of the reasons that that really large proportion of people who might be eligible aren’t accessing those services?

Linos [00:01:39] Yeah, I’m really excited about a new book that came out recently by Pam Herd and Don Moynihan called “Administrative Burden.” And one of the reasons I like this book is because it really helps us conceptualize the different types of barriers that people may be facing. And essentially, they outline three types of barriers. First, you have informational barriers. So it’s possible that people who are eligible for programs don’t know that they’re eligible, or they don’t know what steps need to be taken to be able to access the given benefit. Then you have compliance barriers or compliance hurdles, and these are kind of all the things that are required to access a benefit. It could be time to go into a government office. Sometimes there are fees associated with, you know, getting your documentation ready. So all of those kind of compliance hurdles that we put in place for people to show that they’re eligible or show that they need assistance, end up becoming a barrier for take up for many, for many families. And then there’s this third category of barriers, which they call psychological burdens or psychological barriers that really captures a wide range of potential reasons why people might not take up a program. One that is salient in the social safety net literature is stigma. So the fact that being poor in the United States is still stigmatized and being poor and accessing government assistance is even more stigmatized than being poor. But you could also think about things like trust or mistrust of how your data is going to be used. You might think about fear of discrimination when you actually interact with a government worker. So there are a whole bunch of psychological hurdles that people might be facing that are so strong that they are willing to forgo money, leave money on the table, so that they don’t have to kind of interact with a government service.

Siers-Poisson [00:03:30] In your paper, “It’s Not Your Fault: Reducing Stigma Increases Take Up of Government Programs,” you look specifically at emergency rental assistance and you say that that in particular is a highly stigmatized benefit. Why is that?

Linos [00:03:45] So, Jessica Lasky-Fink, my coauthor, and I did some kind of pre-work to see how people feel about different types of programs. We’re not exactly sure why rental assistance is so stigmatized, but when we looked at a bunch of different programs ranging from the Earned Income Tax Credit to SNAP to rental assistance, we saw that rental assistance is in fact one of the more stigmatized programs – similar to levels of stigma that we associate with mental health or obesity, which are two other very stigmatized characteristics of people in the U.S. One potential explanation is that this really has to do with the American dream and this narrative of housing as being really fundamental to your ability to succeed in the U.S. But there’s still a lot more research we need to do about why rental assistance in particular has been so stigmatized, and stigmatized even during COVID, when a lot of people lost their jobs through no fault of their own as the paper suggests.

Siers-Poisson [00:04:47] I found it especially interesting that you talked about two different kinds of stigma, a societal – or what could be called anticipated stigma – and also internalized stigma. So one more or less from the outside and one from the inside. Can you give us a short description of each and what that looks like?

Linos [00:05:05] Yeah, absolutely. So one thing that we’re looking into now is when we say stigma, what do we actually mean? So there’s a lot of really fantastic qualitative research and quantitative research that talks about welfare stigma more broadly. And of course, this is highly racialized and gendered. There are tropes that go back to the eighties about, you know, who participates in welfare programs. But if you actually take a closer look, there are many ways through which this idea of stigma could affect behavior. So we’re trying to separate those dimensions out to see if we can shift some of them. So if we take kind of societal stigma as this widespread, pervasive thing that exists in our culture, how that then manifests into decision making to go in different directions. On the one hand, we have internalized stigma. So that’s the shame that you might feel for being associated with a group. So think about not wanting to tell your friends that you need assistance or not wanting to admit to yourself that you need help. So that’s really a really strong motivator or a motivator that that sense of shame or guilt that you actually need assistance. That, of course, comes from the larger societal stigma, but is really affecting people internally. And then we separate out a separate type of stigma, which is really about fear of discrimination or anticipated stigma by other people. So there’s a really interesting set of research projects that look at anticipated stigma in health care, where, let’s imagine I come from a group that stigmatized and I’m nervous that my doctor’s going to judge me for my decisions or my life or who I love. And so I avoid going to the doctor because I just don’t want to deal with that level of discrimination. That’s not that I personally feel shame or internalize shame about how I live my life, but I know that other people are going to discriminate against me, so I want to avoid those situations. So we call that anticipated stigma. In the case of the social safety net, you could imagine both playing a role. So there’s both a sense of internalized shame or guilt, but also if you have to interact with a government worker or you have to go into an office, you might worry about how that interaction will go and whether or not that will feel discriminatory or will feel negative in any way. And you might want to avoid those situations.

Siers-Poisson [00:07:22] Your research looks at how stigma can affect whether people decide to access that program or not. And I think much more importantly, if reducing that stigma can actually increase uptake of services. And to determine that you looked at different changes in messaging about programs in two locations Austin, Texas, and Denver, Colorado. Can you describe Study One, which was in Austin?

Linos [00:07:45] Sure. Absolutely. So during COVID, there was this kind of unique moment in American history where a lot of funds were being disbursed to cities to support households that were at risk of eviction. And for a brief moment, there wasn’t enough demand, so there was too much money going into local government. And for those audience members who know local government, they know how rare of an event that is that money is not the problem. So we wanted to work with cities to figure out how we can make sure that people are accessing funds that were sitting there waiting for them in emergency rental assistance. And as you said, we started in Austin and in Denver. The project in the city of Austin really leveraged the fact that the city was already doing a bunch of really interesting outreach and was sending out emails to people they had on their email list to tell them about the program. So all we were testing there is, what is the language that should be on that email? Say you’re a government agency, you’re going to be sending out information about the fact that, you know, rental assistance is available. What should those emails actually say? And what we did is we tested two versions of that email that made slight tweaks to the language. So, you know, each of those tweaks we’re trying to get at a reduction or targeting kind of internalized shame or stigma as well as anticipated stigma, but are really nuanced or light touch if you think about it. So rather than saying “if you need assistance, call this number,” we say, “if you need assistance, it’s not your fault.” And so those very, very slight changes in the language are something that could at least signal to a potential beneficiary that the government doesn’t think that you are a bad person, the government doesn’t think you are immoral, or that you have done something wrong to be in need at this time. And what we find is that if you just tweak the language a little bit on this email, you get much more engagement with the email in terms of click through, if you use the more destigmatizing language. In Denver, which is Study Two, we tried to take that and see how that then translates into actual behavior applying for rental assistance, receiving funds, you know, really just follow people’s journey to see whether or not the destigmatizing language makes a difference in terms of actual dollars in people’s pockets.

Siers-Poisson [00:10:02] And so what did you see? You mentioned that in Austin, it was more about whether people were clicking through or not. And then in Denver, you adjusted the study to see if they were actually applying more. What did you find?

Linos [00:10:14] Yeah. So in Denver, we basically used very similar language but sent postcards to people’s homes. And, you know, this is one of both the beauties and challenges of doing projects with real government agencies is you work with what you can in terms of outreach efforts. So some people, you know, just had the status quo outreach efforts that existed throughout the city, including nonprofits reaching out, you know, things that were available on the website. But we wanted to test sending postcards that were kind of more targeted to people’s homes. So, one group got what we call the information postcard that provided information in simple and clear language about the fact that the program existed and how to apply. So, what number to call, in both English and Spanish. And then some people got this postcard that we’re going to call the destigmatizing postcard that again, looks very similar, but includes language that really pushes this idea that, you know, you shouldn’t feel shame and you’re not going to face discrimination. So rather than saying, you know, “are you not able to pay your rent?” we say, “Are you not able to pay your rent? It’s not your fault, it’s due to COVID.” Rather than saying, you know, “Call this number and we’re going to determine if you’re eligible,” we say “Call this number and we’ll help you determine if you’re eligible.” So really trying to bring the agency and dignity back to the household as opposed to making this kind of an anxiety-inducing interaction. And again, we find that the destigmatizing postcard ends up doing better. So not only does it do better than sending nothing at all, it does better than just sending information, although that’s directional. And what’s really exciting from my perspective is that then things progress in the same direction all the way through funds. So you get more people to ask for an application, you get more people to complete the application, and then you get more people actually receiving funds just by tweaking the language and outreach that was already happening. So the dollar value or the ROI on these changes is really, really high.

Siers-Poisson [00:12:14] What kind of percentages increase are we talking about?

Linos [00:12:18] So it depends how you look at it. The kind of destigmatizing postcard increases interest by about 40% compared to nothing at all, and about 11% compared to just giving information. But given that these are kind of zero-cost changes, these are really large and meaningful effects in terms of outreach and success. One thing that we looked at in Denver, which I only mention as an exploratory and preliminary finding, is that we also see a change in who applies, and it’s too early to see if that would be true in other places. But we see that receiving the destigmatizing postcard increases the number of applications from households that are headed by a Black or Hispanic family or household member. So to me, that’s really interesting and something that would require further research to kind of confirm. But what we see is that if you’ve received this destigmatizing postcard and you come from a household that is already facing other forms of discrimination, the rates go up. And that to me, is something that we should think about more as we think about how we do outreach and what works for home.

Siers-Poisson [00:13:29] So the next stage of your research, Studies Three and Four were online surveys. What were you trying to fill out in the story after the emails and then the postcards and then these online surveys?

Linos [00:13:42] Yeah. So one thing that is always true for those of us who do experiments in the real world with real governments, is that you can get a really good sense of what works in a given context. But you have to make some assumptions about what the underlying psychological mechanism is. And so what we tried to do with our online studies is confirm whether or not what we think is happening in the field is, in fact, happening. So in this case, we’re trying to figure out, yes, we see more people applying, but is it really because we’ve shifted stigma? So we go online, we show the same postcards or treatments to a group of online participants. And then rather than ask them to apply for a program, we ask them a series of questions about how they feel about shame. Both internalize stigma and anticipated stigma, how they feel about discrimination, as well as some additional questions about how they would experience receiving this postcard. And what we find is that the postcard that we call destigmatizing does in fact reduce internalized stigma significantly, and it does so without changing people’s beliefs about how easy it would be to get funds or without change in kind of broader comprehension, which could be some other mechanisms at play in this case. And so I think of the online studies as confirming the mechanism. But what I’m most excited about as a person who sits at a policy school are the field studies, because we can see that this actually works in practice with real households and real government agencies.

Siers-Poisson [00:15:18] So looking at the four different studies and the results from each, what would you say are some of those most important takeaways, especially in a policy context?

Linos [00:15:28] Yeah. So, you know, one thing that is true for any experimental research is that we shouldn’t really convince ourselves too quickly about what works and what works for whom, just based on one or two studies. My sense is that this opens the door for a lot of future research about the role of stigma in the social safety net. There has been a wide range of really interesting research that describes the existence of stigma in this space, but we don’t actually have a lot of studies to turn to where we were able to reduce stigma, and that had a significant effect on decision making. So, I see this paper as kind of opening the door for more research in that space to figure out how do we reduce stigma and whether or not reducing stigma matters compared to other types of efforts to improve the take up gap. The second thing I think that is always interesting from a policy perspective is how these kind of small decisions about how we describe programs can make a big difference. And I always say this to my government partners, you know, these are people who are motivated by public service, who are trying to do what’s best for the communities that they’re serving. And so if you know the choices of words end up keeping people away from programs or re stigmatizing a group that’s already being stigmatized, that’s something we should know and fix quickly because we can, because it’s a low-cost shift. So that’s not necessarily an answer, but I think it pushes for more A-B testing of how we talk about the social safety net and how that might affect people’s decision making more broadly.

Siers-Poisson [00:17:06] In your paper you mentioned that there have been some efforts to improve the language in communications related to safety net programs, but that at local, state and even federal levels, there’s still a lot of what you call “status quo language” being used. And I can imagine, you know, people are busy. They basically just change the dates on the same mailer that they’ve used for the last ten years. It’s understandable. But what are some key changes that you would hope to see that you think could decrease stigma and increase uptake of these programs?

Linos [00:17:37] Yeah, absolutely. So, you know, one thing I do want to emphasize from your question, Judith, is that, you know, government workers are busy. They have so much on their plate and putting together the specific language, the specific word choice of a sentence of a letter is almost too much to ask of any given public servant. So I think part of the role of academics like myself is to figure out what works best and share that in a way that’s useful and can be adopted by government agencies. I guess there’s kind of two big lessons. One is regardless of what the letter or communication is, we have a lot of evidence that simplifying the language and making clear kind of what next steps are for people is really, really important. There’s great research on the Earned Income Tax Credit, for example, that just simplifying information is the biggest predictor of change in terms of people taking up to see if they’re eligible for it. But beyond that, beyond this idea of just simplifying language and being really clear about what the next step is, my sense is that we need to start shifting away from talking about people in need and start thinking about how we can bring back a sense of dignity and autonomy to the households that we are aiming to serve. I’ve seen a lot of very well-intentioned efforts that focus on neediness or talk about, you know, helping people in need as part of a government outreach effort. And that might be appropriate if you’re fundraising to a philanthropic donor or if you’re trying to increase taxes. But when you’re actually interacting with households, it really is creating an identity for people that takes away a level of autonomy from low-income households. And so anything we can do to clarify that the person in charge is the beneficiary, not the government agency, and that the role of the government worker and the government agencies is to support the household in navigating the systems, the better. Again, we have a lot of work to do to figure out exactly what words work best in what context. But the general idea is to really try to shift away from stigmatizing language or language that implies that people are either at fault or are immoral because they need assistance.

Siers-Poisson [00:20:01] Well, Elizabeth, thank you so much for spending time with us and sharing your research. It’s really interesting. And I think, you know, really has a lot of hope to it, too.

Linos [00:20:10] Yeah, I think so. I think we are at a really exciting time for academia to play a role in making a difference in the world of practice because government agencies are excited about seeing what the evidence is. And there’s a lot of academics who are excited to make a difference in the real world. So, I’m hoping we’ll see a lot more of this kind of research in the future.

Siers-Poisson [00:20:32] We’ll be watching for it and will be interested to talk about it.

Linos [00:20:35] Thanks so much for your time.

Siers-Poisson [00:20:38] Thanks so much to Dr. Elizabeth Linos. She joined us to talk about her paper titled “It’s Not Your Fault: Reducing Stigma Increases Take Up of Government Programs.” You can find a link to it in the show notes for this episode. The production of this podcast was supported in part by funding from the U.S. Department of Health and Human Services Office of the Assistant Secretary for Planning and Evaluation. But its contents don’t necessarily represent the opinions or policies of that office, any other agency of the federal government, or the Institute for Research on Poverty. Music for the episode is by Poi Dog Pondering. Thanks for listening.