How Can Behavioral Science Help Programs Better Serve Clients During the Pandemic?


Why don’t government social services programs better serve families struggling through crises like the COVID-19 pandemic? One reason is that these systems are designed for compliance over access. Many of those who are in need and qualify for benefits are deterred by administrative burdens, including excessive steps and paperwork. Insights from behavioral science can help agencies and nonprofits find ways to streamline their processes and simplify their communications with clients.

In this episode of Evidence First, Leigh Parise interviews Rebecca Schwartz, a research analyst in MDRC’s Center for Applied Behavioral Science (CABS). Rebecca describes real-world examples of how CABS has worked with agencies to improve their service delivery to families. She also highlights how the federal government adopted an important behavioral technique — prospective eligibility — in the recent distribution of the CARES Act stimulus checks.

Read the transcript

Leigh Parise: Policymakers talk about solutions, but which ones really work? Welcome to Evidence First, a podcast from MDRC that explores the best evidence available on how to improve the lives of low-income people. I’m your host, Leigh Parise. COVID-19 has pushed families into deep economic distress. Many people have lost their jobs and are struggling with issues, such as food insecurity and housing instability. There are a number of government programs and services designed to keep people afloat during this time. However, administrative burdens and complicated eligibility processes often make it difficult for families to access these supports.

I spoke with MDRC’s Rebecca Schwartz to learn how insights from behavioral science can be used to strengthen programs and better support those in need during a crisis like COVID-19. Rebecca is a research analyst in MDRC’s Center for Applied Behavioral Science or CABS. Thanks so much for joining me today, Rebecca. It’s really great to have you.

Rebecca Schwartz: Great. Thanks so much for having me, Leigh. It’s very nice to be here.

Leigh Parise: All right. So let’s start off just with the first question for people who are listening and may not know, can you tell us what is behavioral science and why does it matter now?

Rebecca Schwartz: Sure. Behavioral science is really about how and why people make decisions. So we use evidence from economics, psychology and a range of other academic disciplines. In addition, at CABS, we try to incorporate the lens of human-centered design, which really means putting ourselves in the shoes of individuals who are facing a lot of barriers, potentially generally, and especially when we look at a time like COVID-19. So what we try to do is use an empathetic approach, again, putting ourselves in the shoes of the people that we work with and trying to really understand the context. And we use that approach to try to understand what’s working and what isn’t for both clients and social service programs.

Leigh Parise: That, to me in this moment, feels particularly responsive and thoughtful and respectful of the people that the services we might be engaging with are intended for and trying to help. Right?

Rebecca Schwartz: Yes, that’s definitely what we’re going for. Let’s take a working mother in these times. Ideally, we would actually speak with this individual so that we’re not just layering on our assumptions, but at the very least we might try to picture someone we actually know. So going back to this working mother, let’s say she’s trying to balance two Zoom classes at the kitchen table, and she also has her own full-time job. Or perhaps in this economy, she might be one of millions who lost their job and is job searching while also worrying about her kids falling behind in school. Then you layer on top of that, the just incredible stress of COVID-19 as an illness. So she might be worried that she or her kids or her parents might get sick. So you’re looking at just layering on a number of barriers.

Leigh Parise: So, there are government supports out there, like food stamps or unemployment insurance, but I know that it’s often hard for families who are eligible for those benefits to access them. Why is that? What are the barriers that are faced by families?

Rebecca Schwartz: While we haven’t really had a chance to rigorously study programs under COVID-19, we see a lot of common barriers across human services, and we think that a lot of those resonate today. So there are three main buckets that we put these barriers into. One is information. Do families clearly understand what they’re eligible for and not eligible for? And even more than that, do they even know where to start looking? Do they have to read and understand long documents with fine print, as many of us have seen with government documents? A second category is motivation. Why might families feel like they should apply given everything they may have heard from friends or seen on the news about long waiting times or complicated applications. And, finally, if you have the information and the motivation, are you going to really have the tools to follow through? So we want to think about using that empathetic approach. Once a family starts to enroll, how many different steps might they have to complete?

Leigh Parise: Right. So I think all of those barriers make a lot of sense that I wonder if you hear a lot from people, “Well, of course you should do that, and that’s obvious.” Because some of it, it seems really straightforward, but then I think all of us have had an experience where we get something from our kid’s school, or we get something from a doctor’s office and it asks you to go through all of these steps or buries the lede. And it doesn’t actually put right up front the information that is most critical for you to know. I wonder if that’s, “Oh yeah. Some of those things are obvious.” Or it’s something that you hear from people when you talk about behavioral science and insights that we could use in our lives and in our communications?

Rebecca Schwartz: Yes, absolutely. I think that people might say it’s obvious. I prefer the word that it’s intuitive. It makes sense because it’s about human nature and we are human beings. But then there’s a second level, which is that people know that this stuff is important, that clear communications and just putting in a box an appointment date and time, or making the first line clear of like, “This is what you should do. This is the next step you should take.” And people know it, but they don’t do it.

Partially, that’s just because people are busy, and behavioral science would also talk about what we call status quo bias, right? People are just accustomed to doing things the way they’ve always been doing them. And there’s a reason for that. There are even evolutionary advantages. We can’t reassess every little part of our personal and professional life every day or we might go a little crazy. But if we did, we might also actually find things that could be handled in a more efficient and effective way. So we’re not asking people to do their entire personal or professional lives, but we do think it’s worth organizations really printing out and reading from — you brought up an example from a school — reading from the parent’s perspective, anything that parents are getting and really interrogating, “Is it clear what I’m supposed to do? And by when and how?” And so many times those three questions that seem relatively small are just not clear at all.

Leigh Parise: That seems like really good advice. And certainly it’s always people’s intention to get their information across and be clear for people. But I think taking a step back and really thinking through those questions is useful guidance for anyone. MDRC recently co-authored a piece with ideas42 in Governing magazine. In it, you talk about how state and local agencies can use insights from behavioral science to redesign government programs so that they can better support families. Can you talk to me about what some of those strategies are?

Rebecca Schwartz: One of the most compelling strategies I think is to simplify or shorten a process. If you take the division of child support that we worked with in Cuyahoga County, Ohio, we partnered with them to shorten just a three-step process into a two-step process. And by eliminating even one step, more parents decided to update critical information related to their child support requirements. I really can’t say it enough, but one thing that behavioral science shows is that every single extra step or even an extra sentence in a form is just enough of a hassle, then it can and does dissuade some people from applying for a benefit, even if they really need that benefit. Plus every extra step has a burden on staff, who have to renew and process every single form. Now, a second thing agencies can do is to review and revise their communications from letters, emails, their websites, or even staff talking points. Clear, concise communication is really key to promoting informed decision-making.

For example, CABS did a different study with Ohio community colleges that focused on summer enrollment. We learned how to highlight available resources for students in a way that actually led to immediate action. In this study, we used one message to remind community college students of two things at the same time. First, that students had available grant money, always good to hear. Second, that students could actually use that money to sign up for summer courses. Connecting the dots for students in that way in the spring increased enrollment in summer courses. And that can lead to both students and colleges helping boost college completion, which is so important. These are just two examples of really high-impact projects that CABS has done always in collaboration with practitioners. We just feel like we have consistently demonstrated that even really small changes on how you present choices can really influence decision-making.

Leigh Parise: Great. Thank you. I think makes a lot of sense. And I think for anyone in this time, just being able to reduce some level of burden by simplifying messaging or making it so that you don’t need to take that extra step makes a lot of sense. I appreciate that you shared that specific example from community colleges. I know that the postsecondary and CABS teams at MDRC now are thinking about how can we take what we learned about how behavioral messaging can increase summer enrollment through that work that was done in partnership with community colleges, to think about how that can be applied now to enrollment outside of summer sessions — just more generally helping students understand what the current state is at their particular college and what steps it takes to get re-enrolled or to get signed up for courses again. So I love that the teams are really thinking about what’s next and how to take some of these lessons and put them to use given the current context. Another strategy, Rebecca, that you talk about in that [Governing] piece is called proactive eligibility. What’s that?

Rebecca Schwartz: Proactive eligibility is really not surprisingly about being proactive. It’s when an organization actually takes the extra step to let people know that they are — or even probably are — eligible for benefits or services. Any organization we were talking before about colleges, for example, or take a government benefits program, they should be able to use some data they already have to let people know that they might be eligible for something. These notifications help ensure, they really help close gaps, so that individuals know they’re eligible for say, emergency food assistance, rather than just assuming that individuals will discover available benefits and then apply. So with proactive eligibility, a person can continue or decline a benefit, but proactive eligibility can just be so helpful because it shifts the burden away from someone who may be having the worst month of their life, as many of us unfortunately have during COVID-19. And it makes better use of the wealth of data that most agencies collect that oftentimes just sits in the computer.

Leigh Parise: One really interesting example you talk about is the federal stimulus payments that were part of the CARES Act. Tell me about how eligibility was determined for those payments and how that approach could be used in other contexts.

Rebecca Schwartz: With the CARES Act, as many people know, stimulus checks of up to $1,200 were distributed automatically to millions of eligible Americans. From a behavioral approach, this eliminated all of the hassle factors that we’ve talked about, about understanding what you’re eligible for, applying, waiting to see if an application goes through, staff needing to process that application. What the CARES Act did — we were just talking about data ­— they used data that the IRS already had, which is pretty baseline data of what the IRS has: income tax returns and addresses. And they even used direct deposit information if they had that on file. An important caveat is that, did they miss people in need? They absolutely missed millions of particularly very low-income people who don’t file taxes, but we don’t want to let the perfect be the enemy of the good, right?

But at least 160 million Americans received up to the $1,200 without taking a single extra action. I did want to point out that I think two key points from the CARES Act could be applied to any program that’s trying to improve outcomes. So, we’ve talked a bit about data. We do know that federal, state, and local governments have lots of available data that could allow them to actually make things easier for people — determining eligibility more quickly than any application process, which requires time and effort.

Second, governments can make helpful automated assumptions. Again, even if the income data from 2018 is not perfect, it can provide a reasonable snapshot for how people might be faring in 2020. So sometimes the government is spending a lot of time and resources on making sure that the data is current from this month. But then that means they have to update it again next month and then that requires a lot of extra forms and time for people to continually updating. If we accept that in any circumstance where people are dealing with challenges, but especially a time like 2020 with really rough limited economic security, then we should be automating more benefits. Ideally, we should be spending more on improving lives than on the benefits themselves, than on non-compliance.

Leigh Parise: Thanks, Rebecca. I think those are really helpful examples that certainly a lot of people will be able to relate to in this moment in time. Thanks so much for joining me, Rebecca. To learn more about the Center for Applied Behavioral Science or CABS, visit https://cabs.mdrc.org/

Did you enjoy this episode? Subscribe to the Evidence First podcast for more.


About MDRC’s “Evidence First” Podcast Series

Policymakers talk about solutions, but which ones really work? Join Leigh PariseMDRC’s Associate Director of Program Development and Senior Research Associate, as she talks with experts about the best evidence available on education and social programs that serve low-income people.