What do Healthcare.gov, police officer recruitment, and 911 call centers have in common? All can be improved through smart policy design tweaks!

In this episode, Goldman Professor Elizabeth Linos and TPTM host Spencer Bowen (MPP 2020) discuss how small changes in policy design can result in big differences in impact. The first in our ongoing series on Policy Design and Implementation, this episode helps break down the different mechanisms that can be used to influence the direction of policy results. In addition, Professor Linos shares her most recent research, about small changes to improve the day-to-day lives (and then the retention!) of 911 call center staff. Do you want to learn more about the big impact of small design changes? Here’s our top three:

1. Read Elizabeth Linos’ article
2. Check out the book Nudge from your local library
3. Learn about one of the first “nudge units” in government, the Behavioral Insights Team, and what they’re working on today.

Transcript

Spencer Bowen: Remember healthcare.gov? 

Barack Obama: I want to talk with you about a new consumer website: healthcare.gov. It's a good resource for understanding the new law, and it offers a few simple tools to help you take your health care into your own hands. 

Commenter: That awful launch for Obamacare, you've heard about all those problems with the website. 

Commenter 2: Some people are getting incorrect quotes and eligibility information. 

Commenter 3: For this is just the tip of the iceberg. 

Commenter 4: It's a technical or technological disaster. 

Commenter 5: Who is responsible for overseeing this project? 

Commenter 6: The elephant in the room is certainly the White House. The government's role in all of this.

Commenter 7: It makes the administration look incompetent. 

Bowen: It was a disaster, but separate from politics and insurance markets and President Obama, what should we learn from healthcare.gov? 

Elizabeth Linos: Turns out that for the average American, their experience or their understanding of healthcare.gov boiled down to a website that wasn't functioning or a system that didn't get them access to the health care they they thought they deserved. And so that's kind of a typical example where we forgot about implementation because we were thinking about the bigger picture. If you start thinking or rethink programs from the perspective of the user, you end up making different design decisions. 

Bowen: That's Goldman Professor Elizabeth Linos. And she thinks the rollout of the Affordable Care Act was a big swing and a miss, not because the ACA is good or bad, but because we were so caught up in health care reform that we forgot about the end user, the people that the law is supposed to help. We're starting off Talk Policy To Me's third season with a focus on design. Two policies with the same goal can have completely different impacts because of how they're structured. Elizabeth Linos is an expert in this area. This is about human behavior, implementation and nudges. This is about how we make government work better. This is policy design. I'm Spencer Bowen, and this is Talk Policy To Me. 

Linos: I'm Elizabeth Linos. I'm an assistant professor here at the Goldman School. My research is really at the intersection of behavioral economics and public management. So I spend all my time thinking about how we can improve government. 

Bowen: So how did Professor Linos end up at a public policy school? 

Linos: I came to this field because I loved experiments. The idea of really trying to understand what works, both in government, but also more broadly in social programs. And so my way of kind of serving the common good is to say, look, let's use data and evidence when we can to make the world a better place. You know, policy implementation or what we think about in public management to me, is in some ways, the forgotten part of the puzzle. 

Bowen: Professor Linos' work lives at a really cool intersection between implementation, management, and policy design. 

Linos: We can all get really good at data, and I can't emphasize how important those skills are in building credibility. But, within that space, public management and kind of the focus on implementation is really about making sure that our understanding of the data matches people's lived experience. Oftentimes we lose touch with the purpose of what we're doing. And so I hope that, you know, a class like a public management class or policy implementation class will help prepare students for that moment, where it's not just about criticizing other people's policies or analyzing other people's decisions, but really preparing you for when you're the person that everyone's turning to to say, OK, what should we do now? 

Bowen: But a lot of policymakers deciding what should we do now? don't think about this stuff. 

Linos: It frustrates me to no end to hear policymakers say, oh, well, we designed it, but then people don't come. They must be lazy. They must not want the service. That's the part that's frustrating to me, when policymakers or government partners, don't see their own role as important in changing how things are done. And, you know, behavioral science isn't rocket science. 

Bowen: So what the heck is behavioral science? 

Linos: I think of behavioral science as the intersection of what we know from psychology with what we know from economics. And the basic idea is that humans are essentially hardwired to take mental shortcuts when they can. If we had to make a really rational, thoughtful decision about every single decision we make every day, we would basically not be able to do anything else. We create these mental shortcuts to save time and to save cognitive bandwidth. And in the majority of life situations, that's a good idea. What behavioral science has discovered is that sometimes those mental shortcuts leads to cognitive bias in ways that don't lead to ideal outcomes. When it comes to government, that matters because we need to start designing our government programs and our government policies based on the realities that people face when they interact with government rather than, you know, what a simple, rational actor model would suggest they do. 

Bowen: Design stuff based on how humans actually act. Seems like a no brainer. The problem is that this is a pretty new field and it's even newer to think about in government. 

Linos: Sometimes it's a team within a prime minister's office. The White House had a social and behavioral sciences team. That being said, this is very new. But really, it's only since 2010 that we're seeing this explosion of governmental interest in behavioral science. And so your average policymaker in a senior position has never heard of behavioral science or if they have, they don't necessarily know what that means. Some of the most successful partnerships I've seen happen when you take academic expertise and bring it to government in a form that is helpful to government rather than to the academic. It's always interesting to me how quickly we can make a mistake when we're speaking to government partners by showing a graph or worse, a regression table. 

Bowen: To be fair, even us policy students barely understand regression tables. 

Linos: And people will just stop, stop listening. So I think part of it is, you know, on the person presenting the data or talking about data to speak about it in a way that makes sense to a government partner who may not be using data in their lives. Overall, I found that if if we do our job right and if we talk about the data in a way that's meaningful, we talk about the people who are served or the experience of the people who come to their door. The numbers can only be an asset. 

Bowen: So let's say we do look at the numbers and learn from the numbers. How would it change policy design? The great part about design changes backed by behavioral science is that they're kind of small and doable. That's why they're sometimes called nudges. 

Linos: I think we were all trained traditionally to think that if you want to make a big difference in the world, you have to have a big change. And one of the the beauties of behavioral science, at least from my perspective, is, you know, to stop thinking in terms of the size of the intervention, but thinking about the size of the impact. And so if we can find a small tweak or a change or a low cost intervention, that can have a disproportionate impact, to me, that's, It's magical. 

Bowen: We're back to the same key theme as earlier. Think about the impact on the end user, the real people. 

Linos: You make different decisions about how much attention you ask of people, how many steps you require them to go through, what language you use when you refer to programs and services. And all of that stuff might seem small or insignificant compared to the size of the impact of a, you know, a large government program or a cash transfer program. But we've found, through decades of research, that those seemingly insignificant components of design are really affecting who gets access to services. In a lot of these social programs or anti-poverty programs, the hurdles end up looking similar. It seems silly to be talking about a hurdle as simple as you know, don't set up appointments only during work hours for people who work, or provide child care if you're trying to get mothers to show up to an appointment. But those really can make, make or break the difference. 

Bowen: Professor Linos' research has led to some very cool and very applicable takeaways. 

Linos: A lot of my research is on is on how do we recruit for government jobs, especially for frontline jobs. My research suggests actually telling people how hard it is might be effective. We did some experiments in policing where we send people different postcards asking them to apply to be police officers. And we found that a message that said, look, this is really hard. If you're the kind of person who thrives in a challenging environment, you're the kind of person we're looking for. That message tripled the likelihood that people would apply to become police officers and actually quadrupled the likelihood for people of color. 

Bowen: You can imagine some of this research being just as important for a local public works department as it is for a big national agency. 

Linos: And and there's a lot of a lot of credibility, I think, that, you know, government gets just by explaining that they're working on it. There's some evidence in operational transparency that suggests, you know, just showing your mess to a resident, just showing them that you're trying, increases trust and engagement with government. So I think that's a really powerful way of thinking about the relationship between a government program and the people it's serving. 

Bowen: This same finding helped get healthcare.gov back on track. 

Linos: When this kind of rollout debacle happened, a team came together of really thoughtful, excellent design thinkers, computer scientists who focused on the user interface. And one of the changes that I thought was particularly interesting is, you know, as people were going back to that website that was not yet working, they now saw an error message that said, hey, thank you for for coming here. You're on the waitlist or you're in the waiting room to get access to to health care and here are some resources that you can look at while you wait. It gives the resident a sense that there's someone on the other end of this, this interface, that someone's working to improving this process. 

Bowen: These nudges are very attractive. They're easy and can have a big impact. But policymakers shouldn't get so caught up in small tweaks that they overlook big systemic necessary changes. 

Linos: It would be a sad day if everyone decided to go towards incremental tweaks and we didn't have people fighting the bigger fights on large, radical social change. You know, those larger change or larger reform is and should be at the core of how we think about the world. I don't think behavioral science type interventions or nudges, as some people like to call them, necessarily negate that broader mission. So really, the way I think about it is, you know, as we're fighting the bigger fight, as we're thinking about systemic reasons that explain how we got to where we are today, we should be doing everything in our power to tweak on the margin until we get to the point where we fixed all those tweaks and now we really just need large change. 

Bowen: After the break, we'll dive into some of Professor Linos' new research and what made her say this. 

Linos: I sometimes joke that I might quit my job and become an advocate for 911 dispatchers and call takers. 

Sarah Swanbeck: Interested in a career in public policy? Apply to the Goldman school. The deadline for applications to our Master's of Public Policy program is coming up quick on December 2nd. The application portal is open and our admissions team is eager to connect with prospective students. If you're looking to sharpen your application, visit our website at GSPP.Berkeley.edu, where you can learn more about the program and hear from our admissions team about how you can put your best foot forward. 

Bowen: Welcome back. Before the break, we talked about what behavioral science is and how we might use it to make public policy design better. Now we're going to talk about some of Professor Linos' brand new research, the kind of research that changes how policymakers think about the programs they design. 

Linos: I've been thinking a lot about why so many smart, capable, motivated people quit government jobs in the first couple of years. So whether you're looking at social workers or teachers or police officers, there is this bizarre trend in all the data that says, look, most people that are going to quit are going to quit in those first two years. It's also surprising, I think, to a lot of people, that we have such high turnover rates in government. 

Bowen: It's true. And some of the stats are wild. In some communities, over 40 percent of corrections officers quit within two years. 

Linos: That's certainly the case with 911 dispatchers as well. 911 dispatchers are this really interesting group of people that face trauma every day, often, you know, hundreds of times a day as they're picking up calls and dispatching, you know, emergency vehicles or other first responders to people's homes. They're responsible for navigating their own emotions and trauma and navigating the trauma and emotions of the person on the phone. But they're treated like call center workers. 

Bowen: 911 dispatchers intrigue Professor Latinos for a couple reasons. First of all, it's a government job with high turnover rates. 

Linos: And so when you're looking at, you know, 40 or 50 percent turnover, it's almost impossible to keep a 911 dispatch center running. 

Bowen: Plus, 911 dispatchers are like first responders in a lot of ways, but don't get treated like it. 

Linos: Even in government, there are jobs that receive support or there's a lot of status assigned to those jobs, and other jobs that are equally important to making public services work that are somewhat forgotten. And so if there is some sort of major event in a city, a mass shooting or some sort of other large traumatic event, the police officers, the firefighters, the medics, they all get a lot of extra support, psychological and otherwise, to deal with the trauma. The 911 dispatcher has to go back to work the next day. 

Bowen: This high turnover and lack of support made 911 workers the perfect subjects for a fascinating study. 

Linos: We did this one trial with 911 dispatchers in nine U.S. cities where we said, look, you know, we can't change the specifics of the job. We can't change the hours and we can't change the level of trauma that you face. But what we can do is increase the level of social support you feel when you come to work in the morning. Police officers also have each other. There's a very strong sense that this is a collective professional identity. So what we wanted to do is build that same sort of professional collective identity for 911 dispatchers. In practice what we did is we set up a kind of virtual peer support network where we encourage 9 11 dispatchers to share their story. And the focus wasn't really on how they're serving the community. The focus was on how they can help each other. So prompts were things like, you know, what would you tell a rookie about the job,  or how would you support a new person, when they first start this this new profession? And the idea was really to build this kind of collective community that's based on a group of people who face trauma that nobody else understands. 

Bowen: Why go after worker burnout with an online support network? 

Linos: So there's some good theoretical literature that says your levels of burnout or engagement at work are not really about your job demands. They're not really about your workload, what you actually have to do, but your job resources. And job resources is a nice, vague, broad category, which can mean anything from feeling like you have a best friend at work to feeling like your supervisor supports you and trusts you. And so what we're trying to tap into is kind of how can we elevate that perceived sense that you have resources at your work to help you cope? 

Bowen:There's no such thing as a policy silver bullet, but Professor Linos' findings came close. 

Linos: The intervention was a six week long intervention. And then we measured the impact both on burnout levels and on resignations. And we found really striking effects. So even four months later, we found a significant decrease in burnout score and a significant decrease in resignations, post intervention. The reason this is shocking is because, you know, mental health and burnout is a serious clinical problem. I don't think a lot of people expected that you could improve it with something as simple as, you know, a social support intervention. Oftentimes when we talk about these issues, the solutions that people refer to are things like lower workload or more time off. And those are really expensive for governments. So it's encouraging to see that we can move the needle with something as simple as a behavioral intervention. 

Bowen: Now, this is just one study with a specific group of people, but it shows how exciting this work is, how simple policy choices can actually improve people's lives. Like Professor Linos said, sometimes it's not about making a big intervention, it's about making a big impact. Talk Policy To Me is a production of UC Berkeley's Goldman School of Public Policy and the Berkeley Institute for Young Americans. For show notes and to hear old episodes visit talkpolicytome.org. Are executive producers are Bora Lee Reed and Sarah Swanbeck. Music heard today is by Blue Dot Sessions and Pat Mesiti-Miller. Our audio engineer is Michael Quiroz. I'm Spencer Bowen. Catch you next time.