Podcast: Anna Gassman-Pines on Unemployment Insurance Access Disparities

Health Affairs Editor-in-Chief Alan Weil interviews Anna Gassman-Pines from the Duke University on her recent paper assessing disparities in access to unemployment insurance during the COVID-19 pandemic.

FULL TRANSCRIPT

00;00;00;04 – 00;00;37;06
Alan Weil
Hello and welcome to “A Health Podyssey”. I’m your host, Alan Weil. The COVID-19 pandemic caused major disruptions to the labor market. The Bureau of Labor Statistics reported that total civilian employment fell by 21 million between the fourth quarter of 2019 and the second quarter of 2020, while the unemployment rate more than tripled. Now the unemployment insurance program, or UI, as it’s sometimes called, is designed to provide temporary partial wage replacement for people who’ve lost their jobs.

00;00;37;15 – 00;01;09;08
Alan Weil
Like many other income support programs, UI was expanded during the pandemic. But as important as a program like UI is for its recipients, program rules limit its availability to only some workers and it excludes others. When the program was expanded during the pandemic, it built upon a platform that doesn’t reach everyone. Disparities in access to unemployment insurance and the health implications of those disparities during the pandemic is the topic of today’s episode of “A Health Podyssey”.

00;01;09;29 – 00;01;45;10
Alan Weil
I’m here with Anna Gassman-Pines, professor in the Sanford School of Public Policy at Duke University. Dr. Gassman-Pines and coauthors published a paper in the November 2022 issue of Health Affairs assessing disparities in access to unemployment insurance during the COVID-19 pandemic. Based on data collected from low-wage workers in Philadelphia, they found that Black and Hispanic workers were less likely to access unemployment insurance than White workers, despite evidence that they experienced similar levels of hardship following layoffs or job loss.

00;01;45;16 – 00;02;00;02
Alan Weil
This paper is one of a group of papers we published in November exploring the health effects of economic security policies in place during the pandemic. You can find all of these papers on our website. Dr. Gassman-Pines, welcome to the program.

00;02;00;21 – 00;02;02;00
Anna Gassman-Pines
Thank you so much for having me.

00;02;02;18 – 00;02;29;05
Alan Weil
So I think in order to have the conversation I’m looking forward to having, we need to introduce our audience to the unemployment insurance program. Most people are probably vaguely familiar with it. Some people have probably even relied upon it. But it is not a universal program, and it’s not administered uniformly around the country. So if you could just start by giving a little description of how the system works, who does it cover?

00;02;29;06 – 00;02;33;19
Alan Weil
What are the benefits? Who runs it? Things like that would probably be a good place for us to start.

00;02;34;23 – 00;03;04;10
Anna Gassman-Pines
Sure. So the unemployment insurance system is a partnership between the federal government and the states. So the federal government sets the broad structure of the program and a lot of the rules, but the states are actually responsible for implementing the program. And states can make a variety of choices about what the program looks like in order to be responsive to the citizens in their state.

00;03;05;03 – 00;03;32;09
Anna Gassman-Pines
So the basics of how it works is that when people are working in the formal sector, both they and their employers are paying into a fund, a pool of money that’s that is available for when workers lose employment through no fault of their own. So that doesn’t cover people who quit their job. It doesn’t cover people who are fired for cause.

00;03;32;20 – 00;04;06;25
Anna Gassman-Pines
But if someone is laid off, then they might be eligible for unemployment insurance. And I say might because they have to have worked a minimum number of hours over the prior four quarters and they have to have a minimum amount of earnings in order to be eligible to receive unemployment insurance when they are laid off. And then, as you said, the unemployment insurance program doesn’t provide complete replacement for the lost wages when someone is laid off,

00;04;07;07 – 00;04;15;14
Anna Gassman-Pines
but a partial wage replacement, and the amount of wages that are covered actually varies fairly substantially across the states.

00;04;16;01 – 00;04;37;22
Alan Weil
Well, that’s a great stage setting. And again, I think if people want to become experts on UI, they’re going to have to look somewhere else. But you’ve made clear that it’s it only is partial wage replacement and it’s only if you have certain qualifying characteristics. Similarly complex is what some of the changes were to the program during the pandemic.

00;04;37;22 – 00;04;45;28
Alan Weil
But again, if you can just give us a thumbnail sketch of how it was expanded, that’s probably also important to understanding the results of your work.

00;04;46;22 – 00;05;16;18
Anna Gassman-Pines
That’s right. So the unemployment insurance program was expanded in two key ways. The first was that eligibility for the program was broadly expanded so that people didn’t have to meet those minimum work hours and earnings requirements in order to be able to get UI. As long as someone had been working in at all in 2020, they were eligible for UI during the pandemic.

00;05;17;09 – 00;05;44;04
Anna Gassman-Pines
The second way that UI was changed during the pandemic is that Congress passed these more generous benefit expansions. So there would be the traditional UI that states provide at whatever their replacement rate is, but then at different periods of time, folks who were receiving UI could get either an additional $300 a week or an additional $600 a week.

00;05;45;02 – 00;05;56;18
Anna Gassman-Pines
So the program was both made more generous in terms of the amount of money that people could receive and also more accessible for a wider range of workers.

00;05;57;18 – 00;06;18;20
Alan Weil
So this is a critical step in trying to address the huge loss of jobs that occurs right away as the pandemic hits. So let’s look at what your study did. You examined a particular population. I think it’s really important to understand who they are and then tell us what you were looking for and we’ll start talking about what you found.

00;06;19;25 – 00;06;44;25
Anna Gassman-Pines
So we have been following a sample of low-wage, hourly workers in service jobs. So all of these folks were working in either retail, food service or hotels before the pandemic was even a glimmer in anyone’s eye, and all of these people that we had been working with also had young children between the ages of two and seven

00;06;44;25 – 00;07;15;26
Anna Gassman-Pines
at the start of our study in the fall of 2019. So we focused on that group for two reasons. The first is that hourly service workers have particularly unpredictable and unstable work schedules. And second of all, having young children makes balancing work and family particularly difficult. So as you can imagine, these families were dramatically affected by the pandemic.

00;07;15;28 – 00;07;42;28
Anna Gassman-Pines
When the pandemic first started and restrictions were put into place, people– we saw people stop working almost right away with the data that we were collecting. And we quickly pivoted to start asking folks about their experiences during the pandemic, about whether the policy supports that were put into place, were helping them and in what ways.

00;07;43;16 – 00;08;13;21
Alan Weil
So part of what I think is so important about the work you’ve done here is that we have lots of understanding of people losing jobs in the economic downturn. You focused on an incredibly economically vulnerable group and as you say, you see the consequences of the pandemic on their employment immediately. So let’s start talking about what you found, what happened to them, how did they… how did their work situations change?

00;08;13;21 – 00;08;19;21
Alan Weil
And then we spend a lot of time in the introduction talking about unemployment benefits. So did they get those benefits?

00;08;20;21 – 00;09;06;16
Anna Gassman-Pines
So unsurprisingly, given what we all know about how challenging that economic circumstance says were during the pandemic, we see that about a quarter of the people in our sample were laid off during the year and a half. The first year and a half of the pandemic. So that is a lot of job loss. We also see that even though all of these people were working in the same kinds of jobs, in the same location before the pandemic started, the workers of color and in particular the Black and Hispanic workers in our sample were more likely to be laid off than the White workers, even though they were all again starting in basically the same place.

00;09;07;17 – 00;09;44;11
Anna Gassman-Pines
So first of all, there was a substantial amount of job loss that was unequally experienced by people based on their racial and ethnic identity. Second of all, for a variety of reasons, not everyone who was laid off was able to access unemployment insurance. So and in particular, there were substantial racial gaps in receipt of unemployment insurance. So in our sample, the folks who identify as white, just over 50% of them

00;09;44;11 – 00;10;02;07
Anna Gassman-Pines
so a little bit more than half, received the full unemployment insurance to which they were presumably eligible. So that includes those extra 300 or $600 payments that Congress had authorized that looks like more of a third of the laid off workers who identify as Black or Hispanic.

00;10;03;20 – 00;10;31;09
Alan Weil
So you have two separate points of disparity. You have higher rates of job loss. And then when job loss occurs, you have lower rates of benefit receipt. And the two combine to well, they add to each other, they multiply with each other, and you have a much less robust financial safety net for people of color than you do for white workers.

00;10;32;04 – 00;11;03;13
Alan Weil
Now, we’re a health journal, and so we’re interested not just in people’s finances, which is critical, of course, but we’re also interested in how this ties to health. We’re going to take a short break. And when we come back, I want to talk to you about the linkages to health and what maybe we can do to reduce some of these health disparities.

00;11;05;01 – 00;11;29;11
Alan Weil
And we’re back. I’m speaking with Dr. Anna Gasman Pines about racial and ethnic disparities in access to and receipt of unemployment insurance benefits during the pandemic. Before the break, we got sort of the top line findings of these disparities in receipt of benefits. But as I teased, we are a health journal. And so we’re interested not just in income, but also health.

00;11;29;12 – 00;11;35;10
Alan Weil
So how did you look at the question of health? What was the health dimension of this study?

00;11;36;19 – 00;12;05;28
Anna Gassman-Pines
So we were asking folks questions about their mental health in particular. There’s a lot of social science and health research that shows that when people lose jobs, their mental health is strongly affected. And we see that with our sample as well. So when people were laid off compared to those who maintained employment, when people were laid off, they were much more likely to be depressed and also feeling anxious.

00;12;05;29 – 00;12;09;23
Anna Gassman-Pines
And that was true across racial and ethnic groups.

00;12;11;02 – 00;12;44;14
Alan Weil
So this was something interesting to me as I read the paper. And I want to make sure I understand this correctly, that if you received UI benefits, the positive mental health effects were essentially consistent regardless of race and ethnicity. So the disparities we’re seeing in the effects on mental health are associated basically entirely with receipt of the benefit, because if you got it, it helped approximately equally regardless of those characteristics.

00;12;44;15 – 00;12;46;27
Alan Weil
Am I understanding that finding correctly?

00;12;47;16 – 00;13;17;09
Anna Gassman-Pines
That’s exactly right. So just like getting laid off worsens mental health, our results show that receiving unemployment insurance when you are laid off improves mental health. Folks who are able to get unemployment insurance are less depressed and less anxious than those who aren’t able to get unemployment insurance. So this policy is strongly connected. This policy support is strongly connected to people’s mental health.

00;13;18;11 – 00;13;38;10
Anna Gassman-Pines
But because we see these differences in access to UI and receipt of UI, what ends up happening is that the workers of color and the White workers benefit equally when they are able to get it, but the workers of color are less able to access this important policy support.

00;13;38;24 – 00;14;16;03
Alan Weil
Yeah, so it’s like we have the right idea, the intervention makes a difference. Now we just have to get that intervention to apply equally to those who need it. Now, there’s a large literature about racial and ethnic disparities in receipt of myriad economic benefits government, forms of government assistance. Before we talk about policies to close some of these gaps and improve the health outcomes, I wonder if you could just help me understand is what you found with respect to receipt of UI similar to what you find with respect to other benefits?

00;14;16;03 – 00;14;31;00
Alan Weil
Or is it different? And just so that we have an understanding of whether we’re solving or need to address sort of a huge problem that affects everything, or if this really is about unemployment benefits and some of the design features that you explained at the beginning.

00;14;32;06 – 00;15;05;18
Anna Gassman-Pines
So that’s that’s a really great question, and there are a few things to say about it. Our own analyses and the work of many others who have looked at other kinds of income supports generally find that providing low-income folks in the United States with additional money improves both material hardship and aspects of social determinants of health, such as food insecurity, and also health outcomes themselves.

00;15;06;06 – 00;15;47;13
Anna Gassman-Pines
So there’s a pretty robust literature that connects providing additional resources to low income folks to improved health and social aspects of health. Unemployment insurance, though, is somewhat unique in the sense that, especially here in the United States, work is so central to our identities that losing a job is harmful not only because of the loss of income associated with losing that job, but also because of the way that our work is connected to who we are and our and our identities.

00;15;47;14 – 00;16;07;14
Anna Gassman-Pines
And so there is research, literature in psychology that shows that part of the reason why job loss is so harmful to mental health is because of how strongly we connect who we are to the kind of work that we do. And if we’re laid off, that can can harm our well-being through that channel as well.

00;16;08;01 – 00;16;39;02
Alan Weil
So economic supports can offset the material hardship side, but obviously economic effects are, as you just said, economic supports don’t offset the the social side. It’s probably a topic for a different day. But I do have to wonder if some of those dynamics were different during the pandemic when job loss was so extensive, that when you think about sort of the social stigma associated with not working, for example, presumably that was lower during the pandemic when it was a much larger phenomenon.

00;16;39;02 – 00;17;04;13
Alan Weil
But again, that’s getting into the realm of speculation. I am interested. So that’s that’s a really critical point. Now, that said, you did find mental health benefits associated with receipt of UI, which is just the cash side. So even without the other mental health benefits of working, even just getting the money in those instances seems to have been helpful or protective.

00;17;05;23 – 00;17;15;16
Alan Weil
So you identify these racial and ethnic disparity in receipt of the benefits. How might we think about trying to close some of those gaps?

00;17;16;10 – 00;17;46;24
Anna Gassman-Pines
There are a number of ways we might think about closing those gaps. One is to recognize that in many states, the UI systems are quite antiquated, which makes access particularly challenging. That was definitely true during the pandemic, when some of the traditional ways of getting this support were closed to people, as in the offices were actually physically closed.

00;17;47;10 – 00;18;33;16
Anna Gassman-Pines
And there were all kinds of stories about the phone lines then being overwhelmed and and there just not being enough staff able to answer the phones, for example, right? So these systems are quite antiquated and and that makes access more difficult. That was certainly true during the pandemic. So one thing we could think about in terms of reducing disparities is to invest in upgrading our systems so that they can be more nimble, so that they can be used on different kinds of platforms, on mobile phones, for example, on the Internet, in-person as as a way of reaching people more effectively wherever they are in whatever way makes accessing the program easier.

00;18;34;11 – 00;18;47;13
Anna Gassman-Pines
We could also think about sometimes there can be this tension between employers and the agency around whether someone really was eligible or not.

00;18;48;01 – 00;19;17;16
Alan Weil
Just to to put a note there. I mean, you describe this as an you know, we use the term insurance. So at the end of the day, employers are paying into this. And if they have employees who successfully claim the benefits, that there is an experience rating process where their premiums are going to go up. So there is a built in tension in the program design where employers are sensitive to whether or not their employees are taking advantage of the benefit.

00;19;17;22 – 00;19;24;13
Alan Weil
Just want to get that in there so that your comment made a little bit more sense to those who haven’t worked with the program.

00;19;24;28 – 00;19;55;17
Anna Gassman-Pines
That’s exactly right. Thank you for that important context. Context. That’s exactly right. So thinking about reforming the financing of of the unemployment insurance system so that employers don’t have that incentive to potentially block worker claims or otherwise try to prevent their employees from succeeding in getting access to this benefit, which we know improves health and provides this important financial support, could be another way of reducing these disparities.

00;19;56;04 – 00;20;39;24
Alan Weil
I am curious about the design element around experience. I’ve always understood that the criteria for participation have disparate effects by race and ethnicity. The provisions put into place during the pandemic, as you noted, stripped out that requirement. But I wonder if its historical presence just was a barrier to people even thinking they were eligible. You mentioned the gap between the receipt of the full benefit between unemployed people who are white and unemployed, people who are identified as black or Hispanic.

00;20;39;24 – 00;20;51;03
Alan Weil
And that might also suggest sort of a legacy of of program rules that were changed but maybe not fully communicated to the applicants.

00;20;51;20 – 00;21;19;23
Anna Gassman-Pines
That’s exactly right. So there’s really good evidence that one of the main ways that people learn about public policies that they may be eligible for and that could help them is through their social networks, through informal channels. It’s not necessarily through a government website or an ad on the radio, but it’s through, you know, sort of trusted people in your circle who can tell you, oh, I went down to the UI office and it wasn’t that bad.

00;21;19;23 – 00;21;51;22
Anna Gassman-Pines
Here’s how to do it, you know? And people do learn a lot about how to access public benefits through their networks. And so if you are in a group that has been historically excluded or who has had trouble accessing benefits, or in the case of, for example, folks who identify as Hispanic, who may be immigrants and concerned about public charge rules, which would be not being able to have a path toward citizenship if you’ve received public benefits.

00;21;52;01 – 00;22;13;12
Anna Gassman-Pines
And that’s something that’s a conversation in your network, right? That historical understanding from the people that you know well and trust could have certainly spilled over into how people thought about accessing benefits during the pandemic, even when eligibility was broadly expanded.

00;22;13;19 – 00;22;48;18
Alan Weil
So as we come toward the close of our conversation, when we think about the policy options here, you described modernizing systems, better access points. You also mentioned changing some of the financial structure so that employers are more supportive of people receiving the benefits. Just wonder if there are any other policies that come to your mind, particularly with a focus on the racial and ethnic disparities that you saw that we ought to be thinking about so that we’re better prepared if we want to build on this program for the next economic downturn or maybe not see these kinds of disparities that you found.

00;22;49;12 – 00;23;18;18
Anna Gassman-Pines
So one of the main things we could do in addition to what I’ve already mentioned, is to permanently extend eligibility for UI to lower wage earners. Folks who have more tenuous connections to the labor market, more unstable work schedules, who might not have been able to meet those minimum thresholds that existed in the past, and that kept them from getting access to these crucial economic supports.

00;23;18;18 – 00;23;37;19
Anna Gassman-Pines
If we permanently extended benefit eligibility to the folks who got that eligibility during the pandemic, many, many, many people of color in our country would be able to access unemployment insurance and therefore to reap the health benefits of getting that important public support.

00;23;38;19 – 00;24;08;14
Alan Weil
And of course, there’s a long history of debate over those provisions because there are those who worry that opening up more benefits to the uninsured will reduce people’s interest in working. And there was that concern raised during the pandemic. And, of course, there’s the racial angle to these benefits as well, that they were created in an environment where there were active desires to keep some populations based on their race or ethnicity out of the program,

00;24;08;14 – 00;24;33;15
Alan Weil
and those rules are sort of baked in. So what I hear you saying is maybe it’s time to reexamine both of those and ask the question, what do we really want this program to accomplish? And think about the design that would do that and maybe move away from some of these legacy decisions based on a time and and considerations that maybe we could move past. Does that seem about right?

00;24;33;15 – 00;24;35;23
Alan Weil
And I don’t want to put words in your mouth here.

00;24;36;10 – 00;25;18;08
Anna Gassman-Pines
That’s exactly right. These are policy choices. Yeah. I just to say that these are policy choices that we are actively making. They are following from, as you said, as a historical legacy of certain groups, people of color in particular, being excluded. That in many ways continue to inform the choices that our elected representatives are making now, and we can all think about whether we might advocate or encourage them to make different choices that make this program more accessible and reduce disparities in doing so.

00;25;18;23 – 00;25;50;04
Alan Weil
So before we finish, I do want to ask, you said at the outset that you started surveying these folks before the pandemic was even on the horizon. You must have a lot of information about them that goes well beyond the pandemic. I wonder if there are any other major areas of examination or findings from this group, which, as we noted, is so economically vulnerable, that go beyond what you referred to in the paper that we published.

00;25;50;22 – 00;26;16;11
Anna Gassman-Pines
I really appreciate you asking that. The original purpose of the study was to understand how unstable and unpredictable work schedules affect worker health. And what we find by getting very fine grained information on a daily basis, asking folks, did you work today? If so, was the schedule that you worked what was originally posted, or was it changed in some way?

00;26;17;07 – 00;26;46;10
Anna Gassman-Pines
And what we find is that these service workers with young children have incredibly unstable work schedules, where on 10% of all days their work schedule was not what they originally thought it was supposed to be. Whether that means getting sent home early, having a shift canceled, having a shift added on at the last minute with very little notice, and that those daily work schedule changes have consequences for these folks’ health.

00;26;46;14 – 00;27;13;16
Anna Gassman-Pines
So their daily mood is much worse on days when there’s been some kind of unexpected change to their work schedule compared to days when everything went smoothly and as planned. And as you can imagine, when this is happening on 10% of days, these daily stressors are accumulating and and having real consequences for worker health and well-being.

00;27;13;23 – 00;27;29;17
Alan Weil
Well, I’m really glad I asked the question. And Dr. Gassman-Pines, thank you for the work you’re doing here, the focus on a particularly economically vulnerable population, the insights that only a survey like this can garner. Thank you for being my guest today on “A Health Podyssey”.

00;27;30;14 – 00;27;34;10
Anna Gassman-Pines
Thank you so much for having me. I really appreciate it.