It’s been a little over a month since the unemployment benefits programs that were established by the CARES Act expired, so we’re taking a look at how well they worked. Washington Post writer Amy Goldstein and Elliott Morris, a data journalist at The Economist, deliver the facts to Jessyn and Paul.
Amy Goldstein is a staff writer at The Washington Post, where much of her work has focused on social policy. She is the author of Janesville: An American Story.
Twitter: @goldsteinamy
Elliott Morris is a data journalist at The Economist.
Twitter: @gelliottmorris
Further reading:
Poverty fell overall in 2020 as result of massive stimulus checks and unemployment aid, Census Bureau says: https://www.washingtonpost.com/business/2021/09/14/us-census-poverty-health-insurance-2020/
Welfare rolls decline during the pandemic despite economic upheaval: https://www.washingtonpost.com/health/2021/08/01/welfare-roles-during-the-pandemic/
Why now is the time to fix the UI system: https://www.epi.org/publication/introduction-why-now-is-the-time-to-fix-the-ui-system/
The racial disparity in unemployment benefits: https://www.rand.org/blog/2020/07/the-racial-disparity-in-unemployment-benefits.html
Unpacking Inequities in Unemployment Insurance: https://www.newamerica.org/pit/reports/unpacking-inequities-unemployment-insurance/introduction/
Ending pandemic unemployment aid has not yielded extra jobs—yet: https://www.economist.com/finance-and-economics/2021/08/28/ending-pandemic-unemployment-aid-has-not-yielded-extra-jobs-yet
Janesville: An American Story: https://bookshop.org/books/janesville-an-american-story-9781508283966/9781501102264
Website: https://pitchforkeconomics.com/
Twitter: @PitchforkEcon
Instagram: @pitchforkeconomics
Nick’s twitter: @NickHanauer
Jessyn Farrell:
The unemployment system that we’ve been relying on through these last 18 months was designed and implemented during the Great Depression, so it’s really old and out of date.
Amy Goldstein:
The number of people in poverty, the proportion of people in poverty actually got better in 2020 than it had been the year before. That was really because so many people we’re getting various kinds of aid.
Paul Constant:
I think the question is why can’t we have a robust safety net all the time, not just during the pandemic?
Speaker 4:
From the home offices of Civic Ventures in downtown Seattle, this is Pitchfork Economics with Nick Hanauer, the best place to get the truth about who gets what and why.
Paul Constant:
I’m Paul Constant and I’m a writer at Civic Ventures.
Jessyn Farrell:
I’m Jessyn Farrell, and I’m senior vice president at Civic Ventures, and a former state legislator.
Paul Constant:
The three key pandemic era unemployment benefit programs that were established by the CARES Act last year expired on September 6th, so we’re going to take a look at how they went today.
Jessyn Farrell:
Yeah, I think the spoiler alert is that we think they went as well as they could have, right? That it was really important to put money and support into place in this massive economic crisis that we’ve experienced in the last 18 months. I mean, I think what is really interesting about this is that the unemployment system that we’ve been relying on through these last 18 months, was designed and implemented during the Great Depression, so it’s really old, and I think in a lot of ways, somewhat rickety and out of date for this moment. Especially given that in the Great Depression, the typical worker was a White male, head of household in manufacturing, and obviously so much has changed about the workforce since then.
Paul Constant:
We’ve gone from a manufacturing economy to a service economy where workers who used to be in factories 10 hours a day are now working at Walmart 10 hours a day. There’s a lot of change and that is not necessarily reflected in the system.
Jessyn Farrell:
Yeah, and in a brilliant report published in June by Rebecca Dixon and William Spriggs, they point out that the revenue arrangement for UI hasn’t been updated since the 1930s either. So the way it’s funded, it was set up to tax companies because swings in inventory used to cause the bulk of unemployment and now it’s recessions that are caused by financial markets. The companies that do create excessive churn in the labor market, like companies that employ temporary contract staffing, part-time, low wage workers don’t pay more into the system because those workers usually don’t even qualify for unemployment insurance.
Paul Constant:
Yeah, and we’ll link to that report in the show notes, but basically it’s just a long way of showing that unemployment does not reflect the realities of the workforce today. Then to complicate matters even more, it’s not just one program, it’s 53 different programs. It’s every state plus Washington DC, Puerto Rico, and the Virgin Islands, and they all have their own ways of defining what workers’ past earnings were, determining how much they should get.
So basically the federal government hopes that a state will replace about half of the workers’ lost wages, but due to different benefit caps and measurement systems in different states, outcomes for workers vary wildly. Like in Mississippi, which is perhaps unsurprisingly the least generous state, it caps unemployment payments at $235 a week, so for someone earning $50,000 a year, that would only be 24% of their salary.
Jessyn Farrell:
Right, and so here you have this really significantly complicated system that did shore up a lot of folks during this time, but I think it’s also really important to talk about how steeped in racial disparity, and really overlaid the racist underpinnings of a lot of the new deal programming, which even though there are a lot of good things, it really was overlaid with this racism. It was created in 1935 and from its inception, it excluded 65% of the Black workforce at the time from receiving benefits, compared to 27% of White workers.
And the administrative details of implementing the unemployment insurance program, Paul, as you just talked about its complexity, it was left to the states and this was a really overt concession to the Southern Democrats. Which meant that there were going to be very few federal controls to ensure that there was equity in who got UI benefits. That racial disparity really persists to this day.
Paul Constant:
If anybody wants to quibble with calling that a racist disparity, unemployment insurance was created in 1935 and from its inception, it excluded 65% of the African American workforce at that time from receiving benefits, compared to 27% of White workers. So leaving it to the states was a concession to the Southern Democrats, providing very few controls for ensuring equity and UI benefits and duration between states. It was designed from the outset to have different outcomes for White Americans and Black Americans.
Jessyn Farrell:
Right, and those outcomes still persist. [Kathryn Edwards 00:05:46] at Rand studied the racial disparity and unemployment benefits last year and found that, because states with more Black workers have less generous benefits overall, White workers have an average maximum unemployment benefit of $463 and Black workers have $423, which is a difference of $40, and that’s over a thousand dollars over 26 weeks of benefits.
Today’s episode, we’re really excited to look at how unemployment insurance worked during the pandemic. The basic takeaway is that it did work but what was really important is that there were some really significant emergency modifications, more money was put into it, more flexibility, and that really had a positive impact on people’s ability to get by.
Paul Constant:
To learn more about that, we’re going to talk to Amy Goldstein, who’s a Pulitzer Prize winning reporter at the Washington Post. I’m a huge fan of her book Janesville: An American Story, which is about the Wisconsin industrial town where the country’s oldest General Motors assembly plant shut down at the end of 2008 in the middle of the Great Recession. If you haven’t read this book, I highly recommend it, it’s really one of the most informative and moving stories about what’s going on in small town America.
I think it’s really pertinent to what we’re going through today because when the GM plant closed in Janesville, it just basically shut off all of Janesville’s economy. Everything relied on that plant, it was a company town. In that way, it’s sort of a microcosm of what happened in the United States when our entire economy shut down because of the pandemic, people lost jobs en masse, there were cascading effects throughout the economy. I think that by looking at what happened at this one town when the unthinkable happened, it provides a useful, admittedly rough, but a useful analogy for what happened last year when coronavirus started spreading across the country.
Amy Goldstein:
My name is Amy Goldstein, I’m a staff writer at the Washington Post where I write mostly about healthcare policy and other kinds of social policy issues. I’m the author of the book, Janesville: An American Story.
Paul Constant:
Obviously the conditions that led to unemployment in Janesville and the conditions that led to unemployment during the pandemic were very different, but the people you wrote about in Janesville actually faced a similar question that many people faced during the pandemic. Which as you said, is what choices does one make when there are no good choices left? In Janesville, did unemployment insurance factor into the decisions that the affected workers made?
Amy Goldstein:
In working on this book, I was really interested in the question of what happens to people when good work goes away? That was a question I was thinking a lot about during the Great Recession of 2007, 2009 and the years immediately afterwards. But it’s the question that’s become newly relevant as so many jobs went away at the beginning of the recession that began with the coronavirus pandemic. In Janesville, it was partly unemployment benefits that affected people’s post work trajectory, but a lot of it was more where people had worked before their jobs went away.
If you had been a long time worker at the General Motors plant, you were eligible for union benefits called sub pay, and that kept people with some money coming in for a year or two. So they had some time to figure things out or many people thought, “Well of course this plant’s going to reopen, it’s been here for nearly nine decades and it’s just a matter of time before the company gives a new product to this plant.” Well, that didn’t happen. But if you were somebody who worked at one of the many supplier companies that had fed to the General Motors plant, you didn’t have that cushion of that union benefit. Those were the people who really had to figure out in a hurry what to do.
Jessyn Farrell:
For those folks who didn’t have union benefits and couldn’t find another good job after the plant shut down or after their suppliers were impacted, what did they do?
Amy Goldstein:
Well, some people retrained, a lot of people retrained. One of the things that I was interested in exploring was if the idea of retraining, or upskilling as it’s sometimes called, is a very popular idea with both Democrats and Republicans in this country, how well was it working in this context? It turned out that some people got jobs in the field that they retrained in. There was a technical college in town where a lot of factory workers tried to become students, but both the college and these laid off factory workers often learned that it was a very hard thing to do to go back to school when you were in your thirties or your forties, and you were worrying about how you’re going to put food on the table for a family, so many people didn’t finish their programs.
And because in the Great Recession the period of joblessness lasted for longer than in most recessions, even if people succeeded in what they were studying, succeeded academically, there often weren’t jobs for them to go to after they were students. So some people took bad jobs. A lot of people took jobs that paid much less than the factory work they had been doing.
Paul Constant:
We’re starting now to look at the pandemic unemployment insurance in the rear view mirror, so let’s start by talking about the scale of job loss last year. How much change did we see in household incomes in other job-related systems like health insurance?
Amy Goldstein:
Well, every year the US Census Bureau puts out a report on income and poverty and health insurance. It comes out in the fall so it just came out a matter of weeks ago for the previous year, so it was looking at 2020, the first year of the pandemic. It had a lot of things to say that were relevant to what we’re talking about. In terms of the total number of workers in the United States, the total number decreased about 3%. The number of people who had full-time year-round jobs declined by nearly 14 million from 2019 to 2020. Now you can’t infer that all of those jobs were lost because of the pandemic, but the pandemic was certainly a contributor to them.
Jessyn Farrell:
Were there any racial disparities among the changes in health insurance that you can tell us about?
Amy Goldstein:
Well, overall health insurance declined a little bit. It lost about 2 million people who were insured, so about 28 million people last year had no health insurance at any time during the year. The fact that it was declining is something that’s been happening for the last few years, so that wasn’t so startling. In terms of the racial and ethnic disparity you’re asking about, there has for a lot of years been a big gap between the number of White people in this country who are insured and the number of Black and Latino people in this country who are insured, and those disparities widened last year. Overall there’re about 8% of White people in this country who had no health insurance through the year. For African Americans, it was a little more than 10%, and for Hispanics, it was about 18%. Those aren’t new disparities, but they’re wider than they had been before the pandemic.
Paul Constant:
What are the big economic takeaways that you see regarding the pandemic? What was the effect that COVID had on poverty in America?
Amy Goldstein:
Well, what effect COVID had on poverty really depends on how you count poverty. There’s an official poverty measure that’s existed for a long time, and by that measure there were more people who were in poverty, about 3.3 million more people. But starting a little over a decade ago, there was a new measure that was created, called the Supplemental Poverty Measure. Now that’s getting a little nerdy sounding, but what it basically means is who was in poverty even if you take into account all the kinds of aid programs that people might qualify for?
It was a poverty measure that was created during the Great Recession and it’s become very relevant during this pandemic. If you look at that supplemental way of thinking about poverty, the number of people in poverty, the proportion of people in poverty actually got better in 2020 than it had been the year before. It declined from almost 12% in 2019 to just over 9% in 2020 and that was really because so many people were getting various kinds of aid.
Jessyn Farrell:
That’s to me, one of the most astonishing takeaways of the pandemic, that poverty actually went down. And that we had this bipartisan experiment in our country of putting a lot more resources into these social programs that had been starved for so long. Maybe you can talk about how do we know there’s a direct, causative link?
Amy Goldstein:
Well, we don’t really know exactly what’s causal, but one of the things that the Census Bureau did in this year’s report, because as you and your listeners doubtless know, Congress created several waves of supplemental aid because of the pandemic. The census found that if the stimulus payments hadn’t existed, the official poverty rate would’ve been 12.7%, which is higher than if we go back to the official poverty rate of about 11%, it would’ve been higher. So at least that’s one way of thinking about what difference the stimulus payments made.
Paul Constant:
You found that welfare roles declined during the pandemic, and I was wondering if you could tell us why they declined? And also any differences and changes on reliance on [TANF 00:16:00] for food assistance or Medicaid that occurred during that time.
Amy Goldstein:
Well, this was some work that I did for my newspaper. I actually was thinking about, what do we know about what’s been happening with welfare? Because going back to the 1930s or the Great Depression, welfare had been a big part of this country’s social safety net that the federal government provides and states provide. I knew that if you look at Medicaid, the federal state hybrid health insurance program for low income people, if you look at the program called [Snap 00:16:36], which used to be called food stamps, food aid for poor people.
Of those programs, the caseloads, the enrollment went up substantially during the pandemic. And I had no idea what was going on with welfare itself so I actually got some figures from the federal government that the Department of Health and Human Services maintains, looking month by month at welfare caseloads from February of 2020, which was the year before the pandemic was officially declared, to the end of that calendar year. To my utter surprise, it turned out that nationwide, the caseloads actually declined just a little bit over the course of those months.
Jessyn Farrell:
How did the job seeking requirements contribute to that decline?
Amy Goldstein:
Well, the welfare program now, you mentioned TANF it’s called, Temporary Assistance for Needy Families, and that version of the program has been around for about a quarter of a century. It replaced something that had predated it, called AFDC. AFDC had been just a welfare program, if you’re poor, you got money. One of the ideas with TANF was that people should have to work or be in work preparation of some kind, in order to get these benefits. Another thing that happened when TANF came along was that states got much more latitude over how they wanted to run their programs. So it turns out that there were some states that kept these work requirements, even when the pandemic began. I mean, states have discretion to say, “You have to work, or we’re going to suspend these work requirements,” they can ask the federal government to just drop work requirements for a while.
The federal government had made clear that they were willing to make those changes because the economy was so bad because of COVID. So it turns out that in the 13 states that left these job requirements in place in 2020, 10 of those had caseloads that fell. Fewer people were getting welfare in 10 of those 13 states. If you look at the rest of the states that either softened or temporarily got rid of their work requirements, not many of them had such kinds of welfare declines. It looks like what people’s requirements were in terms of having to work, even at a time when jobs were hard to find, at a time when there were all kinds of warnings that you should be careful about leaving your house because of the odds of getting infected, that these various states’ decisions actually made a big difference.
Paul Constant:
Are there any other outcomes for families in need of welfare that changed by state to state, that you thought were especially interesting or…
Amy Goldstein:
Well I tried to take a look when I was working on this story, at what a few different states did that seemed as if they were at different ends of the continuum of what state policies were like. I looked at Kansas where that state already has a two year time limit for getting welfare benefits. Now the federal limit is five years so Kansas is much more stringent. It was more stringent before the pandemic and it didn’t change that stringency during the pandemic. People there who I talked to, had a very hard time qualifying for welfare cash benefits.
On the other hand, a state like Minnesota, which I also took a look at, made a lot of adjustments during this time. They lifted their work requirements, they dropped their time limits altogether. They said that you no longer, if you’re on welfare, have to show up and meet with a case worker every month, you can do it virtually. In that state, as you might predict, more people were able to get and stay on cash assistance.
Jessyn Farrell:
Let’s talk a little bit about the Build Back Better proposal and what it might include to offer something like some of the extensive finance support for lower income Americans, that was made available during the pandemic.
Amy Goldstein:
Well, this is a bill that people in Congress are hot and heavily fighting about at the moment, and something that is being pushed pretty much solely by Democrats, who obviously have the majority in both the House and the Senate in Congress. There’s disagreement within the party over how far to go. But the basic idea for what you’re asking me about, is not in this bill to reinstitute stimulis payments, but there are a number of other features that would basically strengthen the social safety net in other ways.
For instance, the bill contains better subsidies, higher subsidies I should say, without being normative about it, for childcare. It would have universal preschool paid for. It would, for the dozen states that have not expanded Medicaid under the Affordable Care Act, it would help people who would qualify for Medicaid if they were in other states, get health benefits. So it’s not so much direct stimulus aid, but it’s help for people who are lower income.
Paul Constant:
The question we like to ask all of our guests is why do you do this work?
Amy Goldstein:
Ah, well that’s a good question. I’ve been a journalist for a long time and I’ve written about all kinds of social policy issues. The book that I wrote was basically looking at the ground level effects of one bad economic time, and how policy matters or doesn’t matter to people. What I think of what I do, is I like to find stories that are right at the border of policy and politics, and look at what effect they have on people in this country. I think that that’s a useful thing to do, or at least I hope so.
Paul Constant:
Well, you certainly succeeded with your book, which I think it was one of the most important non-fiction books I’ve read in the last, I don’t know, five, 10 years. I had a book club pre-pandemic where we talked about political books and it was unanimously loved in that group. It was really informative and elegantly constructed and I really appreciated it. So I just wanted to thank you face-to-face for writing it.
Amy Goldstein:
Well, thank you for the kind words.
Jessyn Farrell:
Well thank you for joining us, Amy. We really appreciate it.
Amy Goldstein:
Good to be with you.
Elliott Morris:
I’m Elliott Morris, I’m a data journalist for The Economist. Most federal unemployment benefits ended on September 6th when the federal deadlines for pandemic unemployment insurance ended federally, but some states rolled back these benefits early, as early as June. There are 25 of these states that ended early, most of them were governed by Republicans. In fact, all except one had a Republican governor, and surprise, that’s in Louisiana where the governor is a mysterious Democrat named John Bel Edwards. So in these Republican states, the conventional wisdom was that cutting unemployment benefits would force soon-to-be workers back to work. By depriving them of their hundreds of dollars of stimulus, they would need to recover this income elsewhere, like traditional economic theory would posit, and so they would go back to work.
This jives with some of the more conservative ideas about welfare, unemployment, and big government over the last 50 years. The fact that different states have different governors, offers just a really interesting opportunity to test whether or not that’s the case. Unfortunately for the theorists and for some of these more conservative state governors and legislatures, their theories have not panned out. Or at least they hadn’t when The Economist published an analysis of self-reported working on August 20th.
Up until August 20th in these states that cut their unemployment benefits early, self-reported working rates as measured by the US Census Bureau, and as analyzed by us, matched the self-reported working rates in states that kept their benefits, almost perfectly. We use an analysis called difference in differences to measure basically the rate of decline or increase in working on average across the 25 states that ended their benefits early, and in those that kept their benefits through just September 6th, and self-reported working rates are almost the same on every single day, relative to a benchmark in April. There was absolutely no difference in self-reported working measured by the Census Bureau in those states that cut off their benefits early, the theories backfired.
But what we did find, again as measured by the Census Bureau, states that ended their unemployment benefits early, saw an increase in the share of families who said they were having difficulty meeting typical expenses in their household. And respondents are allowed to define typical expenses however they want, I imagine it’s stuff like putting food on the table, paying for your kids’ lunches at school, that sort of thing. The transport for your car, gas, household repairs, that sort of thing. So in states that ended their unemployment benefits early, the increase in this difficulty meeting expenses was about four points, three or four points higher than in the states that kept paying unemployment benefits. In aggregate this picture we’re painting with the data, is that cutting off on unemployment benefits did not, at least up until the end of August, increase working, but they did increase financial hardship for families.
Now there are some weaknesses to self-reported spending or self-reported working so we can rely on actual data on spending. Luckily economists have sorted that out for us, right? Thanks to the economists. One analysis from researchers led by an economist from Columbia found that workers in the cut off states who lost their pandemic aid, cut their weekly spending by $278, relative to workers in other states or workers that didn’t lose their spending. So that would implicate the policy choice by Republicans and conservatives a little further, but it’s worth mentioning that this $278, if you extrapolate it, is only about a $2 billion decrease between June and early August in aggregate consumer spending. Whereas aggregate national consumer spending is like $1.3 trillion, so a thousandth of that.
Now everything matters at the human level, right? We don’t want to abstract too far and just look at this at the economic level, but we would say that there could be labor supply effects from this, there could be downstream effects in retail or heavily consumer financed industries if spending were to continue to decrease, or continue at the decreased level, we just can’t observe that yet. While there aren’t any immediate answers after reanalyzing data, we’re lucky in that the Census Bureau is continuing these biweekly household pulse surveys, continuing to ask Americans if they’re working, if they’re spending, if they had pandemic unemployment insurance and lost it, when they lost it. So we can reanalyze these trends over time, we can keep looking at the differences in differences, and we can see if these conclusions stick in the coming months.
Jessyn Farrell:
We heard from Amy that poverty fell overall as a result of the pandemic aid programs, and we heard from Elliott that cutting unemployment insurance did not yield increased employment. So Paul, what do you make of this?
Paul Constant:
Well I mean, I think the question is why can’t we have a robust safety net all the time, not just during the pandemic, right? I think that pandemic assistance programs really challenged our ideas of what the social safety net is supposed to do and why we do certain things the way that we do. I think that it’s taught us a lot about means testing and why that gets in the way of a lot of programs. I think that it’s demonstrated that people don’t use unemployment the way that a lot of trickle down politicians have trained us to think they do. That it’s not an excuse to be lazy and that it’s not a hindrance to hiring, right? There are a lot of issues at play with unemployment.
Jessyn Farrell:
Why people are seeking or not seeking jobs in this moment is really complicated, right? There are real problems around finding childcare, there’s still really significant concerns about health and safety in particular workplaces, like the restaurant industry, other service industry jobs. Then there’s just the fact that a lot of jobs just are not good paying jobs that provide the economic security that people need.
But I think our experiment with really investing in unemployment insurance during the pandemic has really gone straight at a lot of these myths that you point out, right? That if we have an ample safety net, people are not going to go back to work. But I think what we’ve really found is that it just means that people are able to not slide into poverty. They’re able to pay their rent and buy food for themselves, and make it through a really hard time. I think the things that we need to really be asking ourselves going forward then, is what are we going to do to really fix unemployment insurance so that it really does have a meaningful impact in people’s lives?
Paul Constant:
Yeah, and I think we’ve learned that having a strong social safety net, isn’t just good for the people who are on the safety net, it’s good for everyone, right? Because people who are getting those unemployment payments were able to spend their money locally and that kept other businesses from laying off workers, their consumer demand kept the economy flowing. Is there a way to fix unemployment so that it operates that way all the time? So that it holds our communities up, not just families, and keeps businesses running?
Jessyn Farrell:
Yeah, that’s right. And again, I just want to point out that this was a bipartisan experiment, especially in those first stimulus packages, right? You had Republicans and Democrats voting to put a lot more money in the system, to create a lot more flexibility in the system, and to remove a lot of those obstacles that had been put into place actually in another era with both Republicans and Democrats, namely the mid nineties, to really squeeze and starve the unemployment system. I just think it’s really worth noting that people across the political spectrum, shoring up this system, and as a consequence, poverty rates actually went down during this massive economic dislocation that we experienced. I just think that is really astonishing and something that we need to take going forward as we are figuring out how to really create an economy where everybody’s prospering.
Paul Constant:
Next week on Pitchfork Economics, Nick and Goldy are talking to Vote.org CEO, Andrea Hailey.
Speaker 4:
Pitchfork Economics is produced by Civic Ventures. If you like the show, make sure to subscribe, rate and review us wherever you get your podcasts. Find us on Twitter and Facebook at [Civic Action 00:32:10] and Nick Hanauer. Follow our writing on medium at Civic Skunk Works, and peek behind the podcast scenes on Instagram at Pitchfork Economics. As always, from our team at Civic Ventures, thanks for listening. See you next week.