A Discussion of the Changing US Labor Force

5/7/2024

Tom Heintjes:Hello, and welcome to another episode of theEconomy Matters podcast. I'm Tom Heintjes, managing editor of the Atlanta Fed's digital magazine,Economy Matters, and today we're sitting down with Nye Hodge, an analyst in the Atlanta Fed's Center for Workforce and Economic Opportunity. Today we're going to discuss recent research by Nye, and some coauthors, into changes the workforce has seen. And so, without further ado, let's say "hi" to Nye Hodge. Nye, thanks for being with us today. It's great to have you on the podcast and to talk about your work.

Nye Hodge: Tom, thanks for having me. It's a pleasure to be here with you.

The Atlanta Fed's Nye Hodge. Photo by Stephen Nowland

Heintjes:Thank you. Nye, in your research, which you titledThe Labor Market, Then and Now: The First Two Decades of the 21st Century, you and your coauthors, Stu Andreason and Carl Van Horn, discuss how the workforce-and the job market overall-has evolved in recent decades. Before we get into your findings, I wanted to ask you: What led you to undertake this research? It's really quite an ambitious and deep dive into data and other information.

Hodge: Absolutely. It was quite an ambitious undertaking, as you stated. We analyzed a heap of data that covers the last two decades of the labor market, and we did our very best to distill it into a cohesive overview that looks at how things have improved, stayed the same, and how they've worsened for workers. But to your question-Carl Van Horn is distinguished professor of public policy at Rutgers University, which is also my alma mater. Carl is the director of the Heldrich Center for Workforce Development at Rutgers, and they released the first iteration of this report-called "The Labor Market, Then and Now: Changing Realities in the 21st Century"-back in 2010, which looked at the first decade of the labor market. We got involved with the second iteration that looks at the last two decades of the labor market. This time around, we took a much more expansive look and turned it into a three-part series rather than the single release, which was the first iteration.

Heintjes:Right. Well, first of all: go Scarlet Knights!

Hodge: Go Scarlet Knights!

Heintjes:And Nye, obviously, technology is a big factor in changes in the labor market. What are some other changes?

Hodge: This is a pretty loaded question, Tom. Technology is certainly an obvious factor. It's tangible, it's traceable-meaning we can see when X technology came into play, then Y happened. Technology is comparatively easier to understand, compared to other factors-things like policy changes, economic shocks, access to education, cost of living, child care and elder care, the aging population, mental health crises, for example. There's certainly a protracted list of items that are intangible and also more abstract than technology but still have a very measurable impact on the labor market.

Heintjes:We'll get to some of those factors shortly, and I don't want to give them short shrift here, but first, let's talk about the main reason people work, which is to earn money. How have wages evolved over recent decades?

Hodge: Absolutely-employers would have a hard time getting people to show up for free, right? Part one of the report opens with a look at inflation-adjusted median earnings for the years 2000-22. And there's a lot of illustration throughout the report, which I think helps make the report more pleasurable to read and much more digestible. We're using charts and tables as our graphics, and the graphics in the earnings section show that earnings really ebb and flow over the years. Year to year, median earnings go up, they go down, they stagnate. The positive news, though, over the long-term, earnings do rise despite those ups and downs that we see in the charts. For example, from 2000 to 2020 median earnings increased 15 percent, from about $45,000 to $52,000. That included some big gains from 2018 to 2020. In fact, 2020 earnings hit a two-decade high-but that was knocked off, though, with a roughly 6 percent decline from 2020 to 2022, due in part to inflation.

Heintjes:Your research makes it clear that we can't look at workers as a homogeneous bunch. We need to look at subgroups, like genders and ethnicities. What are the trends you see there? Are some groups faring better than others?

Hodge: Absolutely. It's so important and imperative to do crosscuts of data, especially breaking things out demographically. There's always so much to be revealed as we go down the mountain that certainly will be missed if we stay at the peak of the data-so I'm glad you asked that question, Tom. One of the biggest improvements we've seen at the demographic level is the gender pay gap between male and female workers. In 2000, female workers earned about $76 for every $100 male workers earned. By 2022, that improved to $83 for every $100. Of course, we can't ignore the slow burn of it taking 22 years to see a $7 improvement, but the demographic data also reveal why it's been such a drag. So, female workers who identify as Native Americans saw a 15 percent decrease in earnings over that same period, 2000 to 2022. Female workers who identify as Hispanic or Black/African American saw relatively modest gains of about 11 percent to 8 percent, respectively, over the same period. Driving much of that improvement to close the gap were gains by female workers who identify as Asian and Pacific Islander, or White. Their earnings went up 45 percent and 18 percent, respectively. So without doing those demographic breakouts, you don't see that full narrative. This indicates that to get to parity, factors that are holding back groups have to be addressed, or we really won't see pay equity between male and female workers.

Heintjes:Nye, another facet of the labor market you look at in your work is labor force participation, or what economists call LFP. We heard that LFP declined during the pandemic. How have those trends looked in the last decades, and what differences do you see between men and women participating in the labor force?

Hodge: As you mentioned, pandemic-related conditions certainly had a negative impact on labor force participation. We saw retirements spike, folks leaving the workforce due to health concerns for themselves and their loved ones. Child care was also a factor that caused people to leave the workforce. But what was being talked about less during this period was the fact that the labor force had been declining since the year 2000, really. The pandemic certainly accelerated things, especially with early retirements, but the declining workforce was not a new phenomenon. As far as gender differences go, participation among female workers is declining at nearly half the pace of male workers. From 2000 to 2022, participation among female workers dropped about 3.5 percentage points, compared to 6.3 points for males. And there's a couple of factors at play here, and I'll cite just a few of them. We have one of the largest generations, by number of people, retiring-that is, the baby boomers. The baby boomer workforce is going to be overrepresented by males, just by virtue of social conditions workers faced in that generation. So we have a larger number of males retiring compared to females. Fast forward to today-we have fewer people to replace the baby boomers, despite females who have made historic gains in educational attainment and are entering the workforce at historic rates, which is contributing to the slower decline in female participation rate. However, younger males are increasingly detaching from the workforce and even from pursuing higher education, which is contributing to the higher decline in male participation because of the lower replacement rate.

Heintjes:Nye, you mentioned the pandemic, and I want to touch on that a little bit. That's among two great economic shocks during the period you examined in your work, which is the Great Recession, and the resulting global financial crisis, and the COVID-19 pandemic. Those events obviously brought great changes on the labor force. What did you see as a result of both of those events? Were the effects the same, or were they different-and if so, in what ways?

Hodge: Both of these events caused recessions that were similar in many ways-recessions are always going to have some similarities-but there are also some differences in considerable ways as well. There are certain metrics that are just synonymous with recessions, right? You're going to see a spike in unemployment, which we saw. You're going to see labor force participation take a negative hit, which we saw, and usually you'll see earnings take a hit as well. I would venture to say, inarguably, the biggest difference between the two recessions were the policy responses coming down from Washington. During the Great Recession, federal dollars came down to offer economic relief to corporations. MIT did a study that puts that number at just under $500 billion. During the pandemic, Washington provided relief not only to businesses but also to individuals, which was an unprecedented approach. It was a massive undertaking. The New York Times put out a nice, graphical breakdown of this last year. They've estimated that about $5 trillion was spent on pandemic relief efforts-to individuals and families, to businesses, state and local governments, health care, and other miscellaneous expenses. But the exact figure to individuals and families was about $1.8 trillion, and about $1.7 trillion to businesses, and I imagine that in the future, these measures would be considered a blueprint for managing future recessions. Of course, no two recessions are exactly alike, and this one was largely tied to whether or not it was safe to return to work, and to be outside and going out and doing things. But evidence suggests that these steps that Congress took prevented a long, drawn-out recession. Also, poverty actually went down during this time. Imagine that! There's a recession, and poverty went down, despite the economic shock.

Heintjes:Nye, there's more I want to discuss with you, but we're going to take a break and listen to a few words from the Atlanta Fed.

[advertisement]

Heintjes:Welcome back. We're talking to the Atlanta Fed's Nye Hodge about his recent research into the changing labor market. Nye, I want to pick back up by asking you about the low unemployment rate, which we hear a great deal about right now. Once you looked more deeply into unemployment, did anything in particular catch your attention?

Hodge: Yes. Standing at the peak of the proverbial data mountain-that's when we're looking at aggregated data-we see unemployment at healthily low levels. At the height of the pandemic, unemployment spiked to around 14.4 percent in 2020. It recovered in about two years to 3.3 percent, which is considered a solid economic footing. By contrast, it took several years for unemployment to recover from the Great Recession. When we go down the data mountain, though, what we see is Black, Hispanic, and Native American workers, young workers, workers without a four-year college degree, really bearing the biggest brunt of unemployment shocks. Those groups tend to see longer recovery periods compared to the average.

Heintjes:Nye, in your report you look at unemployment insurance-which, as you mentioned, was a lifeline for many people during the pandemic. How would you describe the role of unemployment insurance during that period, compared to other historical periods

Hodge: I'd like to take a quick step back on this piece, just to go into unemployment insurance just a little bit. Unemployment insurance was part of President Roosevelt's New Deal program to mitigate the economic hardships caused by the Great Depression. It was established in 1935 as part of the Social Security Act, and it's a joint program between federal and state governments to provide temporary financial assistance to unemployed workers who lose their job through no fault of their own. So, think recessions-a company downsizes and lays people off, a merger happens and people get laid off, etc. It does not cover if someone up and quits or is terminated as a disciplinary action. There are some narrow qualification guidelines that people have to hit to be approved for unemployment insurance. Usually there's things like minimum number of hours worked, minimum number of earnings, you have to be an employee as opposed to a contractor or self-employed or a gig worker. Those are some of the basic qualifications, and states do have some slight variations, state to state. But generally, Congress does expand eligibility during recessions. But the pandemic-era expansions were the largest expansions of the program since its inception: 78 percent of people who applied for unemployment insurance received it during the COVID-19 recession era, compared to 40 percent at the height of the Great Recession. And experts are attributing the quick economic recovery from the pandemic-related recession to measures like these, which you did not see during the Great Recession.

Heintjes:Well, of course, we can't talk about a pandemic without talking about health insurance, and I want to ask you, what trends have we seen among workers and health insurance, and how have these trends changed? Obviously, the ACA plays a big role here, but is there anything else at work?

Hodge: Yes, absolutely. The ACA-which is the Patient Protection and Affordable Care Act, which was enacted by President Obama in 2010-has the goal of expanding health care coverage and reducing out-of-pocket costs. The fact that the law passed in 2010 was actually perfect for this analysis, because we looked at the decade before the law was passed, and then the decade after the law was passed. During the first decade of the 21st century, the medically uninsured rate was actually increasing. The percentage of Americans without health insurance increased, from 14 percent in 2000 to just over 16 percent in 2010. Five years after ACA passed, only about 9 percent of Americans were uninsured by 2015. And then during the pandemic, Congress passed measures to further expand access to health insurance and to ensure that people who are losing their jobs wouldn't also lose health insurance, because a lot of times, health insurance is tied to work. So, in 2020 we actually saw the uninsured rate go to under 9 percent.

Heintjes:You also looked at trends in retirement benefits, and I thought that was fascinating. How have those benefits changed during the first part of this century?

Hodge: Actually, I also thought that this particular section was quite interesting. Workers today have more access to retirement benefits than in the past. In 2005, 60 percent of workers had access to retirement benefits. By 2020, that increased to 70 percent of workers. But the interesting part to me was that uptake only increased by 5 percent. In 2005, 50 percent of workers participated in a retirement program. It increased to 55 percent in 2010, and it has remained unchanged since then. Research attributes the stagnation to the kinds of retirement plans that are available. Employers have mostly moved away from what's called "defined benefit retirement plans," which are generally employer-funded and often don't transfer if a worker changes jobs-think pension plans, for example. Instead, more employers are now offering defined contribution plans, where individuals generally have to sign up, and then both they and their employer contribute-think 401(k)s, for example. And generally, these plans are movable, so they stay with the employee when they leave their job.

Heintjes:This has been a fascinating conversation, Nye, but before we say goodbye, I want to ask what's next on your plate. Is it too early to talk about upcoming research? I mean, you never run out of things to talk about with the workforce.

Hodge: Absolutely, you never run out of things to talk about with the workforce. There's always research to try to help uncover solutions to some of the problems that workers are facing. We don't have any research coming out-at least, I don't have any research coming out-in the short term. Definitely some things probably coming out towards the end of the year. But I will point listeners to research that came out early last year called Closing the Digital Skill Divide, and that research has really been making its rounds and picking up traction. It looks at employer demand for digital skills-and really what the research found-is that 92 percent of jobs require some form of digital skills. Workforce groups and practitioners have really been using this research to advocate for workforce funding to ensure that workers in their states have the skills they need for the jobs of today, and tomorrow. That research can be found just by googling "closing the digital skill divide."

Heintjes:Okay. Well, we'll have to have you on another episode to talk about that, for sure.

Hodge: Absolutely-I would love to.

Heintjes:And I just want to say thanks for taking time to talk with us and our listeners today about your work. I hope this is only the first of your visits to the recording studio.

Hodge: I'm looking forward to coming back, Tom. Thanks for having me.

Heintjes:And I should note that we'll have a link to the research we've been discussing today on our website at atlantafed.org, and I encourage you to take a look at it-it's very interesting and accessible material. And that brings us to the end of another episode of theEconomy Matters podcast. Again, I'm Tom Heintjes, managing editor ofEconomy Matters. Thanks for spending time with us today, and let's meet up again next month.

Public link

Please use the above public link if you want to share this noodl on another website.