Regional President Speech

Challenges to Women’s Labor Force Participation Thomas I. Barkin President, Federal Reserve Bank of Richmond Charlotte Economics Club Charlotte, North Carolina June 3, 2019 Good afternoon. Thank you very much for inviting me to join you today. I’d like to tell you a little bit about my background and my perspective, and then I’d like to share some thoughts on growing the economy. Before I say more, note that the views I express are my own and not necessarily those of my colleagues on the Federal Open Market Committee (FOMC) or in the Federal Reserve System.1 I joined the Richmond Fed last January after a 30-year career in consulting at McKinsey, where I was the chief financial officer and led the offices in the South. I’ve spent my professional life helping firms make decisions about hiring, compensation and prices, and I’ve made a lot of those decisions myself. So I’m trying to bring a different perspective to the FOMC. My colleagues on the committee are some of the most talented macroeconomists, bankers, academics and financial regulators in the country. But as the only committee member coming from management, I can approach things as a practitioner rather than a researcher. At the same time, I have greatly enjoyed the opportunity to dig into economics. In particular, I’ve been studying long-run economic growth. At some level, the math of growth is pretty simple: We need more people working, and we need them to be more productive. In recent years, however, both population and productivity growth have been slow. I’m hopeful that productivity growth will improve, but in the meantime, it’s worth exploring how we can draw workers off the sidelines. We need that to happen — for the health of our economy, for our fiscal soundness and to lift the fortunes of those left behind. Trends in Women’s Labor Force Participation A lot has been written about the declining rate of male participation in the workforce, but I’d like to spend today focusing on the participation of women. In the United States, women’s labor force participation increased steadily throughout the second half of the 20th century. But around 2000, it leveled off and then began to decline. This is a divergence from other developed countries. In the late 1990s, for example, the labor force participation rate for women aged 25- 54, whom I’ll refer to as prime-age women, was the same in the United States and Canada. Since then, however, the rate has continued to increase in Canada while it has decreased here, leading to a 7.9 percentage point gap between the two countries in 2017. If the United States had the same prime-age participation rate as Canada, an additional 5 million women would be in the labor force.2 That would be a significant boost to our potential growth. What is behind the gap in women’s participation? First, we need to calibrate in the context of demographics. One difference we see in the United States is a much larger population of Hispanic women, who traditionally have a lower labor force participation rate than non-Hispanic women, by around 9 points.3 The share of the female labor force that is Hispanic has increased from 8.7 percent in 1997 to 16.4 percent in 2016, a trend that accounts for just under a point of the participation gap. But note that this trend is changing. The participation rate for all prime-age women in the United States has increased since 2015, likely due to the strengthening economy, and some have found that Hispanic women accounted for almost two-thirds of the increase.4 -2- Another demographic trend to consider is educational attainment. Higher levels of education are associated with higher labor force participation, and in both countries, an increase in the number of women with a college education boosted their participation. But the increase was larger in Canada, where the share of prime-age women with a bachelor’s degree increased almost 17 percentage points from 1997 to 2015. In the United States, the share increased 11.4 points. Increasing levels of education accounted for an increase in participation of almost 4.5 points from 1997 to 2016 in Canada versus just under 2 points in the United States. Finally, let’s look at the effect of an aging population, which tends to lower participation rates. In this case, the relative demographics actually helped the United States, although the overall effect was fairly small. Between 1997 and 2016, an aging population cost Canada a quarter of a point; during the same time period in the United States, the aging of the population had very little effect. If you are following the math, we started with a 7.9 point gap, plus a quarter of a point due to an aging population in Canada. Then, we subtract 2.5 points for education and around 1 point for the Hispanic population. That leaves us with roughly 4.5 points unexplained by demographics. -3- Challenges Facing Less-Educated Women What’s behind the remaining gap? You might assume, as I did, that parental obligations drive the difference, but the number of women not working for this reason has not significantly increased over this time period. Of course, participation does vary for women who have children, but the decline in participation does not fall along those lines. Instead, the decline is most dramatic for prime-age women with less formal education — regardless of whether they have children or not. Between 1997 and 2015, the U.S. participation rate for prime-age women of all education levels dropped almost 3 percentage points. But for prime-age women with a high school education or less, participation dropped 7.4 points. It fell even further — almost 9 points — for women aged 25 to 44. Surprisingly, that is not the story in Canada. In Canada, the increase in women’s participation is driven primarily by women aged 45 to 54, who experienced an increase of almost 10 points. Within that group, the gain was largest for women with less formal education, roughly 8 points. So what is going on with less-educated women that is driving down their participation in the workforce in the United States yet driving it up in Canada? And more important, what can we do about it? With this population, and this decision, I want to make the case that money matters. So let’s look at the fundamental economic equation for these women. The Economic Equation The Earned Income Tax Credit (EITC) is a great program that lifts millions of people out of poverty each year.5 But its phase-out can create barriers to working.6 Imagine a woman in her -4- mid-30s. She has two children and lives in northern Virginia, where she works at a day care center and is married to a construction worker. If they both work full time, they earn around $55,000 per year — at which point they no longer qualify for the EITC and stand to lose $3,700 at tax time.7 And because our system taxes the first dollar earned by the second earner at the same rate as the last dollar earned by the primary earner, they are pushed into a higher tax bracket. Add to this the cost of child care and other work-related expenses, and the family’s monthly take-home pay might be actually higher when the woman does not work or works informally in the shadow economy. Similarly, disability insurance is a vital safety net for many people. But there are aspects of our disability insurance program, including the fact that you instantly forfeit life-long benefits if you engage in the workforce, that create disincentives to working. The number of prime-age women in the United States who received Social Security disability insurance benefits increased from 1.2 million in 2000 to 1.9 million in 2014, increasing from 2.0 percent to 3.0 percent of women in this age group.8 Let’s imagine a single mother in her mid-40s, working as a retail cashier in Columbia, South Carolina. She struggles with a chronic illness that makes it difficult for her to work. Disability insurance provides her with a lower gross income than working full time, but it lasts until retirement, comes with Medicare eligibility and doesn’t entail costs such as child care. If she takes a full-time job, she has to assess the risk that — if that job goes away for any reason — she loses these benefits that bridge to retirement. -5- Finally, let’s imagine a married woman who works in a poultry processing plant in West Virginia. If she and her spouse both work, they earn an additional $600 a month, but there are a lot of nonfinancial incentives — what I’d call the “hassle factor” — that motivate not working or working informally. If she works informally, for example, by babysitting for friends or family, she has a more flexible schedule and her day-to-day environment may be more attractive. What Can Be Done? How do we make it more likely that women like the ones I’ve described participate in the labor force? Employers play a big role. We hear a lot these days about companies struggling to find employees. They can break down barriers to women’s participation by offering more stable schedules, better working conditions or on-site child care. For example, we know of a furniture maker in West Virginia who’s been able to attract and retain a majority female labor force by offering a single shift, replacing educational requirements with apprentice-style training and aiming to provide employees with a better lifestyle. And, while I mentioned earlier that the most dramatic decline in participation does not appear to be among women with children, employers could also consider paid maternity leave and high- quality affordable child care. Some research has found that over a quarter of the increase in women’s labor force participation in other OECD countries relative to the United States is due to the expansion of these family-friendly policies.9 -6- Policymakers can help. I’ve discussed policies that may discourage women from entering or staying in the labor force, such as the joint tax system, the phase-out of the EITC or the structure of disability insurance. Fiscal policies are out of the Fed’s purview, and therefore I’m not taking a stand on any particular policy nor suggesting the incentives have changed, but I do think that the total incentive structure merits reexamination. Work has been done that shows that a Canadian tax reform that effectively reduced the tax rates of a second earner increased the labor supply of married women.10 And finally, solutions can come from these women themselves. As I noted earlier, we know that education increases women’s participation in the workforce. Workers with more education also tend to have higher earnings, greater wealth, more resiliency to downturns and better health. But before we turn this problem back to the workers, we have to remember that the same barriers that make it hard for many women to be in the workforce make it hard to go back to school. In our district, we’ve seen some innovative solutions to breaking down these barriers, such as the Workforce Investment Network (WIN) Program in Charlotte, a collaboration between the Carolina Fintech Hub, community partners and employers. The WIN program targets women and minorities, provides technical training and soft-skill development and guarantees program graduates a full-time position at participating financial services firms. Importantly, in addition to paying students during the training, the WIN program provides wraparound services, such as transportation and child care assistance, that make it easier for students to complete the program. -7- Conclusion To sum up, women with less education in the United States face a number of obstacles to working, which could be contributing to a decline in women’s labor force participation in the United States relative to other countries. As a policymaker, I’m concerned not only about the well-being of these individuals, but also about our country’s long-term economic growth. Employers, policymakers and workers can invest to remove these obstacles; doing so would benefit them and our nation as a whole. Thank you, and now I look forward to your reactions and questions. 1 Thank you to Nina Mantilla and Jessie Romero for assistance preparing these remarks. 2 This figure is based on the prime-age participation rate in Canada in 2017. 3 Richmond Fed calculations based on Bureau of Labor Statistics data. The 2016 labor force participation rate for prime-age Hispanic women was 66.7; for non-Hispanic women it was 75.8. 4 Data is for 2015 through 2018, see John Robertson, “X Factor: Hispanic Women Drive the Labor-Force Comeback,” Federal Reserve Bank of Atlanta macroblog, March 6, 2019. 5 In 2016, it was estimated that the EITC lifted between 4.1 million and 5.8 million people (including 3 million children) out of poverty, according to a 2018 report by the U.S. House Committee on the Budget. 6 Nada Eissa and Hilary W. Hoynes, “Behavioral Responses to Taxes: Lessons from the EITC and Labor Supply,” Tax Policy and the Economy, 2006, vol. 20, pp. 73-110. 7 Author’s calculations using TAXSIM v.27. Daniel Richard Feenberg and Elizabeth Coutts, “An Introduction to the TAXSIM Model,” Journal of Policy Analysis and Management, Winter 1993, vol. 12, no. 1, pp.189-94. Available via the NBER website. 8 Social Security Administration, “Annual Statistical Report on the Social Security Disability Program, 2000,” September 2001; Social Security Administration “Annual Statistical Report on the Social Security Disability Program, 2014,” November 2015. 9 Francine D. Blau and Lawrence M. Kahn, “Female Labor Supply: Why is the United States Falling Behind?” American Economic Review, May 2013, vol. 103, no. 3, pp. 251-56. 10 Thomas F. Crossley and Sung-Hee Jeon, “Joint Taxation and the Labour Supply of Married Women: Evidence from the Canadian Tax Reform of 1988,” Fiscal Studies, 2007, vol. 28, no. 3, pp. 343-65. -8-

Tom Barkin (2019, June 2). Regional President Speech. Speeches, Federal Reserve. https://whenthefedspeaks.com/doc/regional_speeche_20190603_tom_barkin

@misc{wtfs_regional_speeche_20190603_tom_barkin,
  author = {Tom Barkin},
  title = {Regional President Speech},
  year = {2019},
  month = {Jun},
  howpublished = {Speeches, Federal Reserve},
  url = {https://whenthefedspeaks.com/doc/regional_speeche_20190603_tom_barkin},
  note = {Retrieved via When the Fed Speaks corpus}
}