Skip to Main Content

Tom Barkin

Getting People off the Sidelines: The Ultimate Workforce Development Plan

Tom Barkin
Nov. 12, 2019

Tom Barkin

President, Federal Reserve Bank of Richmond

Danville Pittsylvania Chamber of Commerce
Institute for Advanced Learning & Research
Danville, Va.


  • Helping recruit people who have left the workforce into jobs is important for future economic growth.
  • At the same time millions of prime-age people are not working, millions of jobs are going unfilled.
  • The disconnect likely stems from several sources: the attractiveness of available jobs; a mismatch between where people live and where jobs are located; barriers such as poor health or addiction; and incentives to work.
  • Community groups that focus on population health are part of the solution. It’s also important to think carefully about tax and benefits structures. In addition, the tightness of the labor market is motivating some employers to re-examine the jobs they offer.

Thank you for inviting me to speak with you today. It’s a pleasure to be here in the Dan River region, and I’m looking forward to a series of discussions about the impressive suite of actions you’ve taken to rebuild your economy.

I came to Richmond almost two years ago. In that time, I’ve been very much “on the ground” to learn about the communities in the Fed’s Fifth District, which spans from Maryland to the Carolinas. As I’ve traveled, I’ve seen areas that are thriving and areas that are struggling. I think the Fed has a real opportunity—and responsibility—to make a difference, particularly with respect to labor market outcomes. That’s why I’m excited to learn more about what you’re doing.

The focus today is “what’s around the corner” in workforce development. I’ll leave to our distinguished panelists the topics of the jobs of the future and how to train people for those jobs. Instead, I want to talk with you about what has to be “around the corner” if we are to grow our economy—and that’s helping recruit into jobs people who have left the workforce. Maybe they lack the skills they need. Maybe they’re in poor health. Maybe they’re making ends meet through benefits they’re scared of losing. Maybe they’re struggling with addiction. Maybe they’re discouraged about the jobs they’re qualified to do. What can we do to bring these people off the sidelines and back into the game? That would be the ultimate workforce development plan.

Before I say more about my perspectives on this, note that the views I express are my own and not necessarily those of my colleagues on the Federal Open Market Committee or in the Federal Reserve System.1

Who’s on the Sidelines?
Let’s start with looking at who these people are.

It won’t be news to you that the labor market is tight. The unemployment rate is near a 50-year low, at 3.6 percent. The rate is the same here in Pittsylvania County. Across our country, there are currently 7 million jobs available—and only 5.9 million people looking for work. In fact, there have been more job openings than job seekers since March of last year. It’s the first time we’ve been in this situation in the nearly 20-year history of this data series.

But the unemployment rate only tells you about people who are looking for work. More than six times as many people are of what we’d consider the relevant age for work but aren’t looking. So let’s dig into those numbers. According to census data, around 3.5 million prime-age people (that is, people between the ages of 25 and 54) are looking for work and are therefore counted as unemployed. But 22.4 million prime-age people are not working and are not looking for work. When you add it up, we have 25.8 million prime-age people, or 20 percent of that population, who, for some reason, are not working; the rates are about 10 percentage points higher in smaller towns.2 About two-thirds of those not working are women and around half have a high school education or less. People not working also are disproportionately people of color.3

The numbers look different in other countries. For example, if we had the same percentage of prime-age people working as in the United Kingdom, we would have around 5 million more people in the labor force. That would imply 5 percent more growth potential in our economy.4

Not every person out of the labor force wants to be in it, or should be. About 10 percent of this group is in school, gaining skills that will help them succeed in the workforce in the future. Another 35 percent are caring for children or other family members, a personal choice they’ve made—and by the way, a contribution to the economy that our metrics don’t measure.5 But that still leaves more than half of this group, or more than 14 million people.

What’s holding them back? Pulling together analyses from multiple economists gives us a hint, recognizing these numbers overlap. Just under half of the remainder (24.4 percent of the total, or 6.3 million) report that they have a health problem or disability that keeps them from working or limits the work they can do. Four million prime-age adults are receiving Social Security Disability Income benefits, and over 2 million receive Supplemental Security Income.6 In addition, there is research suggesting that the increase in opioid prescriptions between 1999 and 2015 drove down the labor force participation rate, particularly for prime-age men.7

There may also be some people who are not counted as working by the official statistics but who are working informally. Boston Fed researchers have estimated that in 2015, 28 percent of “unemployed” workers and 17 percent of those not in the labor force were working in the gig economy, although they typically only worked a few hours per week and earned less than $350 per month.8

What Jobs Are Available?
So we know there are millions of people not working—but there are also lots of jobs going unfilled. In every town I visit, I hear stories about restaurants and homebuilders and nursing homes that can’t find workers. Where’s the disconnect? 

Part of the issue might be that many of the available jobs aren’t that attractive. According to data from the Bureau of Labor Statistics (BLS), 63 percent of employment is in occupations that are low skill, meaning they don’t require more than a high school education, and 45 percent is in jobs that pay a median wage of $35,000 or less per year. These jobs, such as food service workers and personal care aides, are predicted to have high growth. We see this in the Dan River area as well. Last year, the majority of job openings in the Dan River region required a high school degree or less and over 60 percent paid less than $13/hour.9

These jobs often have other shortcomings in addition to low pay, such as unpredictable schedules, difficult working environments and hours that can fail to meet employees’ needs. In addition, the industries in which these low-skill jobs are the most prevalent are the least likely to offer employer-based health insurance. For example, only 35 percent of firms in the retail industry offer health insurance benefits, compared with 57 percent of all firms. It’s even worse for part-time and temporary workers. Only 28 percent of firms offer health insurance to employees who work less than 30 hours per week and, not surprisingly, just 7 percent offer it to temporary workers.

Even when employers do offer health insurance, many workers don’t take advantage of it—often because it’s more than they can afford. In 2019,the average premium at firms where at least 35 percent of workers earn $25,000 or less annually was $515 per month for an individual and $1,469 month for a family.10 For a family of four with an annual household income of $50,000, that’s 35 percent of their monthly income.  

Another shortcoming is that many low-wage jobs aren’t very stable. Industries such as leisure and hospitality, food services and retail trade have high turnover rates.11 There’s also a question about the long-term stability of these jobs, as automation may displace many low-skill workers in the coming decades. It’s already started—in many fast food restaurants, for example, it’s possible to place your order through a kiosk. My former colleagues at the McKinsey Global Institute predict that by 2030, 38 percent of food service jobs and 24 percent of customer service and sales jobs could be displaced by automation.12

Identifying the Mismatches
What would it take to put these on-the-sidelines workers into the available jobs?

Part of the story is skill mismatch, as my colleague Mary Daly of the San Francisco Fed recently discussed.13 Twenty-six percent of employment is in higher-skilled jobs, meaning they require a bachelor’s or higher. Employers are finding it difficult to find employees with the relevant talent in fields like technology and nursing. Improving K-12 education, leveraging community colleges, increasing college completion rates and classic workforce training concepts, like those being discussed today, are necessary.

We also hear a lot about location mismatch. The McKinsey study I just mentioned also predicted that the most job growth will occur in the nation’s 25 most prosperous cities, while smaller towns are at risk of losing more jobs. It’s easy to say, “Well, people should just move to where the jobs are.” But rising costs of living in big cities, especially for housing, have often more than offset the urban wage premium for low-skill work. So for many nonemployed workers, it may not make sense to leave a hometown that has a relatively low cost of living, not to mention a support network of friends and family, for a low-pay job in a high-cost city.

I also want to talk about three other issues that don’t get as much attention: health, incentives and job attractiveness.

When it comes to health, I’ve mentioned the role that opioid addiction may play in labor force participation, and I’m intrigued by innovative addiction-to-work programs. But there are many other health issues going on as well. The relationship between unemployment and poor health is well-documented, and prime-age individuals who are not in the labor force are far more likely to report poor or fair health than those who are employed or even unemployed. More than one-third of prime-age men out of the labor force report a disability, compared with 6 percent of prime-age unemployed men and 2.6 percent of prime-age employed men.14 And more than one-third of nonworking, nonelderly Medicaid recipients report living with multiple chronic conditions.15

At the same time, hospitals are closing, especially in rural markets, and health care is becoming ever more expensive. How can we return people to health and to the workforce? There are community groups that recognize that you can’t grow a talented workforce without a healthy population, such as the Health Collaborative here in Danville and Active SWV, a nonprofit building an ecosystem of physical activity in the community. The Health Collaborative has volunteers focusing on healthy eating, active living, access to health care and creating healthy spaces. It currently has more than 90 individuals who represent more than 50 organizations spanning local government, education, health care, business and nonprofit sectors. The key for these local success stories is scaling.

Next I want to talk about incentives. Our benefits structure, which provides critical support for people in need, sometimes discourages working. For example, if you aren’t working and are receiving Medicaid, it doesn’t sound very appealing to take a low-wage job that doesn’t offer affordable health insurance and that also makes you or your children ineligible for Medicaid or the Children’s Health Insurance Program (CHIP). This situation, which we call a “benefits cliff,” is far more likely to occur in states that have not expanded Medicaid.16

The structure of disability insurance also creates disincentives for work. That’s because you forfeit lifelong benefits if you engage in the workforce—so if you’re on the margin of being able to work and your choice is between an unstable, low-paying job and a stable (albeit also low-paying) benefit, it wouldn’t be surprising if you chose the benefit.

Our tax system can also make working relatively unattractive for certain segments of the population. For example, we have “joint taxation,” which means that the system taxes the first dollar earned by the second earner at the same rate as the last dollar earned by the primary earner, pushing the household into a higher tax bracket. Add to this the cost of child care and other work-related expenses, and a family’s monthly take-home pay might be actually higher if the second earner doesn’t work or works informally. In addition, the Earned Income Tax Credit phases out at higher income levels, which could lead someone to conclude that working isn’t economically worth it.

Overall, we have a population with health problems, and they have incentive to stay where they are to maintain eligibility for Medicaid, CHIP or disability insurance. Then, they look at the jobs available to them and see unstable jobs, unpredictable hours and low pay. Working doesn’t seem like a very good trade.

What can policymakers do to make it a better trade—for the good of the individuals and for the good of our economy as a whole?  Fiscal policies are out of the Fed’s purview, and therefore I’m not taking a stand on any particular policy, but the incentive structure warrants re-examination. We do see that Medicaid expansion, for example, which Virginia has already done, changes the incentive equation.  

Finally, some jobs are going unfilled because they’re viewed as unattractive, for the reasons I’ve discussed. The pay, the working conditions, the unpredictability or the perceived stability can be deterrents. Or, workers’ prior jobs could have had higher pay or higher status, and they don’t want to take a role they view as a step back. Or embedded perceptions could play a role. More than 60 percent of workers in four of the top five occupations with the most projected growth, such as personal care aides, are female. Some men might be choosing not to pursue jobs in those fields.

The good news here is that the tightness of the labor market is motivating employers to re-examine the jobs they offer. We see significant wage increases for many entry-level positions; a major bank, for example, recently announced a $20 minimum wage. We’ve talked to a steel company air conditioning its plants to make the environment more comfortable, a furniture company offering hours tailored to primary caregivers and companies being more flexible on teleworking. That’s good news. Many economists believe that if we can get people into the labor force, they are likely to stay in the labor force.

Job training and job placement really matter. But in today’s economy, with such a large percentage of the potential workforce not working, I also want to make the case that job design, workforce health and incentives to work matter too. As I said at the beginning, that’s the ultimate in workforce development.

Thank you, and now I look forward to your reactions and questions.


Thank you to Abigail Burns, Nina Mantilla and Jessie Romero for their assistance preparing these remarks.


Prior research we’ve done has found that employment rates are around 10 percentage points lower in smaller towns, based on census data for ages 16-64.


Though the majority (approximately 51 percent) of prime-age nonemployed individuals are white, non-Hispanic, people of color are nonemployed at higher rates. For prime-age individuals, we calculate that approximately 25 percent of the black, non-Hispanic population is nonemployed versus only 18 percent of the white, non-Hispanic population.


Hours worked varies across countries and is not included in the calculation.


We estimate 8,947,983.20, or 34.66 percent of prime-age nonemployed (2019), are stay-at-home parents.


SSI provides disability benefits to people with limited work history.


Alan B. Krueger, “Where Have All the Workers Gone? An Inquiry into the Decline of the U.S. Labor Force Participation Rate,” Brookings Papers on Economic Activity, Fall 2017, pp. 1-87; Dionissi Aliprantis, Kyle Fee, and Mark E. Schweitzer, “Opioids and the Labor Market,” Federal Reserve Bank of Cleveland Working Paper No. 18-07R, May 15, 2018.


Mary A. Burke and Anat Bracha, “How Big is the Gig?” 2018. We note the gig economy is difficult to measure and estimations vary. A recent BLS study compares employment numbers from the CPS with time spent in gainful employment from the American Time Use Survey and found smaller differences than Burke and Bracha.


Based on data prepared by Chmura Economics for the Dan River Region Collaborative.


Statistics on health insurance from the Kaiser Family Foundation, “Employer Health Benefits, 2019 Annual Survey.”


Data from JOLTS via Haver Analytics. Turnover rates: retail trade: 5.2 percent; leisure and hospitality (includes accommodation and food services): 6.7 percent; total private industry: 4.2 percent; financial activities: 2.4 percent.


McKinsey Global Institute, “The Future of Work in America,” July 2019, Exhibit 12.


Mary C. Daly, “Beyond Fairness: The Value of an Inclusive Economy,” Speech at the Los Angeles World Affairs Council & Town Hall, Los Angeles, California, Oct. 16, 2019.


See Tables 4 and 5 in Alan B. Krueger, “Where Have All the Workers Gone? An Inquiry into the Decline of the U.S. Labor Force Participation Rate,” Brookings Papers on Economic Activity, Fall 2017, pp. 1-87.


In Medicaid expansion states, 2019 income limit is: $17,236/year for an individual or $29,435/year for a family of three. In non-expansion states (16 states) median eligibility limit for a family of three is $8,532/year and childless adults are not eligible (except in Wisconsin). CHIP eligibility: Median income limit for family of three is $55,452/year.

Subscribe to News

Receive an email notification when News is posted online:

Subscribe to News

By submitting this form you agree to the Bank's Terms & Conditions and Privacy Notice.

Phone Icon Contact Us

Jim Strader (804) 697-8956 (804) 332-0207 (mobile)