Banner image for Uneven Outcomes in the Labor Market: Understanding Trends and Identifying Solutions - February 1-4, 2021

Day 1: State of Labor Market Outcomes

The opening session of the Uneven Outcomes in the Labor Market: Understanding Trends and Identifying Solutions conference features remarks from Reserve Bank presidents Raphael Bostic of Atlanta and Eric Rosengren of Boston. In this session, panelists discuss new research on long-term labor market trends and examine the impact of the COVID-19 pandemic on the labor market.


Stuart Andreason: Hi. I'm Stuart Andreason from the Federal Reserve Bank of Atlanta. I want to welcome you all to day one of the Uneven Outcomes in the Labor Market conference. The conference is organized by community development staff at the Federal Reserve Board and the Federal Reserve Banks of Atlanta, Boston, Cleveland, and Philadelphia. We welcome you today. I want to go over a few, quick housekeeping items before we launch into the conversation. To manage conversation and questions throughout the conference, we are going to be using a tool called Slido. You can access this by going to Slido, using the hashtag Uneven outcomes, or you can simply press the join the conversation on the Slido button below. Please submit questions and comments. We will be responding to questions throughout. We have staff that are going to be answering your questions, both logistical and topically. We will be using it to queue up conversation with our presenters. And we will be using it to follow up; if we can't get to your question, we will do everything we can to respond to it. Now I encourage you to open Slido now because I'm going to introduce a poll. We want to learn a little bit about you.

So, you will see our first poll if you go to Slido now. Over the next four days, we really hope to convene researchers, policymakers, and practitioners to examine disparities in the labor market and explore policy solutions to address these disparities. We aim to deepen understandings of disparities in employment, labor force participation, income and wealth, to learn about their implications for economic growth, the health of communities, and individual well-being. We want to learn a little bit about you. What type of an organization do you represent today? I encourage you now, and I'm going to pause for just a second, as that poll populates, to learn a little bit about what is happening with you all. And who you are. Very interesting.

I hope that you can all see the results changing live. We've got a real mix of people, which is exactly what we were looking for. We were looking for a mix of people in academia and think tanks and policy organizations, workforce practitioners, Federal Reserve staff, and others. We have financial institutions, employers and some foundations as well. Which is really what we were looking for. We wanted to come up with a mix of people.

Now I encourage you to look at the conference website, to see the overview of the next four days. We are really excited about it. And I want to introduce a little bit about what's happening today. Today's conversation is going to be on the labor market. We will discuss long-term trends as well as recent changes brought on by the pandemic. We are really excited that today, Bill Rodgers from the Heldrich Center, the chief economist at the Heldrich Center at the University of Rutgers and a professor of public policy, will be joining us. Bill is also a visiting scholar for 2021 at the Federal Reserve Bank of Atlanta, where he will be exploring many of these same challenges. He's got a long career in exploring challenges and disparities in the labor market, and we are excited to have him lead our discussion today. He's going to start with some opening comments. Now, it's my pleasure to turn it over to Bill to introduce our day. Welcome, again, to the conference. Thanks so much.

Bill Rodgers: Stu, thank you so much, and again, thank you for inviting me to spend at least remotely, the year with you and your colleagues at the Atlanta Federal Reserve Bank. As Stuart said, before I introduce our researchers, I wanted to share some thoughts on kind of what motivates me for my research and the policy work that I've been doing and also give you my perspective on these issues around uneven labor markets, uneven economic outcomes. And to start, I think we want to go back in time, just over a year ago, prior to the pandemic. It seemed like a lifetime for many of us. But there were three big labor market issues of concern. One is what many of you have heard of is called the clamshell chart. That is basically the historical divergence between productivity and wages starting in the late 1970s. A second key factoid that many of us labor economists have been focused on has been the labor share. That is where we talk about the fraction of output that is accounted for by employee compensation. Since 2000, that share has been trending down. And then the third has been the overall expansion in income inequality, and also expansion of persistent inequality by race, ethnicity, and gender. Now, the identification of the various contributions focuses on supply, demand, institutional features of the economy, a lot of these explanations have focused on globalization, particularly immigration, trade.

Other explanations have been on technology, information technology in the 1990s, and now robotics and AI. And then the third broad category is public policies such as minimum wage, the erosion in worker voice, and agency, criminal justice policy and there's a litany of other policies. But the bottom line is, what's happened is that many of the policy choices that we have made or have been made over the last few years, last few decades, have shifted the risk, have shifted the risk of illness, job loss, or natural disasters from society, back on to individuals, their families and their neighborhoods. And in many cases, there's been a growing inability to handle these new risks that families are facing. These are important, especially the latter, because even though we are experiencing or we were experiencing the best economy since World War II just prior to the pandemic, economic security for many Americans was not achievable. For example, 40 percent of adults didn't have enough savings to cover a $400 emergency bill or expense. Forty-three percent of households can't afford the basics to live. They aren't earning enough to cover the combined cost of housing, food, childcare, healthcare, transportation and even a cell phone. More than 25 percent of adults skipped necessary medical care last year because they couldn't afford it. Again, this is prior to the pandemic. One-fifth of adults aren't able to pay all of their bills every month. Thirty-eight of nonretired Americans think their retirement savings is on track. Two-thirds of Black Americans, two-thirds of Latinos, say they are doing OK financially versus 77 percent of whites.

The problem in my mind is that we have deferred maintenance on the nation's human priorities, and that's what the United Nations likes to call investments in human and social capital. Now, the pandemic and murder of George Floyd exposed the consequences of not investing in the nation's new priorities, exposes the uneven outcomes and access to quality health care, wealth, income, employment, and earnings for all low- to moderate-income families. It also exposed systemic racism and sexism. Today, we will hear new research on both the uneven, long-term trends, and the uneven impacts that the pandemic has had on workers and their families. We will also, and I'm really excited to be on the fireside chat with two the presidents of Federal Reserve banks, look forward to that. These presentations and conversations will not only document the unevenness but also advance solutions for improving the economic security and all Americans, but especially for those that have not historically had safe workplaces, fair workplaces, which can then lead to secure retirement. So with that, I want to introduce our first panel, and their full bios are available on the conference website. There is a button on your screen that will take you there.

Our first researcher, Laura Choi, she is vice president of community development at the Federal Reserve Bank of San Francisco. She will be presenting coauthored research on the economic gains from equity. The presentation represents Laura's views and not those of the Federal Reserve. Niki vonLockette, associate professor of public policy of African American studies at Penn State University will be presenting research titled, The Impact of Systemic Inequality on Income. Brad Hershbein, senior economist of the Upjohn Institute; his paper on the pandemic's evolving impacts on labor market. The discussant for these pieces will be Rebecca Dixon, and she is the executive director at the National Employment Law Project. She will discuss how this research can be used in policy and practice. So enjoy the presentations, and now we will transition to the various videos.

Laura Choi: On behalf of my coauthors Shelby Buckman, Mary Daly and Lily Seitelman, I would like to thank the conference organizers for hosting the event and the chance to present our paper. We document persistent gaps in labor market opportunities and outcomes by gender and race and ask how much these inequities have hindered U.S. economic output. This paper is built on a growing body of research that shows that closing gaps in areas such as employment, education, earnings, and occupation can lead to significant gains in GDP. We contribute to the literature by estimating the aggregate economic losses associated with these gaps and decomposing the aggregate to understand which disparities generate the largest costs. Our analysis focuses on differences by race and gender and employment, hours worked, educational attainment, educational utilization, and occupational distribution. We also examine gaps in the returns on these variables and find combined economic losses measured in the trillions of dollars. We begin the paper by plotting trends and labor opportunities and outcomes for adults age 25 to 64 to observe differences by race and gender.

Given time constraints, I will only share the charts by race but will comment on gender trends as well. On slide two, the left panel shows trends in the employment to population ratio, and you can see that whites have higher employment rates than Blacks and Hispanics. There's also a persistent gap by gender, with men having EPOP rates on average of 14 percentage points higher than women. There is some evidence that the gaps narrow during economic expansions, but it's not sufficient to close them. The right panel shows the share of individuals with a BA or higher. And a larger share of whites has a BA compared to Blacks or Hispanics, reflecting underlying structural barriers and gaps in opportunity for people of color.

The trends by gender are different. As men and women were similar to the '90s, but the share of college-educated women surpassed that of men in the early 2000s. But even when people have a BA, they may end up in jobs that do not fully utilize their education, and this resulted in underutilization for the entire economy. On slide three, the chart on the left shows educational utilization, measured as the share of BA holders who are in jobs that require that degree. As you can see, utilization rates are higher for whites, and the gaps have grown over time. The pattern is different by gender as women are consistently utilized at a higher rate than similarly educated men, and that gap has narrowed over time.

The final trend we highlight is opportunity across industry and occupations. And as an example, the graph on the right shows the share of individuals who have a professional occupation in the durable goods manufacturing sector, as these are among the highest paid workers in that industry. Whites are consistently much more likely to hold these jobs than Blacks and Hispanics, and this is true to an even larger degree for men in comparison to women.

With an understanding of these historical gaps, we turn to estimating their aggregate cost. On slide four, we use a basic growth accounting framework and focus on the labor contribution to aggregate output. This includes labor quantity, which is a function of the number of workers and average hours worked, and labor productivity, which is a function of education, utilization, and industry occupation distribution. To conduct our experiment, we first create mutually exclusive groups by age, gender, and race. And then compute their input values using current populations of survey data for the variables in a framework. So for example, we could examine average hours for black women age 25 to 34. We then imagine a counterfactual in which labor market gaps do not exist and use a shift-share technique to estimate the impact of eliminating these gaps. We replace the employment rates and hours worked of women and people of color with those of white men, who are the default group in our analysis, because they have historically faced fewer systemic barriers in the labor market. We also close gaps in the shares of educational attainment, utilization, and industry occupation distribution. And to examine the returns to opportunity, we measure labor productivity using average hourly earnings. This comes from the theory that an individual's marginal product is equal to their wages. And it's not a perfect measure, but we can account for differences in wages due to a given group's education, utilization, and occupation distribution, referred to here as shares.

Finally, we estimate the impact on aggregate output. So let's turn to the results on slide five. Our first experiment asks how much more output the U.S. economy would've had if historical gaps in employment and hours had been closed. The top row of the left panel shows the actual values for EPOB and weekly hours. Below that are the counterfactual adjustments by shifting race only, gender only, and gender and race together. In 2019, closing gaps by both race and gender would have increased EPOP by eight percentage points and added 2.4 hours to the average work week. The right panel summarizes the dollar value of closing these gaps, and all values are presented in 2019 dollars. Equalizing employment by race and gender would've boosted GDP in 2019 by 0.63 trillion dollars, and equalizing hours would've added 0.37 trillion dollars.

We next examine closing gaps and labor productivity, which we measure with average hourly earnings or AHE as shown in slide six. The left panel shows the impact to AHE of changing education shares alone as well as changing shares of education, utilization, and industry occupation distribution, all at the same time. It is important to note that in this exercise, we only change the shares of opportunities but hold the returns constant at the prevailing rate for each group. So for example, we would close the gap between the share of Black and white men that have a college degree, but that share of Black men would still maintain the AHE for college-educated Black men. Adjusting the determinants of productivity changes aggregate AHE only slightly. For example, closing gaps across all shares race and gender in 2019 increases average hourly earnings by 62 cents. This is due in part to the fact that people of color historically made up a smaller share of the population than whites, and that gender gaps in education shrank over time to become nonexistent. The right panel shows the impact of these shift shares on GDP. In 2019, removing all gaps in labor productivity shares by race and gender would've increased GDP by 0.17 trillion dollars. We also conduct an additional counterfactual exercise to acknowledge that the returns to opportunity can vary by race and gender in ways that aren't fully explained by differences in skills, utilization, or other measurable variables. And it's likely that discrimination and bias play a key role as unobservable barriers.

For that exercise, we keep the determinants of labor productivity fixed at their actual values and only adjust the returns by race and gender. So for example, we would keep the share of college-educated Black women age 25 to 34 at their actual value, but we would assign this group the average hourly earnings of their white male counterparts. And the impact to GDP is significant; eliminating the earnings gap would have increased GDP nearly six times more than the impact of our previous exercise, where we adjusted labor productivity shares but kept AHE at their actual values. So putting all of this together, slide seven summarizes our final experiment, which closes all gaps in labor inputs, determinants of productivity and the returns. We acknowledge that this is not a general equilibrium model, but the bottom line of our analysis shows that the U.S. economy would've had about $2.6 trillion more output in 2019 if gaps in labor market opportunities and returns were eliminated. And our results show that the gains have been rising over time as the U.S. population becomes more racially diverse. So when we add it up over 30 years, we estimate that equalizing shares of labor market opportunities would've added $34.4 trillion to GDP. And if we equalize both opportunities and returns, we estimate aggregate gains of $70.8 trillion.

So what does all of this mean? In this paper, we focus on closing the gaps, but it's important to acknowledge that these gaps are rooted in systemic inequities. Explicitly racist policies like Jim Crow laws and redlining have had long-lasting effects on residential segregation and wealth accumulation for communities of color, especially Black communities. And in addition, the existence of labor market discrimination at the individual level is also well documented. And together, these have implications for policies that address both structural racism and workplace discrimination. There are also important structural impediments to full economic participation for women, and we must discover the distinct labor market barriers that women of color face. Our decomposition reveals that the gaps cannot be fully explained by observable measures of talent or skill, indicating that gender and race are still meaningful predictors of labor market outcomes. And we are seeing firsthand the implications of these inequities in the context of COVID-19, which is causing disproportionate harm to the labor market outcomes of women and people of color. These issues are becoming only more important as demographic shifts change the racial composition of the labor force. The findings should compel us to move with urgency to eliminate inequities, both because it's the right thing to do and because it will be critical to maintaining global competitiveness. In short, achieving gender and racial equity is crucial to our shared economic future.

Thank you. And with that, I would like to turn it over to Niki vonLockette from Penn State.

Niki vonLockette: Thank you. Today, I want to propose a different way to think about racial economic inequality. I've always been baffled by while we expect racial inequality will decline over time, there continues to be mounting evidence to the contrary. The wage gap between whites and Blacks has actually been increasing since 1979, even though disparities in educational attainment and test scores have been actually decreasing, those things we can account for it. Wealth disparities between Blacks and whites have quadrupled since the '90s; it's just not going in the right direction. So instead, I want to turn our attention to systemic racial inequality. Things like residential segregation, which has not dropped significantly since the civil rights era. In some cities, it's not changed at all; it's truly Black-white segregation. School segregation. Our public schools have re-segregated to the levels that existed at the height of the civil rights movement. There currently stands a $23 billion, billion with a B, funding disparity between white and minority schools.

I want to propose a different way to measure and think about the role of systemic inequality and sustaining inequality. I suggest a systems approach to assessing the impact of systemic racial inequality. This idea suggests that racial inequality is in fact a system or a set of dynamically related domains of subsystems like segregation, school segregation and that the disparities across these domains are usually reinforcing, they are interdependent, depend on one another to function, so that we cannot know the effects of one of these domains—say education—without in isolation, by taking it out of its relationship and understanding its interdependence on the other factors. This suggests that we look at these things as a set or a system, rather than individual. So I have built a metropolitan structural racial inequality data set. I have been passionate about understanding this for years. So, I have built for over 200 cities in the U.S., I have constructed a longitudinal data set of the structural measures of racial inequality, measured from 1970 to 2017. This gives us a way to assess the cumulative effects of inequality over time, and it also emphasizes by doing this at the metropolitan level, cities as a site of the construction of inequality; there's a lot to suggest that this happens at the local level.

So this data set includes a number of measures; I don't have time to go into all of them, but residential segregation, school segregation, net worth, arrest rates, policing, political ideology, racial attitudes, all at the metropolitan level over time. So I use a different approach for, and I want to recommend a different approach and I use it here to actually analyze systemic racial inequality. I use structural equation modeling. This is what I want to propose and without too much technical detail, I simply want to talk about what the advantages, what using this different modeling will give us in terms of better understanding more comprehensively how systemic racial inequality works. It allows us to simultaneously model multiple outcomes at the same time rather than just one. We can also estimate the complex relationships among the independent variables among each of these structural factors, on our outcome of interest. And we can estimate both direct and indirect effects on inequality over time. So first, I start here with this, an actual conventional analysis. This is just to give us a baseline. This is the fixed-effects analysis, and we are more familiar with the work of the Fed and here we are looking specifically at—this is the first take of looking at how changes in our structural measures that we have affect Black-white income inequality over time.

So as these factors change, how does that affect the income inequality over time? I will just pull out a few, highlight a few. So for example, as the education ratio, the education disparity between whites and Blacks increases over time, so does the income ratio over time, the income, inequality between whites and Blacks, favoring whites, as school segregation increases over time. This also has an effect on the income inequality that I mentioned. So these, and there are several of these relationships, this shows us there is a cumulative effect over time of structural inequality in cities. OK, so let's a look at structural equation model. So for the first part of SEM is an exploratory analysis that allows us to identify the associations among the structural measures. I've been interested in how—so we know there are a lot of the sort of domains and substances, but how they work together? Which ones are more likely to cluster together? This first analysis, I found there were two factors, two clusters that emerged. The first is residential segregation, school segregation, occupational segregation, and educational disparities and income disparities; those all hung together in a way that made all kinds of sense to me. The second cluster or factor that emerged is one that I call a policy one. It included political ideology, state political ideology, racial attitudes, minimum wage for the city or state and levels of ..., so those all sort of came together as a sort of policy factor. I use these clusters to build the main structural model, which I turn to now.

So here you see the structural equation model where we have a number of paths from each of the structural variables, factors among themselves to each other, and to the outcome of interest, income disparities between Blacks and whites. So this is the type of modeling that I suggest. This is the way in which I think we should model structural and systemic inequality. It's a bit overwhelming and complex, but it can really sort of help us understand how these [factors] work together to create inequality. I will point out a few highlights here. One, as expected, segregation affects net worth or wealth. School segregation affects educational inequality, which then affects income ratio, so does joblessness and so do policing rates with respect to income inequality. So putting all of these together, looking at how as a group, they work together to maintain inequality, to me is the next step, and really to comprehensively understand how systemic racial inequality persists.

These findings offer several implications for policy in my mind. One, we clearly need to shift our focus from individual level programs such as training which are important but to begin to really incorporate system-level inequality into our analysis and into our policy. Change or understanding perspective of the problem, rather than expecting, continuing to expect a natural decline, to seek to understand a system that was put in place a long time ago that was designed to actually maintain inequality and maintain the stability and do so effectively. So when we have that perspective, we understand we need to design interventions that themselves that are disruptive, that are designed to dismantle system-level inequality. I also suggest that these policy interventions that are at the local level may be more effective, as we know that the mechanisms that create inequality operate in the local economy. That gives a more important role for community development policy and regional policy. I can see that diagram I did before, I can see that diagram shifting and looking different in different regions of the U.S. I also want to point to the sort of system of what I call silos of policymaking that are built into our federal and state agencies that are designed to address solitary problems like housing and employment, may be less effective at addressing the complex problem that has these multiple domains. So here, a phrase many of you have heard, really the importance of interagency ordination. But I want to suggest even further that we actually create a new agency that's designed to actually help with interagency coordination and collaboration. You know, the racial justice agency ... and Department of Justice that is really aimed at collaborating efforts across the federal agencies, state agencies, and local, and even across public and private sectors. I thank you for your time today. I will turn it over to my copanelist, Brad Hershbein at the Upjohn Institute. Thank you.

Brad Hershbein: Thank you. I'm pleased to present new research, joint with my coauthor, Harry Holzer of Georgetown University, on how labor impacts of COVID-19 have evolved in recent months for different groups and what policy responses we should consider. First, COVID-19 has clearly decimated the labor market. Economically disadvantaged groups, the young, the less educated, racial minorities, workers in leisure and accommodation, women, have been especially hammered, even relative to past recessions. Moreover, as of late last year, and still true today, cases and deaths have been rising and the employment recovery has stalled. To illustrate, this graph shows the prime age employment rate for 25 to 54-year-olds. As of last December, 76.3 percent of this age group were employed. While this marks a jump from the trough in April, it is still down more than four percentage points from February of 2020 and is about the level reached in early 2014. Moreover, the recovery is flattening.

That picture, however, masks substantial variation across places and groups. Neither the decline nor the recovery have been even, and not all employment measures give the same picture. Moreover, we find that the COVID mortality rates matter more for employment recovery than state economic restrictions as mortality matters in future months while the impact of restrictions is only temporary. This lingering effect of mortality is a prime reason why the recovery has likely slowed. And despite new financial relief and beginnings of vaccination, a full recovery may take a while. We believe such recovery could be accelerated through more aid for state and local governments, many of which have already announced budget cuts; national guidance on vaccine distribution; resolving debt overhang for renters and businesses; and enhanced sector-based training for permanent job losers and temporary essential workers.

To show how the different measures of employment matter, this graph plots three national measures for people age 18 to 64, all drawn from the Current Population Survey, the source for monthly employment statistics like the headline employment rate. The blue line shows the employment rate, the share of people with jobs, but excludes individuals who are classified as away from work for the week for non-specified reasons, something other than say, vacation, temporary illness, childcare, etc. The Bureau of Labor Statistics believes many of the people on the latter group, many of whom are actually self-employed, should properly have been classified as unemployed. The redline further excludes those who are working part-time involuntarily. And this red line is our preferred measure going forward as it captures both overall employment and reduced hours. The green line shows the aggregate number of weekly hours worked, which further captures changes in hours, but it's also a somewhat less intuitive measure. However, all the measures show similar patterns of huge drops in the spring and a slowing recovery since. Although it's worth noting that the gap between the blue and red lines has widened during the pandemic, illustrating how involuntary part-time employment matters.

The graph on the right here shows another important measure, unemployed people with permanent job loss. The red line here shows this indicator relative to all unemployed as is typically reported and thus dips in April, when a huge influx of temporary layoffs occur. We thus prefer the blue line, which shows permanent job losers among all people, regardless of employment, and this measure has increased steadily over time. As of October of last year, it reached about 1.6 percent of the population. As I mentioned earlier, however, these employment impacts vary tremendously across groups. These sets of graphs show the adjusted employment rates, excluding the involuntary part-time, for four occupation-based wage quartiles on the left, and for Whites, Blacks, and Hispanics on the right. Low-wage workers have clearly fared the worst, and if anything, their recovery has been the most aborted in the more recent months not shown on the graph, as leisure and accommodation is what drove the job losses in the December jobs report. Similarly, racial minorities have been hard-hit, although it's worth noting that Hispanics have recovered faster than Blacks. Yet, it's important to understand how much of these racial gaps are due to the types of jobs workers have. The left graph here repeats the previous graph, except each racial group has been normalized to its own January 2020 level. So they show percentage changes. The graph on the right statistically controls for individuals' occupational-wage quartile and level of education in each month. These controls dramatically shrink the racial gaps and implement changes, especially for Blacks and in the spring. They matter less by the fall. This suggests that the slower recovery for Blacks and Hispanics is not due only to the types of jobs they are more likely to hold.

Additionally, employment impacts have varied across states. These graphs show employment rates and permanent job loser shares for three groups of states based on the timing of the first peak in COVID caseloads. Spring, summer, or fall. The fall peak states have had less employment disruption than the other two state groups. And in the paper, we show that this is not due to the timing of state economic restrictions, but is rather accumulating impact with mortality, which seems to make people more reluctant to move about or engage in normal spending behavior, and other researchers have found this using cellphone mobility data. Despite some prognostications for a rapid employment recovery in 2021, we are not so sure based on the direction we see in the data. In fact, our estimates imply that the average cumulative mortality rate growth, just since October, predicts a 1.5 percentage-point drop in employment rates. The relief bill passed last December should help, but it doesn't address many of the looming bankruptcies, evictions, food insecurity, and other social problems. And slow vaccine rollout to date is likely to make them worse by extending the time before things can get back to a new normal. Thus, we believe more targeted aid is needed, especially if we want to preserve any of the recent gains made by economically disadvantaged groups. We detail in the paper several of the most promising research-backed approaches, such as sectoral job training, wage supplements like the earned income tax credit, specific employer hiring credits, and others. Thank you very much, and I believe Rebecca Dixon is next.

Rebecca Dixon: Creating an equitable society, one with a large stable middle class where families are thriving regardless of race and other constructs and characteristics is not the natural outcome of market forces. Equity is rather created by society and by the institutions, laws, policies, and practices that govern society, its economy and especially its labor market. Creating a just and inclusive economy means designing institutions that support the creation of quality jobs, and decent wages, working conditions, as well as enacting policies to support those who are out of work.

A clear example of how our policies and practices support or undermine equity is redlining. There is a pretty clear understanding among policymakers and the that public redlining is an instance where the government had a choice to make that could either increase equity or undermine it. The choice made there increased inequity, and its impact is still being felt today with the direct connection to the racial wealth gap, since homeownership is a big asset builder in this country. Labor policies are still submerged for policymakers and the public. For example, occupational segregation, which confines women, Black, Latinx, immigrant and other workers of color to the lowest-paying, dirtiest, and most dangerous jobs is not a part of our discourse on segregation.

Employment discrimination is often overlooked as a contributor to the racial wealth gap, even though employment discrimination has been legal in this country for far longer than it has been illegal. Laura and her coauthors highlight the opportunity to participate in the economy and succeed based on our ability and effort, and this is at the foundation of our nation and economy. A recent Pew Research Center survey on American values finds that the majority of Americans, 60 percent, say that most people who want to get ahead can make it if they are willing to work hard, while 39 percent say hard work and determination are no guarantee of success for most people. For self-identified Republicans, 78 percent believe that you can get ahead if you work hard. But as the presenters have noted, structural barriers and embedded inequities in policies and practices have persistently disrupted this narrative for many Americans. Arguably, this mismatch between our beliefs and reality is detrimental, not only to economic growth but to the stability of our democracy as we have seen insurrectionists trying to find a narrative that explains their own challenges.

As our population grows more diverse, the need to disrupt these inequities will only become more urgent. The diversity is rapidly increasing in the younger portion of our population. In fact, in 2019, for the first time, more than half of the nation's population under the age of 16 identified as a racial or ethnic minority. Among that group, Latino and Black residents comprise nearly 40 percent of the population. That is compared to 1980, when 80 percent of the population was white. Niki makes it clear to us that the passage of time alone will not lead to a reduction in inequity. She noted that the wage gaps have increased since 1979 and the racial wealth gap has quadrupled since the '90s. We have seen worsening outcomes on average for everyone who's not in the top 5 to 10 percent. An analysis of Census household income data shows that the share of adults who lived in middle-class has eroded over time from 61 percent in 1970, down to 51 percent in 2013. And Brad explains to us that due to COVID-19, we have a potential tsunami of debt, eviction, and food insecurity on the horizon, with the weight of that being uneven and borne primarily by women and people of color. This will worsen the inequity in our society, pulling us farther away from our ideas of equitable opportunity and self-made outcomes. By some estimates, more than 30 million people in the United States could be in danger of losing their homes, with nearly 12 million entering the new year behind by almost $6000 on rent and utilities, and millions of others falling behind on their mortgages.

So what are the applications for policy and practice? I have two points to make on this. The first point, the American dream is calibrated to individuals, but our policy practice must reach below the surface of the individual and address systems that maintain these inequities and allow them to worsen. I believe opportunities for individuals to access education and training are vital, but we also have to have an option where we can muster the political will to use our labor market institutions and levers of government to improve wages and working conditions and eliminate baked-in wage penalties for women and people of color. Some examples of systemic policies are raising the minimum wage, restoring the ability and right for workers to join together collectively to have a say in their working conditions, addressing the unfairness of a two-tiered system that supports lower pay for equal work in the case of temporary jobs. We can also institute fair scheduling and stop the erosion of labor standards, tied to worker status, such as for gig workers and independent contractors. We could take seriously the need to tackle employment discrimination, continuing the work of enshrining equal opportunity as the law of the land. The second point: We have evidence, and we have seen the results of race- blind and race-neutral policymaking. Opportunity gaps and income gaps have widened, and those at the margins are often excluded from the policies outright, or those policies don't consider their unique needs and so they are defacto excluded. I think it's time we consider how to craft policies that are inclusive and intentional about targeting inequity with clear goals and measurable progress. As Niki mentioned, this means looking at the levers of change, including state and local policy and practice. We may be moving in the wrong direction there as well, with state and local solutions being preempted and overruled in the current environment.

I want to leave you with some common-sense ways to center equity and policy design. First, representation matters. The people who are making the decisions about policies should look like the country and include folks from directly impacted groups who can bring their passion, wisdom, and insights. Second, we must consider historical and present-day inequities. What does the data tell us about who is most advantaged and disadvantaged by specific policy recommendation? Likewise, we should consider the intended and unintended impacts and lastly, power and participation. By engaging workers in their communities, we can craft more effective policies, while we honor another American value, that of self-determination.

Now I would like to invite the panelists to join me for our Q and A. And as they are coming on, I'm checking out our questions here. So our first question ... it is for Laura. The model that you use seems highly expensive and it seems that Blacks and women will add to, let me switch to another question momentarily. Given the importance of occupational segregation and the disparate impacts we've seen in last year, which policies do you all as a panel feel are best suited to address this problem?

Choi: This is for all of us, Rebecca, just want to confirm?

Dixon: Yes, yes for all of you; sorry about that first one.

Choi: No problem at all. I can jump in. First of all, thank you, thank you for the question and just this chance to reflect on policy. You know, I think any question around what's one policy, I want to kind of reframe and put us back, I think, to Rebecca, your comments, and I think what Niki raised in her point as well, which I think all of the research and all of these discussions are pointing to how interconnected and systemic these policies are. When we think about one policy solution, really keeping that in mind of its interaction with all the different policies we talk at, so when we think about occupational segregation and specific workforce policies, I push us back even further to thinking about what underlying that, what's the trajectory to get people into the occupations they are in.

So I think we have to think expansively about residential segregation, wealth accumulation, education opportunity, the entire pipeline of what it takes to get people ready for the workforce. And I would say even broader than that—I hope you don't mind I'm going kind of big here-but thinking about policy across the life course. So, you know, we jump right away to higher education policy, but I think we have to look all the way to healthy babies, early childhood education, thinking about what it takes to remove inequities across the policy spectrum, across the life course, to really address these downstream effects, when we are seeing occupational segregation and adulthood, recognizing that there is a whole host of root causes that go back for generations. So when we think about policy, we need to take this very big systemic look. I will leave it at that kind of big macro-level, but that's my response to the question.

Harry Holzer: Let me jump in. All of Laura's solutions sound great; they are very, very big and would take many decades to do. So what Brad and I sort of look at is the opposite, things we can do in the next year to both recover employment, and then to give people the skills to sort of move up in the labor market. And giving people better skills would not eliminate occupational segregation, but of course it would help for them to be trained. So on the demand side of the labor market, Brad and I embrace things like creating infrastructure, which will create a demand for higher skilled construction work and that would go hand-in-hand with supply policies, to train people including more people of color, more women, etc., for those jobs. Then we focus on the issue of education and training more specifically. And there are a lot of different ways you could subsidize what we now call the WIOA system, the Workforce Innovation and Opportunity Act, pump more money into those streams. You could give more money to community colleges, targeted to workforce development in those areas as well as four-year colleges as well. So while we support everything Laura says, we think there are practical things you could do just in the next year, not only to speed the recovery but to enable people to have the education and training to get into better jobs. At the same time, we create more jobs through some of the measures we outline.

Hershbein: If I could just jump in, sorry, to add to Harry; I will be brief. But we can do both. And I think we need to do both. And when there is a time of crisis, it is a terrible tragedy in many ways, but it's also an opportunity if we take advantage of it, of beginning to dismantle some of the long existing barriers that many people in our country face in terms of economic opportunity, and that's a combination of making these long-term investments, and shorter-term investments that can help people right away and have payoffs within a matter of weeks or months as well as payoffs that are going to affect children now, even babies now, decades later. We need to do both.

Dixon: Niki, did you want to address this one?

vonLockette: Thank you, I appreciate it. Really quickly, I was reminded Bloomberg Foundation just made a $100 million pledge to the four Black medical schools at HBCUs. So the percent of black doctors today is about five percent of all doctors. In 1970, it was three percent. Unless we have some major kind of, you know, input, we are not going to change that. That addresses both skills and the structural piece around, what we think around some of those disparities in higher education are really about assets and the ability to send kids to college and to continue to pay those—medical school has one of the largest—but this idea, and I like Laura's point, this idea around, across the spectrum. So I will sort of, in advance of one of the questions I heard was around skills. I think skills is about school segregation; we have people going to different schools, we're going to have different outcomes in terms of quality.

If we could, addressing school segregation is one of those structural things that's really important, but Laura's point around the pipeline, right? So that's going to address that. But that is something that we can do. I've put in the presentation, $23 billion disparity in school funding. That is something that we can do, it's something we would see changes sort of, you know, sort of immediately with younger children. But we would see such changes not only down the line in terms of inclusion and access, but imagine if we now began to provide quality education to whole segments of populations that we have categorically excluded, imagine what type of innovative capacity is just waiting for our country down the road. Anyway, I wanted to mention Bloomberg. And that was a check I assumed got cut pretty recently. So I mean, there is some immediate kind of things that can be done. And I think from the policy perspective, both approaches seem really wise. Thank you.

Dixon: Thank you, and when I saw that number, the disparity number, I did a double take. I was like, billion? Yes, billion with a B. Our next question, and to any of you who would like to take it—what is the one policy that we can have that can have the greatest impact addressing structural racism? I know you all have said one policy is not enough, so please just chime in with your best answer.

Holzer: Probably desegregating the K-12 school system, since that leads to so many long-lasting inequities down the road. Again, it's not a quick solution, but it would have very major implications for future employment outcomes.

Hershbein: I would agree. And one thing I would like to point out is that I happen to live in Kalamazoo, Michigan, the home of the Kalamazoo Promise. The Kalamazoo school district is incredibly racially, ethically and socioeconomically diverse, but it had been leaning toward getting more and more segregated due to white flight. And a free college program, the Kalamazoo Promise, stopped that. It arrested the change. So now it is roughly racially balanced, and it is getting more desegregated. And so there is the potential for higher education programs to actually change that by making it valuable to stay within a school district and fight for more resources together. It's unfortunate that has to happen this way in some cases, but it has led to a lot of constructive change. So there is this integration between higher education and K-12. That opportunity can affect K-12 schooling as well.

Dixon: Did either of the other panelists want to address this one as well?

Choi: I'll jump in really quick. And rather than specify one specific policy again, I think maybe one idea to bring to our policymaking that I would like to put out, and I'm borrowing from legal scholar John Powell, which is this idea of targeted universalism. So when we think about our policies, finding a way to ensure that they are, although universally applied, are targeted to, you know, creating the most benefit for those who have been disproportionally harmed in the past. Things like this school desegregation, things around, you know, wage increases that would impact, minimum wage or others that would disproportionately impact those who have been most disadvantaged, so rather than any one policy, I would like to propose a carrying that spirit forth when we make policymaking.

vonLockette: I want to add really quickly, that your first question was sort of like which structural approach, residential segregation because residential segregation undermines school segregation. It's been a real pickle, putting it lightly. It's been a real pickle from a policy level to address in this issue of sort of come even if we sort of, you know, investment in certain areas, this issue of white flight. So, you know, ways to sort of incentivize it. But some of the ideas around investments in communities, you know, particularly Black and Latino communities that were really heavy hit as a result of the Great Recession and the kind of, the devaluation of homes. And ways in which, you know, institutions like the Fed, and whatnot.

And other types of both financial institutions, intermediaries can play a role and again sort of investment in communities and investment in neighborhoods. And looking at, we have already seen, there's already evidence as well around looking at the reintroduction of toxic loans and toxic financial products to targeting Black and Latino homes after people got that sort of slap on the wrist, that they are back at it. There is evidence of that already. So anything we can do them in terms of regulation to both, to minimize and negate that kind of, those kinds of dynamics, but also to increase and incentivize investment, that again segregation seems to be, from my work with systemic, it seems to get this sort of heart of so many of the other pieces. So if I were to start with only one, it would be residential segregation. I think you would get so close at nipping at the others as well, because they are so interconnected.

Dixon: And so you all have given us lots of food for thought, in terms of solutions and policy options. So one question for you is we know what to do, but how do we start to build the political and public will to make some of these policy changes and to sort of make these long-term commitments to actually monitor whether these things are improving?

Holzer: I can take a quick crack. On the policies that I mentioned originally, around workforce development and higher education, there is both a moral case and a practical economic case for doing that. You know, the moral case is a justice case that I think lots of people have alluded to. There is also a practical case. As baby boomers retire, it becomes harder and harder for employers to find the skilled employees they need. Of course, they could raise wages more; that would help attract more workers, but a really well-functioning workforce system is really in the economic interest of the country. We can't afford this massive waste of talent that occurs right now, and we have a system that simply doesn't allow so many people, people of color, low-income people, to reach their full, productive potential. I think the case has to be that combination of moral and pragmatic economic.

Choi: I will jump in here. Because I agree, I think that was really the genesis of the research paper that my coauthors and I put together, was really wanting to make the economic case and demonstrate the economic gains from equity. I agree. You know, we spent a lot of time researching the size of the gaps, how big they are, but we wanted to flip the script with this research and show how much do we have to stand to gain from actually closing these gaps and making that really economic case that addressing gender and racial inequities is something that stands to benefit our entire country, our entire economy.

vonLockette: I'm really looking for, like I like the example that Brad mentioned. I'm in the process of sort of jotting down notes, I'm in the process of finding where these little sort of programs or initiatives, demonstration projects, are working, where people are sort of trying some daring experimental things and they are actually working. I'm excited about sort of collecting those, and then thinking about, you know, we are looking at sort of either big institutions like government or foundations, to then begin to sort of unfold those in different, you know, in other places. So one of the things I thought was so interesting is what happened in Georgia, where there's the sort of, now this sort of idea, what's most interesting to me that didn't get as much play is that some folks are using that model and applying it and others are trying to sort of think about applying it in other states that have similar kinds of characteristics, in terms of, you know, the demographics, etc. So this idea of finding something that works, and not reinventing the wheel, but there's so many bold, brave things out there, and really sort of getting on the ground and finding those things. The community development aspect of what the Fed is doing here, I think that sounds like a really great place to start, doing that sort of fact-finding.

Dixon: Brad, did you want to weigh in on this one?

Hershbein: Yeah, thanks. I will agree with what everyone else has said. To me, I think one of the most elusive but perhaps most rewarding targets is getting better alignment in a lot of communities, and there are so many different groups all trying to do relatively similar ends, and trying to get them to gain and join forces and figure out how to most effectively make their message heard, and so that they don't re-duplicate, so they are not repeating what's been done and they can actually use their forces most effectively. That is something where training future generation of leaders, whether those be community activists, whether they be politicians, whether they be business leaders. We have to have more programs to train people so that they can help bring other people together to bring the cohesion that is needed for a lot of this change to happen.

Dixon: Thank you all. Another question has come in, very timely; what policies should the Biden Administration enact to respond to racial inequity gaps that were exacerbated by the coronavirus and recession? That is open to any of you.

Holzer: Well, I think, again, if we focus on the short-term as opposed to these long-term structural problems, this combination of, in the recovery and the rebound create more—not just more jobs but more good jobs through things like infrastructure spending, and then to really, really make our education and training our workforce system more inclusive, more equitable. Actually following up on what Niki said, there are all of these training programs, sector-specific training programs with very, very good track records and evaluation; we really need to have to think about how to replicate and really scale those and make sure they are inclusive, make sure they are accessible for the least, people with the least opportunity. And I think that combination could do the best to address that exacerbation we've seen in this past year.

Hershbein: I will agree with my coauthor, but I would also add there is a public health aspect of that and the recovery, in terms of employment, in terms of economic opportunity, is not going to happen until we have solved the virus problem. The vaccination rollout, although better than some other countries, has still been pretty atrocious compared to the potential that we could've done, and that is disproportionately affecting people of color, and people in lower-paying jobs. And the sooner that we can get that done and revitalize the sectors that are being most badly affected, which are relatively low-paying, that will help, but we also need to have greater investments in those sectors. Part of that may be through a minimum wage, part of that may be through additional skills training, to increase the productivity there, and naturally produce higher wages, so that there's market pressure to do that. This would cause a better feedback loop, but really, we need to have the public health aspects of it much better solved than we have done to date.

vonLockette: You know, the idea, I really like the sort of, the sort of conclusions around the workaround, what have we sort of anticipated impacts of COVID, and ways in a moment of potential intervention, right. So we are already seeing early data that the vaccine rollout is already having racial and ethnic disparities in who has access to the vaccine. So one of the things, there's two things that pops to mind, one is deal with that. And one of the ways thinking about in terms of work, is really pushing on the category of essential workers so that we have ... making sure that category is expansive, we are talking not just about, you know, medical workers who are incredibly important; we've done that, but also now let's do grocery store workers, let's do bus drivers. And some places we're doing that, some places we aren't.

So much of this has devolved to the states, and we have seen that, in terms of inequality, we have seen what happens. I would really sort of, I would really advocate for a stronger set of federal guidelines and possible regulation on the vaccine rollout, particularly with respect to the notion of communities that are affected by this, not only in terms of their health but in terms of their economic viability, and the ways in which those two are so compounded with one another. And then the stimulus and relief, that this is a way to minimize the blow and minimize the impact, and if we do it and we do it right, and we make sure that again, when it goes to the states first [that] there's a lot of federalism. There's a lot of room for states to make all kinds of decisions, and if there is a stronger sort of guidance from the Biden administration, that will help with some of these, with some of these issues, I think.

Dixon: Laura, let me just squeeze in the last question and you can answer either or both. The last question here is, do you believe that the current disconnection between the labor market and the capital markets is concerning, and will be the new norm? Either question you would like to address.

Choi: I will actually go back to the former question, and I want to, in addition to everything my copanelists shared, one thing I would like to drop for consideration is the importance of childcare. I think we are seeing particularly in this moment. I'm definitely feeling it personally, with the closure of schools and childcare and knowing that disproportionately, people of color are in essential jobs; they are not able to telecommute and work from home, the importance of childcare is so vital in this moment. As we think about getting people back to work, getting people back to education and training, the need for childcare is so critical and especially seeing how this had a disproportionate impact on women being forced to leave the labor market, I think childcare is such a huge part of our recovery.

Dixon: OK, we are at time. I want to just thank our panelists for this really exciting and interesting discussion—lots and lots of meaty ideas here. This concludes our Q and A, and I'm going to turn it back over to Bill Rodgers. Thank you.

Rodgers: Rebecca, thank you so much. Panelists, thank you for your presentations and thank you for a thoughtful conversation. I'm now joined by President Raphael Bostic of the Federal Reserve Bank of Atlanta and President Eric Rosengren, president of the Federal Reserve Bank of Boston, for a fireside chat. For so many of you that are on the Northeast corridor, what better day for a fireside chat? I'm looking out my window and seeing nice, big, puffy wet snowflakes coming down, which will get shoveled in a few hours. But as I said, welcome. What I want to do is pose a, you know, series of questions and hopefully it'll be more conversational over the next I guess 20 to 25 minutes, and clearly I will say at the offset, this is going to feel woefully too short ... to talk about these issues, but I'm so excited that we have two of our nation's Federal Reserve bank presidents willing to have this conversation. So the first question I want to ask you two is, during the Great Depression, we talked about the three Rs: relief, recovery, and reimagination. Where do you think we are on that spectrum of relief, recovery, or reimagination?

Eric Rosengren: I will go first, I guess. Thank you for the question. Hopefully the parallel to the Great Depression isn't something that –

Rodgers: There was no intent there, that concept.

Rosengren: I would say we are still on the depths of the reception. So while a lot of part-time layoffs have come back to work, we are still talking about 6.7 percent unemployment. We are still talking about having three million people on temporary layoff. So we are really in the depth of it right now. I hope over the course of the spring, we are really talking about a significant recovery. But I think it does depend on some of the things that were raised in the previous session, particularly some of the public health outcomes. The other thing I would possibly focus on is a lot of the work that the previous session was focused on was thinking about what employees can do. I think one of the aspects I hope we can have a conversation on is what more employers can do. So as important as workforce development is, I think it's really important to think about how we structure jobs, so that it is attractive to a workforce where they are and how they live, and not assume that the workforce is going to adapt to what the employee, the employer, is looking for.

Rodgers: Thank you. President Bostic?

Raphael Bostic: First of all, good to see you, Bill. Thank you for agreeing to do this. I actually agree with Eric. And I didn't know you were going to do the three Rs, so I had to write them down. But I actually think we are in the midst of all of them at the same time. So as Eric noted, there is a lot of relief that is still required because a lot of people are out of work. A lot of people are in a precarious situation. You heard the discussion in the previous panel about childcare and the stresses that the lack of that imposes on people, introduces to people. There's a lot that needs to be done and that relief is going to be critical. But at the same time, there has been a considerable amount of recovery, and you are seeing this in a number of sectors where business leaders that I'm talking to have actually started looking to the future and not being so immediate in just trying to deal with their crisis mode. And what that then means is that the reimagination becomes critical. Because as we look to the future, we need to be thinking about what things might we do differently to position people to be better able to participate in our labor market, so I think we are in the midst of all of these things, I really appreciated a lot of the solutions that were raised earlier. Many of them—actually you heard kind of a little debate among the panelists, around should we be focusing on the recovery stuff, like the short-term stuff, or the re-imagination stuff? And the back-and-forth was actually quite interesting.

Rodgers: Thank you. Thank you. So let's, we are coming up on the one-year anniversary or one year of being in this pandemic. So let's kind of take a little bit of a step back. What are the most important things that have been revealed in the past year about the pandemic labor market in your eyes? President Rosengren?

Rosengren: Some of the problems were well-established before the pandemic but really just became obvious during the pandemic. Let me give you one concrete example. We are doing a lot of work in East Hartford, Connecticut. And in East Hartford, there is a low-income neighborhood, mostly Black and Latino. It's within a five-mile radius of all kinds of manufacturers including defense manufacturers that had lots of jobs open. But many of the people that were in that low-income neighborhood were not working at those high- paying manufacturing jobs. They had never spent the time asking why it was that there were unemployment rates in the double digits right next to their buildings and they couldn't fill their jobs. And when you dig into it and asked the people, there are a variety of things they said that don't come up in just skills training. So bus line access; if you are poor and don't have a car, you need access to mass transit. Bus lines didn't go by any of the manufacturing firms, so somebody who didn't have a car couldn't work at those manufacturing plants. Shift times—many of the shift times were designed so that female workers couldn't work because they started well before kids went off to school and frequently, they were looking for people in late shifts, where there was no mass transit and much more difficult for working women who have children to do. Regular hours that were not erratic. Effective childcare, which came up in the previous panel. So if you are a single parent dealing with kids, you need to have effective childcare, something that we don't have, and finally, sick days. Now all those aspects were a problem before the pandemic, but they've all become much more apparent with the pandemic.

So with the pandemic, what we are calling essential workers are those people that were living in East Hartford, because their situation has gotten much worse. They still don't have access to transportation. There is still not accommodation in terms of hours. Childcare is even more of a problem as the previous panel highlighted because public schools in the United States are our main means of childcare. That's quite different from what happens in Europe and many other countries. And many essential workers didn't get sick days or not very limited sick days and didn't get paid for it. As a result, they came to work sick and got others sick as well. So all the aspects of the essential workers that we are seeing with very high layoffs in areas that require people to work in a non-socially distanced setting, were really problems that were here before but now are much clearer and much more public.

Rodgers: President Bostic?

Bostic: I don't have much to add on top of Eric's comments, which are all on point. I guess the thing that in thinking about this that I hope people have become much more sensitive to is that all employment is not the same employment. And that there are some forms of employment that really allow for more adaptability, more flexibility, some jobs you could actually work from home and be close to as productive, whereas others you didn't really have those choices. And when you don't have those choices, the types of things that are introduced through a pandemic put extreme stress on you in ways that others have the luxury of avoiding. And so when we think about resilience or economic mobility, I think that that is part of the dynamic here.

One thing I talk about a lot is that because of the different character of employment and the different character of labor markets, recovery is not going to come at the same pace, and so we really need to think hard about how do we make sure that we don't declare victory too soon, and acknowledge that there are still many sectors and many communities where recovery is going to be a long way off. We do a survey of lower- and moderate-income communities that started right before the pandemic, and one of the things that was so striking was that their horizon for recovery was a year, a year and a half longer than what you heard, or what we heard when we surveyed people in other communities. So we need to have that longer-term vision, and a lot of that is because those communities started more precarious to start, and so there's going to be a lot more recovery that's going to be needed to get them and make them whole.

Rodgers: Yeah, I think, when listening to both of you, what comes to my mind is a phrase that I had developed when I was a professor at William and Mary, back in the 1990s and basically we created these two realities economy, right? This is two realities economy and in particular, at the beginning of this, we talked about is it going to be a V-shaped recovery or is it going to be a Nike swoosh recovery, and I think what you all are describing is that it is now the letter K, that is the operative moment, it's a K style recovery.

Bostic: You know, Bill, I actually use the less-than symbol. But it's the same basic principle, that depending on what your station is, you are on the up part or you are on the down part, and the stresses that we have, have really caused all of our inequities to be exacerbated. And the road back is just going to be very long if you are on that down part of the symbol.

Rodgers: We will add another tool. So, for me, it also harkens back to, I think you and I and President Rosengren, we are old enough to remember the ... commercial with the sage old mechanic who advised his client; he said you really need to change that air filter. Or you can pay me now or you can pay me later. And as I said in my opening remarks, we've really slowed our investments in human priorities, the human and social capital. And it's so important that the two of you are willing to speak about these kinds of issues and talk about inequality. I think my experience, you know, for my career, is members of the Federal Reserve have at times been reticent to talk about inequality. I appreciate so much that you are willing to do this.

So my next question then is, what is needed to recover stronger, and as President Bostic, you said, to help Americans return to that, having that resiliency. President Rosengren?

Rosengren: I think we need to rethink jobs as we are reemploying people. Though, we've seen some aspects where jobs have become much more flexible. So if you had told me that I would be outside the office for nine months and all my employees, for the most part, would be outside of the office for nine months, and yet we are operating pretty well, I wouldn't have expected that to have worked nearly as well as it has. As Raphael highlighted, for those workers that can do that, but we didn't allow people to work from home to a large degree. What does that mean when you don't allow people to work from home? Well if somebody is taking care of an ailing parent or has young children, it means that they may just drop out of the workforce, because they may not have the childcare or elder care that they need. So we need to think about how to add flexibility into our labor market. We have learned that technology, not for all jobs, but for many jobs, can be much more flexibly integrated into how people live. So ideally, if we can use this technology coming out of this, it'll be much easier for people that need flexibility in the workforce to be able to do that. Again, it's not for all jobs, but I think there are a lot more jobs than what we had before.

I think employers have to be much more focused on employee needs. So thinking more broadly about how we think about, how do people get to work? So one of the challenges right now, I was walking around downtown Boston last week. It's an empty place. That was incredibly vibrant in February is completely desolate right now. The stores are empty. The restaurants are empty. There are relatively few people on the street. So how do you get people to go back into the downtown? First of all, you have to take care of public health. The second thing you have to do is take care of the public transportation. So if everybody is afraid to go on a commuter rail, on a subway, on a ferry, then it's going to be very difficult to rejuvenate inner cities, particularly those inner cities that are reliant on mass transit, which tends to be particularly on the East Coast, that's incredibly important. So I think we have to continue—right now it's very expensive for state and local governments to be funding mass transit, but it's critically important that we keep mass transit going because that's the lifeline for many low-income workers who don't have cars. That's particularly true in a place like Boston, where there's no place to park when you get into downtown Boston, so you almost have to commute by mass transit or live downtown, which is quite expensive. So I think we need to rethink jobs in a flexible way, and ask employers to be thinking about how they can bring employees back with real job ladders. So we have seen a disproportionate amount of the layoffs that have been in places like entertainment, retail, hotels, restaurants. Many of those jobs were designed to be shiftwork, inflexible schedules, not a lot of flexibility. So I think we need to think about those kinds of jobs very differently as we start reworking the workforce.

Rodgers: Thank you. President Bostic?

Bostic: I'm going to take this a slightly different way. I agree with everything he said. So yes, yes to that, but I would also say that I think that we need to think about or rethink how we invest in the infrastructure that gets workers skills. And we are doing a lot of partnerships right now where we are trying to marry together employer needs, where employees are, and training programs that can cause people to or allow people to get reskilled to fit the needs of tomorrow. For example, we are in a partnership with the Markle Foundation and the National Urban League, among others, called the Rework America alliance, where we are trying to create local conversations so that there can be that kind of coordinated response and create a focal point for what jobs need to be trained for and what people are available to get the skills to do those jobs. We don't do a very good job of that in this country. And for many, many years, the notion of coordinated job training has not been something that we have been willing to invest in, but as transitions in the economy are happening more rapidly, and as the skills for the new jobs are not matching the skills that you needed for the old jobs that are being disrupted, we need to have a much more nimble infrastructure that allows for that sort of thing. And what that will do is then allow us to think more holistically.

Much of the discussion today has been about holistic thinking. You need the childcare; you need the transportation. You need the affordable housing, and you need the skills development. If we can get that kind of systems approach considered more broadly in more places, I think we can make a lot of progress. Then there's one other thing I think is really important here. Which is that, for many people who we want to be in the workforce, they don't have information about how they might do that. And you know, one of the things that I try to do all the time and say all the time is, we can't just sit back and hope that they find us. We've got to go find them and engage and have that conversation that's active, that there are opportunities, it doesn't cost $500,000 or four or five years of work, there are jobs with good wages out there, and we need to make sure that we make it easy for people to see that, and then have the infrastructure in place to help them get the skills that allow them to fill those positions.

Rodgers: Right. Yeah, I think the, before we go, what you just said there, I'm sorry, segues into the next that will probably the last question. But before I go there, I just want to add, I think it relates to an earlier comment about employers. And that there has to be possibly a new version, a new type of an ADA, Americans with the Disabilities Act, but in the context of, if we are going forward where pandemics are going to be a recurring major public health issue, that's not only all the items that you all are mentioning but we also have to get workers or employees back safely in their workplaces. And one of my colleagues and I are working and thinking about these kinds of issues, beyond just PPE, but really a much broader set of conversations on what employers have to do and hence what they need to be provided.

So, that is a whole obsession in itself, ... but the final question I want to ask you, and this is where I said it's a good segue from what you had said, President Bostic, and that is really, what can the Fed do? There is the typical view is you all are participating in FOMC deliberations and that's all very important and is needed and will continue to be going on, but what other things are you all doing or can be doing? President Rosengren, I will start with you.

Rosengren: Thank you, Bill. I think there are number of things the Fed can do. The first thing is do everything in our power to get back to full employment as quickly as possible. There is a big role for fiscal policy there, so I don't want to say that the burden is all on monetary policy. I actually think most of the burden is actually on fiscal policy when interest rates are as low as they are. The second thing is, during these kinds of problems, when interest rates get to zero, we have to do better about designing programs for small businesses that hire an awful lot of employees. So the Main Street lending program was an attempt to do that. I hope we can learn from the Main Street lending program and actually have much broader access to small and medium-sized businesses, something that typically is not what the Fed does because we focus on markets, and when we focus on markets, we are focused on some of the largest companies in the country. They are the ones with access to the bond market. The third thing is, I think we have a reasonably large community development group. I think working with employers for a flexible job market, as well as getting skills for workers, I think we need to do more. And I think Raphael is exactly right, that it needs to be coordinated, but all the onus can't be on the employees; it has to be getting the employees and the employers working together. And, finally, I just think that we need to be thinking differently about how these all fit together. So you mentioned public health, and I think that is critically important. And it's not really just about PPP; it's about many of the jobs that essential workers have. People didn't really spend much time worrying about safety. And we have to think differently about these jobs and what the roles and obligations of the employers are.

Rodgers: Thank you so much. President Bostic?

Bostic: I think four is the magic number, because I've got four things as well. One is just to convene and try to create public conversations among people who have various perspectives and ideas that need to be heard across and beyond the usual suspect type of conversations. So sessions like this are important; we now have our Racism and the Economy webinar series that we are doing, it's a System program. And what's been interesting is that it sparked a bunch of conversations among people who hadn't sometimes ever spoken with each other. And allows and it creates new opportunities and possibilities for solutions moving forward. The second thing that we do, and this program has been an example of that, is really promote ideas for how you can solve and address some of these problems. We need to be in the idea space. We need to be operating much more on the solution space.

When I was an assistant secretary at HUD, I used to go out and get yelled at all the time. And what I used to tell people is, don't yell at me that there is a problem, that's why I'm here. I know there is a problem. Let's spend our time talking about what a solution looks like, and I think that our promotion role can really be successful in that. A third thing that we can do is advise. And, you know, you just heard Professor Rosengren or President Rosengren, I will give you a lot of titles, sorry. Here's some advice about areas where policy and attention needs to be given, and we have the luxury of having large staffs who are out talking to business people, talking to community folks and being able to identify where the holes and where the need that need filling? And we can provide advice on that. The fourth thing we can do is research. You heard in the video earlier today, research by the San Francisco Fed, shedding light on and trying to quantify what the cost of not having full and even participation in labor markets can do. Those sorts of efforts can really help shape the conversations, so I'm going to remember, this is trillions of dollars we are talking about, up to $70 trillion of output, that is real money. And that should hopefully get people's attention and allow us to have entrée into considering things that we might not have so easily been able to broach without that kind of effort. So there is a lot we can do. And I think this is a magical time for us in the Federal Reserve System. We've really stepped up. And I'm looking forward to that continuing moving forward.

Rodgers: Thank you. The Federal Reserve System is a very important institution. And institutions matter, institutions breaking roles and how the economy works. So before I let you go, thank you all for being so generous with your time. But what I typically do when I interview folks, I end with kind of a speed round, where I mention a word, and you get only a word to respond to what comes to mind when you think or when you hear that word, when you think of that word. So we'll just do like two or three, then we will call it a day. The first one is, we will go to President Rosengren, the pandemic. One word.

Rosengren: Essential.

Rodgers: President Bostic?

Bostic: Tragic.

Rodgers: Monetary policy? President Bostic, you go first on this round.

Bostic: Evolving.

Rodgers: President Rosengren?

Rosengren: Supportive.

Rodgers: The new administration? President Bostic?

Bostic: Hopeful.

Rodgers: President Rosengren?

Rosengren: Raphael took mine. I was going to say hopeful.

Rodgers: There you go. And structural racism?

Rosengren: Change.

Rodgers: President Bostic?

Bostic: That is a hard one. Burdensome.

Rodgers: And then finally, today's conference or the conference today and tomorrow and the next day. President Rosengren?

Rosengren: Illuminating.

Bostic: Exciting.

Rodgers: Thank you, gentlemen. Appreciated, thank you.

Rosengren: Thank you, Bill.

Bostic: Thank you.

Rodgers: Now I will turn it back over to Stuart.

Andreason: Thanks, everyone. Thanks for such a rich discussion. I will say that we had many more questions to talk about. We were looking for a one-word answer from the presidents on their thoughts about Tom Brady's move to the Sixth Federal Reserve District recently. We will have to follow up with that some other day soon. And just on behalf of the organizing committee, I want to acknowledge all of our participants for the wonderful conversation that we have had today, and the discussion that we have had.

Just a reminder, this really continues. We are back tomorrow at the same time, 2 o'clock Eastern, to continue the conversation on the Uneven Outcomes in the Labor Market. We hope you will join. Tomorrow, we will focus on the future of work and changing technology and alternative work arrangements. We hope that you will join. The day will begin with remarks from President Loretta Mester, the president of the Federal Reserve Bank of Cleveland, followed by David Weil from Brandeis University with a discussion of new research, the same that we had today.

Registration and a full agenda are available on the conference website. There is a button on your screen right now that you can press to go to that website and see the, tomorrow and the following days. In a moment, you will be, you will get a link to the survey for today's conference, and we hope you will fill it out. We hope you will join us tomorrow, and we thank you for joining us today. We hope that you learned something and that you will share ideas going forward. Thank you so much.