Joe Fuller: Workers on the bottom rungs of the employment ladder are especially vulnerable to economic shocks. Events like the 2020 coronavirus pandemic hit low-wage workers especially hard as industries like retail, restaurants, and hospitality cut back their workforce. Reskilling is the key to economic and social mobility for those workers. But in these circumstances, it’s more difficult than ever. Welcome to the Managing the Future of Work podcast. I’m your host, Harvard Business School professor and visiting fellow at the American Enterprise Institute, Joe Fuller. I’m joined by today Marcela Escobari, senior fellow at the Brookings Institution and former Latin American director for the US Agency for International Development. She is also the former executive director of Harvard’s Center for International Development at the Kennedy School of Government. As head of Brookings’ Workforce of the Future initiative, Marcela focuses on data-driven approaches to boosting opportunity for America’s 53 million low-wage workers. Economic mobility is one of the toughest nuts to crack. Every economy relies on a large number of low-skill jobs. But how do you prevent them from becoming career cul-de-sacs? Today, Marcela will share with us her work on “upskilling paths” and “reskilling infrastructure”—how economic development planners, business leaders, and educators can help low-wage workers improve their prospects. Marcela, welcome back to Harvard.

Marcela Escobari: Thank you, Joe. Great to be here.

Fuller: In your report from last fall, Realism About Reskilling, you count 53 million low-wage workers in the U.S., more than 40 percent of [all workers between 18 and 64] years old. How do you define low-wage workers, and in which kind of job categories do they fall in?

Escobari: Yeah, so the numbers are quite large, as you note. And here we count everybody who is low wage as anybody who is making less than two thirds of the median salary of a full-time worker. And we account for differences in place. And what we find is this staggering number of 53 million workers who are low-wage workers. And if you think about the median of that group of people, they're making less than $18,000 a year. This is less than $10 an hour. So there's not a lot of wiggle room with these kind of numbers.

Fuller: Interestingly, it seems that those workers are really concentrated in a few actual job categories. You identified 10 which are a very, very large percentage of those lower-wage workers.

Escobari: Right. So, they're very concentrated, and 47 percent are in 10 categories. And what we see with these categories is not just that they have low wages, but that their situation is not stable—where, for these people, a very small disruption can mean financial ruin. So these are workers that are not only experiencing a lot of churn, and tend to switch from one low-wage job to another, but they're also more likely to be hourly workers. They're more likely to be contractors, which, as you know, contractors by different estimates today make up over a third of workers in the American economy. So, if you think about three of some of the largest categories like retail, travel accommodation, food service—in these three industries, more than three fourths of workers are paid hourly. And in these three sectors alone, we're talking about 22 million workers that are probably at higher risk of layoff, higher risk of getting hours reduced, in any small crisis.

Fuller: One of the things that was very interesting in your research was how you demonstrated that at both ends of the income spectrum, low income and high income, people changing jobs disproportionately end up in that same income quintile. That low-wage workers tend to move to other low-wage jobs, just as high-wage workers tend to do the same thing, kind of puts the lie to the Horatio Alger story about America that you start off as a bootblack and you end up CEO of the company.

Escobari: Yeah. I think this is the malaise that we're most worried about, right? This idea of the erosion of the American dream that if you work hard you can get ahead. And now the truth is that you can work hard and still make $9 an hour, work three jobs, and not being able to get out of that cycle. What we did is get a data set from the [US Census Bureau and US Bureau of Labor Statistics (BLS) Current Population Survey] CPS of job-to-job transitions to understand the churn—who's moving and whether they're moving upwards. And what we found is that, as you said, first, something that is not surprising, which is people in the lowest quintile churn move from job to job the most, but they tend to see very little wage growth. So 52 percent of people who make $10 to $15 an hour tend to move and change to jobs that keep them in that wage segment. And probably more interesting, or disheartening, is that people in the low-middle quintile, the second quintile which make $15 to $19, who switch also have a 55 percent chance of transitioning downward or remaining in the same wage bracket. So, little by little we are shrinking that middle class in each of those churns.

Fuller: You introduced the concept of a job-to-job quality indicator in your Realism About Reskilling paper. Could you elaborate on that? It looks like not all jobs are created equal in terms of the possibilities for upward mobility that they open.

Escobari: Yes. What we were trying to do is understand, in a world of increased displacement, if I am a steel worker who loses his job, am I likelier to become an electrician and make $100,000 a year than become a software engineer? Can I understand what are the actual transitions that people are making and try to maximize those upward transitions. So what we did is we ranked every job and looked at every transition out of an occupation to say which transitions tend to be upward, which transitions tend to be downward. And we created an index for every occupation. And we focused on these low-wage jobs to say are there “good bad” jobs and “bad bad” jobs? And that's what we found. There's a lot of low-wage jobs. Low-wage jobs are going to be here to stay. But, are there jobs that are steppingstones toward higher-paid jobs and others that are stuck not only in low wage but with very little mobility? And we find that there's a group quite surprising, in some ways, others not so surprising, of occupations where we see both low wage and low mobility, which I think is where our focus should be. And that includes occupations like cashiers, food preparation workers, cooks, but also preschool teachers and housekeeping [staff], health aides, people who, many of them, have higher degrees. But just the structure of those jobs means that once they're stuck in them, they don't tend to be able to move upwards.

Fuller: So, Marcela, in talking about these steppingstone jobs, you really emphasize the importance of transitions, and historically a lot of the discussion has been about “what are the enabling skills that qualify someone to move beyond their current role?” And, of course, that begins to start getting reflected in what we think about in terms of lifelong learning opportunities and getting credentials and whatnot. Talk about your focus on transitions, and what have you learned from that that's original?

Escobari: We were very purposeful in trying to look at transitions versus skill similarity to understand what actually are the possibilities of upward mobility for people. And this is similar to the methods that we use in understanding industry diversification and what's likely to grow and decline in places. We wanted to look at reality in predicting the future. So we took this data set of 8 million observations, over 100,000 actual transitions, to try to understand what are the moves that are actually happening and trying to understand why some people are getting stuck and why some transitions that you wouldn't expect are actually happening. And if those transitions are upward, how can we then understand how to maximize those transitions. And just to give you an example of the difference between looking at skills and, of course, skills are the embedded qualities that allow for these transitions. But sometimes we don't know how they might show up, right? If I look at the skills of a janitor or an office repairer and I look at what other occupations [are] closely related in terms of skills, you might see dish washing, housekeeping, food preparations. But we've also seen that janitors can move into maintenance and repair work, which is a decidedly upward transition. So, if we only looked at skill similarity, we might miss transitions that are realistic and upward and run the risk of improperly funneling janitors to other low-wage occupations. And what we're trying to do is find ways in which we can break that cycle and then understand the mechanisms by which we speed them up.

Fuller: So, in fact, you may see some non-intuitive paths that you could help accelerate by creating better training opportunities or resources for people to understand that they could make those transitions.

Escobari: To give you an example, if you take an occupation like computer network administrators, one that is growing and in demand, it might seem obvious and clear that many computer scientists move into that occupation. But what my you might be surprised to find out is that you also have office repair people and some retail workers that have entered into that occupation. So knowing that kind of information can help you know and invest in workers that you wouldn't have expected could grow, learn, and grow into a certain occupation that's in demand and make you more willing to invest in reskilling and give opportunities for mobility and promotion for some of these workers. And I think this kind of data that we have at a national scale can be supplemented with firms’ own information around mobility, and we're hoping that that can lead to more internal bets and more opportunities for promotions within firms as companies are trying to fill these very hard to fill tech jobs.

Fuller: Marcela, you referenced to how employers are seeking talent or thinking about who's a candidate qualified for consideration leads us to this broader question you raised about the skill or reskilling infrastructure in the country. That's been spoken about widely about all the many presumed deficiency of it. What did your work tell you about it and where we should be thinking about investing more or changing our pattern of spending?

Escobari: Well, let's start with different actors because I think if we are going to tackle the skills gap, it requires work from every actor in society, not just the community colleges and the post-secondary infrastructure, but also within companies. Because as you've seen, they are feeling this skills gap. Technology is moving fast. More jobs require high tech skills, and these are hard and expensive to find. Turnover is expensive. Hiring and firing is expensive. Yet the response hasn't met this need necessarily. Accenture actually surveyed 1,400 CEOs, and almost half of them lamented that they can't find the talent that they need. But only 3 percent of that group said they were planning to increase their training budgets. So, the question is “why?” That's, I think, a puzzle. And I think at a time that we need this very work-based specific knowledge, firms have been decreasing their training budgets and slashing entry programs. And I think that we have a tragedy of the commons problem, and that has landed us in this low-skill equilibrium. I'm going to give you an example. I started my career as an analyst at J.P. Morgan in investment banking, and at the time the training was six months long. And probably a couple of years before me that training used to be nine months. And the idea was that you could get philosophy majors to crank out discounted cash flows. This training today is two weeks. So, I think the issue has been that in every recession these were the programs—and maybe it's got to do with our accounting rules that [treat] both training and people as expenses, that these are the first to go. And we get legitimate answers [from companies, to this question of investment in training], right? If you're the last firm standing providing these kinds of trainings, with workers moving and companies poaching each other, and you're the only sucker who trains— [you don’t] see the benefit of this. Similarly, some firms think they're not in the business of training. But I think this short term view, which again can make sense in the short term, if everybody takes it, we end up in this place where we're in a low skill equilibrium, and everybody's suffering from the skills gap.

Fuller: It's an interesting question. I mean, we know we have a large population of people caught in low skills jobs, and we know that there are a large number of low skill, low paying jobs. Do you view that as a demand or a supply problem? What I mean by that is, yes, currently there's a significant number of employers that have these low skilled jobs. But when you think about some of the emerging business models we've seen, especially in the gig economy, it's almost as if the business model is intended to build off of the large supply of people who are available at reasonably low wages and are only viewed as qualified to do reasonably low value-added tasks.

Escobari: I think this is an interesting question, and I do think while there is a vast need for reskilling, we have oversupplied bad jobs. So what am I saying? Of course, that low wage work will always exist and has existed, but what you want is two things. One is that low wages or low wage jobs do not need to be bad jobs. They can have the minimum protections, but they can also deliver what this promise of automation was supposed to deliver, that the job utilizes our skills and our uniquely human capabilities. And that's a job design question, and I think very few companies are taking on the challenge of automation as an opportunity to improve job design with the training that that implies. But second, entering a low wage job should not be a life sentence. These jobs can be starting points for young people, for those who haven't yet invested in education and up-skilling, for people who are working multiple gigs. But they need to actually offer prospects of mobility and allow for the disposable time and income to reskill and move upwards. And I think the question is can we make sure those paths exist and help people navigate them? And our research shows that those paths are limited and decreasing in many occupations.

Fuller: Your paper is kind of provocatively entitled Realism About Reskilling. Why did you call it that? And when you're talking to a policymaker or an executive or an educator, what is it that you want them to understand to have them engage this issue with a greater sense of realism?

Escobari: I think what we were trying to show is that obviously reskilling is crucial, and I think it's vastly under supplied right now, but it's not a silver bullet. If the problem that we're trying to solve is increased inequality, a lack of mobility, then more bootcamps are not going to bring back the American dream. It needs to link to solutions that encompass a variety of actors where firms are able to change the way that they hire and the way that they help and contribute to the reskilling equation. Reskilling organizations are able to provide the kind of skills and credits and majors that actually make a difference and provide mobility. And policy makers need to create all the other set of infrastructure that allows for people, particularly those that are most vulnerable, to access and benefit from these opportunities. And what we find is that particularly these workers, low-wage workers and vulnerable populations, the current reskilling infrastructure is vastly inadequate in serving their needs.

Fuller: You talk about the kind of idea of surrounding solutions, and this of course invokes research that now goes back 20, 30, 40 years about issues ranging from childcare to transportation and how that contributes to this kind of lock-in effect that you described, this life sentence of low skill work. What are some surrounding solutions that you think can work and maybe get beyond that kind of a traditional list that's been bandied about for a long time?

Escobari: What we tried to do in this final chapter of our report Realism About Reskilling is bring a bit of that qualitative aspect of understanding what's missing in the current infrastructure. And we did it using a lot of the design principles that many consumer companies use to understand user needs. We called it the “reskilling user journey.” [We] say, "Let's look at how it's being offered from the perspective of this user because then we will be able to see the friction points much more clearly." And what we find is that indeed these workers find friction points in places that you and I wouldn't necessarily predict because it's not our reality, and sometimes it's the most basic thing, entry. When promoting actually amazing content that is out there, we can't assume that if you build it, they will come. These users require extra work to recruit to bring them into the reskilling infrastructure. Second and one that is even harder to talk about is this sense of self efficacy. Many of these workers have had consistent negative educational experience throughout their schooling. They've been told they're the C, D players, and they carry with them that kind of disempowerment, which is very hard to reprogram. And many are going to stations when they say, "We're selecting for people that are motivated," they are purposely selecting for those that can complete and succeed with less scaffolding and not because they're evil. This is because that's the incentive to try to find the people that can most quickly be taught a certain skill to become useful in an employment situation. But I think we are missing an untapped talent pool and missing the people who might need it the most. And to give you just a couple more of these friction points, they require more career navigation. What is actually going to make an upward trajectory? Sustained support. It's not just an intervention and a six month checkup. They're going to go up and down, and they need that support through time. And also, they face the obvious things, so logistical barriers, the childcare, the fact that they can't get to the centers because transport is expensive, and their job is already an hour and a half from their homes. And, in some cases it's cost, but like I said, cost is not the only issue. And lastly, that you want content that is targeted to them. And I'll give you an example. We thought that MOOCs, these massive online courses, were going to be the boom for this population. They're online, they're accessible. But what we've found is that they're actually geared toward college graduates. Seventy percent of the ones who take them are college graduates. Then the algorithms are actually optimizing on that population, making it much harder for somebody with a fourth grade literacy level to actually engage and feel successful. So I think this idea of entry, understanding the economic barriers, creating self-affirming content, understanding how to navigate and understand what will be a realistic path for them and giving sustained support has to be part of our infrastructure or these workers will continue to be left behind.

Fuller: We're in a time, of course, of significant economic uncertainty and dislocation with the onset of the Covid-19 situation. This is one of a number of kind of periodic disruptions to the labor market. Any insight into in how that's going to play out for this population and what it's going to suggest in terms of programs that people, whether they're employers or policymakers, are going to want to consider for getting people back to work after this disruption?

Escobari: When you put stress in a system, it breaks in its weakest points. We've known that the increased precariousness of people's work arrangements has existed for a while. This crisis has just brought it front and center. We know that precisely in the kind of occupations that have the most vulnerable workers, hourly workers, contract workers, travel, hospitality, food services, those are the ones that are going to be hit the most, and we don't have a great safety net for that. I think what's going to make this crisis different than others is that one, it's not a slowdown. It’s [that] activity has come to a screeching halt. While we will end up with consumption boosts, unfortunately we can't spend new money in the places that might need it the most. So, I'm hoping that at least on a second phase that we can think of solutions that can help workers navigate displacement, because that's a reality that's not going to change. I think there's some principles that I think will be helpful, particularly to local leaders that are going to be thinking about how to reignite their economies. One is to create systems that can help in the long run, too. For example, setting up a system of portable benefits, which has been tried quite successfully in France, which can be set up by employers. And now we can use it for sick leave but later can be utilized for re-skilling, for unemployment insurance and whenever we want to deliver cash to workers for very specific goals. And we could create that infrastructure now. Second, it's targeting solutions to the right people. We know who is the most vulnerable and who is going to be hit the hardest from the industries to the contract workers—those who work for temp agencies, those without health coverage, women who tend to be overrepresented in certain roles, and targeting to those kind of occupations like which we said are extremely concentrated in low wage workers will make a difference. Lastly, to make it easy to access, which at times might seem like a trade off in targeting, but I think this is where we can be a lot more innovative. We need better safety nets from unemployment insurance in the short term to massive reskilling. We've been cutting these programs since the '70s. We're making it more difficult to access. We spend half of the unemployment programs compared to OECD countries. We've cut our workforce programs by a fourth since the '70s. But, many of these programs that actually 'have been quite well-designed end up failing because either there's meager funding or because they create such complex qualification criteria for people that make it very difficult to navigate. I'd say let's create systems that we can use for the future, target the right people, and really make them easy to target. I think that will be helpful in this crisis and beyond this.

Fuller: Marcela, as we think about that, how should we be thinking about it in terms of geography? Obviously the impacts so far have been asymmetric across sectors and you've done a lot of interesting research about how economic activities at the level of of cities as well as states differ and have different implications for the available skills mix and the projected skills mix. How does that all factor into the way we should think about a response for this crisis?

Escobari: I think the principles apply to all regions, but whenever in doubt, I do think that a one size fits all rarely fits this country or economic policy. I think in this particular example, every city is going to be affected differently based on the industry structure that they have and the kind of sectors they are going to see an immediate contraction. Just to give you an example, Las Vegas, 17 percent of their workforce is involved in travel accommodation. In Houston, that's less than 1 percent. So, when we emerged from these immediate challenges, it's going to be up to local leaders to kind of get their own economy up and going again. This is where I hope that some of our place based insights can be helpful.

Fuller: Who should we be looking toward? Mayors, governors? What do you think of the right unit of analysis?

Escobari: So, I think regional leaders from mayors to economic development shops, chambers of commerce, work for boards, they're all reading the same headlines. “Automation is coming. The middle class is shrinking. Workers are ill prepared and either you're a superstar tech city or you have no chance of catching up.” All of them are asking, what does this mean specifically for me? This is what we're trying to answer. I think everybody can benefit from place-based insights that are specific to their city and play a part in linking the economic development and the demand side and the creation of good jobs and the supply side. How do you get workers to take advantage of those opportunities and maximize the upward mobility for your workers?

Fuller: Well, that does sound a little bit like the list of usual suspects. What needs to change to get them to move beyond this kind of skills infrastructure we've got now, which importantly does not abandon everybody to a terrible fate, but it does regularly create this very large population of people with a few skills of real market value and, worse yet, almost no opportunity to add the skills necessary to get out of those economic circumstances?

Escobari: I think part of the issue is not looking at this problem in silos, and this is where linking the demand with the supply is key. But many of these policymakers need to be thinking about demand. How is it that I am going to understand what's growing and contracting in my economy, that I have a set of choices on the kind of capabilities that I build to attract and grow industries that are going to create stepping stone jobs? Those jobs and those industries are going to look different for me if I'm St. Louis, or San Francisco, or Milwaukee and understanding that particular path. Then the strategic choices that you can make to accelerate the growth that you want will be key to then understand, “Okay, as I know which occupations are likely to grow and decline in my city, how can I create a responding re-skilling infrastructure that matches that growth potential and it's not in isolation?”

Fuller: So it sounds like you're talking about a much higher level of integration across these independent bodies and aligning them on a new type of common objective, which is the percentage and growth rate of good jobs in a locality. Marcela, how should, how should cities think about creating those good jobs and how should they be measuring progress?

Escobari: I think this is where we are trying to bring very place-specific insights, because while we are hearing those national headlines, the response is going to be very different. There's not a set of five tech industries that every city should try to grow. What we found, and this is with methods that have been tried in the international scope by, actually Harvard economists, is that growth is path dependent. That if you tell me what kind of industries are thriving in your city, that implies a set of capabilities that are existing that can tell us what kind of new industries are likely to thrive. And it also tells you which industries are both feasible in your city and strategic. Meaning, if these industries grow, you'll have the capabilities to attract other complex industries, which means that city leaders have, they're making choices every day on the kind of capabilities they build. And they can be very purposeful based on their unique capabilities and the opportunities of growth that are available to them to both change their trajectory if they've been contracting or to accelerate it. And that set of very strategic choices can translate into the kind of occupations that are needed to grow these industries. And that's a key piece of information where you're linking not only the national trends of what's happening to every occupation, but local staffing patterns and local industry makeup that can define how your trajectory is going to be different than other cities.

Fuller: So, it seems to me that you're saying we need to integrate the way we think about local economic development, what industries we're supporting, nurturing, trying to attract, through the lens of the jobs they create and the skill base they cultivate and to tether those to creating an infrastructure that supports a workforce that allows people to get and keep these better paying jobs.

Escobari: Yeah. Yes, absolutely. And in a way, as a policymaker, you have limited resources and of course you want all flowers to bloom, all private enterprise to thrive. But as you're thinking about where to invest, what capabilities to build, then you're making choices all the time. And in deciding if you want to be carbon neutral and invest in green industries and, likewise, invest in industries that create us the kind of jobs that are a stepping-stone for the workforce that you have. These are choices that you make by the capabilities that you built. And in a way, where we're definitely proposing is that that should be the mentality. It's not about attracting industries, those incentives, but it's building the capabilities that make certain industries thrive because then those capabilities, of which talent is the most important, stays with you.

Fuller: So, I think what you're saying, and certainly what your research suggests, is that local geographies should be really trying to think of entirely new things to do. And in thinking about the data on which they're making those judgment on a fundamentally different basis than they have historically.

Escobari: Yes. I think policy makers and regional leaders have to trust that they have a deep understanding of their own communities, of their capabilities and that their solutions are going to look different and to be bold when thinking about them. You know, we're hoping that our analysis can provide a bit more nuance and place-specific insights, but, but if anything, the message is for them to be bold and not necessarily expect others to lead, particularly in the wake of this recession. I think it's a time to be bold and not narrow in the way that we think about solutions. So, I think we can't let this crisis paralyze our thinking and lead us to look at responses in a very narrow term. There's a lot that we know and that has helped vulnerable workers and if anything, this is a time for both cities and firms to be bold in their responses. I often get asked by city leaders, “okay, can you tell me the best practices? Who's doing things that work?” And in a way, we don't have a lot of these examples, so we need policymakers and firms to try new things, to use their understanding of their unique place and opportunities and be bold in finding solutions that can help workers transition and that can bring dynamism to their economies. Just to give you an example of the current moment. Costco has started using their employees from their automobile division and retrained them as checkout counters, because, they've seen a surge. This is going to happen throughout the economy where we're going to have a surge of need in certain parts of the economy and vast numbers of workers that are going to be let go, and if we're able to be creative, both within firms to see how we can re-utilize workers and maximize their potential, their creativity, I think firms that can act creatively in this time of crisis, they are going to see the payoffs, not only in the communities that they live in, but in their business models.

Fuller: Marcela, thanks for sharing your thoughts and your findings from your very original and interesting research, and thanks for joining us on our podcast.

Escobari: Thank you, Joe.

Fuller: Thank you for listening to this episode of the Managing the Future of Work podcast. To find out more about our project on the future of work, visit our website at hbs.edu/managing-the-future-of-work and sign up for our newsletter.