Bill Kerr: It’s early 2022 and the labor market remains in unchartered territory. We’re two years into the Covid-19 pandemic, amid a worker shortage, and record numbers quitting their jobs. In the U.S., stubbornly high inflation and recent increases in pay have raised the specter of a wage-price spiral. A popular narrative holds that, after many years of declining leverage, labor is newly empowered. Whether this assessment is true or not, workers at the low-end of the pay scale don’t seem to be securing significant and lasting gains. Low-wage work continues to be precarious and often risky, with little prospect of upward mobility. Moreover, the damage is on both sides: employers pay a price in lost productivity and expensive churn. What accounts for these inefficiencies, and what would it take to make low-wage frontline work less of a dead end?

Welcome to the Managing the Future of Work podcast from Harvard Business School. I’m your host, Bill Kerr. I’m joined today by my Managing the Future of Work Project co-chair and podcast co-host, Joe Fuller. Joe is the lead author on the project’s recent report on upward mobility, entitled Building From the Bottom Up. The research sheds light on a dynamic that predates the pandemic—the high turnover and limited opportunity confronting the more than 4 in 10 American workers in low-wage jobs. We’ll talk about the disconnect between workers and employers, barriers to advancement, and importantly, how to break the cycle. Welcome to our podcast, Joe.

Joe Fuller: Bill, it’s a pleasure to join you and be on this side of the table.

Kerr: Joe, in light of the pandemic, the timing of this upward mobility research was either spot on, problematic, fortuitous, maybe some combination of those. So start with some background about what led to this particular research project.

Fuller: There is this growing percentage of the workforce that are seemingly permanently mired in low-wage jobs, so that we have 44 percent of workers making less than 200 percent of the poverty line. And I was interested in that, because if you look at the data that’s provided, for example, by the Bureau of Labor Statistics, it characterizes them, essentially, by their employment status. Are they long-term unemployed? Are they short-term unemployed? Are they in part-time work and want to have more hours? And so I was curious us about, let’s learn a little bit more about why they’re there, not just who they are and what the absolute magnitude of the population is. And I also thought in the back of my mind that there might be some learnings here about America’s very, very unsettling workforce participation rate. Twenty or 25 years ago, the U.S. workforce participation rate was 10 points higher than Western Europe; now it’s two points lower. Women’s workforce participation has been going down since 2003; men’s very steadily. And this affects not just older workers, but we have almost 10 million prime working-age males who are neither in employment, education, or training. So I thought those two phenomena would be probably related to each other.

Kerr: So given that the data that’s publicly available or typically collected doesn’t serve this purpose—either because it’s not longitudinally tracking individuals or, as you’re just describing, maybe they’re not participating in the workforce—how did you go about identifying the context of low-wage work and then building some evidence around it?

Fuller: Well, we set out to think about it through three lenses that were supported by extensive interviewing, but also three separate surveys. One of the surveys was of people who have been in low-wage work for three consecutive years but have not received a material raise or change in employment status or title—so, essentially, been in the same job more or less for roughly the same wages for three consecutive years, prior to Covid. We worked overtime to make sure that we didn’t get people confused about what was happening in Covid. We had a parallel population of workers who had broken through that 200 percent of the poverty line barrier, that had gotten substantially more pay, that had been advanced. And so, that allowed us to ask both populations questions for the first population about, “What do you think explains why you’re stuck? Why you haven’t advanced? What’s standing in your way?” versus those who had, “What happened? What allowed you to get this advance?” and also capture data on their demographics— by race, by gender, what industry they worked with, had they changed jobs. The third survey was of employers, how they viewed the prospects of low-wage workers, what stood in the way of their advancement, what they were doing about it. Because we wanted to be able to contrast the way companies explained or describe what they’re doing to support upper mobility, versus how their low-wage workers or those who had been advanced explain their experience of employment. What help was effective from employers? What were employers actually doing in their view, versus areas where the employers might be either fooling themselves or overstating what they were doing. So, it was a complex set of data to gather.

Kerr: I’m not aware of a similar data collection effort that has spanned thousands of workers in low-wage settings, also the employers on the opposite side in the interview. So tell us just a little bit about the scope and how HBS, the Managing Future of Work Project, but also the Gates Foundation that was a partner in this, how you came about that as the appropriate lens to look at. And then, what were the challenges that you confronted in actually carrying out that level of research and identifying people to interview and similar?

Fuller: As has happened in some of our prior research, there aren’t that many studies that simultaneously look at the way workers or the supply side of the labor market and the demand side employers view the same phenomena. And so, really, as a regular principle, we always want to explore whether or not that’s going to be a rich area of inquiry and have essentially a presumed stance that we’re going to do both in parallel. As you mentioned, we were generously supported by the Gates Foundation in the research. They have been increasingly interested in economic upward mobility. It was a difficult sample to assemble. The workers, particularly, because the low-wage workers are harder to find, less motivated to participate in these surveys, and also, we had to take great care to succinctly explain what we were looking for in the candidates, so that when we were qualifying someone as a potential subject of the survey, we didn’t get someone into the sample who actually didn’t represent the group we were looking for. So that was a challenge. And we had to augment the survey with some targeted populations when our first couple of pools underweighted various subpopulations. So it was an involved process. Another thing I’m excited to share with you, Bill, is because there is this gap in the research, as we gathered the data, we decided what we were going to do is have the survey instruments and all the data be accessible to others through our Managing the Future of Work website, so that if other researchers want to try to cut the data differently or want to try to do some longitudinal research, they have much greater ease of extending the research in this area, because it is completely virgin territory, and we think there are a lot of additional questions that would really benefit from research and will yield actionable results for employers with favorable impacts on low-wage workers.

Kerr: So you do all this hard work, and you’ve earlier pointed out that one of the main outputs of this is to show these disconnects between how employers are viewing the labor market in the environment, and then how workers are viewing low-wage work and its potential. So walk us through some of the key differences.

Fuller: Most companies—and it’s particularly true of large companies—believe that they are doing things that advance the interest of their workers on a widespread basis and effectively. And we looked at 60 different instances of what we call “touchpoints”—important events in someone’s work life, whether it was the onboarding process or the performance review process, or whether skills training was offered, or even offboarding process, the exit process. And we evaluated how employers described their performance at each of those touchpoints with the way the two different worker populations did. And we looked for asymmetries, obviously. And what we found with employers is that grade inflation is alive and well, that most large companies gave themselves pretty high marks, especially in those areas that workers said were particularly important to them and to their success, looking at the differences between workers who had been advanced and how they explained why that happened, versus those that hadn’t been, zeroing in on four or five different touchpoints. The first is, the workers that got advanced said, “I had someone who took an interest in my advancement. I had some mentorship, someone who’s rooting for me, someone who’s giving me some advice outside the normal parameters of supervision or the workplace.” Workers who weren’t said, “I don’t really have anybody who’s doing that.” The workers who got advance said, “I got pretty regular feedback.” The workers who didn’t, often—up to 50 percent of them—said, “There’s never any regular feedback here. Or, it’s only infrequent.” And interestingly, regularly reported that they had the skills to be advanced, even though the employer said workers who weren’t being advanced didn’t have such skills, which strongly suggests that there’s a communications breakdown between the employer and the employee about what are relevant skills, where do you stand against that? Where do you need to improve? And a huge one was that employees who advanced said, “There were pretty clear pathways. It was known to me what I needed to do. And there were resources available to help me to do it—to cultivate a new skill or demonstrate a skill—so that I could be advanced.” Now, another variable, which I’ll add to this, is that there were asymmetries—unevenness in the evaluation of companies—in terms of their effectiveness in supporting the advancement of lower-wage workers through the levels of management in the organization. We created three tiers of the people we interviewed in the companies, and we surveyed in the companies: a senior management tier, a middle-management tier, and a frontline supervisor tier. And as you move down those three tiers, with each of those steps, there is a substantial reduction in the degree to which companies said, “We are doing those things.” The C-suite was saying, “We’re providing good feedback. We provide clear pathways. We make them comprehensible to workers. We’re providing mentorship.” Whereas, the middle tier was saying, “Well, yeah, we are trying, but it’s hard.” And the frontline supervisors’ evaluation of the way the company was performing were quite close to the workers’ evaluation and rather distant from the C-suite’s evaluation.

Kerr: Is it just that they’re not getting the right information, is it they’re trying to paint for themselves, maybe, a rosier picture of where they could be in the future? Or was there something else that was likely to be this internal gap?

Fuller: This is a classic problem of general management thinking that, because they have a policy and they have a procedure and they have a manual, that it’s actually being followed and executed against. I think the second thing is, many of these types of outcomes aren’t actually linked to the performance system for the supervisors. It’s not something that’s tracked. So if you look at the business intelligence systems that so many companies use to track key performance metrics, they’re looking at their open-to-promise inventory. They’re looking at their standard costs, versus their realized costs. They’re looking at their supply chains and their performance of supply chains. And they’re evaluating managers up and down the line, operating managers, on those bases. But they don’t have a metric for which managers are associated with the highest levels of turnover by job description in the company. And they don’t have a measure of managers who seem to regularly be the bosses of low-wage workers who get promoted into better jobs, or have higher retention, or higher engagement scores. So I think that’s part of it. I think the other of it is that, being a frontline supervisor of low-wage workers is a really demanding job, because, you have to translate that into what type of working environment you’re talking about. So the average number of people that work in a manufacturing plant, under a frontline supervisor, is about 30. So that foreman has got 30 workers on the shift that they’re responsible for. They’re going to give two 45-minute personal feedback sessions a year to 30 people? That’s asking a lot, and particularly if I’m not being measured on that, I’m being managed on making the production ticket, I’m being managed to getting the product to flow, I’m being managed on quality of the product and scrap rates.

Kerr: Yeah. If in this vicious cycle, if the employee will leave relatively soon anyway, that can reduce my near-term incentives to be making those 45-minute feedback sessions.

Fuller: Exactly. I think one of things that’s most interesting and provocative—for me at least—in this research is that I think there’s a self-fulfilling prophecy going on here. Our research said that low-wage workers, their first preference, over 70 percent of workers said was to stay where they’re currently employed if there were opportunities for them to advance and grow in that workplace. I think that immediately defies a bunch of conventional wisdom. But why do they leave? Because there aren’t opportunities for them to grow. The employer sees that cycle as, “This job has high turnover. Therefore, it doesn’t make sense for me to train people who are going to leave. And the job’s got to be relatively straightforward, not being that productive a role, because a new person has got to be able to come in and do it relatively quickly because I’ve got eye-watering rates of turnover.” And that becomes a self-fulfilling prophecy. “If it’s going to be low-wage and not have training, I’m going to have turnover.” And this creates this flywheel of turnover. And no chief executive says, or no CHRO says, “Oh, I really want to have a high-turnover strategy.” But employers describe this cycle as if it’s some law of Newtonian physics or divine mandate, as opposed to there are ways to interdict it. And I think we’ve given people some insight into how to interdict it and improve their economics and competitiveness.

Kerr: This first conversation point about the gaps has been execution, where there’s a gap between what was set out as the policy or the hope, and what’s actually being delivered. You also in the report note that employers can lack insight into some of the personal circumstances of employees that help upward mobility. So it’s not about execution, it’s things that they don’t know. Tell us a little bit about that part.

Fuller: One of the very interesting insights I thought was that many of the workers that do advance do it by changing not only employer, but industry. And that when we asked why people change their jobs in both the two worker populations, things like “ease of getting to work” were right at the top of the list. And that caused us to ask, if employers believe they’ve got these pathways set and do have opportunities for people, why is it that workers either don’t take advantage of those things, or are unaware of them, or whatever? And, when you look at it, basically, a lot of employers are very focused on what we call “UX,” the user experience. But they’re not particularly focused on the “EX,” the employee experience, particularly the “LWEX,” the low-wage employee experience. Why? Because once again, they’re going to leave in large numbers, consistently. And so when an employer starts understanding what’s motivating somebody to leave, or what’s motivating turnover or failure to engage in those programs or pathways and opportunities the employer does offer, they tend to have a stylized explanation for that. And by looking at what’s actually influencing, for example, the ability of a worker to take advantage of an education benefit, or to stay after their shift to work, or to leave in the first place, the penetration of offboarding was much lower than we anticipated. And so you’re not gathering that data. “I left because the bus schedule changed, and now I have to leave my home at 6:00 AM to make sure I’m on time. And there’s a job 15 stops closer for the same amount of pay. I’m going to leave.” By the way, one thing about that cycle of changing employers, which is corrosive for the worker, is that each time they do that, they go back to the starting line in terms of the opportunity to advance, particularly if they shift industry, or they have to prove themselves, get comfortable in this workplace, understand what it’s going to take to advance here. And constantly going backwards to the starting line makes it very, very hard to complete the race successfully in a good time.

Kerr: Joe, the sample that you’re looking at is purposefully set around the low-wage workplace. Overlay for this, education, and also just, either, credentialed education or other forms of skill. How does that map into the low-wage workforce sector?

Fuller: Well, many people are surprised that there are people with post-secondary degrees—including college degrees—in the low-wage sector. There are often some collinearities there, like they have some other activity they’re very interested in—the person that’s trying to make it on Broadway, or is an artist, or a potter, or something in their spare time. But here we are generally talking about a population that is high school graduate or less. They skew to women. They skew to racial minorities. Many are English as a second language speaker. And each of those, of course, also correlate to those personal circumstances that might limit someone’s ability to take advantage of opportunities that employers are accessing. So this cohort of workers is highly likely to have significant caregiving obligations that they are personally obligated to meet. A lot of single parents, for example, in this population, although the largest part of the population were people who live by themselves. Often, that’s, of course, going to be a young person that’s recently just out of high school or has some college but no degree.

Kerr: And given that the population skews toward people of color and women, there’s clearly a connection that can be drawn from upward mobility and also diversity and inclusion. So walk us through how the report brings those pieces together for employers.

Fuller: When we think about whether or not we want to invest in a worker and advance a worker, it strikes me that it’s always easier to make a confident decision there if you’re making it on the basis of a personnel file and a history with that person. And the companies that employ large numbers of low-paid workers have a big pool of diverse talent. And rather than going out and playing the spot market for diverse talent for jobs higher up the hierarchy, which is the default strategy of most companies seeking to improve their performance and diversity and inclusion—I think, you’ll agree with me that, we’ve never seen more widespread, sincere, determined efforts by companies to improve on those dimensions—but they tend to go to the old playbook, which is, “We have an open position. Let’s go into the spot market. And we have a preference for a diverse candidate.” The marketplace for diverse talent for middle-manager positions and technical position is white hot and very, very competitive. What about growing your own? Your lower-wage workers are disproportionately diverse, but they know about the company’s operations. They know how we do things around here. They’ve demonstrated the commitment to work. And by the way, all our data suggests that when colleagues see colleagues get advanced, they feel better about their employer, even if they don’t get advanced. So we think there’s a real opportunity here for particularly larger companies and what passes for middle-sized companies. But we could be talking about companies with hundreds of millions of dollars of revenue and hundreds, if not more than a thousand, employees, to see a golden thread running through this. Let’s have a more determined effort to add skill-building opportunities that our lower-wage workers can access. So we’re going to qualify them to move up the organization chart. Let’s make those programs, let’s design them for a better EX, a better employee experience, so it actually takes into account the personal circumstances that might be inhibiting their performance. And let’s anticipate where we’re going to need talent, so we’re less in this cycle of, “I need this person now. I have no choice but to compete with everybody else who’s looking for people with the same talents and are going to bring more diversity to your company.” So we think there’s a really exciting opportunity to use the real commitment that companies are expressing—to higher performance in DEI—to break this low-wage trap cycle we describe in the paper.

Kerr: Joe, you have noted how the report was designed to understand mobility challenges before the pandemic and in an environment without it, but the pandemic has also been very disruptive to the labor market over the last couple years. Tell us a little bit about the particular experiences of low-wage workers during Covid.

Fuller: Well, I think it’s certainly varied by industry, because obviously there are asymmetric impacts, in terms of furloughs and whatnot. We have seen the low-wage workers, significant migration across sectors in Covid—so sectors like hospitality and food service, non-durable manufacturing have been losing headcount, just partially ending up in fields like transportation and logistics. And that’s essentially workers jumping from one not-great job with some pretty unattractive attributes, to jobs where there’s more security and employment, more likely continuity, and, maybe you’re training some negative attributes for some new ones, but the net is you have a more pleasant job experience, and you’re making more money. Many of these workers also experience what a lot of us have in their industry, which is, a sudden acceleration of technology deployment, as people work to remote, or what had been four-wall retailing goes to remote retailing, et cetera, B2C online business model. So there’s been an infusion of the need to learn new skills in certain job classifications. But Covid also, of course, exacerbated a lot of the challenges workers had in staying employed or taking advantage of upward mobility opportunities. So those category of things we talk about under the employee’s life—their life conditions and the associated EX considerations—many of those seemed to have contributed to people essentially remaining in their previous job and not having as many advancement opportunities.

Kerr: Well, maybe a final question. Tell us a bit about the reaction or the response to the report and, perhaps, a glimpse is to what you see coming up next.

Fuller: The vast majority of the people we’ve talked to, companies we’ve talked to, including one of America’s largest employers, this week, is we need to understand this better. We are prepared to accept that there’s a much higher yield loss, if you will, but we’re trying to do what’s actually being experienced. Tell us more about idea of, how do you link this to our performance management metrics? Are there opportunities to use emerging offers, for example, in credentialing and certification—whether that’s provided virtually or augmented training, with things like virtual reality—that we can look for? So I think there’s a genuine spirit of inquiry on this, and not the shield arms popping up and people saying, “We don’t suffer from that problem.” I think though, it’s going to take us to some interesting places, and an area of research that we’re going to turn to now is coming back to this question of, “What are the attributes of a good job?” For the majority of people in the United States that don’t have a four-year degree—people that have certificates or middle-skills jobs—there are many operative definition of what’s a good job. And many of them focus just purely on the economics. It pays this much. It has health benefits. It has, maybe, paid leave, or a couple of other principal benefits. And those are very important indicia of what constitutes a good job. But thinking about attributes of a good job through lenses of, “Is this a job that’s cultivating transferable skills, so if from my employer goes out of business, or if I move my location, or if a new company comes to town, I might have an opportunity to take advantage of that? Is this a job that provides the opportunity for me to consistently grow my capacity, to be productive? For example, my capacity to learn new technologies and skills as that technology innovation cycle continues to spin at the rapid rate, so I’m not obsolete, or left behind, or left in that bind of, I’m a low-wage worker, I haven’t been being trained. So now my employer’s going to go into the spot market, take a good job that I might have occupied had I had few more skills or been directed to get to a few more skills earlier in my career.” What’s the lifetime value of a job? So there are a lot of attributes that I think are more structural that we’re going to try to understand better and that might bring a little bit more rigor to defining what constitutes a job, enabling us to talk to employers about, “Here are the ways that you could think.” Many of us are familiar with green certifications for buildings, that they’re standards that you can assess a building on that objective, the rationale that applies to everybody. A really sincere hope of mine is we could get that far, but I’m describing the victory parade and we haven’t fought any of the battles yet. So we’ll have to see how we do.

Kerr: Well, I look forward to that podcast when you have a few more of those battle scars and also victories to share with us. The report’s title is Building From the Bottom Up. Joe, thanks so much for coming and joining us today.

Fuller: Bill, it’s great to have a chance to talk all this through with you.

Kerr: We hope you enjoy the Managing the Future of Work podcast. If you haven’t already, please subscribe and rate the show wherever you get your podcasts. You can find out more about the Managing the Future of Work Project at our website hbs.edu/managingthefutureofwork. While you’re there, sign up for our newsletter.