- 01 May 2024
- Managing the Future of Work
Wharton’s Peter Cappelli on changing the talent equation
Joe Fuller: There’s a tendency to view certain HR and labor market practices as an expression of some form of economic law, but confirming their validity requires examining their origins and the current context. Such an examination suggests that some common labor practices are, at the very least, passe if they were ever valid in the first place. Why don’t employers invest more in their workers when doing so materially reduces turnover and yields an attractive return on investment? Why don’t business performance metrics encompass the value of human assets? Fundamental questions like those can get lost in news headlines and in the normal course of business.
Welcome to the Managing the Future of Work podcast from Harvard Business School. I’m your host, Harvard Business School professor and nonresident senior fellow at the American Enterprise Institute, Joe Fuller. It’s my particular pleasure to welcome a fellow academic, Peter Cappelli, to this podcast. Peter is a professor of management at the University of Pennsylvania’s Wharton School and a scholar whose work I’ve long admired. His latest book is Our Least Important Asset: Why the Relentless Focus on Finance and Accounting is Bad for Business and Employees. He argues that companies consistently underspend on their employees, and he goes on to suggest that such a strategy is actually a false economy driven by short-term considerations rooted in financial accounting. We’ll talk about how cost cutting in recruiting and training has raised costs by driving up voluntary turnover and leading employers to rely increasingly on the spot market for labor. We’ll also consider how changes in the C-suite and the human resource function have narrowed the basis for decision making. And we’ll talk about possible solutions, including accounting rules that take stock of how investments in human capital actually affect the bottom line. Well, Professor Cappelli, welcome to Harvard’s Managing the Future of Work podcast.
Peter Cappelli: Thank you.
Fuller: Peter, it’s a real pleasure for me to have you join us as a guest. I’ve admired your work for so long. You’ve been someone who’s pursued a very wide range of research in the general area of human assets, touching on lots of different subjects. What do you think are the most pressing issues for business leaders today as it relates to their personnel?
Cappelli: I think the most pressing issue, which is also the most pressing issue for the country—and one of the biggest ones—is that everybody wants to hire somebody with three years’ experience, and nobody wants to give them three years’ experience. If you look at the evidence about training, it’s just been going down, down, down, down. Continuing to go down. We’re not helping people learn how to get proficient in their jobs, and for sure we’re not helping them develop. We’re focused entirely on talent acquisition. You’ll remember the field of human resources shifted maybe 15 years or so ago. The hot topic was all about talent management. And talent management was making decisions about planning, what are you going to need, and how are you going to meet it with hiring, but also with training and internal development and things? And now it’s all about talent acquisition, which is just about hiring people from each other. And it just does not make sense. It doesn’t make sense.
Fuller: What are your thoughts about what triggered that change? It does seem like we’ve moved more and more decidedly toward strategies for playing the spot market for labor—rather than growing your own talent or investing in upstream talent pipelines so you’re trying to secure a robust pool of candidates—but rather just beggar thy neighbor, outbid the other people and, strangely, just outbuild on a comp[ensation] basis and a reputation basis, as opposed to putting forward robust opportunities so people have confidence they can grow with a new employer relative to what they had on offer in their incumbent employer.
Cappelli: Yeah. No, it’s a great phrase to say it’s about the spot market, because that illustrates what some of the problems are. And the problems in a spot market is, you sometimes don’t find what you need. And if you find it, sometimes the price is unbelievably higher than you think you should be paying. But I’d say the answer to the question as to why that happened was partly because, starting in the ’80s, companies began to dismantle their internal development systems. And the reason for that is they were downsizing. And it had some logic to it. There are some pretty big fixed costs to starting and running a good internal development program, but they were terrifically useful. One of our strategy colleagues documented a few years ago that the entire disk-drive industry could be traced to people from IBM who left the company. They were all extremely well trained, they knew the industry inside and out. So that’s one part of it. But I think the other, which is more pernicious, is that executives started to believe that this was simply the smart thing to do. I had a CEO tell me—this is maybe 20 years ago now—say, “Why should I train my own talent when somebody else will do it for me?” And then you started hearing companies say, “We want to get better at retaining our own people and hiring from our competitors.” Well, you could do that, but you can’t collectively do that.
Fuller: I’ve certainly heard also many times the inverse, which is I don’t want to spend a lot of money training people for them only to go off and work for my competitors.
Cappelli: Right. And if you look at white-collar jobs, the way you train people and develop them the best is through stretch assignments and giving people opportunities, which cost you nothing. Actually, it’s like an apprenticeship model. One of the reasons business education exploded was because companies stopped putting people into classrooms and spending a year or two developing them. Apprenticeships pay for themselves. So we could certainly do that, but we can’t even get resources to do that.
Fuller: It is, it does seem like a self-fulfilling prophecy, where companies create a shortage of skills that they then proceed to complain about. And you touched on something earlier, which also is confirmed by our research, which a lot of larger companies, frankly, they’re almost anorexic. They’ve been de-layered so much, they’ve been cost-reduced so much. And that brings us to the topic of your recent book, Our Least Important Asset. In that book, you make a very provocative argument that the way financial accounting practices treat human assets and account for the cost of human assets actually would reflect and reinforce this tendency you’re talking about. Could you explain your hypothesis to our listeners?
Cappelli: Yeah, it was inductive. I started out to write a book that was going to describe how we actually manage people, as opposed to what the textbooks say. Most of the textbooks about how employees are managed and the practices of things like hiring and training and promotion and things still look like they’re from the 1950s. If you think about hiring, the typical person who changed jobs at least a year or so ago, according to census data, said they were not even looking for a job. The job came at them, which is completely different. So I started to go through, practice by practice, and you saw what looked like the bones of the old model. There were things that were kind of recognizable, but it was all so stripped down that you just couldn’t see how it would even work, which ends up being incredibly penny wise and pound foolish, as they say—companies that don’t spend any effort or money to see whether the people they’re hiring are good or not. We focus on cost per hire, but we don’t look at quality of hire. It doesn’t make any sense at all. Why would you do that? And then I started to learn more about the CFO’s role and financial accounting and where these arguments come from. You think you have the ability to hire, and you’ve offered somebody a job. And then you hear this, “Well, we got a hiring freeze.” And you say, “Why are we doing that? The work is not getting done.” “Well, it saves money.” “Well, how does it save money to not fill vacancies?” I’m always reminded of this quote: “I finally weaned my mule off of food, and then he up and died.” I started to think about, where is this coming from? And then you start looking into financial accounting and how financial accounting treats labor. And maybe just to back up for a second, financial accounting has become much more important, because investors have become more important over the last 30 years or so. Shareholder value drives everything companies do, which means thinking about what investors are interested in. And financial accounting was set up in a period when human capital was not thought to be of much value; it was all about physical capital. And so financial accounting—which is based off of regular GAAP accounting set by the Financial Accounting Standards Board, FASB—does not see human capital as existing. Capital has to be physical, and assets can only be things you own. So if you don’t own something, can’t be an asset. If it’s not an asset, you can’t invest in it. So employees, I’m sorry, can’t be assets. You can’t invest in your employees, because they’re not owned by you. Even people under contract can’t be assets. Even a football team, with all those players under contract, no assets there. So you think about training, why don’t employers want to train? Well, first of all, investors can’t see your training; it’s buried under administrative expenses. And then if you think about how you could get better, one of the problems in terms of performance is that we measure lots of things on a per-employee basis, like profitability and revenue and costs, that sort of stuff, per employee. Which makes sense, except that you could have workers now without having employees. So you could have contractors or you could have leased employees. Some measures say 11 percent of workers are leased. They’re not employed by you, but they work for you. You become maybe legally a co-employer, but they don’t count in accounting as your employees. So suddenly my value of my company appears to just skyrocket, because I’ve cut my denominator on my revenue, or cost per employee. And so, wow, we look remarkably better. So there’s a series of enormous distortions that come out of the fact that financial accounting really just can’t deal with human capital and, frankly, has not seen any particular need to do anything about that. They’re perfectly happy with what they got.
Fuller: Peter, how do you account for the fact that, if we work at two of the world’s most distinguished business schools, we have lots of alumni who are CEOs in C-suites, renowned investors. And many of them, certainly the public company CEOs, I believe, will pass a lie detector test if you ask some questions like, “How important are people in your strategy?” And you’ll get pious comments that absolutely they’re our most important asset. Are they fooling themselves, or do they just not see this dissonance?
Cappelli: Well, that’s why the title of my book is Our Least Important Asset, is to play on the fact that so many executives and boards and things used to say, “People are our most important asset.” I think they understand it, that good employees would really matter. I think they also understand, though, that investors don’t care about that. And investors would really like to see your employment costs as low as possible. And as a result, you see all these really ridiculous penny-wise-and-pound-foolish practices. Here’s maybe one of the most egregious ones. If you think about what line managers have to do, line managers now do hiring, not recruiters. Most companies just have trimmed down their recruiters to almost nothing. Why would you have line managers, who may be paid many hundreds of dollars per hour, doing recruiting, when recruiters are paid maybe $40 or $50 an hour? It’s the exact opposite of industrial engineering efficiency stories. Well, the reason is because cutting headcount is really important, independent of efficiency issues. Back to the CEOs, I think they are the answer here, because the CFOs are pushing the financial accounting solutions because they are playing to what the investors want. That’s their job. The CEO is the person in the position of saying, “Well, wait, that approach, which makes us look better temporarily on financial accounting, is going to hurt us in the long run.” They’re the ones that have to think about efficiency. Some of the problem, I think, has been on the human resource side. There’s been nobody who’s willing to make that case and say, turnover costs, for example, incredibly expensive. This is one of the great scandals, is very few employers even know what their turnover costs are. And if you don’t know what your turnover costs are, and you think it’s cheap to replace people, you don’t care if people quit. You don’t care if you have to fire lots of people. It just seems trivial. Some of the problem has been the HR community has not been willing to step up and at least present gently the idea that what we’re doing is really not efficient. There’s all kinds of reasons why not, but they’re no longer very credible. HR people should be able to speak simply in the language of finance. It’s not that hard to generate measures of these things, even if they’re not perfect. But for some reason it just never happens.
Fuller: Do you attribute that to the fact that they’re not all that facile in the language of finance, that they’re really not given an opportunity to air their concerns? Because one thing that I find mystifying about what we see in big, sophisticated companies—I’m not talking about modest, midsized companies that are worried about just continuing their basic operations—that the logic of phenomena like Total Quality Management or supply chain management, where the company has configured its processes purposefully to identify systems effects, has never seemed to penetrate HR. That not only will they not know their turnover rate, they won’t do exit interviews to know why people are leaving. All the recruiter sees is more requisitions for more level-A pipe fitters, as opposed to what happened to the last 10 we sent down the line to operations. Why is this so constipated?
Cappelli: Yes, it’s a very good point. And it really is a puzzle. I think in the 1980s or so, the human resource function found themselves fighting a rear-guard action as restructuring advocates took down one component of HR after the other, after the other, after the other. And those people were not really prepared to defend it. They didn’t know much about finance, they couldn’t articulate these arguments. Their staff didn’t know much about it either, they never had to justify their practices and positions. And in terms of fields, HR was largely a preserve of psychology, which was not a field that thought about these issues or knew necessarily how to address them. I think, too, though, the way we think about CEOs is quite different in the last 30 or years or so. In the early ’80s and before then, CEOs delegated lots of stuff. And decisions would come to them to bless or not bless. But now it seems to me that mainly the way things work is people try to guess what the CEO already thinks, and then they go with that. Very little is really delegated.
Fuller: This seems to be doubly confusing, because we use popular terms now, like “post-industrial society,” “information society,” “technology-driven economy,” where one would assume that the premium placed on nurturing talent, growing talent, would be going up, not down. But it’s just been overwhelmed by, in your estimation, this laser-like focus on economic performance.
Cappelli: Well, financial performance is the big thing, right?
Fuller: Yes. Better turn of phrase. Have you seen companies beginning to confront this, or examples of companies that are breaking out of this cycle?
Cappelli: Well, you probably remember this, too, that a few years ago there was an awful lot of interest in alternative measures of accounting—trying to find outcome measures beyond simply what GAAP would allow you to do, and in a way to try to get at intellectual capital and other things like that. The problem with accounting is that, unless everybody does it, it’s not useful. Fortunately, there is a big important group of players who see the same issue from a different perspective, and this is the investor community. And the investor community says, because we can’t learn anything about human capital, we don’t know how to value all these tech companies. It’s not their equipment, but that equipment is the only thing you can actually see in financial accounting. It’s all human stuff. And we can’t see any of that. And as a result, it makes this impossible to value. Now, I have some sympathy for that, but honestly, I don’t care much about the investor side of the story. I think they can probably take care of themselves. But I am worried about the economic efficiency issues, because of these penny-wise-and-pound-foolish issues. We’re focused on cost per hire, but we’re not paying any attention to whether our hiring causes us to get better people or not. We can’t train people because that’s an expense that will hurt us in financial accounting, and there’s no way for investors to see that’s where we’re spending our money. We’re not blowing it on coffee. The good news is, the investor community has been on this for a bit. They have pushed the Securities and Exchange Commission to take more serious looks at it. Some of our accounting colleagues, some of yours and some of mine, are involved in that effort. And it’s a smart one, and it’s a reasonable one. The proposal out for comment is simply to require them to add a couple of measures. One of the measures is how much you’re spending on training. Another is to look at your total labor spending, not just your employee spending. But if you’ve dumped your employees and you’ve got a bunch of leased employees, that’s actually way more expensive than having employees, although it looks way cheaper in terms of financial accounting. There are two or three things they’re asking that the SEC make companies report that would change a lot of this.
Fuller: And companies have certainly been reluctant across the board to accept more reporting requirements. They talk about, it’s releasing competitively sensitive data, it’s useless, it’s not going to do anything but put more pressure on budgets, because now we have to report something that’s got no managerial import. There’s an awful lot of skepticism about it. As you look at the HR function, one phenomena that we’ve observed in our research is that there seems to be a growing trend with leaders of human assets in companies actually coming from the operations side of the business. They’re no longer people who came up on the talent side or came up on the total rewards benefits and compensation side, or even on the labor relations side. But rather, they have run operating units. And are you also picking up on that? And what would you think explains it? And is that a reason for hope or a reason to think that thinking more creatively about HR will become even harder to do?
Cappelli: Well, it’s a really good question. I think new thinking would go a long way. The canonical HR models are not particularly relevant anymore—how do you select people and craft selection tests, and all that sort of stuff. It just doesn’t seem to work anymore. Probably the biggest skill that HR leaders need now is managing vendors, and so much of everything is outsourced in various ways. So different thinking would probably be really important. I think it does matter a lot for somebody to understand human behavior. And there’s a long history of problems with having engineer-driven people handle employee management. And business schools in the executive programs have thrived for decades simply explaining to engineers that employees are not robots. But somebody who has an engineering background and has understood people issues would be really quite useful. Now, we used to see this all the time, because we would see executives who were groomed from within, they’d go through rotational assignments, and they would know about the whole organization and every function. But those programs are largely gone. And particularly, the early management training that people used to get when they left college and went to a big corporation, you get a couple of years of training. GE just dismantled its last big program, the financial management program, which was five years of training and development inside GE. So those are all gone. And I think the other problem is, we have now some behemoth tech companies that are started by people who had no management experience, and have never really run anything before, and no real understanding about people. If they had any discipline, it was engineering. And it takes them a long time to realize that you can’t run employees and manage them as if you’re a start-up when you got 100,000 employees. And not every solution is an engineering solution. So it takes them a long time. The problem is, the perception, that because these companies are so wealthy and been so successful, they must be doing everything right. They succeeded because of the opportunity was so great, and they got there first, and all kinds of other reasons which are sensible. But that doesn’t mean that everything they do makes sense or would make sense for other companies.
Fuller: You’ve done recent work on the evolution of the C-suite. I did some research with my colleague here, Professor Raffaella Sadun, about how what boards and executives were looking for in C-suite officers—in terms of the content of the job descriptions, the dossiers they were giving to headhunters—had changed over time. You’ve looked carefully at the work histories and the demographics of the C-suite. What you found there, could you share that with us? But also, do you think it offers some hope that there’ll be new types of thinking in the C-suite, or do those demographic trends and those work histories suggest we’re continuing to arc down this pathway you so articulately described earlier?
Cappelli: We started looking at the Fortune 100 top 10 executives of those 1,000 people in 1980, and then we followed them every 10 years to see who held those jobs and what changed. And not surprisingly, companies turned over their executives much faster. It peaked in 2001, actually, where these CEOs and top executives were younger than they had been 20 years ago, by about maybe eight years younger. Big numbers, seven or eight years younger. But that’s reversed now. Actually, they’re about the same age now they were in the 1950s. You see a lot more experience across industries, which sounds like a good thing. No longer thinking that you had to be an oil guy to run an oil company—and it had to be a guy, too, because now there are at least more women in that industry. It’s still not representative of the whole industry or economy on that one. So some of that is good. I think one thing we also see, though, is less experience across functions. So people tend to be more functional in their career path, which is probably not a good thing, since they don’t have a breadth of understanding how the whole organization works. One thing, which is a good sign, is that we’ve seen quite an expansion of foreign-born executives. It’s now up to about 14 percent, which is now the population estimate for the average number of people who are foreign born in the U.S. And, of course, there are more women in these top jobs, too, but the progress to the very top is still pretty slow. We did another study, a couple years ago now, where we found that companies really appeared to give women a push. They got there faster than their male counterparts to the top jobs when they, to think cynically, they really needed some women. So they had none in the top 10. The first couple that would get into the top 10 got there really quickly, but that didn’t last. After that, there was no more push. I think what we’re seeing, which is not so helpful, as we said before, is this notion since Lee Iacocca, that the CEO ought to be the person who is making all the decisions. They’re responsible for everything. In a legal sense, I guess that’s true. And Sarbanes-Oxley and other legislation literally makes that true. But the idea that they should be calling the shots on every aspect of the operation of a company seems really like a bad idea. Especially with respect to people management, where they typically don’t know much.
Fuller: Well, it’s certainly the case, as you point out, if someone in one of these anorexic organizations has basically performed brilliantly in one silo, marched all the way to the top, become the primus inter pares of their colleagues, but don’t have that significant understanding of the workings of other key functions. And also this trend toward hiring people across industries. I confess that historically I’m suspicious that they’re just great managers can go from one setting to another. But it certainly suggests that if you’re an effective manager in terms of managing company financial results and performance, if you know how to do that, that’s a skill that travels. It doesn’t matter if you’re selling hotel rooms or consumer goods or airline tickets, if I get the way the ratios work, if I know how to deal with the capital markets, if I know how to deal with a board. We’re speaking in the, I guess we could call it early spring, late winter of 2024, we are living in the age of generative AI. How do you feel human capital plays out in that era? And an era where it could be that the half-life of a lot of technologies, which is shortening dramatically, is about equal to the time it takes to master a new technology, which is a juxtaposition we’ve never seen before, where through things like software as a service, that many more technologies are almost disposable. New products are designed to be able to pick up almost seamlessly from the incumbent products, which really changes the composition of work, the pace and nature of skills acquisition. How do you see that all playing out?
Cappelli: Well, I think there’s long been a confusion in the pundit class between the complexity of technology and the complexity of the skills required to use it. It’s incredibly complicated pieces of technology, which are not all that hard to use, like a cell phone, for example, even though they’re getting hard to use well. I think that the question maybe is better for the future of managing people in the sense that we can’t expect to simply run to the local university and hire people who have the skill we need that is hot this year, which is what the tech world has been doing. And that’s why they’re constantly complaining there’s nobody to hire is because they want a Java programmer with mobile app knowledge to graduate from MIT or Caltech this year when we want to hire them. So we have to maybe start thinking of ways to, as they now say, upskill, or we would just say train people so that can continue to figure out these new technologies. I think though that the technologies that change, most of them are iterative. I think machine learning and generative AI have been an exception. If we look back over time, we don’t see the technologies that people in management would’ve used really change all that much in the period before. And, of course, generative AI and machine learning are quite similar. It’s just the data that they’re using, which is different—text versus numbers. And I think the other thing we need to think about, and maybe you and I especially on this, is to tell the story from the perspective of what work requires rather than letting technologists tell the story about their guess as to what technology will do. My colleagues, Sonny Tambe and Valery Yakubovich, and I have written one thing a couple of years ago about machine learning, and spelling out what some of the issues were for algorithms. And we’ve just finished one on generative AI. And we’re much less persuaded that these are going to be radical changes. I think with machine learning, what we have seen is it really hasn’t penetrated very far because it’s quite expensive to develop, and it’s not very flexible. Generative AI in the form of large language models (LLMs) is the opposite. It’s really easy to use, and you don’t have to develop it, because it’s already developed. But we haven’t thought very much about how it could actually be used. The thing that it is absolutely best at is writing college term papers. And the reason for that is because the questions we usually ask are ones that have been asked a lot already. And the answer, in terms of the paper, just has to be good enough. And the people who are grading, which would be us, have to know whether it looks good enough or not. But lots of things that we think ChatGPT and others could be used for, if you listen to the technologists, are really not that useful. For example, they say, “Well, it could write customer letters back, and it could do managed correspondence really easily and quickly.” But nobody writes those letters now, they’re all form letters. They go through legal first. You don’t get a unique response from a company if you write to complain about lost bags. And if you let ChatGPT or these others write it, you see the problem that Air Canada is in right now. Air Canada had some LLM-powered chatbots that were giving customers fares and discounts that the company didn’t want given. They were solving the customer’s unique problem, but the company didn’t want it solved. There’s lots of things that they’re not going to do and lots of additional work that they’re going to generate. Somebody’s got to figure out, for example, every time you use one of these LLMs to generate a report, is it a good answer or not? Is it good enough? Is it full of hallucinations or not? And that’s not an easy thing to do. If it’s easy for your organization to know the answer, you probably didn’t need it in the first place, because there are people there who already knew it. So if you’re looking for something novel and new, finding somebody who can tell you, “This looks sensible,” versus “This is weird” is a pretty difficult thing to do. There’s some others too. But here’s the last one that we think is kind of important. It’s also easy to drown in competing reports from the large language model. You produce one telling us that here’s what the curriculum should look like in five years. I don’t like it because it cuts my department out. So I go to a different LLM, and I put in the same prompt there and I get a different response. And not only that, I could go back to the same LLM a couple days later, put in exactly the same prompt, I won’t get exactly the same answer. So somebody has to adjudicate these competing answers from large language models. You start to see, if you’re thinking about it from the perspective of work, that even though it’s possible for LLMs to generate reports quickly and easily, it doesn’t really answer the problems that organizations need answering, not without a lot of help. I think that’s our kind of view. Start with the work first, and then you get a quite different answer.
Fuller: It’s evocative of some of the other technologies, spreadsheets, PowerPoint, that came in and seemed to have actually done more to paralyze and confuse decision-making processes in companies than contribute to them. Or at least, the scales were close enough in balance to make it an arguable proposition that these technologies that are supposed to improve analysis and communication and alignment did as much to create transactions and inhibit progress as they did to advance them. Especially when you’ve got this phenomenon where, by definition, if there’s no one in the company that can really contest the results of generative AI, then you get dueling positions from equally ignorant people. And this touching faith that somehow better prompt engineering will get the answers opposed to having management judgment and deep knowledge of how markets work and what organizations are capable of.
Cappelli: That’s a good point, because it just means if you get better prompts, you get more precise wrong answers. And does that help you much? Probably not.
Fuller: Well, Peter, as someone that follows your research closely, what are you excited about? What are you working on? What can we expect to hear from you in the coming months and year?
Cappelli: Well, I can’t say I’ve got a big plan. I’ve never been a person that was particularly structured, which is why my research has been all over the place. We do have an article coming out shortly about some of these questions, elaborating on the problems that CEOs have by ignoring people issues. The fact that morale is a problem, health of employees is a problem, and we’ve got tight labor markets. None of these things mattered very much when you were in a soft labor market. But it’s not a soft labor market, it hasn’t been for about five years now. And I guess if you were to bet, it looks like it’s going to stay that way. I think that’s the thing which is most immediately on our plate right now.
Fuller: Well, that’s a terrific question. Because certainly the supply of people with the types of social skills, technical skills, numeracy, that are so much in demand, that supply is very constrained. And unless we get a sensible bipartisan approach to immigration, which if you believe that’s going to happen, Peter, then I’ve got some bridges over the Charles River here that I’m forced to part with, I’d be happy to sell you one. The supply side is completely contrary to the way demand wants to work, which is to play the spot market and don’t invest in breadth of skills. Don’t invest in cultivating your own talent supply chains.
Cappelli: No, that’s a very good way to put it, that spot market just is not very good when, as a buyer, you don’t have bargaining power. And in a tight labor market you don’t. And that’s the game they’ve been playing, and I think they’re going to regret it going forward.
Fuller: Well, as a longtime devotee of my friend and business partner, Mike Porter, I have to embrace that. That’s a classic industry structure analysis point to wrap up our conversation. Well, Peter Cappelli of the Wharton School at the University of Pennsylvania, it was such a pleasure to talk with you and trade ideas about managing the future of work.
Cappelli: Thank you, Joe.
Fuller: 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.