Bill Kerr: Few institutions are more closely associated with society-changing technology than the Massachusetts Institute of Technology. But don’t look to the school for reflexive techno-optimism. Amid rising inequality and predictions of a robot apocalypse, the university recently undertook a deep examination of the impact of emerging technologies on the world of work. The MIT Task Force on the Work of the Future spent two and a half years delving into the history and economics of technological change. The group’s November 2020 final report is titled “The Work of the Future: Building Better Jobs in an Age of Intelligent Machines.” The report draws on the expertise of MIT faculty and colleagues at other universities, graduate students, and an advisory board representing business, government, education, and the organized labor sector. The Task Force notes that automation isn’t displacing humans wholesale. Indeed, the report’s authors state, “History and economics show no intrinsic conflict among technological change, full employment, and rising earnings.” They caution, however, that the impact on jobs and income has to be more equitable going forward if we’re going to avoid negative societal consequences.

Welcome to the Managing the Future of Work podcast from Harvard Business School. I’m your host, Bill Kerr. To discuss the sweeping scope and findings of the report, I’m joined today by Task Force co-chair David Autor. David is the Ford Professor of Economics at MIT and the associate head of MIT’s Department of Economics. Welcome to the podcast, David.

David Autor: Thank you very much, Bill. It is a delight to be here speaking with you.

Kerr: David, you’ve been a prominent and path-breaking labor economist for a few decades. And you were, for our listeners, my graduate adviser a while back at MIT. So I’m going to ask you a little bit of an unfair starting question, which is to summarize for us just a few key themes of your work before we get into the “Work of the Future” report. What have been some of the highlights of how you’ve been approaching labor economics over the last few decades?

Autor: Thank you for the question, and it was an honor to have you as a student, Bill. And now I feel we’re peers and more. Two bodies of my work have focused, one, on the impact of technological change on labor markets, and the other, trying to understand how globalization— particularly China’s rise—has changed job opportunity and employment. Both bodies of work emphasize that these forces, which tend to raise productivity and raise national wealth—both technological advances and trade with other nations—have very serious distributional consequences. Although they are good for nations as a whole, in the sense that they raise GDP, they’re not necessarily good for everybody in a nation. Technological change, as you emphasized in your introductory remarks, does not eliminate work in net, but it changes work a lot. And that creates winners and losers. And it makes certain skill sets obsolete, it causes needs for people to continually retrain, and it complements some skills and talents very strongly, while substituting for others. So it’s a mixed bag. Same with trade. In the work that I’ve done with Gordon Hanson of the Harvard Kennedy School and David Dorn of the University of Zurich, we have found very clear evidence that China’s rise—although great for world welfare, great for the world middle class, good for US GDP—had really adverse consequences for manufacturing employment in a lot of communities in the South Atlantic and the Deep South. And those were bitterly painful and politically consequential as well. So to summarize, I think economics teaches us that we should embrace technological change and trade as ways of expanding the production possibilities frontier—increasing the set of things that we can do with our resources. And that’s true. But we need to recognize simultaneously that that doesn’t intrinsically benefit everybody in a country. And without appropriate compensatory policies or policies that distribute some of those gains, some people are going to get much worse off, even as countries get wealthier.

Kerr: Some of your prominent work has been even measuring what is the degree of the distributional differences that we have in society—how has that changed over time. David, we have a course called Managing the Future of Work for our MBA students. And we teach in the second day of the course a paper of yours that has one of the greatest titles of all time. It’s “Why Are There Still So Many Jobs?” So I’m going to ask you to give the answer to our listeners.

Autor: I have to credit my wife for that title. I was struggling with what to call the paper; she proposed that, and I thought it was brilliant. So why are there so many jobs? Why—after two centuries of incredibly rapid automation that has moved us physically out of agriculture, in terms of huge portions of labor force, and out of manufacturing—why are we still doing so much work? So part of the reason that technological advance has not eliminated work is that it has created new demands. As we become more productive, as we become more affluent, we consume more. That’s it. I think that’s kind of the Economics 101 answer. There’s no question that technological change eliminates large sets of tasks and passes some work that was done previously by humans to machines. And so we think of technology as often just eliminating jobs. Simultaneously, a lot of what technology does, or automation does, is it gives us new tools to make us more productive in the things that we are still doing, the things that are not automated. So roofers use pneumatic nail guns to hang shingles. Doctors deploy batteries of tests to make diagnoses. Teachers deliver lessons through telepresence. Long-haul truck drivers upload their route parameters to cloud dispatching platforms to ensure that they never ride with an empty load. In those ways, technology doesn’t just eliminate part of our work, it makes us much better at the part that remains. And so that increases our value and increases the demand for our services. It doesn’t make them less necessary; it makes them more. And the third part of the answer is on the creation of new work. It turns out that a lot of the work that we do at present is work that did not previously exist. If you look at the set of jobs in 1940 and compare them to the set of jobs in 2018, you’ll find that only about 35 percent of employment in 2018 is in job titles that were present in 1940. The other two-thirds are in new titles that have been created in your intervening eight decades. What I mean by new titles, I mean things that people report themselves doing the Census Bureau had never heard of, and so had to create a new entry into its alphabetical list of occupations. So you might see a pediatric surgeon or an artificial intelligence specialist or a remote drone operator. As we become more technologically advanced and as we become wealthier, we proliferate new activities, many of which require specialized skills and expertise that didn’t previously exist. We actually create new types of work that require people to learn, to educate, to specialize, and to become fluent, to become talented. And that is an important source of the growth of work over time.

Kerr: Two and a half years ago, MIT was going to create this Work of the Future group. And now I want you describe for us a little bit the origin and the mandate. And with this two centuries of having plenty of jobs remain despite technological progress, was there a worry that this time was different?

Autor: MIT’s President Rafael Reif really set this into motion. And what President Reif was responding to was a sense in the kind of popular press that things are changing very rapidly—robotics, artificial intelligence, autonomous vehicles. And so a great deal of public discussion and a certain amount of public alarm about this, in a sense, that MIT had a role, a responsibility. MIT is innovating a lot of these technologies. Shouldn’t we be thinking hard about what they’re doing? How? And how worried should we be? And if so, what should we do about that? The mandate was open ended. But one job was a positive one—descriptive—what is happening, what is changing, how fast? What are its implications for what machines will be doing, for what people will be doing—and the challenges and opportunities that come along with that? And two is, what are the sets of policies and institutions and investments that will help us to make the most of this opportunity and to mitigate any possible downside consequences? And our report is simultaneously very optimistic and very cautionary in pointing out the many possibilities that are open to us, and the many ways that we squandered possibilities over the last four decades. And one conclusion we reach is, if we continue on our current trajectory, we’ll have lots of productivity growth without a lot of shared prosperity. And it’s not a direction we want to continue in. We want to change direction—not technologically, but in terms of the economic consequences of some of those technological changes.

Kerr: One of the framings of the report is around the great divergence that you just described, where technology has helped increase productivity and raise the overall GDP [and] has not led to mass unemployment. But it also has not been shared equally. Can you break this down a little bit? What are the sources of that inequality? And was it inevitable? Was it something that the technology and globalization had to produce?

Autor: As shown by the Nobel Prize-winning work of MIT Professor Emeritus Robert Solow, a lot of productivity growth comes from technological advances that allow us to be more productive—i.e., to get more out of the same set of inputs or get the same amount of stuff with less work. And so it is what historian Joel Mokyr calls “the lever of riches.” With the end of the Second World War, productivity growth in the US and throughout much of Europe was extremely rapid. It was also extremely evenly shared. The average wage of workers rose in lockstep with productivity growth, as did the median wage workers. If you look in the data after the mid-1970s, the big change relative to the prior three post-war decades is not the end of productivity growth—it definitely grew less rapidly over the four decades and before then. The big difference is how that productivity growth is being distributed. In general, the average worker compensation has moved upward with productivity growth—not fully, but mostly. However, the median has largely stood still. So another way of saying that is, the distribution of those gains has been so unequally shared, so much inequality, that most of those gains are going to the top 1 percent, to the top 10 percent, to the top 25 percent. And very little is reaching to the middle or below. And that’s what people call the “Great Divergence.” And I think it’s an important factor underlying why people are so anxious about the technological change that is ahead of us. So then you ask the second question: Is that inevitable? And we would say the answer is “no.” How do we know that? Well, if we look across the countries to which we compare ourselves—typically, many European countries, some Asian countries—all of them have faced the same set of challenges and opportunities as has the US: changing technologies, rising globalization, demographic changes. And many of them actually have faced them more strongly than the United States, because these are more-open economies. And yet, very few have seen this very strong divergence between the path of productivity growth and the path of median wages. And that’s because they’ve pushed back in various ways. So they have managed to take the same forces of rising productivity and change their distribution—not radically differently, but enough that there’s more shared economic growth, and as a result, more economic security, and in many cases, less fear of what’s to come. So if you talk to the typical Swedish worker and say, “Are you worried about robots?”—and Sweden has a lot more robots than does the United States per capita—they would say, “No.” Paraphrasing the survey of it, it says they would say no, because the view is, “Sure, those robots may take some jobs. But if they make us more productive, then we’ll be wealthier. And if we’ll be wealthier, then I presumably will get higher pay.” And I don’t think there’s any presumption of that sort in the United States. The fact that your company is getting more productive does not necessarily mean that your pay will be rising. In fact, it very well may mean that your job is no longer needed. So we would say that this divergence wasn’t inevitable, but it’s very costly. We pay a high price in economic insecurity and people’s fear of the future, because they’re not afraid of the technologies, per se. They’re afraid of what those technologies will do to their ability to earn a secure living.

Kerr: David, sometimes it’s said that for the US to really be the technology leader, to really be at that frontier, even relative to Sweden, we’ve got to have a greater amount of inequality, that we have to provide these amped-up, big entrepreneurial incentives for the innovation that happens. Is there anything to be said about that being just at that very leadership position, that there has to be some amount of trade-off?

Autor: That’s a question that we grappled with. Let’s put it this way. In the first four decades after the Second World War, the US was the world’s unquestioned innovation leader—the one that had the most startups, the most new ideas, the most new technologies. We were a much more economically equal country at that time. In the subsequent years, we’ve become much more unequal. Yes, we’ve remained innovative. But it’s not obvious that our current innovation is superior to our prior innovation—arguably, it is not—or that the current era of innovation stands on the shoulders of all this inequality that has come in in the interim. I would argue that it doesn’t. I should caveat that, when I speak of greater equality in that era, I’m referring, unfortunately, only to the subset of workers who were allowed full participation in the economy. Unfortunately, at that time, that did not include many minorities, particularly African Americans, and women. So, greater equality only by one limited metric. There are several ways to see that the US seems to be paying a high price for its level of inequality. One is, we have fairly low rates of labor-force participation. The fraction of working-age men and women who are actually working is lower than in most other rich countries. You would think, if we had a great or extremely well-functioning free market, that fraction would be higher. A second is, we have low intergenerational economic mobility, meaning the likelihood that a child from a poor family becomes a middle-class or high-income earner in adulthood is much lower in the United States than in many countries to which we compare ourselves, including Canada. So as your colleague Raj Chetty likes to say, “If you want to realize the American dream of going from rags to riches, the best thing you can do is be born in Canada.” Seems ironic. Finally, the US does not have substantially more rapid economic growth than all of these other countries that we look at. Over the last five decades, the US has grown about as fast as you would expect it to, given how wealthy it was initially. And wealthier countries actually grow more slowly, because they’re already out at the frontier. So I agree with the thrust of the question—that the US does have a uniquely entrepreneurial economy, we have a great venture capital system, we have an ability to scale ideas, we have a lot of the world’s leading tech firms. And it would be a terrible tragedy for us to lose those things. But I do not see any evidence that those things are an outgrowth of the very high levels of inequality that have mushroomed over the last four decades, because we had things like that previously. And I would argue that, in fact, we are squandering opportunity by failing to invest in a good chunk of our citizens and by putting up with high rates of societal dysfunction that inhibit people from realizing their potential and for contributing to that mission of innovation and growth.

Kerr: And, David, if you had to give a couple of recommendations that the US could learn from Europe, from the European workforce model, what would those be?

Autor: One is a little more worker voice in setting the conditions of pay and work. In most other countries, there’s a much more harmonious relationship between labor and management that often involves workers in the boardroom helping set decisions to some degree, or at least having a voice. And also bargaining is often more sectoral—meaning it’s not one firm at a time, but it’s the wage setting or working conditions in an industry or in an occupation. So one is, we need to innovate in worker voice. Another is that our tax policy excessively subsidizes automation relative to investment in workers. This is something that my colleague Daron Acemoglu—and your frequent co-author and also another of your esteemed MIT advisers—has worked on a lot—that our tax code is increasingly skewed in a way that causes firms to distort their decisions in the direction of automating rather than investing in workers. It’s like in a condition where a firm would say, “I could do this with labor, I could do this with machines, [it would] be about equally costly.” Then they look at the tax code and say, “The federal governments can go in with me 20 percent if I do this with machinery, and not so much with labor. So I’m going to move in that direction.” So changing our tax code would help. A third, of course, is education and training. And a fourth is actually a continued commitment to innovation. The US has actually dialed way back on public investment in innovation. When the government invests in innovation, it does two things: One is, it stimulates a lot of follow-on innovation by the private sector, which is good for growth, good for opportunity. Also it shapes the direction of that innovation toward productive purposes. And often what’s most productive over the long run is not what’s commercially most attractive in the short run. And so the public sector can play a very constructive role in steering innovation in useful directions. And currently, the US has stepped back from that—a long way.

Kerr: You advocate a lot of reforms in education and training programs, with the goal of supporting working lifelong skills and their upkeep. What are you highlighting as the main pieces that you need to make that work?

Autor: It’s no longer the case that a high school degree is really a marketable credential. It’s a foundational credential to get some other substantial skill set that then allows you to get a reasonably well-paid, reasonably secure job. But if you’re not going on for college, it’s just the Wild West out there. It’s very difficult for people to navigate. What skills should they learn? Where should they get them? Should they go to a community college, to a for-profit college, to some other type of nongovernmental organization that provides those skills? And how should they fund that? So we want to see education improved for people heading into skilled vocational work and adults who are reinvesting, reinvigorating skills, or changing careers throughout their working lives. That’s not easy to do. It involves a mixture of private and public actors. Sometimes we have really excellent examples, like the community colleges in South Carolina working with government and industry to do skilled training for and feed the growing industrial manufacturing sector there. Some of it will involve new forms of education delivery. So one thing that is very clear to us, as we read up on this topic and study what was going on is, we are in an age of educational innovation, where access to education should be getting much cheaper, much more convenient, and a lot more engaging than it has been historically. For most adults, going back to school is just deadly. They’re too old to be sitting in classrooms all day. They have family responsibilities; they can’t just go and be in a classroom the exact same time every other day, or every week. And their attention spans are not such that they really want to be watching people writing on blackboards. But none of that is going to be necessary going forward. More content will be delivered to your home, to your screen. It hopefully will be better produced than your typical college lecturer. And on top of that, it should be more engaging, because it will be done not just through reading and watching TV, but hopefully through virtual reality and augmented reality, where people can have many of the experiences of doing something, even without that thing physically in front of them. Right now, the greatest strength of the US educational system for adults is that it’s flexible—there are many, many points of entry—but it’s also disorganized. And so providing the venues, certifying their quality, providing certified credentials that allow people to invest in skills and use them elsewhere, and also providing more economic incentives for employers to make some of these investments in workers—this can’t be fully an employer’s responsibility. But we, for example, give employers a tax credit for investing in R&D. You could provide a tax credit for firms to make investments in workers if those investments led to and paid for recognized credentials.

Kerr: I’ll give a plug for our podcast series. We have a lot of innovators in the apprenticeship education space that are trying to take their experiment and now scale it up. So this is a great place for us to try to collaborate to help build this education of the future. But I want to start by asking you to share some of your recent work about cities and how you have emphasized that big cities are no longer that escalator in terms of moving up the income distribution or getting the better job that they used to have. Can you talk us through what changed? And then, how has that impacted our economy?

Autor: I was looking at the hollowing out the well-known polarization of work—meaning the decline in middle-skills, production, and clerical administrative jobs—and the simultaneous growth of, on the one hand, high-wage, high-education, professional, technical, and managerial jobs, and other hand, low-paid, typically low-education, personal services, food service, cleaning, entertainment, recreation, security, also home health care. And I was wondering how that polarization has played out across space, geographically. And I’d assumed it was happening everywhere, uniformly. And what I was really amazed to discover is that a lot of this decline in middle-skills jobs reflects the unwinding of a specific feature of dense urban areas that was present some decades ago and has been disappearing—from essentially in the 1980s to the present. And that is that cities offered higher-paid, more-skilled occupations to people without college degrees. So as you moved up the population gradient for our density gradient—from rural areas to suburban areas to metro areas to cities—you would see people without college degrees moving out of manual work, moving out of agriculture, and moving into offices, moving into factories. They were doing skilled work alongside managers, alongside professionals, alongside engineers. And they were earning much more in those jobs. So in other words, there was this set of opportunities that were found in cities—that were not found elsewhere—for workers without college degrees. And that’s, I think, where the idea comes from that there’s this escalator of urban opportunity—that if you go to the city, that’s how you get on that escalator. What has happened over the last several decades is that those middle-skilled jobs—those office jobs, those production jobs—have eroded. Obviously, we have very few factories left in cities, very little manufacturing work left over all the United States. But also a lot of that office work has now been automated, and those who remain doing office work are those with four-year college degrees. They do much more complex jobs. And so cities have bifurcated, where they are very heavily populated by, of course, professionals at the frontiers of their industries, whether that’s in design, whether that’s in health, whether it’s in research, whether that’s in finance. And then, most everybody else is basically doing food service and cleaning and security and basically seeing to the comfort and the safety and the health of the more affluent. And as that has occurred—as the occupational structure in urban areas has bifurcated—the wage premium that non-college workers receive in urban areas has very steeply contracted. So not only is there no longer this differentiated occupational structure between cities and non-cities, but also there’s no longer a differentiated wage structure between urban and non-urban areas for those without four-year college degrees. And this is particularly pronounced among Blacks and Hispanics, who are overrepresented in urban areas.

Kerr: And so, David, let’s now turn also and incorporate the R&D theme, the public money and federal investment in R&D. In your report, you talk about the work of Simon Johnson and Jonathan Gruber, who are both at MIT and authors of Jump-Starting America. They were on our podcast last year. Do you advocate that approach where it’s not only about increasing the R&D level, but attempting to build more-effective regional cities throughout the country that can be new innovation hubs, not just the superstar cities that had been dominating patenting and venture capital and so forth for the last couple of decades?

Autor: So I found Simon Johnson and Jonathan Gruber’s work very thought-provoking and influential. And they suggest that a lot of benefits of innovation aren’t just technologies to create, but the type of work centers that are associated with it. Universities are a great example. Universities—not only do they employ scientists, but they bring students into town, they create all kinds of demand for services. They actually impose a perpetual age structure, where they never get older—the faculty get older and retire, but you’re always replenishing them with the young. And so there are all kinds of local economic benefits of innovation clusters. And they advocate spreading those out more. So I like the idea of trying to use innovation money to create broadly shared benefits beyond simply the innovations, themselves.

Kerr: As we record this, it’s December of 2020, and we’re still in the midst of Covid and lots of remote work. There are stories about Goldman Sachs contemplating moving some of its divisions from New York to Florida. Rents are coming down in many of the older, high-priced locations. Have you thought about how overall are we doing on the superstar city phenomenon beyond investment into the heartland with R&D? Do you think we’re going to have a new normal that’s going to emerge about the benefits of co-location versus being able to tie in remotely?

Autor: I do think, when we come out of the Covid crisis, we are going to be in a new era where people choose to locate their work. And the most critical piece of that change has to do with telepresence—the idea that we can be in two places at the same time—one physically and the other cognitively. The difference is not the technology, per se, or even our knowledge of how to use it, but the coordination problem that we’ve solved. Because it wasn’t enough for one or two people to say, “Hey, I can just connect remotely.” We need to all simultaneously agree that that’s a permissible and normal form of doing our work—is not being physically present but just being telepresent. The Covid crisis has solved that collective action problem for us. It’s caused us all to invest in technology, invest in skills, and all agree that this is a feasible way to work and to learn that actually is relatively productive. And I think that has lots of long-term benefits. One of them is that I hope it will allow for some of this urban de-densification. It’s not a healthy thing when so much of the GDP growth in the country is coming from just a few superstar cities—coming from New York and Los Angeles, San Francisco and Houston and Seattle—and comparatively, so little of the growth is coming from other places. We would like people to be able to spread out. And I think the reason so much of the wealth is in those cities is because people feel they have to be in those locations to do productive work, to be highly paid. When someone that spreads out, I think it could actually be politically and socially healthy for there to be more mixing of education and income groups. And so I’m optimistic that there are big productivity consequences to come and other social benefits. But it will be a painful transition. Obviously, it will be hard on those cities. It will also be hard on many of the service workers we were just speaking of who perform a lot of the support jobs in those cities. When you decide to telecommute to your job instead of driving to your job, that means you’re not buying coffee along the way, you’re not going out to dinner at that expensive downtown restaurant, you are not demanding the work of building security services and of building cleaning. And so, as in many cases—the decisions of the highly paid workers about what to consume, where to live, what to spend their money on—drive the employment opportunities of the less affluent, who essentially provide the ancillary support services for that type of work. And so we will be asking them implicitly to make large adjustments in where they are going to locate, where they’re going to look for work, and even what type of work that will be. So in the long run, I think this is going to have many unanticipated and certainly unseen benefits. But in medium run, it’s going to be a challenging adjustment, not only for cities but for many of the less-educated service workers who are living in those cities and make their living from the work patterns and lifestyles of highly paid professionals.

Kerr: David, maybe one final broad content question: We hear a lot about gig work. What did you guys see as gig work’s role in the future of work?

Autor: So that’s a question that we did not fully address. We talked about how, in many ways, the structure of benefit systems has frayed and decayed as work has evolved. So for example, in the United States, only about a third of people who are unemployed actually collect unemployment insurance benefits. And that’s because, in many ways, our system has become less comprehensive and less well-geared toward the type of work people do. So for example, in many states, if you’re a minimum-wage worker and you work full time for half a year, you still haven’t earned enough to qualify for unemployment insurance benefits should you lose your job—even though, if you were paid twice as much, you would qualify within three or four months. So that’s a problem. Gig work is a little harder, because when we don’t have an employer-employee relationship, it’s much harder to regulate and even know what people are doing. So what that means is that we need to think expansively about how to tailor the social safety net in a way that it provides both security and appropriate incentives for people who are not tied directly to a stable one-to-one relationship with an employer. I think it’s almost inevitable that will be a larger fraction of workers going forward, especially in this now telepresent labor market. Many people will be providing services to more other businesses as freelancers, or “e-lancers,” from their homes. One of the first really important steps we could take to facilitate that process would be to decouple the provision of health insurance from direct-hire employment. The US has a uniquely dysfunctional system of health insurance provision that is an artifact of historical events during the Second World War and has made many people dependent upon their job to have access to health care, so that when you lose your job, the insecurity is greatly magnified. There’s no reason for us to do it that way, and in many ways, it imposes costs on employers and creates a wedge between the cost an employer pays and the wage a worker actually sees. So, that one step alone would help a lot with the movement into gig work, because it would cover one of the fundamental needs of every individual and household, which is guaranteed health insurance security.

Kerr: David, this’s been quite a journey over topics, and maybe I can end with a question that’s about how Covid inserted itself a little over halfway, maybe three-quarters of the way, through the development of this impressive report about the future of work. Did the pandemic shift the priorities? If so, how? And now that we have the announcement of some very needed vaccines, how do you anticipate this evolving the future of work conversation over the next year or two?

Autor: When the pandemic became very severe, we met as a task force and said, “Look, should we just hang up our skates here for a year and wait and see what happens? Maybe what we’re doing is not so relevant.” But then we looked around and we looked, for example, at the passage of the CARES Act, which was the relief act that Congress passed that basically gave money to businesses, to households, and to the unemployed at an unprecedented scale, and not only generously but comprehensively to even people who wouldn’t previously qualify for unemployment insurance. And we said, “Well, this really shows how much institutions matter, how much governance matters in responding to challenging circumstances.” And that had been the theme of our interim report, which was called “The Work of the Future: Shaping Technology and Institutions.” And the thesis of that report had been how critical collective social choice and governmental institutions and societal institutions had been in shaping outcomes. And so we felt we were actually going in the right direction, and that our final report would acknowledge the Covid crisis and say that it reinforced the lesson that the choices we make collectively as a society are at least as important as the technological and global forces that we face in shaping the outcomes we want to see. So, that actually gave us renewed conviction that we’re going in the right direction. Now in terms of looking forward in the post-Covid labor market, I do think it will look different for some of the reasons we just spoke about. One, of course, is this change in the character of cities, the rise of telepresence, we will see an enormous decline—permanently, I think—in business travel, people just no longer going to cross continents for 90-minute meetings. And so that means, also, a decline in all kinds of hospitality services, and in limo rides, in expensive hotels, where people stay on weekdays and go out to fancy restaurants and expense accounts. And it’s also going to move, it’s going to pull technologies from the future into the present. Things that we would have discovered or been using widely 10 years from now are now already coming online as a result of adaptation to this challenge. So I expect the pandemic will have shifted both the type of work we do and the way we do it and, hopefully, our priorities about the importance of assuring economic security, of responding to big societal challenges with big public efforts to collectively produce outcomes that we want to see. So I’m cautiously optimistic—not certain at all—that these events of the last year will reinforce for people the sense that we have a lot of agency in the way the future unfolds. It’s not happening to us; it’s something that we create. And by recognizing that, we can take the possibilities and invest wisely to create a set of outcomes that we view as desirable and just.

Kerr: David Autor is a professor of economics at MIT and was the co-chair of MIT’s Task Force on the Work of the Future. We highly recommend looking at their report “The Work of the Future: Building Better Jobs in an Age of Intelligent Machines.” Thanks so much, David.

Autor: Thank you so much, Bill. That was a pleasure speaking 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/managing-the-future-of-work/. While you’re there, sign up for our newsletter.