Member Spotlight

David Autor and Michael Stepner

Last updated: February 19, 2021

Date of recording: December 15, 2020

Michael Stepner, who completed his PhD in economics at MIT, received the Academy’s 2020 John Heinz Dissertation Award for his “Essays on Health and Social Insurance.” Later in 2020, Academy Member David Autor, MIT Ford Professor of Economics and one of Stepner’s dissertation advisors, was recognized by the Heinz Family Foundation with a special 25th Anniversary Heinz Award. The Academy’s Distinguished Visiting Fellow, Fay Lomax Cook, spoke with both scholars to learn more about their work on inequality and social insurance, their influence on each other, and big research questions for the future given the impact of COVID-19. Listen to the recorded conversation (30 minutes) and read the full conversation below.

(The winner of the 2021 Heinz Dissertation Award will be recognized at a ceremony as part the Academy’s March 2021 policy conference, Pathways to Economic Security.)

About this Interview

David Autor is Ford Professor of Economics and associate head of the MIT Department of Economics. He is co-chair of the MIT Task Force on the Work of the Future, an Institute-wide project examining technology, employment, and labor policy. His scholarship explores the labor market impacts of technological change and globalization, earnings inequality, and disability insurance and labor supply. Autor is an active contributor to the policy discussions around human capital and labor force participation. He has received a number of awards, including the National Science Foundation Career Award, the Alfred P. Sloan Foundation Fellowship, and the Sherwin Rosen Prize for outstanding contributions in the field of labor economics. David Autor earned his Ph.D. in public policy from Harvard University’s Kennedy School of Government. He has been a Member of the Academy since 2018.

Michael Stepner Michael Stepner is an economist studying public finance, health, and social insurance and will be an assistant professor at the University of Toronto starting in July 2021. Stepner received his PhD in economics from MIT and is currently a post-doctoral fellow at Harvard University. Stepner is collaborating with a team of researchers at Opportunity Insights, studying health inequality around the world. (Learn more about his current research on his website.) He became an Associate Member of the Academy after receiving the Academy’s 2020 John Heinz Dissertation Award.

Fay Lomax Cook is professor emerita of human development and social policy at Northwestern University and faculty fellow of the Institute for Policy Research (IPR), where she served as director for 16 years. From 2014 to 2018, Cook was assistant director of the National Science Foundation (NSF) and headed the Directorate of Social, Behavioral, and Economic Sciences (SBE). Her research focuses on the interrelationships between public opinion and social policy, the politics of public policy, public deliberation, and the dynamics of public support for programs for older Americans, particularly Social Security. She has been a Member of the Academy since 1990.

About the Academy’s Heinz Dissertation Award

The John Heinz Dissertation Award was established in 1993 in honor of the late Senator John Heinz, who was a founding Member of the Academy and an advocate for healthcare reform and social insurance.  With support from Teresa Heinz and the Heinz Family Foundation, the Academy’s annual Heinz Dissertation Award recognizes and promotes outstanding doctoral research by new scholars focusing on policy questions in social insurance and related areas, including health, aging, and economics. It is one of several Academy programs designed to promote scholarship in the field of social insurance and encourage a new generation of scholars, administrators, and other professionals essential to a strong social insurance infrastructure.

Senator John Heinz was an early advocate for health care reform, seeking to control costs while improving benefits for all Americans. When the National Academy of Social Insurance began operating 30 years ago, founder Bob Ball saw a policy environment filled with misinformation and confusion. He reached out to others to form a bipartisan effort to counter the confusion. His first Board of Advisors consisted of Senator John Heinz (R-PA), Senator Patrick Moynihan (D-NY), Lane Kirkland (AFL-CIO), and Alexander Trowbridge (National Association of Manufacturers). Today, the Academy is a community of over 1,100 of the nation’s top experts on social insurance.


Fay Lomax Cook:

David Autor:

Congratulations to both on your Heinz awards. I think this is the first time the winner of the Heinz Award was on the dissertation committee of the winner of the Academy’s Heinz Dissertation Award!

You both know about Senator Heinz, who was concerned about the state of working conditions and how they were affecting workers and families, and how social policies can really make a difference. To begin our conversation, could you talk about some of the key takeaways from your research that you think Senator Heinz would appreciate?

I know Senator Heinz was concerned about forces affecting the working class, individuals who did not necessarily have a college degree and were highly exposed to the forces of technology and globalization.  I think the work I’ve done that would most resonate with Senator Heinz would be the impact of rising trade pressure with China, especially after China’s entry into the world trade organization in 2001, on the plight of workers in manufacturing. Pennsylvania, which Senator Heinz represented, has and had many manufacturing workers, and in my research I pull out specific towns that have been directly affected.

An important piece of my work I’m particularly proud of is calling people’s attention to the fact that international trade, which economists favor and which I favor as well, has aggregate benefits, but very strong and stark distributional consequences. Economists have not paid enough attention to those distributional consequences. As a result, they may have helped spearhead policies that while making  some nations wealthier have made some people poorer, without preparing adequately for that. I think that Senator Heinz would very much want us to take that into account and think about the policies that insulate and provide social insurance to workers even as we embrace the benefits of full trade policies.

Fay Lomax Cook:

Michael Stepner:

I totally agree with you and think that he would resonate with that research. Michael, what about you? What do you think Senator Heinz would especially appreciate about your work?

I think in the most recent months Senator Heinz would probably be concerned with precisely the same problems so many economists and truly everyone is concerned with, which is the state of the economy in the midst of the COVID-19 crisis. Particularly when thinking of the state of working conditions, and how that’s affecting families. I think about how, as the economy has started recovering from the depths of the recession that we saw in mid-April (2020), we have really seen a bifurcation of economic activity.

The recession has ended for rich Americans. For the upper middle classes, employment and income has returned to normal levels, but for people at the bottom quarter of income distribution, about one in five are still out of work. There’s been no recovery in that group since mid-July (2020). Large portions of American families are still suffering, while the upper classes are doing just fine. One real concern for me is what that might portend for the future.

Take retail workers as one broad category, for instance. David Autor’s work has shown that, increasingly, the economy for low-income workers has been concentrated in the service sector. Despite the fact that retail spending was up toward the end of the summer and into the fall—actually higher than a typical year—only high-income workers have returned to their normal levels in the retail sector, while low-income workers are still out of work in disproportionate amounts, about one in five. I wonder whether this is going to be a short-term adjustment, and as the vaccine rolls out and the economy returns to a new normal we will see a return to full employment. Or will this reflect a more long-term technical change as businesses shift to new modes of production? That open question is one I’m especially interested in and I’m inspired by my advisor David’s work on.

David Autor:

Michael Stepner:

I share Michael’s concern and I do think that we’re in for a big change that will be highly disruptive. Although there has been some growth of work in services such as food, restaurants hospitality, and business travel, will they ever return to “normal”? Why would they, given how we’ve all embraced telepresence?

We’ve had a lot of growth of low-paid work and that, ironically, is beneficial because it creates competition for those workers. When the pandemic is over, more workers will do their jobs telepresently, fewer workers will commute to offices, and fewer people will take business trips to marque cities. That’s going to put downward pressure on demand, wages, and employment opportunities for workers who already were towards the bottom of the labor market performing personal service jobs.

That really highlights the importance of social insurance, because this is a group for whom work has disappeared and no matter how hard they want to find a job there are none available. At the beginning of the crisis, we saw massive expansion of social insurance; in fact, the rate of poverty went down in the United States. But now we’re in a period where not only has the expanded social insurance expired, but regular social insurance—unemployment insurance—is also starting to expire for workers who were laid off at the beginning of the pandemic even though their jobs haven’t recovered. For those workers, the social safety net has been really important in 2020 and the need might extend into 2021.

Fay Lomax Cook:

Michael Stepner:

That’s a depressing story you’re telling, but it’s so important…For a moment though, I’d like to go back to the two of you. You were both honored by the Heinz Foundation, but on the surface the research each of you is doing now seems a little bit disparate. What are some of the things you each see as underlying some of your different bodies of work?

Comparing my research work to David’s work, I think back to the first time I sat in David’s office before I even started graduate school and we chatted about our shared interest in the effect of global trade on the American worker. Through all of our conversations over almost a decade now, we always seem to return to the topic of big economic forces that are buffeting American workers.

One concrete example of how David’s work has affected mine is I remember reading a working paper that he wrote with Jon Gruber and Mark Duggan focused on how short-term disability insurance provided by employers is affecting social security disability insurance provided by the federal government through the Social Security Administration… Focusing on the big questions has led both of us to study disability, and in recent months David and I have spoken about the effects of the recession and some of the social policies the government has passed to try to help support American workers through that.

David Autor:

Fay Lomax Cook:

Michael’s work on short-term disability and how it can provide an onramp to long-term disability, not always in a way we’d be happy about, is terrific work. I think something we really have in common is zooming out and looking at major forces and demographic and economic trends that are highly consequential. There are a lot of economics that look at very fine details, and that can be valuable, and you have to look there. But Michael’s work—on the inequality of mortality, on disability, on social insurance provided by the Canadian tax system and health care system—focuses on big questions.

Michael and I have spoken about collaborating on another big question, which is changes in the urban environment and how that is affecting employment. I come to economics from a policy background, and have worked directly in education and public service, so I’m drawn to the questions that have a concrete effect on people’s well-being, livelihood, and opportunity. That’s why I admire Michael’s work and I feel it’s something we both aspire to.

We’re always interested in what a graduate student might learn from his advisor, but we’re also interested in what advisors learn from graduate students. So David, you’ve worked closely with Michael when he was at MIT, and I’m wondering if you could share some of the things you’ve learned from him.

David Autor:

Fay Lomax Cook:

One of the things I’ve learned is just watching the way Michael works. He’s an incredibly skilled as a leader of research and as a technician. For instance, I was reading some of Michael’s notes on how to organize projects with large teams, using software and coding standards to systematize the process. I was really inspired by his examples. I was thinking, I need to catch up with the way Michael works. He’s leading a multi-country project on mortality inequality, and to tackle these big questions you need to have the ability to manage a big team and assemble ideas and facts and put them all together. Michael is extraordinarily good at that, especially for someone so early in his career.

What about you, Michael, during your years at MIT, what did you learn from David?

Michael Stepner:

Fay Lomax Cook:

I could spend the rest of the conversation talking about aspects of the research process that I’ve learned from David. So often when I met with David and we talked about research, he would encourage me to “zoom out” and think about the big picture and what it is about my data that can tell us something about the big picture.

Perhaps more importantly than what I learned from David about research, I think that David models the value of kindness in his advising…Today we are talking about highlights from our research, which is always fun, but masks the times when research wasn’t going well. Certainly, there were plenty of times in my graduate experience when I was frustrated and feeling down about my research. David is unfailingly kind and encouraging to his graduate students and that is a characteristic I found invaluable in motivating me to do my best work. And one that I hope to apply as I move into being a professor and training my own graduate students.

What a wonderful thing to say. For all of us, in our work with colleagues, graduate students, or wherever we’re working, the notion of kindness is really important. And there’s another quality too, and that’s gratitude.  I’d like to go back to something you started to talk about, and that’s the impact of COVID-19 on the economy and the workforce. This is something everybody is thinking about. Michael, I know you’ve been doing some amazing work on tracking the economic impacts of COVID-19 on businesses and communities in rural towns. This is really unusual research; can you tell us a little more about how you’re doing this tracking, how it works, and what you’re learning from the tracking.

Michael Stepner:

This is joint work I’ve been doing with Raj Chetty and Nathan Hendren at Harvard, as well as John Friedman  at Brown, and an extraordinary team of young researchers who are working for Opportunity Insights at Harvard. We are working in partnership with private companies. Every private business has a whole wealth of data that they are using every day to make business decisions. They are analyzing the data that they gather and are planning accordingly.

We typically work with government statistics that are gathered from surveys. The U.S. government, Census Bureau, the BLS, and the BEA all run representative surveys, which are absolutely invaluable in providing a representative picture of the economy. However, they have two limitations: One, is they tend to lag, so as things change quickly in the period of COVID-19 the government statistics tell us what’s happening last month or last quarter, rather than what’s happening right now. The other limitation is that the government surveys make it hard to see what’s happening in specific sub-groups. It’s always important to look at sub-groups, but right now it is especially important because the pandemic has affected high-income workers so differently from low-income workers. It’s also affecting some states much worse than others.

What our work does is gather this data from a slew of companies who have volunteered to share their data and put it up on the web so that policymakers, journalists, researchers, and the public can learn from this anonymized data.

One of the things I’ve learned, which I already mentioned, is how differently this pandemic is affecting the employment and income of high-income workers compared to low-income workers. One other aspect that really came through was the importance of social insurance. When we look at the different policies that state governments and the federal government applied early on in the pandemic, one of the biggest was the Paycheck Protection Program.

David and I have both done separate analyses on the effects of the Paycheck Protection Program using different sources of payroll data.  Our results are quite similar, showing that the Paycheck Protection Program preserved quite a large number of jobs—millions of jobs—but that was a fraction of the number of people who lost their jobs.  For example, as we spent $500 billion dollars on the Paycheck Protection Program, the cost per job saved is over $100,000.

For all of the people who still lost their jobs, handing money to businesses who aren’t hiring hasn’t helped them. The extended unemployment insurance benefits and the stimulus checks sent early in the pandemic provided huge boosts to the income of low-income Americans who were suffering.

Fay Lomax Cook:

David Autor:

David, you’ve done a different kind of tracking, tracking the growth rate in cities over many years.Even before the pandemic, based on the work you’ve done, it was said that big cities became inequality machines. Can you expand on that a bit and tell us how COVID-19 has affected cities?

Yes, but first I’d like to add that the work that Michael and his team are doing with data collection is very cool and a new model for a type of research I’d like to see more of and I’m sure we will see more of.

The work I’ve done, which started with my Ely lecture for the American Economic Association in January 2019, looks at the change in the urban landscape of employment. It showed, quite to my dismay when I discovered it, that although historically we think of urban areas as escalators of opportunity for people of all education levels, that has changed dramatically. Looking at the data over the last six decades, during the 1970s and 1980s it was still true that non-college workers in urban areas earned a lot more and did different work, rather than just doing services or manual labor. They tended to be in factories and offices working alongside skilled, highly educated workers and learning skills that way.

 Over the last three decades that has really hollowed out, as factory employment has declined and even more so office work. Urban employment has become highly bifurcated between on the one hand professional, technical, and managerial workers with high levels of education and high pay, and on the other hand large cadres of personal services workers who provide cleaning, security, recreation, entertainment, and home health care mostly for the comfort and well-being of the more affluent.

As a result, the urban wage premium for non-college educated workers has collapsed. This is even more true for minorities. COVID will change that trajectory because the one good thing you could say about that source of low paid employment is that although they were not great jobs, at least there were a lot of them. And the one thing worse than having too many low-paying jobs is having too few, because there’s not enough demand for the workers who are going to wind up in that type of employment.

I think what we’re going to see in cities is a decline in this type of service work, largely because of the decline in business travel and all the hospitality work that goes along with that due to the increase in telework. It’s going to cause a reshuffling of work, which in the long run may be better, but in the short term will reduce opportunity and reduce wage growth among workers who are already in the more vulnerable jobs in our economy.

That brings us back to the critical role of social insurance.  As Michael said, the CARES Act was passed in March 2020, and though it had many flaws, it was incredibly generous, bold, and expansive in a way that U.S. social insurance has not been in decades. I hope we can at least learn from that and continue in that direction, even if it’s not at that level in the next Congress and next presidential administration. Government played a constructive role during the crisis that no private organization could have done. No philanthropy or for-profit firm could have said, “Let’s take ten percent of U.S. GDP and distribute it to households, businesses, and unemployed workers”.  That was a very valuable thing to do.

Fay Lomax Cook:

Michael Stepner:

Yes, I agree, it’s important to learn from things we’ve done so we can make things better for the future. In terms of wrapping up the conversation, people often turn to Heinz Award winners for some direction in terms of the role of research and big ideas for the future. Could you briefly share either the important research question you’re going to pursue or the research idea you hope some of your colleagues will pursue in the future?

One very important question is how the unemployment insurance system might be strengthened and improved. Unemployment insurance has been invaluable during the crisis, but we really saw the limitations of our current system. For instance, at the beginning of the pandemic the only form of expansion available to Congress was a lump sum $600 increase because that’s the only thing the administrative systems were set up to handle. I think we can learn from that and also use that to inspire research, as well as policymaking, around what sort of unemployment insurance system would we want for the next economic crisis.  

I think one of the key issues in that regard is the question about automatic stabilizers. Can we have an unemployment insurance system that’s more sensitive to the local unemployment rates? In times of plenty, when jobs are plentiful, long-term and generous unemployment benefits might be less necessary because you can walk down the street and find a job quite easily. In places where those jobs are extremely scarce, and even during a good economy there can be some places where jobs are scarce, there would be a real rationale for more generous unemployment insurance benefits. So redesigning that system, and using the experience of this crisis to catalyze some research efforts and policy efforts, that where I hope to see things going in the coming years.

Fay Lomax Cook:

David Autor:

In fact, Michael, our next big task force at the Academy will be on Unemployment Insurance, so your comments are perfect for some of the considerations we need to have.  How about you, David, do you have some thoughts for big research questions for the future?

What makes it hard to answer, is that there are so many questions to consider. For the past two years I’ve been one of the co-leaders on the MIT task force on Work of the Future, set up by MIT president Rafael Reif to ask what’s changing in the workplace, how technology is involved, and how we can take advantage of opportunities.

What I’ve been thinking about is the growing disjuncture between the path of productivity (meaning output per hour of worker) and compensation of the median worker in the U.S. and other advanced economies. In the U.S., it’s a particularly large and growing gulf.

We’re looking at factors that help us understand what’s going wrong and how we can course correct. I do think we can course correct because, as we discuss in our task force report, many countries have done much better than the U.S., even though they are facing the same set of technological challenges and opportunities, globalization, and demographic pressures.

I’m focused on the set of institutions that shape the translation from productivity into income distribution, wage levels, and opportunities. Going forward, how can we do things to foment rising opportunity, which is something America has not done well during the past four decades but can do much better. I’m convinced of that.

Fay Lomax Cook:

Well, on that hopeful note, that we can do much better, and that we can course correct, I want to thank you both for this wonderful conversation. It leaves me very optimistic knowing that the two of you are working on some of the most important issues of our time—the future of work and productivity. And congratulations again on your Heinz Awards.