Married couple Anne Case, professor of economics and public affairs, and Angus Deaton, professor of economics and international affairs, at Princeton University in Princeton, N.J. (Yana Paskova for The Washington Post)

Academic research rarely goes viral, but this past year, the work of Princeton professors Anne Case and Nobel laureate Angus Deaton became the center of a national fascination over the woes of white America. It's also landed the two in the crossfire of a debate about race and the lack of attention given to problems faced by African Americans.

In 2015, Case and Deaton pointed out that mortality rates among middle-aged white Americans have been rising, while mortality rates for other Americans and citizens of Western Europe have continued to plummet. Two weeks ago, the researchers released a deeper report probing the underlying reasons. Opioid abuse and alcoholism and a spike in suicides have been the most visible causes, but Case and Deaton say these “deaths of despair” are symptoms of a much larger problem — one that's rooted in the economy, and how forces like technological change have battered less-educated whites in recent decades.

But the economy has been brutal lately for all Americans without a college degree, and this new report kicked off another round of controversy over how the American mainstream often ignores black experiences. Critics have complained that Case and Deaton’s focus on white mortality risks drawing attention away from equally pressing problems, like the persistently higher rates of black mortality.

Many of these objections have to less to do with the actual science and more to do with the context in which it was published. President Trump won huge swaths of voters in 2016 by promising to address the grievances of the white working class, and white nationalists endorsed his campaign. Case and Deaton's research on white mortality seemed to speak directly to that political narrative: Not only were white Americans suffering from their own problems, but they were also literally dying out faster in middle age.

Yet African Americans have long suffered higher mortality rates and lower levels of happiness. In light of the divisive racial rhetoric surrounding the Trump campaign, some worried that the interest in Case and Deaton's research would only heighten the sense that African Americans were being erased from the national conversation.

On Monday, Case, Deaton and I had a telephone conversation about their latest work, which explores the connection between a bad economy, bad personal choices, and these rising rates of death in middle age. Case and Deaton say that understanding these root causes, which are not unique to white Americans, can help shed light on the problems of all Americans. We talked about potential solutions — but spoiler alert: The situation doesn’t look great.

The following is an edited and condensed transcript of our chat.

Guo: I wanted to discuss this question about whether we are neglecting African Americans with all of our focus on white mortality.

Deaton: I talked to one journalist the other day, and she said, “Ever since your first paper, people have been really upset at you for ignoring black people.”

We weren’t trying to insult anyone. We found something that we thought people didn’t know about.

Our first paper, in the PNAS, was very much just the facts. People were upset at us for not putting black mortality on one of our charts, but the reason it’s not there — which we explain — is that black mortality is so high it doesn’t fit on the graph.


Guo: I think the controversy may have less to do with your papers, and more to do with how the media seized these facts and amplified them — in a way that I’ve not seen done with science research in a long time.

Case: It’s not as much news if people’s mortality rates are falling the way you would hope they are falling. What seems like news is when mortality has stopped falling, and no one has noticed that it has stopped.

White Americans had just flatlined where the European countries continued to make progress, and where other groups in this country — African Americans and Hispanics — continued to make progress. So what the heck is going on here? We weren’t making progress anymore. That, to us seemed like the bigger story.

Deaton: Anne presented the first paper once and was told, in no uncertain terms: How dare you work on whites.

Case: I was really beaten up.

Deaton: And these were really senior people.

Case: Very senior people.

Guo: That’s incredible to me because I think one reason these questions are important is that what we learn about rising mortality among this group of white people might also help us understand health problems that are affecting other groups.

Case: Absolutely. In this one group, we might be able to isolate why this is happening, and that’s probably going to be relevant for all of us.

The other part is that if some groups are more resilient in some ways, what can we learn from that? The press often gets it wrong and will say “blacks and Hispanics have higher mortality than whites” — which is not true. Hispanics have lower mortality. What makes Hispanics resilient and can we use what we learn from that group to help all groups?

Guo: Yes! The Hispanic mortality paradox. I don’t know if we’ve gotten to the bottom of that though.

Deaton: It’s only a paradox because people are thinking about it wrong! I don’t think income solely determines health. I think lots of other things determine health. There’s this narrative, that is entrenched in some of the professions, that there’s this mysterious thing called “socioeconomic status” that is immutably correlated with health. And it isn’t.

Case: I think our Figure 1.2 is very important because it seems to suggest that among people with less education, black or white, mortality rates are high and they’re converging. So perhaps it’s time to also be looking at slicing this in a different way, if we want to talk about the people most at risk.


Guo: Figure 1.2 in your paper had me intrigued — both black and white Americans have seen this uptick in mortality rates in recent years. Do you think that’s a cyclical thing, caused by the economy getting better, or is it something else?

Case: I think it’s something else. What I think is happening is that in the very last couple of years, deaths from drug overdoses started to increase for African Americans in all of these five-year age groups. So this drug epidemic that swept in, making it a kind of perfect storm, seems to be now be bleeding into the African American deaths as well.

It’s a little early yet to see what’s going to happen, but it’s certainly something that should cause alarm bells to go off.

Guo: In your paper I kind of see the mirror image of a debate that happened in a more antiquated time. Forty, 50 years ago, when posed with the question of “What’s wrong with African Americans,” most people weren’t blaming the lack of jobs. They weren’t blaming the economy. They were blaming a lack of values. What your research shows, at least the way I interpret it, is that the underlying economic forces — not just the forces that determine your income, but that determine your career and your life trajectory — are much more powerful than we thought.

Deaton: I think there was some argument about, when working class jobs went away in black communities, that it tended to precipitate these other dysfunctions of various sorts. But yeah, there is this parallel here.

There was this terrible drug epidemic among blacks 30 years ago, and people used to say, “well that was because black culture or black society is terribly deficient.” One way to look at these figures is to see that this has nothing to do with being black. It can happen to whites too.

Case: The data are all consistent with it starting with a bad labor market. In a previous generation, you had the chance of getting a good job with only a high school degree — a job where you could move up the ladder, where there would be what we call “returns to experience”: that as you grew older, your wages would rise.

A lot of those jobs are gone — jobs with health benefits, jobs with on-the-job training. And without that kind of a job, it’s much harder. Your marriage prospects are much poorer. So you might move in with a woman or with a man, but unlike in Europe, where cohabitation is quite stable, in the U.S. cohabitation is quite fragile.

So you not only don’t have a job that gives you structure, you also don’t have a relationship that gives you structure. And this is all happening in a point in time where religions that are being chosen also don’t give you much structure.y

So all those things that would be helpful in building a stable life, a middle class life, those are all turning to dust.

We think that it’s consistent then, in response to that, people take to drink, or people think “I don’t want to be a burden on people” and they kill themselves, or start taking drugs.

Guo: When I read your latest paper, I immediately thought of some of the recent work that Autor, Dorn and Hanson have been doing looking at the impact of the “China Shock,” which was such a nice instrument that they have been using to illuminate the different effects of unemployment and job loss. One of their most recent papers showed how poorer job prospects decreased the marriageability of men.

Deaton: We do cite that study, and in fact Gordon Hanson was there when we presented our paper. We regard that as an important piece of confirmatory evidence. But remember that these techniques that use instruments like that, like the China Shock, they actually tell you what’s happening for the people who were displaced by this shock at that particular time. It’s a local result, and it’s arguably a clean result, but we’re after this longer term story.

Guo: I feel like the biggest question that your paper leaves open is: What’s next? Treating the opioid epidemic would definitely help, but that would be treating a symptom, and what are we going to do about the underlying economic causes?

Deaton: It’s incredibly difficult. We’ve been saying that ours is a very pessimistic conclusion. I mean, treating the opioid epidemic would be the easy bit, but treating the opioid epidemic is not at all easy!

Okay, so you build a huge wall around the country and you stop foreign trade and there’s no China anymore — maybe that would have short term effect in creating some jobs? But then the technological change that’s going on at such a rapid rate would still start whittling away those jobs.

There are longer-term things, like people talk about training programs, and that’s obviously good, though it’s not entirely clear how much that applies to people in their mid-50s.

Anne and I, I think, differ a little bit on how much education is a solution for this. But it’s certainly clear there are lots of people who are not getting BAs who are capable of it. So we need to do a much better jobs of getting these into school.

Case: But it’s also the case here that there are people who don’t want a four-year BA. We’ve been around this block many times: We do need to think about how we want to train people to enter the 21st century labor force. What kinds of skills should they have that they’re not currently getting in high school?

Deaton: When you talk to CEOs, to the business community, they say, “We have lots of vacancies, but the educational system is not training people up to do those jobs.”

To which the counter answer is: “Okay but you just told me five minutes ago that the speed of technological change in your industry is so rapid it’s terrifying even you.” So do you want to train up people for something that’s not going to be there two years from now?

Case: The other thing that makes it harder in America rather than Western Europe is that there really is a difference for a large swath of the population in how they feel about receiving government transfers. We’ve all been trained up on the idea that we are individuals and we take care of our families and our neighbors take care of theirs, and that’s the way we like it. It’s very hard to give somebody something when they see it as handout that they don’t want.

Deaton: We haven’t really talked about how none of this is happening in Europe.

Guo: Absolutely. That’s one point that I made a week ago when I wrote about this. It’s not just a lot of these death rates are rising, but that a huge chasm has now opened up between us and Europe.

Deaton: The obvious difference is that the safety net is enormously more generous in Europe. And lot of people in their 50s who lose their jobs can go on retirement. You get a doctor’s certificate and you get paid pretty much your salary until you die.

There’s one other policy recommendation that I’ve been pushing. We’re spending about three trillion dollars a year on health care. And our life expectancy is going down. Whereas all these other countries are spending way less, and their life expectancy is going up. For me the implication is if we implemented single payer, we’d get rid of a lot of these costs. Not without screaming and yelling, of course, and not without goring a lot of oxen.

But the crucial thing is recognizing the extent to which these rising health care costs are responsible, at least in part, for the stagnant wages for people without a college degrees. If they’ve got an employer and they’ve got health care, their wages are getting pushed down by the employer paying for that health care. People don’t even realize this. They think it’s for free.

I’m not a left-wing nut pushing for single-payer! It’s not because I like socialized medicine. It’s just because I think this is eating capitalism alive, and if we want a healthy capitalist society in America, we’ve got to get rid of this monster.

Guo: I want to return to the college or no-college debate, which to me seems like a little bit of a false dichotomy right? What we really should be thinking about is flexibility, and what kinds of skills we can impart to people — whether it’s in a college context, or a vocational training context — that can be flexible enough to weather the kinds of changes we’re going to see.

Case: Absolutely.

Deaton: Well you know, that’s the argument for a liberal arts education. Do we know how to identify and teach people these skills? I’m not sure.

Case: That doesn’t mean we couldn’t learn.

Deaton: We could learn.

Guo: Well I think about all the manufacturing work that so many people did. Being handy, or being good with a machine tool, that’s a set of talents that for a while was very flexible. You could stamp sheet metal, you could operate a bandsaw, and it all kind of fell under one rubric — shop skills. I wonder what the modern equivalent to that is. I don't know, maybe it’s programming?

Deaton: It’s hard to know what’s going to be replaced by technology tomorrow. It feels like we’re all at risk. I feel only safe as an emeritus professor!