Knowledge@Wharton – Income inequality has been in the spotlight as a key force behind Brexit and even the election of President Donald Trump, who promised greater prosperity for certain groups that were left behind. Income inequality has many facets, but one that is probably not at the forefront is the effect it can have on public health. Could you give us some statistics about these ill health effects in the United States among certain groups?
Angus Deaton – Let me preface that a little bit. I really, really don’t think this has much to do with income inequality, and so I resist. There’s literature out there claiming that income inequality is bad for everything, including health. I’ve argued against that for many years. It’s not clear why Mark Zuckerberg, or someone who develops Facebook or does some other thing that benefits many people and gets very rich in the process, is responsible for the poor health of people at the bottom who are not doing as well as he is. That’s really important because otherwise you get led to the thing where the cure for the bad things that are happening to health is higher taxation and redistribution, and I don’t endorse that for that purpose. I might endorse it for other reasons, but that’s not so. With that preface I can answer your question.
What we discovered in our first Case-Deaton paper was that if you look at white, non-Hispanics in midlife, in their early 50s for example, their mortality rate after 100 years of declining had turned the wrong way or at least flattened out. This is not happening to other groups in the U.S. It’s not happening to Hispanics. It’s not happening to African-Americans. And it’s not happening in any other rich country in the world. This is happening to both men and women. Perhaps the most shocking thing is that a lot of the deaths come from what you might think of as behavioral factors, which are alcohol – alcoholic beverages – from suicides and from drug overdoses. Many of those drug overdoses are accidental overdoses from prescription drugs. People often think the health system is responsible for our health. In this case, the health system is responsible for killing people, not actually helping them.
In the more recent paper, we spent a lot of time dividing up people by whether they went to college or not. It’s like there are two Americas out there: the people with a B.A., and people without a B.A. The mortality rates of white non-Hispanics without a B.A. are going up faster than the average. They’re much more subject to opioid abuse, suicides, alcohol-related liver disease and heart disease, which has been a major cause in mortality decline. Mortality from heart diseases stopped declining and started rising. There’s a lot of really bad stuff going on, especially for this group without a B.A.
There’s a quote in The Atlantic where you said that if you had to choose between living in a poor village in India and living in the Mississippi Delta or a suburb of Milwaukee in a trailer park, you’re not sure which would be the better life. Could you elaborate on that?
Yes, I did say that. Part of this – it was designed to shock a little bit, which it appears to have done. And part of it has to do with how difficult it is to measure poverty anywhere. If you use something like the World Bank’s global poverty line, it is $1.90; let’s call it $2 per person per day. That takes no account of all kinds of living — in a cold climate, it takes no account of health care, it takes no account of housing, which is incredibly important.
So, $2 a day in India would probably be $4 a day, just to take a rough number, in the U.S. There’s been a fair amount of documentation by the people who do ethnography that there are quite substantial numbers of people in the U.S. living on very, very small amounts of income. There’s some dispute as to how much they get in government benefits, partly because the service may under-report what they received and partly because some of those benefits are in kind and you can’t use Medicaid to put food on the table or pay the rent. There are a lot of things — the social life in an Indian village may be more protective than it is in some places in America. Then people always say, if you just look at infant mortality and health, we do much better and that’s not even clear. So that one of the things I noted was that there are counties in Appalachia where life expectancy is lower than it is in Bangladesh.
This probably comes as a surprise to some Americans who would be first hearing about this. I wanted to get back to something you said at the outset. Some might suspect that many of these problems in the U.S. could be attributed to slowly growing, stagnant and even declining incomes. But you found that that cannot provide a comprehensive explanation. Rather, this has to do with “a cumulative disadvantage over life in the labor market, in marriage, in child outcomes and in health triggered by progressively worsening labor opportunities at a time of entry for whites with low levels of education,” which in this case means without a college degree. This is where you are making that distinction: the problems are not caused by income inequality per se but by this cumulative disadvantage over time. Could you elaborate on that a little bit?
You said income in there … inequality depends on the top too. So you can reduce inequality by hurting the rich, and that’s not anything obviously to do with this…
Can we focus on this idea of it being a cumulative disadvantage over time.
Yes, absolutely. This is something that our sociologist friends have been telling us for a long time and we probably haven’t been listening until recently, which is that the consequences of the labor market are very profound in other aspects of people’s lives.
One of those things is that if you’re a young man, in the old days you could get a job and expect your wage to increase steadily over time as you acquired skills, experience and tenure on the job. You could walk into any job starting at maybe $15 to $20 an hour with General Motors, and you could build a middle-class life out of that. You would get married, have kids and by the time you were in your 50s, you’re thinking about having grandchildren. Your life has unfolded in a fairly positive way, both materially and spiritually.
What’s happening now is those jobs have progressively gone away, so you get a situation in which these people are coming out of high school and don’t get such good jobs. They get sort of crappy jobs where there’s no real commitment by the employer or the employee. Now your girlfriend refuses to marry you. That’s something that sociologists have been thinking for a really long time — when jobs go away so does marriage. One of the things that has changed, of course, is that 50 years ago if your girlfriend didn’t marry you, you didn’t move in with her and have children. That is what is happening now. That’s a change in social mores about what is acceptable and what is not. So, there’s a big rise in cohabitation. They live together, have a kid, and then some better opportunity comes along and the girlfriend says, “OK, I don’t like you so much anymore. I’ve got a guy here who is earning $10.50 and not $7.50. And I’m going to move in with him instead.”
So then you lose your kid. If you’re the man, you move out. The kid may have two or three dads by the time he or she is a teenager. Some of those dads may be very unsatisfactory as fathers, and the ones that are good fathers, they get lost, too. So, it’s very bad for the kids, terrible for the fathers, and not so great for the mothers either. You reach your 50s, and you haven’t got all of those experiences and tenure and skill bonuses that used to [go with it]. You’ve had a series of not very permanent jobs. You’ve got your midlife crisis, but it’s terrible because you don’t have all of the support from marriage and everything else.
So, the labor market is the fundamental thing, but there’s no short-term correlation between income earnings and people’s health.
One of the interesting things is the psychology behind all this – you write about this idea of despair that the groups you’re talking about experience over a long period. Is it correct to say that this is part of what leads to the overuse of alcohol and opioids and other unhealthy practices – obesity and that sort of thing — which then translate into the declining statistics on mortality?
I think that’s right. If I had a setback — if I was expecting a pay raise and didn’t get it, or I was in an occupation where my employer signaled his displeasure with me by cutting my wages — I might be really disappointed but I don’t think I’m going to kill myself. [But] if at 50 I lose my job or I have setback like that, and I look back and I don’t know my kids anymore – I may have a kid who is drug-addicted too — then life gets really, really, really tough.
I also think — and this is something we haven’t looked at really seriously in the data yet — that if you are switching out of these long-term jobs into these more casual jobs, you’re more likely to get injured. There’s a huge increase in pain reporting. If you just asked people, ‘Do you feel pain?’, and they said yes, you might believe maybe they’re making it up. But if they report sciatic pain, which requires them to say it’s shooting down their leg, I don’t think people make that up so much. There’s a big increase in that, and a big increase in neck pain and so on. So there’s an epidemic of pain underlying this too, and the most plausible story that the sociologists and others have been writing about, Andrew Cherlin has a couple of terrific books on that, is that people’s lives are coming apart.
Unemployment in the U.S. has improved since the financial crisis, but more people are going out on disability. Some would say that they’re gaming the system and hurting the employment stats. What you’re saying argues exactly against that idea, it seems.
Well “exactly against” is too strong. I don’t think we would ever claim there aren’t people who take advantage of the system. But there are two things to say about this upsurge in pain.
There’s a very nice piece by Jeff Liebman in the Journal of Economic Perspectives, which looks at the increase in disability. A lot of it is just demographic, the changing composition of the work force. Even if the disability rates are not changing, you get a big increase in disability. Our work would suggest that important parts of that are real. And much of the residual is in exactly these sort of issues like pain, and non-specific soft tissue things that are not like a broken leg or something. And our work would suggest that important parts of that are real, as you just suggested.
Another thing worth noting is that disability systems in Europe are much, much more generous than the ones we have here. In a lot of European countries, if you get a doctor’s certificate in your mid-50s saying you suffer from intermittent headaches or something, and you can retire on something not far away from what you were earning before. There was a time, certainly when the average retirement age in one or two countries was lower than the minimum possible retirement age. So, there were just a huge number of people on disability, and that’s clearly to do with having an over-generous disability system. But you don’t see any of these people killing themselves, either quickly or slowly, in Germany or in France or in the U.K. in the way that we see it in the U.S.
The psychology behind what you write about, and this idea of despair, is especially interesting. There is a cumulative effect as we have discussed. It’s not just that someone looks back and understands that they weren’t able to have the life — economically speaking at least — that they were hoping to have, or they expected to have when they were younger. It’s that they aren’t having even a life as good as their parents, or in some cases their grandparents. It is that realization that they find crushing.
Well, you could certainly put it that way, yes. There are several papers showing that it’s right about in these generations where that switch happens. So if you were in my cohort, if you were born in 1940, 1945, something like 90% of these people were better off than their parents by age 30. Whereas if you were born in 1960, which is the sort of people we’re looking at here, less than two-thirds of them were better off than their parents at age 30, and it’s probably gotten worse since then. This is because the experience premium — or the return to experience in the labor market, for people who only have high school degrees — has fallen so rapidly that the gap between them and their parents is probably widening with age.
Maybe that is an especially difficult a thing to live with when there’s conventional wisdom that suggests that each generation expects to be better off than the one that went before. So when that doesn’t happen, it really disrupts foundational beliefs.
That’s certainly one hypothesis. And it’s plausible, but there are other hypotheses. Going back to some of the stuff that Andrew Cherlin has written about, the working-class family was one in which the man went to work in manufacturing, the woman stayed at home and looked after the children. That was very much the basis for self-respect and for meaningfulness and for understanding yourself and your life. That may be more important than what your dad earned, though that could be important too.
In thinking about what has led to this cumulative downward economic spiral, why aren’t jobs as good as they used to be? We know that globalization has hurt the labor market to some degree. A lot of people would subscribe to that idea. What other things are contributing to the more difficult labor market for folks and the cohorts that are having such problems as you write about?
I’m not the best person to ask about that, and the conversation you had before with Bob Shiller and Jeremy Siegel about some of those issues. There are some externals — robots and globalization — but robots are not all of it. There is this skilled-bias technical progress, too, and the gap that has opened up between people with and without an education.
My own view is that these people have probably been hurt more by technical progress and by the upscaling of jobs. This refers to the replacement of some technologies by others that don’t require so many people. But globalization has certainly played a part too. There’s a much larger effective supply of labor in the United States. That is certainly contributing to putting downward pressure on wages.
So for the purpose of these deaths, I’m not sure it matters all that much, which of those [factors are most important]. It certainly matters politically because people identify scapegoats like immigrants or globalization rather than the technical change. Technical change is much harder to think about. And that’s true for economists too, I mean, you can build a wall around the U.S. but that wouldn’t stop technical change. It might slow it down. We could turn ourselves into Albania, but still people would be thinking of smarter ways of doing things, and we would still have these issues.
You have talked about the idea of inequality and rent-seeking by companies in the U.S. Could you discuss that?
Well, it’s just that, like a lot of people, I’ve been trying to ask myself the question, what exactly is wrong with inequality? People are jumping up and down and saying inequality is a really terrible thing, and the question is why? Why is it such a terrible thing? I think I’ve made a little progress inside my own head; I’m not sure I’m persuading other people. But what I take is the European view, which is the one I grew up on in Britain.
The first seminar I ever attended in my life was Tony Atkinson talking about his 1970 paper on the measurement of inequality. That approach to the world is what philosophers have subsequently rechristened prioritarianism. That means they are declining the social marginal utility of income so that people who have a lot of money are not worth as much at the margin for public policy as people at the bottom. In that sense, inequality is inherently evil because if you have more money than I do, taking money away from you and giving it to me will make the world a better place, subject to incentive constraints that, if we take too much away from you, there won’t be any left for me.
That’s been the European tradition. That’s the problem that James Mirrlees got the Nobel Prize for solving, which is the optimal income tax and so on. There is a lot of economy theory centered around that set of ideas. I think economist [Thomas] Piketty’s book [Capital in the Twenty-First Century] is based on that, that it’s just the scandal that the rich are rich because that money could be much better used by poorer people.
I’ve come to not believe that, partly from reading various philosophers and other people.
I think that if you’ve got two people, one of whom is richer than the other, and neither is in distress in any way, I don’t see why it makes the world a better place to bring them closer together. Put another way, if you get rich, it doesn’t do me any harm at all – any material harm or any harm in any other way. But if you get rich and nullify my vote by lobbying Congress to close down the schools I depend on, that’s clearly a bad thing. There’s a nice little book by the philosopher Harry Frankfurt, for instance, arguing this. I think it’s quite convincing.
I just don’t think inequality by itself is bad. That puts me at odds with a lot of economists, a lot of people on the left, a lot of liberals. But that doesn’t mean the inequality that we have is a good thing. The issue is if instrumentally inequality is bad. So, if someone gets very, very rich and other people don’t, that person might use that wealth to hurt those people, and it might not even be in an income space. It might be what I said about schools, or they might undermine the health system or nullify your votes, or that Congress only listens to rich people. That’s a concern that goes back to the Greeks, which is that rich people might effectively take over the state and, at worst, enslave poor people or have poor people acting totally in their interests. That’s the sort of thing I think is really bad about inequality.
You asked about rent-seeking, and that is part of it. I think a lot of the inequality that we get in the U.S. today comes through people seeking special favors from the government by lobbying, by getting special deals, getting the rules changed. A lot of that is going on now. We’re supposedly having a bonfire of regulations, but a lot of these regulations are to prevent rich people stealing stuff from poorer people or from the nation as a whole, which is sort of the same thing. So if you just contrast what Mark Zuckerberg did with rent-seeking, I think, as I’ve said elsewhere, I think it’s okay to get rich by making things. It’s not okay to get rich by taking things by theft, as it were.
Can you give an example of rent-seeking?
On the health care sector, one of my friends explained that the health care sector is a giant bait ball, which everyone is feeding off. According to various calculations, we’re spending about $1 trillion a year more on health care than we would need to, relative to other countries. What’s more is our life expectancy is among the lowest of all of those countries and is actually falling. So, it’s not clear what we’re spending that $1 trillion on – or it is clear what we are spending that $1 trillion on — a lot of people are getting very rich off that $1 trillion. There’s lots of rent-seeking by pharmaceutical companies, by device manufacturers, by hospitals and all the rest of it. That maintains the system in which they get enormous sums of money and deliver very poor health care.
To connect back to the groups we were talking about earlier, the middle-aged whites with lower levels of education, what could help improve their lot?
That’s a really tough one. We think the opioids are a terrible scandal that should never have happened. The mass over-prescription of these drugs has killed a lot of people and perhaps reduced some pain, perhaps not. There’s low-hanging fruit there, though that is hard enough, which is to put more pressure on doctors not to prescribe these things except under very carefully controlled circumstances, which is what happens in other countries. In Britain, for instance, opioids are rarely prescribed outside of hospitals where they can be controlled and watched. Of course, I once saw people turning to cheap street heroin or switching from legal Seconal to illegal Seconal [a barbiturate derivative drug].
Let’s come back with the underlying despair that we discussed earlier. Not just about use of opioids but the alcoholism, the high suicide rates that are outlined in your studies. What can be done about that?
Well, nothing very quickly. Because if you believe the cumulative disadvantage story, you could turn off the tap of current disadvantage, but the remnants of all of those years is still going to be there. It’s something you would only chip away at.
The obvious place to look is in education. But I’m somewhat of a skeptic of just saying if we could only educate everybody, if everybody had a B.A., we would all be okay. Everybody doesn’t want a B.A., and also it’s not like the B.A. is just like a new drug or something which everyone should have.
The people who have bachelor’s degrees are different in all sorts of other ways from people who don’t have them. That said, there’s good work out there showing that lots of people who could benefit from a university degree are not getting it. To the extent that our school systems are failing kids, and bright kids who could get out of this are not getting out, that is certainly something we should work on.
Some would argue for training programs, which are more like German training programs, and it is certainly true that CEOs complain all the time about the educational system not generating the sort of workers they need. I worry a little bit about that because with all of this technical change, the workers we need today are not necessarily the workers we need tomorrow. But it might be better to train them in something for which there is a demand right now. Maybe community colleges and other places could play a larger part in that.
In your book, you have this really interesting analogy. You make the point that income inequality is kind of a dance. If one group gets ahead, that’s good. They’re blazing the trail and showing others the way to improve their life. As long as there is a way for the groups that weren’t involved in that initial spurt ahead to move up, then that’s progress. But you liken it to a situation with miners underground who are trying to escape. They dig the tunnel and escape from a world where health care is not so good and the economy is not so good. They find a way to make things better, but then they can either turn around and reach back to bring others up, or they can fill in the tunnel behind them so that they maintain their status.
I think that’s why you don’t want to take too simplistic a view of inequality, when a lot of inequality actually is progress. To be against that sort of inequality is to be against progress and what philosophers call leveling down. Let’s keep ourselves all miserable because there won’t be any inequality then.
But the guys who dig the tunnels, who are part of the fellowship and help us escape, they can very easily feel like it might be a good idea to fill in behind them and change from one end to the other. So, the entrepreneurs in Silicon Valley whom we all admire so much, if you don’t watch them like hawks they’re going to turn in to rent-seekers five minutes later to protect their innovation.
Do you think that we have too much filling in the tunnel behind us in this country today?
Yes, well that’s what we have anti-trust and things for. We have to enforce that stuff, and I get the sense that there’s not been a lot of enforcement of that in recent years.
You noted that there is a high correlation between counties that are suffering the kinds of problems we talked about earlier with health and mortality, and the counties that voted for Donald Trump. There’s a very high correlation showing a level of dissatisfaction that led to a reaction similar perhaps to what we saw in Britain with Brexit. Is that correct?
Yes. The Economist even had a picture, but it’s a little too easy because you can put almost any dysfunction on the right-hand side and it predicts the Trump vote. The correlation is usually about 0.4, so there’s a strong relationship there. But you could put obesity or people dying or all that sort of stuff. So, it’s not really a great mystery.
This article was published by our partner Knowledge@Wharton. Copyright Knowledge@Wharton. All rights reserved. Translated and reprinted by permission.