Romesh Vaitilingam:  Welcome to Vox Talks, a series of audio interviews with leading economists from around the world. My name is Romesh Vaitilingam and today's interview is with Professor Peter Diamond from MIT, co‑recipient of the 2010 Nobel Prize in economics. We met in Lindau, Germany in late August 2011 at the Fourth Lindau Meeting on Economic Sciences, an event which brought together 17 of the 38 living economics Nobels with nearly 400 top young economists from around the world. Peter had just spoken at the panel discussion on the impact of demographic change on our retirement and healthcare systems and he began by explaining why the good news of rising human longevity has raised such big long-term issues for the advanced economies.

Peter Diamond:  We have seen in the advanced countries a steady trend of longer lives. What has made adjusting to that such a big deal right now? I think there are two answers to that that come together. One is that the baby boomers and particularly the drop in the fertility rate after the baby boomers is a relatively rapid move in the demographic structure compared to the way the trend looked otherwise. Of course, not all the countries experienced the baby boom ‑ that is not a worldwide phenomenon ‑ but many did. And so, the process of adapting slowly to change in demography is not such an option when the demography is changing more rapidly. The second element is, the US has no automatic adjustments relative to demography. We have adjustments relative to wage growth and price growth. We don't have any adjustments for demography and this was the way the systems were set up when they were started at different times in different countries.

It was probably not a good idea. The systems were much smaller. So, the extent to which they weren't such a good idea wasn't a big deal. They're all so big now, they’re expensive and in terms of the impact on economy, obviously, a problem in a big programme is a bigger deal than a problem in a small programme. And so, what we have seen ‑ Sweden and Germany are two examples ‑ is building demographic developments into the automatic structure of the response to what's going on. And that's one of the things I urge for the US. I did that in my book with Peter Orszag back in 2005 Saving Social Security.

Romesh: Can we talk about the issue of retirement if that's the big question with the ageing of our society, people living longer and an increasing slice of the population being of the older generation. How do you think about it? Do you think that is a relatively easy thing to solve compared to the healthcare, which we’ll come to in a moment?

Peter:  First, let's note that we had two things going on simultaneously. People were living longer but secondly, men were retiring younger and younger. If you look at the 20th century, the work day has shrunk, the work week has shrunk, the work year has shrunk and the work lifetime has shrunk, not in percentage term, but in absolute terms, they were shrinking, at the same time as people were living longer. In the US that shrinking stopped in the mid-80s and has reversed already. People work longer. The pension system doesn't control when people retire. What a pension systems, if it's well designed, does is to set up a menu of alternatives of: You can retire on this age and you get this pension. You can work a little longer and get a bigger pension! Here's how much bigger it would be. And for some people, it will make sense to retire early.

Increasingly, I would expect people wil respond to this by working longer. I talked about men here ‑ not because only men are important in the labour market ‑ just to be clear; but because the combination of factors about women had led to a much more complex picture. But the growing importance of genuine careers for women has meant that there are more things going on there. So, the point is that the system will work well if, to a significant extent, it succeeds in inducing people to retire later.

And it will do that by doing, in the US context, one thing, which is lowering benefits as a function of age in the way that can be offset by working longer. And in some other countries where the increase in benefits, when working longer, is not large enough, they also need not just to change the level but to change the tilt so that working longer does more for you to encourage an appropriate but diverse pattern of when people retire.

In the US context, where the replacement rates are quite low by international standards, where the tax rate is low by international standards, we do not want to do the entire adjustment on the benefit side. It is totally appropriate in the US that the tax rates go up as part of the solution.

Romesh:  If we could turn to healthcare, I’d be interested to get your perspective on that, thinking about how that relates to demographic change. In your talk, you said, basically, demand for healthcare is unlimited. It could be all if us wanted to get a doctor to check in on us once a day to see how we are doing.

Peter:  That’s is such a wonderful example.

Romesh:  Why don't you recount that example again?

Peter:  This actually came from Eliot Freidson. I believe, it was his PhD dissertation written in the 50s. He's interviewing somebody in the Bronx which is where I started growing up before my parents moved out to the suburbs, part way in my elementary school education. And he says to this man, "If price were no object, what would you like for medical care?" And he says, "It would be nice each morning as I have breakfast, if the doctor stopped by and asking, ‘How are you feeling today?’ and I could say, ‘Fine, thank you.’" And the point is that, there are a huge number of things where some value can come from more tests, from procedures. There are many, many ways to spend money. So, we have to find methods of deciding what you do and what you don't. For many factors, we do that by saying, "You want it, you buy it." Healthcare is – and ought to be – a heavily insured activity because it is such a major risk for people.

And the insurance will of course have a moral-hazard dimension, which insurance almost inevitably does. But that doesn't mean you don't want the insurance. It just means you want to work on the design. You want to work on the scale. The insurance will recognise that and so you need to design a mechanism for limiting demand. We can do it in part through prices, in part integrated with insurance. We do it in part by relying on the professionalism of doctors to say, “That's not worth doing.”

We do it in part by having systems where you don't see the doctor everyday and so there are delays. There are lags. You have to wait for things to come. So there are multiple ways of doing – to use the abominable word – ‘rationing.’ Rationing to an economist is something that is inevitable as a function of limited resources. And so, the key question is how to do it and as part of that, in the current US context, we need to recognise we have a budget problem for the federal government associated with healthcare.

It is the elephant in the room. But we have a budget problem for state and local governments relative to healthcare and it’s the elephant in their room. And we have a healthcare problem for businesses for their employees. It may not be quite as large a problem but it's still pretty big. And we have a problem for lots of individuals to the extent that they are paying for healthcare.

So, what is important here is not to focus on how do we shift the cost from this level of government to somebody else. That may work for the budget, but it doesn't work for the American people. We have to say, how do we get better value out of the healthcare in a way that will limit the cost growth and hopefully, at the same time, improve care. And it seems to me there are some things right now we know that, say, Medicare can change and make things better.

But to a large extent, the medical care is a very complex animal and we need to do experiments, evaluations, a further round of attempting to do things, in order to figure out what to do. This is very much a process. To say we know now exactly how much we should spend on Medicare eight year from now is silly. It doesn't mean it's silly that the legislative process has a default which is different from not having a default. But it seems to me it's a default. It's something that ought to be changed as we see what evolves in our ability to learn about how to do a better job of healthcare.

So I think, while it's appropriate to be working on some of the healthcare issues, I think it's also important to recognise that there are other elements in the pension systems that I talked about, and where our ability to do things that work for the long run is greater than our ability in medical care. And we need to be doing more than just looking at the elephant in the room. We need to do more than thinking in terms of the spending limit rather than thinking about how do we get it better. And once we are getting it better, how much do we then want to spend?

And it will also depend on things we can't predict. The appearance of new diseases, continuing problems from existing diseases that could get worse... There are all sorts of things that make that much more uncertain than the pension area and therefore, a genuinely hard problem that needs to be approached with attention to detail rather than a blunderbuss.

Romesh:  You're done. Thank you very much.

Romesh Vaitilingam interviews Peter Diamond for Vox

August 2011

Transcription of a VoxEU audio interview [http://www.voxeu.org/index.php?q=node/6925]