Modelling climate change and the world economy

Per Krusell interviewed by Romesh Vaitilingam, 15 April 2011

Unfortunately the file could not be found.

Open in a pop-up window Open in a pop-up window


Download MP3 File (13.0 MB)



Topics:  Development Environment

Tags:  climate change; environmental economics and development

Romesh Vaitilingam interviews Per Krusell for Vox

April 2011

Transcription of a VoxEU audio interview [] 

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 Per Krusell from the Institute for International Economic Studies in Stockholm. Per and I met at the European Economic Association's annual meeting in Glasgow in August 2010, where we spoke about his research, modeling the impact of climate change on the world economy, I began by asking him how this project began.

Per Krusell:  I mean, it started, I guess, because of the Stern Report, and reaction to it, or to the ongoing work, actually, by some of my colleagues in Stockholm. And, that's a group of people that mainly consists, actually, of scientists. They wanted economists in on this project. And, at the time I was a Princeton, but Torsten Persson, he asked me if I was interested in taking part. And, I think it seems like an exciting area. It seemed like it, at the time. I still think so. So, I think what's been fun for me in this project is that I'm, comparatively speaking, good at modeling, macro‑modeling. And, I think there's a great need for that in this area, and I can just, off the shelf, use my tools in this area that I think desperately needs it because I think it's a relatively recent area. Nordhaus has worked on it for a little bit more than a decade, but he's used very old methods, I think. He's done a good job, and he's a pioneer, and it's, you know, great to have his work. But, I think there's so many ways in which we can improve on his modeling of the world economy on modern macro‑tools that are, you know, being discredited on a daily basis.

You know, you don't have to like these methods for business psychoanalysis, I think, to support the methods for this use. I think it's more of macro‑growth analysis and thinking about what makes countries rich and poor. And then, the energy market and the connection to the climate system and the damages and so on.

For me, it's been very enjoyable, because we find that the returns from doing research in this area are so high, so quickly. And I think it's just because it hasn't really been populated by people like me. And, there are all these people like me who could have higher returns by doing a little bit of work in this area.

Romesh:  Why do you think the climate change scientists were so keen to get an economist involved? What is the difference that economists make in the way that Stern almost revolutionized the world discussion of this?

Per:  I think there are various dimensions here. One is, in general, that economists are important for getting support. More generally, I think for the climate scientists to have an impact, they have to influence economists in leading positions. And, that partly happens through us, through the academic economists. So I think that's important.

But they have a strong need on the more scientific level. They have a need for projections of energy use in the future, and what determines the oil price? And what about the different markets for energy? And, you know, how does that work? How do we think about it in a systematic way?

So, actually in their models of world temperature, there is a key input, which is the world's use of fossil fuel. And, they just take that as given and they ask economists, maybe, "What do you think it will be?" But, they realize that it's really important to think about that more carefully. And also, obviously the two systems interact. Economy influences the climate. The climate influences the economy back. And, they think about it the right way, that they need some of our knowledge.

Then, I think, they don't necessarily realize that we're also important for systematic welfare evaluation, because they think they can evaluate welfare just based on, you know, I don't like heating, or I like heating. But, what we do is we translate it into consumer welfare with our standard methods. I don't think they appreciate that that's a useful method.

But, we've found that when we have educated our meteorologists and so on, that they really started respecting what we do, and they're understanding us much better than, I think, before. So, I think it has side effects, in that, these scientists that we talk to actually understand that we are very serious and that we're not... Well, you know, economists are not always the most popular people. So, in that sense, I think they're getting more out of us than they thought in the first place.

Romesh:  Can you explain the significance in your modeling of the regional differentiation, how you have different impacts around the world? Some parts of the world perhaps even benefiting from climate change?

Romesh:  Right, exactly. And, I mean, that's an issue that comes up a lot in discussion. "Well, you know, isn't it better with a little bit hotter climate?" And it is better in some places, but it could be quite disastrous in other places. And, I think they have a strong interest. The scientists have a strong interest in figuring out damages in different parts of the world, I think because they are concerned with development and with Bangladesh being hit hard by flooding, et cetera. So, we felt a need to develop something much more regional than Nordhaus's model. Nordhaus has a big model with eight regions. The world is divided up based on some different characteristics. Some are geographical, but some are rich and poor.

We just divide the globe up into small cells, one by one degree or five by five degree on the map. And, we'll be able to, we think, make predictions as to how different parts of the world are likely to be affected by climate change.

Obviously, we need different methods or different for getting this to work, but I think that one of the main issues here is distribution of welfare across the world. So, it's somehow a part of development economics as well.

Romesh:  What do we know, so far, about that in terms of the differential impacts? There seems to be a sense that it is actually the poor countries that are more the victims of climate change than the rich countries.

Per:  Right. Yes, so we haven't reached the point where we have our independent conclusions on this. But the historical data that people have studied suggests that, for whatever reason... What people do in, like in the work by Dell, Jones and Olken, is simply run regressions, where you see whether output growth is helped or hurt by changes in temperature, or by the level of temperature. They find significant effects in the poor countries, and quite large effects as well. So it seems that it's in the data, and we have to explore it further and look at possible ways of adapting to climate change, etc., in those poor countries.

An interesting aspect is that if you compute the costs, and you sum them up, the sum isn't necessarily so big. Because if you think about growth effects in poor countries, they may be large, but they are so poor that the numbers are not so big if you add them up. So actually, the global sum of the costs doesn't seem so high. It would seem easy for the rich countries to compensate. So, one of the big challenges for politicians is to try to do that.

But until we have good ways of compensating, we really have to study the poor countries because the effects there are large, and they are close to the point where people die if temperature goes up by three more degrees. I think it's right now. I think it's super important, but it also seems like there could be policy change that allowed some of these countries to cope much better. I think the kind of work we do will help in convincing policy‑makers that it is really important.

Romesh:  How does your analysis think about the impact of sudden unexpected crises or unexpected disasters? In a way, in the economic world, we've had the financial crisis, but we've had things like big hurricanes devastating the Haitian economy, floods affecting 20 million people in Pakistan recently. How does it think about impacts of sudden weather events that might be the result of climate change?

Per:  The kind of models we use are not detailed enough on the timescale, actually, so we measure damages as average damages over relatively long periods. A year would be absolutely the shortest, but I think our main models will have one time period be five or ten years. I think that's something that requires a separate type of economic analysis. I don't think we'll provide that, at least not in the short run, detailed analysis of that. But that's an important question.

Unfortunately, what we do... We have models whose horizon is more like 50 years, and think about what has to happen with policies over the next 50 years, and 100 years too, for that matter. But 10, 20, 30,... I don't have a good answer to that, unfortunately.

Romesh:  Well, perhaps finally, we can go back to the policy issue. We talked about policies over periods of 10, 20, 50, 100 years. What is the range? How do you describe the taxonomy of different policies that national and international policy‑makers might implement?

Per:  Well, we have focused mainly on taxes on fossil fuel use. It's the most obvious instrument. I think, of course, there the competitor, cap-and-trade. Have caps, quantity restrictions on how people trade in emission permits. I don't have a strong view on whether one is better than the other. I think the arguments that are put forth in favor of one as opposed to the other are more of a political nature, in the sense that some are easier to implement politically. I know Mankiw thinks that taxes are good because that brings in revenue, so that one can lower taxes on labor income, for example, because he likes such a policy.

I can sympathize with that or not. But in some basic economic sense, they are very similar, so we essentially think about trying to have world taxation—and it has to be pretty much used everywhere, so it's not enough that a few rich countries use taxes. It has to be used everywhere, so the issue is how high these taxes have to be do the right thing about the climate. That's where we're at right now.

I think one aspect of policy that we are not so good at yet, but some other people have worked on more, like Philip Aghion and company, is exactly how taxation interacts with support for alternative energy technology. But that's an important area, too, of policy. It's not clear how much they interact. There are arguments back and forth ‑- the green paradox.

But for the moment, I think we just look at the most obvious, easiest to use policy in principle. If you could get politicians to just implement it, because it's a pretty easy thing to do. Our main questions are how high does it have to be.

Actually, the stuff we presented yesterday, which was based not on this multi‑regional model, but a one-region model, suggests that Stern's numbers were at a pretty reasonable level. You don't have to assume extreme discount factors. So the numbers are pretty big. You need pretty high taxes, and you need to use high taxes now. That's what we find so far. But I have to say it's preliminary, because we haven't been working on this for very long.

Romesh:  How does the research program develop from here, in terms of building the model and pulling out the answer?

Per:  Well, we keep plugging on main aspects of the modeling, trying to understand energy use in the past, to be able to predict it in the future on a global scale. I think that's hard. Trying to understand what determines oil prices is hard. I think we're doing this work on multiple fronts at the same time, a bunch of people. We're hoping more people will be interested and just do work in other places on these same issues. I think competition would be great in this field. It will push the frontier so much faster.

We are basically doing a bunch of parallel projects, and they involve everything from relatively pure empirical studies to theory to solving computational challenges. I mean numerical methods. I couldn't say I'm working on something very specific. Right now we have all these different aspects, I think are all interesting and fun.

Romesh:  Per Krusell, thank you very much.

Per:  Thank you. 


Savings Banks Foundations and Swedbank Chair in Macroeconomics, IIES, Stockholm University, and a CEPR Research Fellow.


CEPR Policy Research