Discounting Global Warming
Sir Partha Dasgupta, Frank Ramsey Professor of Economics, University of Cambridge; Fellow of St. John's College

Roberta Balstad: Thank you very much, Joel. That was excellent. We're very fortunate to have Sir Partha Dasgupta as our second speaker on this panel. Sir Partha is the Frank Ramsey Professor of Economics and was the previous Chairman of the faculty of economics and politics at the University of Cambridge. In the 1990s he was Chairman of the scientific board of the Beijer International Institute for Ecological Economics in the Royal Swedish Academy of Sciences, and from 1989 to 1992 he was Professor of Economics and Philosophy and Director of the Program in Ethics in Society at Stanford University. His research interests have covered welfare and development economics, the economics of technological change, population, environment and resource economics, the theory of games, and the economics of under nutrition. He is a member of the British Academy, a member of St. John's College, Cambridge, a foreign member of the American Academy of Arts and Sciences, a foreign associate of the US National Academy of Sciences, and of the American Philosophical Society. Sir Partha Dasgupta.

Sir Partha Dasgupta: Thank you very much.

Orthogonal to most other speakers, I want to not give you data today, but I wanted to take you through a puzzle that has bothered me for about fifteen years or so, the resolution to which I had quite some time back, but it's only recently that I realized that it needs speaking about. So I'll conduct a simple piece of analysis with you to resolve a puzzle that probably has occurred to you, it certainly occurred to me some time ago.

Pretty much a third of the lectures we have heard the last day and a half have been on global warming, and climate change in general and global warming in particular. And speaker after speaker have informed us of the magnitude of the phenomenon, and the difficulties humanity is likely to face almost certainly in the next forty, fifty, sixty years time. And yet if you were to read the economics literature, and I represent my profession here today, you would find very little said about global warming, the problems that global warming might raise and what should be done about it. If anything, the suggestion has been that it is not our problem, it's tomorrow's problem.

Now you'll find that in pretty much everything that has been done in the technical literature in economics is estimating the costs and benefits of investing in projects or programs that could curb the process that we are now experiencing. About two years ago, to give you another example of what I mean, about two years ago eight eminent economists were invited to Copenhagen to discuss the following question: suppose, they were asked, you had fifty billion dollars to spend over a five-year period and here's a list of ten global problems, how would you rank them, where would you spend the money? And of these ten, as reported in The Economist, with considerable amounts of approval, by the way, I should say, the Copenhagen consensus placed global warming and climate change at the bottom of the list. Now that's a puzzle, all right, it's a paradox. The question is is economics beyond repair?

I want to use the fifteen minutes I have with you to show you that the problem is not with economics, but our reluctance to engage with environmental scientists and work through this extraordinary problem. And I say it's an extraordinary problem not because I have any firsthand understanding of it, but I trust my environmental scientists, and many of them are quite outstanding. I tend to hang out a lot with environmental scientists and they have educated me, so I trust them the way they trust my economic analysis, and I don't see any other way forward. So I'm not going to reiterate what you have heard about climate change, but to show you that economic analysis, if pursued honestly, and it requires a certain amount of dispassionate disbelief, intuitions, and I want to share that with you now.

Now the question is why did the eight eminent economists in Copenhagen decide that it was not worth it, that is to say, a project or programs that are designed to curb the process that we are discussing here are not worth their value, that the investment should be made sometime in the future? The reason is that economists tend to use a positive rate to discount future benefits that come from current costs, okay? And a simple intuition of compound interest and compound discounting shows you that small positive rates used for costs and benefits in the future look very small when valued today.

To give you an example, if you were to use a discount rate of 4% a year, a dollar's worth of additional consumption benefits 100 years from now would be worth less than three cents today, or to put it the other way, you would charge more than $30 worth of consumption benefits 100 years from now for giving up $1 today. So you can now see why a project needs to be really productive if it pays off fifty, sixty years down the road then it doesn't look very good. Now having chosen 4% to illustrate, I didn't quite pluck it from air, the custom at least this country amongst my economist colleagues for social projects, now notice we are looking at a global commons issue, so I'm thinking of something like a collective decision as to the amount of investment that needs to be made. For public projects in the United States it is customary to use 4 or 5% discount rate, the argument being that these would be the opportunity cost of the investment that you are putting in, government bonds, safe government bonds, earn approximately that kind of interest, so a competing project ought to bear that kind of a yield, otherwise it's not worth it. And it's that kind of argument which has led most economists to conclude that, as I say, global warming or climate change is tomorrow's problem, not today.

Now I want to remind you that the kind of calculation I want to do in the next five, ten minutes with you is geared to answering the question what should humanity do? We're looking at global commons problem, we're not asking what you as an individual should expect for a dollar's investment when you put it in a bank or wherever, we are asking for a collective decision. So the question is why should we discount as a collective body why should we discount at all?

Now before I give you the full way we economists think of it, and when I say we economists think of it we actually happen to think about it correctly, so it's not a professional opinion, I'll carry the argument with you now so I'm convincing you that that's how you will argue if you thought through it. The societal problem is a humanitarian problem, it's a global commons, but you could use the same argument for any collective decision bearing on your own country, or if you happen to be an African viewing it from Africa's vantage point. And I'll come back to this in a minute. But for the moment I want you to think of it as a global problem. So the question is why do we wish to discount collectively?

Now there are two basic reasons why collectively people would wish to discount future benefits in dollar terms vis a vis today's dollars. First is because it's in the future, and it could be that we are impatient. Certainly at the private level, at the individual level, we do display impatience, we want the benefit now rather than later. So that's the first argument. The second argument is really rather tricky, but it's the one which really carries the day. If you think you're going to be better off twenty years down the road then an extra dollar twenty years down the road will be less valuable when twenty years down the road happens to be today then that same extra dollar today. You need it more now than you would twenty years down the road because you forecast that you'll be richer. So a rising consumption is an argument for wishing to discount at a positive rate. So these are two arguments, and these arguments go back, by the way, centuries amongst moral philosophers. And economists took them up.

Now the first argument has been questioned, and if Peter Singer were here he would say of course the first argument doesn't operate, not for society as a whole. You as an individual may wish to be impatient, but society shouldn't because it's going to be there, or hopefully. I have to say that after day and a half's lectures that I've heard it's not quite clear that that's going to happen, but be that as it may we'll assume that the first one is not operative. So what about the second one? The empirical work, both in collective choices and individual choices, suggest that the rate at which you should discount as a community would be something like three times the percentage rate of change in per capita consumption, if you want to use averages there. It's a very crude number, I can be questioned, but it's in the right ballpark. If you want to take something like seven or eight times that you'll have to justify it. For our purposes here I'm going to suppose that it's something like three times the expected rate of change in per capita consumption.

Now here's the kicker. I spend an awful lot of time in America because it's one of my three favorite countries. The others are England and the United Kingdom. And one thing I've discovered is that my American friends have never, but never, thought about declining consumption, because they've experienced nothing but growth. But let me remind you that Africa, sub-Saharan Africa, over the past thirty years or so has actually experienced declining consumption per capita. Now here is the theorem then. If rising consumption is an argument for using a positive rate to discount future benefits, then declining consumption per head is an argument for discounting future consumption benefits at a negative rate. Okay? It's the flip side of the argument that I've just now given you.

Now suppose you take climate change seriously, as we have been urged to since yesterday morning, suppose you think that there is a reasonable chance that certain parts of Earth are going to suffer significant economic losses. Usually these figures are given in global terms, but if disaggregated, when I've spoken to environmental scientists they suggest that the tropics are going to get hit on average. The temperate zones may experience positive benefits from climate change, so I've been told, but there will be a significant difference in the regional distribution of benefits and costs. So I want you to think about the possible scenario to see how the argument goes. Suppose certain regions of Earth suffer substantial losses in income over the coming fifty, sixty, hundred years, due to the variety of reasons that have been mentioned in a day and a half, since yesterday morning. Then how would that argument run? So I'm going to give you an example of how discounting works in that kind of a situation.

Consider a scenario, it's literally taken from scattered sources, a scenario in which global consumption per capita, now this consumption per capita, it's a weighted average over various regions. So if you think that the really poor if they suffer some more ought to be given a greater weight for their suffering than the benefits that might happen to already rich people, then something like a weighted average is what I have in mind here. Consider a scenario in which global consumption per capita increases at an annual rate of .5% for the next fifty years, and then declines at 1% a year for the following 100 years. What discount rate should the global community use? Using the formula that I've just given you, it means that for the next fifty years we should be 1.5% discount rate, that's three times .5, because it's increasing so it's a positive rate, but then for the following 100 years we should be using a negative discount before the forecast is that the per capita income would be declining, and the rate would be 3% for the subsequent 100 years coming from three times -1. Notice then the discount rate isn't plucked from air, it has to be anchored to the forecast you make because you're asking yourself what would happen, how should I view a perturbation to the forecast? And the perturbation is the project and the programs that you are thinking of for global climate change.

Now if you do this calculation it turns out, it's very easy to do it, that a dollar's worth of additional consumption 150 years from now is worth $9 of additional consumption today, or to put it the other way, the global community should be willing to forego $9 worth of additional consumption today for an extra dollar's worth of consumption benefits 150 years in the future. Of course it's a completely different story from the Copenhagen consensus. Even although the negative discounting comes fifty years from now, so it's a fairly optimistic scenario for the first fifty years, the negative discounting means that future losses are magnified when viewed from today, whereas positive discounting of course attenuates it, and that's the mystery, and that results in the puzzle. In other words, if you take climate change seriously, then it's not possible to take the Copenhagen consensus attitude that we will take a positive discount rate because that's the opportunity cost of capital.

My final comments really are the following. This doesn't mean that the private rate of return on investment, even in the scenario that I'm depicting, will be necessarily negative, not so. You could have the rate of return to private investors being positive while the community discount rate is negative. And the reason is that we are looking at a global commons problem, where on the business per usual we are looking at a problem which is a tragedy of the commons. There is absolutely no reason why the rates of return, the social rates of return, on investment should be the same as private rates of return. And we are used to this kind of reasoning that there is a wedge, there can be a wedge, between private and social. What we are not used to is the fact that the difference can be so great that one can be positive and the other the negative. Now that's a theoretical possibility, it's an actuality. That is to say, if you were a government advisor in some arbitrary country in sub-Saharan Africa in 1970 and you had a reasonable forecast of what was in store for sub-Saharan Africa, you would be absolutely insane to use a positive discount rate to evaluate public projects.

Thank you very much.