Department of "HUH!?!?"

The Stern Review on Global Climate Change Once Again

Bill Nordhaus has placed his critique of the Stern Review on Global Climate Change (on the internet at in yellow covers: it is now NBER working paper 12741 at (available for free at Nordhaus is extremely smart, extremely hard-working, extremely knowledgeable on this issue, and plays it straight. Nevertheless, I find myself dissatisfied.

I am not dissatisfied with where Nordhaus gets: I believe that he is right in assessing that Stern's conclusions depend on assumptions about what we today owe the future that are not obvious, that are at least debatable, and that I do not think I share.

I am dissatisfied with how Nordhaus gets there.

Specifically, the hackles on my back rise when Nordhaus writes:

Suppose that scientists discover that that a wrinkle in the climatic system will cause damages equal to 0.01 percent of output starting in 2200 and continuing at that rate thereafter. How large a one-time investment would be justified today to remove the wrinkle starting after two centuries? The answer is that a payment of 15 percent of world consumption today (approximately $7 trillion) would pass the Review’s cost-benefit test. This seems completely absurd. The bizarre result arises because the value of the future consumption stream is so high with near-zero discounting that we would trade off a large fraction of today’s income to increase a far-future income stream by a very tiny fraction. This bizarre implication reminds us of Koopmans’s warning quoted above to proceed cautiously.... The large damages from global warming reflect large and speculative damages in the far-distant future; the impacts now, as in today, are small; and, as I will suggest below, the 20 percent cut in consumption from global-warming might be reduced by an order of magnitude if alternative assumptions about discounting are used...

The only thing I agree with is the final lines: our conclusions about the steps we should take immediately to deal with global climate change would be reduced by an order of magnitude if alternative (and perhaps, I think, better) assumptions about discounting are used. But the rest:

My first problem is with the framing, which Nordhaus uses to generate phrases like: "bizarre result," "completely absurd," and "bizarre implication." A more neutral way of framing the issue would be to say that the Stern Review estimates that each $1 we invest today in reducing the impact of global warming will improve the state of things in 2200 by $36 (adjusted for inflation) then. Is that a good investment for us to make on behalf of our great-great-grandchildren? Perhaps, perhaps not. 2200 is far away. But a 36-to-1 payoff is a big one. On the other hand, people in 2200 will probably be richer and have better technologies than we do; that is a reason to think that efficiency should not be considered separately from issues of distribution, for we don't usually ask the poor (i.e., us now) to pay to improve the condition of the rich (i.e., them in the future).

The question of whether we today should, or are morally obligated to, make the 36-to-1 two-century investments that the Stern Review is debatable. It is worth doing if we have a hurdle rate for such investments of less than 1.8% per year or so. I think our hurdle rate should probably be bigger than that, but I am not sure. I am, however, sure that Stern's position that we should be making such high-payoff long-term investments is neither "completely absurd," "a bizarre result," nor "a bizarre implication."

My second problem is with Nordhaus's rhetorical attempt to shrink the magnitude of the potential problem by talking about the "large and speculative damages in the far-distant future." The implicit economics underlying this rhetoric is that because investing in controlling global warming is a risky and uncertain proposition the appropriate hurdle rate to apply is greater than the hurdle rate on safe investments like long-term inflation-indexed government bonds. But this economics is simply wrong. As Tyler Cowen puts it, uncertainty is not the friend of doing nothing. Investments in controlling global warming are not risk-increasing but risk-reducing ones: they are more like buying insurance than like speculating on unproven technologies. The appropriate hurdle rate is thus lower, not higher, than that for sure things.

And this leads to my third problem with Nordhaus: his claim that:

The Review’s unambiguous conclusions about the need for extreme immediate action will not survive the substitution of discounting assumptions that are consistent with today’s market place...

I don't see that at all. The 20-year inflation-indexed U.S. government bond interest rate is currently 2.1% per year. That is today's marketplace. The appropriate discount rate to be used for long-term risk-reducing "insurance" investments is less than the risk-free rate, and seems plausibly less than 1.8%. I don't buy the assumption that our judgments should be ruled by the prevailing configuration of financial market prices--I am not confident that the market is without failure along this dimension. But Nordhaus apparently does. And as best as I can see the discounting assumptions in today's marketplace are entirely consistent with the conclusions of the Stern Review.

John Quiggin has an excellent discussion of all these issues and more at:

And my not-completely-informed views about the economics of controlling global warming are:

What Do We Owe Our Great Grandchildren?

What do we owe our great-great-great grandchildren? What actions are we obligated to do now in order to diminish the risks to our descendants and our planet from the increasing likelihood of significant global warming and its associated climate change?

Everybody--well, almost everybody: ExxonMobil, U.S. Vice President Richard Cheney, and their paid-for servants and deluded acolytes are exceptions or pretend to be exceptions--understands that when human burn hydrocarbons carbon dioxide goes up into the atmosphere, where it acts like a giant blanket, absorbing infrared radiation coming up from below and warming the earth.

Everybody understands that we really do not know how much global warming a given amount of extra carbon dioxide produces. We have models, we have forecasts, we have projections, but global warming might be a much smaller and might be a much larger problem than the central-case projections of climate models suggest. Everybody--well, almost everybody: ExxonMobil, U.S. Vice President Richard Cheney, and their paid-for servants and deluded acolytes are exceptions or pretend to be exceptions--understands that here uncertainty is not our friend, and certainly not an excuse for inaction. Uncertainty about its effects should lead us to do more to guard against global climate change than if we knew global warming would proceed exactly as the central-case projections forecast.

Everybody--well, almost everybody: ExxonMobil, U.S. Vice President Richard Cheney, et cetera, et cetera--understands that the world's governments, non-profit institutions, and energy companies ought to be spending a much bigger fortune than they currently are on research: research into technologies that generate power without adding carbon dioxide to the atmosphere, research into technologies that such carbon out of the atmosphere into forests or oceans, research into technologies that cool the earth by reflecting more of the sunlight that lands on us.

Everybody--well, almost everybody: U.S. Vice President Richard Cheney, et cetera, et cetera--understands that the burden of dealing with global climate change over the next two generations should be carried by the rich countries of the world. They got to take an easy carbon emissions-intensive path to industrialization and riches. It looks like China, India, and company will not be able to take such an easy path, and it would be unfair to penalize them for the loss of the easy hydro-carbon burning road.

Everybody--well, almost everybody, et cetera, et cetera--understands that now is the time to build the international institutions that will manage our reactions to global climate change over the next several centuries. Now is not the time to disrupt these institutions, or to prevent their creation.

What there is real dispute about is what else we should be doing right now and in the next decade. We economists like to think of things in terms of prices. And when we economists see something going wrong in the sense of having destructive side-effects, we like to tax it. Taxing it makes the individuals who are undertaking actions feel in their wallets the destruction they are causing elsewhere. Maybe the action is still worth doing, and maybe not. Imposing a tax--imposing the right tax--on those who are, say, driving low-mileage SUVs is a way of harnessing the collective intelligence of humanity to deciding in which case the bad side-effects are a reason to stop. But it has to be the right tax.

An SUV going ten miles in the city and burning a gallon of gasoline pumps about 3 kilograms--6.5 pounds--of carbon in the form of carbon dioxide into the atmosphere. Should the extra tax on this--and on all carbon emissions--appropriate for global warming be on the order of five cents a gallon, fifty cents a gallon, or a dollar fifty a gallon? Our views will change as we learn more, but at the moment whether the tax should be five or fifty cents a gallon hinges on a question of moral philosophy: how much do we believe that we owe our distant descendents?

Australian economist John Quiggin has a very illuminating discussion on his website The Stern Review on Global Climate Change (on the internet at which comes down more on the side of fifty cents a gallon, immediately, does so because they project that spending today to reduce carbon emissions is a very good investment for the future. If the world grows in per capita income at about 2% per year, a marginal expenditure of roughly $70 today in cutting carbon emissions would be worth it if it were to enrich the world of 2100 by about an extra $500 of year-2006 purchasing power, once all the damages to the world economy and environment from global warming, costs of adjustment, and so on are taken into account. This looks like a very good deal to Nick Stern and his team.

On the other hand, critics point out that the world today is poor: average GDP per capita at purchasing power parity today is roughly $7000. We expect improvements in and the spread of technology to make the world of 2100, at a 2% per year growth rate much richer than the world of today: $50,000 per capita of year-2006 purchasing power. We today can use the marginal $70 per capita, critics say, much more than the richer people of 2100 will need the $500 they would gain from not having to suffer from the effects of global climate change.

What critics don't often say is that the same logic applies to the world today. The U.S., Japan, and Western Europe today have average incomes of roughly $40,000 per capita. The poorer half of the world's population today have incomes of less than $6,000 per capita. I believe that the same logic which says that we today need our $70 more than the people of 2100 need an extra $500 also tells us that we ought to tax the world's rich in the OECD more and more to fund world development as long as each extra $500 in first-world taxes generates even as little as $70 in extra poor-periphery incomes. If we in the world's rich now are stingy toward the (likely to be much richer) future and want to leave them our environmental mess to deal with, we should be lavish toward our poor brothers and sisters today. If we today are stingy toward our poor brothers and sisters now, we should be lavish toward our descendents.

If you protest that distributional considerations militate against our spending a lot today to reduce global warming risks to the future, you had better be enthusiastic about massive programs to reduce current U.S. and global inequality. At least, that is what you had better be if your positions are based on some moral principle--rather than on the principle that what we have, we hold, and nobody is going to pry it from our hands.