Thursday, September 02, 2010

Economists' Tunnel Vision On Global Warming

Some economists have studied the return on our investments in controlling global warming by using the typical methodology of their discipline: They assign a money value to the costs of global warming, and then they use a discount rate to determine the Present Value (PV) of these future costs.

The problem is that global warming will have costs that last for many millennia, and this sort of discounting assigns a Present Value of close to zero to any costs that are a few thousand years in the future.

It does make sense to discount future financial costs in this way, because we can earn compound interest on the money to compensate for that future cost. But money will not compensate for the cost of global warming completely, as it would for a financial cost. If we do not control global warming, then even if people living thousands of years from now are very wealthy financially, they will still be living in a world that has lost most of the species that now exist, that has dead acidified oceans, that suffers from widespread desertification, and so on.

The two best known studies of the costs and benefits of global warming are by Richard Nordhaus, who uses a discount rate of 3% and concludes that most investments in controlling global warming are not justified economically, and by Nicholas Stern, who uses a discount rate of 1.4% and concludes that massive investments in controlling global warming are justified.

But even at Stern's low discount rate, most of the future cost of global warming is discounted to a PV of near zero. Using very rounded figures to give a general idea of the orders of magnitude involved:
  • At a discount rate of 1.4 percent, a cost paid 1000 years from now has a PV of less than one one-millionth (10^-6) of the payment.
  • A cost paid 2000 years from now has a PV of less than one one-trillionth (10^-12) of the payment.
  • A cost paid 3000 years from now has a PV of less than one one-quintillionth (10^-18) of the payment.
To put that in perspective, an environmental cost 3000 years from now that is 100 times as great as the total of the world's current GDP (which would involve overwhelming environmental devastation), would have a PV of less than 1 cent using Stern's discount rate.

If people in the year 5000 live on an earth with dead oceans, with most of the land turned to desert, and with most of the species that now exist extinct, they will be shocked to look back and sees that, when the decisions about controlling global warming were being made, 3000 years earlier, even the economist who wanted to do the most to control global warming valued the environmental devastation they face as a cost of less than 1 cent per year.

It does make some sense to use that sort of discounting for flows of money, but it does not make sense to discount massive environmental destruction in a way that makes us conclude that it is not worth spending 1 cent a year today to avoid having a mostly dead earth 3000 years from now.

I sometimes feel that academic disciplines today are a bit like scholastic philosophy during the Renaissance, when the humanists ridiculed the scholastics for debating about how many angels can dance on the head of a pin. In these debates about discount rates, economists are trying to decide exactly how many years in the future it will be before a mostly dead earth is counted as a cost of less than 1 cent.

The problem is that academics have to publish in peer reviewed journals to get tenure, and the best way to get published is by using the conventional methodology of the field and applying it to a problem. You don't get very far by rejecting the conventional methodology. That means economists study global warming by assigning a money value to environmental destruction and using the discount rate to determine its PV.

They do this despite the fact that there is no way in the world to come up with a plausible discount rate to apply over thousands of years. A modern financial system has existed for only a few hundred years, and we don't have the slightest idea of what the appropriate discount rate will be in a few thousand years.

They choose the discount rate based on the conclusion that they want to reach - a low discount rate if they believe we should control global warming and a higher discount rate if they believe we should not. But even Stern's low discount rate gives far too little value to effects that are millennia in the future.

When people think concretely about global warming, about what it would actually be like to live in the worst-case scenario that we might be leaving to future generations, no sane person would say that we should not pay 1 cent per year now to avoid having a mostly dead world 3000 years from now.