The effect of discounting is strongly nonlinear, however. At a discount rate of one percent, the discounted value of $1 million 300 years hence is around $50,000 today. But if the discount rate is five percent, the discounted value is less than a mere 50 cents. In this example, the discounted value changed by a factor of 20,000 when the discount rate changed by a factor of just one.

Economists disagree about what value to choose for the discount rate when determining an appropriate level of investment in climate change abatement. Stern used an unusually low rate and when that figure was plugged into his computer model, unusually high damage figures came out at the other end. Hence, his call for a high level of investment in climate change abatement today.

Among the most prominent economists studying the costs and benefits of climate change is Bill Nordhaus, who has argued for using a higher discount rate and therefore arrives at less startling results with respect to an economic estimate of the damages from climate change, and with respect to the measures we should take in the near term to mitigate negative impacts.

We would point out that most previous investigations (including the Stern Review and those by Nordhaus and others) do not consider the effects of the changing composition of economic wellbeing and changing relative prices. These changes can have an effect on the calculation of the present value of costs of climate change that is as substantial as the choice of discount rate.

Any discount rate assumes a growing economy. But it's unrealistic to assume constant, unwavering growth, equal for all sectors. Both logic and history indicate that growth tends to be concentrated in some sectors, depending on resources, technical innovations, and consumer preferences. If the output of some material goods (such as mobile phones) increases, but the availability of environmental goods and services (like clean water and biodiversity, or rain-fed agricultural production) declines, then the relative prices (or willingness to pay) for the environmental amenities should rise over time, a fundamental point first made by John Krutilla some 40 years ago.

Because of rising relative prices, the environmental sector could see its share of the economy grow in value even as it becomes physically smaller relative to a growing conventional sector. This has consequences for discounting itself that have been overlooked. In a multi-sector model, discount rates will not generally be constant—nor will they be the same for each sector. There will be a change in the relative prices for goods and services from sectors that grow at different rates.

Accounting for relative price changes can dramatically increase the abatement necessary to mitigate climate change. Using Nordhaus's integrated assessment model for climate change, Sterner and Persson show that using relatively high discount rate parameters used by Nordhaus but also modeling changes in relative prices yields results that are similar to the conclusions of the Stern Review and differ greatly from previous work by Nordhaus and others. If one were to use both low discount rates and changing relative prices, one would find even stronger support for strict and immediate abatement measures than did the Stern Review.

We also have a second concern with the Stern Review—that it may not give sufficient weight to nonmarket damages.

The nonmarket impacts of climate change are at center stage, because it is precisely the prices of these goods and services that we expect to rise over time. Nonmarket impacts from climate change include biodiversity and ecosystem loss, the effects of air pollution on human health, and damage from extreme hurricanes, droughts, and floods. The Stern Review does a great job of presenting many of these, the costs of which could be very high over the coming century: billions of people could suffer water shortages, and tens to hundreds of millions are at risk of hunger, diseases like malaria, and coastal flooding.

Those impacts could also have extreme social consequences if droughts force mass migrations, coastal inundation drives environmental refugees inland, and conflicts erupt over increasingly scarce resources. Such social problems have the potential to make the already serious climate damages much worse. However, social impacts are not included in the Stern analysis, nor have they been included in most other economic analyses. To give a full picture of the costs of climate change and the benefits of mitigation, these impacts should also be taken into account, together with their expected increase in relative value over time.

We believe that it is exactly the nonmarket effects of climate change that are the most worrisome. Given the risk of catastrophes, the main effect of climate change will be not to stop growth in conventional manufacturing, but rather to damage some vital ecosystem services, making them relatively scarcer and raising their relative prices.

In a thorough evaluation of the effect of relative prices, one would assess changes by sector. Clean water, rain-fed agriculture, and some other ecosystem services have particular importance for the very poor, and the climate change damages suffered by the poor are particularly important for human welfare. The extent of the price effect depends heavily on the elasticity of substitution, which measures the change in the composition of willingness to pay for goods and services when relative prices change.

In the meantime, analyses of abatement costs and benefits need to take into account the content of future growth. Future scarcities, whether caused by the changing composition of the economy or by climate change, will lead to rising prices for certain goods and services. Escalating prices for environmental goods and services raise the estimated damage of climate change, counteracting the effect of discounting.

Future scarcity values for nonmarket environmental assets are likely to generate high damage figures, even assuming high discount rates. Combining the low discount rates in the Stern Review with rising relative prices could lead to support for even higher levels of abatement than Stern recommended. This would mean that society should consider atmospheric greenhouse gas concentration targets that Stern deems unrealistic: a target below 450 ppm of CO2 equivalents and consequently even more restrictive stabilization scenarios. Emissions targets such as these are a frightening thought indeed.