The debate on discount rates is a long-standing one. As SAR notes (IPCC, 1996a, Chapter 4), there are two approaches to discounting; an ethical, or prescriptive, approach based on what rates of discount should be applied, and a descriptive approach based on what rates of discount people (savers as well as investors) actually apply in their day-to-day decisions. SAR notes that the former lead to relatively low rates of discount (around 2%–33% in real terms) and the latter to relatively higher rates (at least 6% and, in some cases, very much higher rates).

The ethical approach applies the so-called social rate of time discount, which is the sum of the rate of pure time-preference and the rate of increase of welfare derived from higher per capita incomes in the future. The descriptive approach takes into consideration the market rate of return to investments, whereby conceptually funds can be invested in projects that earn such returns, with the proceeds being used to increase the consumption for future generations. Portney and Weyant (1999) provide a good overview of the literature on the issue of intergenerational equity and discounting.

For climate change the assessment of mitigation programmes and the analysis of impacts caused by climate change need to be distinguished. The choice of discount rates applied in cost assessment should depend on whether the perspective taken is the social or private case. The issues involved in the application of discount rates in this context are addressed below.

For mitigation effects, the country must base its decisions at least partly on discount rates that reflect the opportunity cost of capital. In developed countries rates around 4%–6% are probably justified. Rates of this level are in fact used for the appraisal of public sector projects in the European Union (EU) (Watts, 1999). In developing countries the rate could be as high as 10%–12%. The international banks use these rates, for example, in appraising investment projects in developing countries. It is more of a challenge, therefore, to argue that climate change mitigation projects should face different rates, unless the mitigation project is of very long duration. These rates do not reflect private rates of return, which typically need to be considerably higher to justify the project, potentially between 10% and 25%.

For climate change impacts, the long-term nature of the problem is the key issue. The benefits of reduced GHG emissions vary with the time of emissions reduction, with the atmospheric GHG concentration at the reduction time, and with the total GHG concentrations more than 100 years after the emissions reduction. These are very difficult to assess.

Any “realistic” discount rate used to discount the impacts of increased climate change impacts would render the damages, which occur over long periods of time, very small. With a horizon of around 200 years, a discount rate of 4% implies that damages of US$1 at the end the period are valued at 0.04 cents today. At 8% the same damages are worth 0.00002 cents today. Hence, at discount rates in this range the damages associated with climate change become very small and even disappear (Cline, 1993).

A separate issue is that of the discount rate to be applied to carbon. In a mitigation cost study, should reductions of GHG in the future be valued less than reductions today? It could argued that this is the case, as the impacts of future reductions will be less. This is especially true of “sink” projects, some of which will yield carbon benefits well into the future. Most estimates of the cost of reductions in GHGs do not, however apply a discount rate to the carbon changes. Instead, they simply take the average amount of carbon stored or reduced over the project lifetime (referred to as flow summation) or take the amount of carbon stored or reduced per year (flow summation divided by the number of years). Both these methods are inferior to the application of a discount rate to allow for the greater benefit of present reductions over future reductions. The actual value, however remains a matter of disagreement, but the case for anything more than a very low rate is hard to make (Boscolo et al., 1998).

More recent analysis on discounting now examines rates that vary with the time period considered. In surveys of individual trade-offs over time, Cropper et al. (1994) estimated a nominal rate of around 16.8%, based on a sophisticated questionnaire approach to valuing present versus future risks. Most importantly, however, these authors found evidence that respondents do not discount future lives saved at a constant exponential rate of discount. Rather, median rates seem to be decline over time (i.e., a rate is not constant over time but decreases as the time horizon lengthens). Using different econometric specifications that allow the discount rate to decline over time, Cropper et al. (1994) estimate that mean discount rates are greater for short time periods relative to long time horizons. For example, fitting their data to a hyperbolic function suggests that mean discount rate is 0.80 for 1 year and 0.08 for 100 years. While the pattern is consistent, the implied rates using linear discount rate functions are much larger: 34% for the initial period and about 12% for the last period.

Hyperbolic discounting implies that a person’s relative evaluation of two payments depends on both the delay between the two payments and when this delay will occur–sooner or later. For instance, people often have an impulsive preference for immediate reward. Some people prefer to receive US$1000 today over US$1010 in a month’s time, and yet they also prefer US$1010 in 21 months to US$1000 in 20 months, even though both choices involve a month’s wait to obtain $10 more (see Lowenstein and Prelec, 1992). Theoretical support for hyperbolic discount rests on the idea that, while interest rates from financial instruments can be used to identify appropriate discount rates for time horizons of a few decades, they do not apply to future interest rates for far distant horizons. These will be determined by future opportunity sets created by many factors, such as economic growth. The fact that the scope of these future opportunity sets for the far distant future is not known adds another layer of uncertainty into climate policy, which tends to drive discount rates down.

Weitzman (1998) surveyed 1700 professional economists and found that (a) economists believe that lower rates should be applied to problems with long time horizons, such as that being discussed here, and (b) they distinguish between the immediate and, step by step, the far distant future. The discount rate implied by the analysis falls progressively, from 4% to 0%, as the perspective shifts from the immediate (up to 5 years hence) to the far distant future (beyond 300 years). Weitzman (1998) suggests the appropriate discount rate for long-lived projects is less than 2%. Finally, hyperbolic discounting has less support if it leads to time-inconsistent planning, as argued by Cropper and Laibson (1999). Time inconsistency arises when a policymaker has an incentive to deviate from a plan made with another person, say in the future, even when no new information has emerged. Policymakers of today try to commit future policymakers to a development path that is sustainable. But when the future actually arrives, these new policymakers deviate from the sustainable path and reallocate resources that are efficiently based on prevailing interest rates.

Finally the case is made for calculating all intertemporal effects with more than one rate. The arguments outlined above for different rates are unlikely to be resolved, given that they have been an issue since well before climate change. Hence it is good practice to calculate the costs for more than one rate to provide the policymaker with some guidance on how sensitive the results are to the choice of discount rate.⁷ A lower rate based on the ethical considerations is, as noted above, around 3%.

IPCC Homepage