Using a zero-discount rate could help choose better projects and help get to net zero carbon
This page in:
Choosing the right discount rates matters for selecting good projects. The authors of this blog provide strong reasons why a rate of zero is appropriate for discounting costs and benefits of most investment projects - and especially climate-sensitive ones.
Estimating future benefits and costs is fundamental to appraising a project. The discount rate is used to express future monetary value in today’s terms. Using a higher discount rate reduces the value of the future stream of net benefits or costs compared with a lower rate. Therefore, a higher discount rate implies that we value benefits less the further they are in the future.
Doing cost-benefit analyses of projects is a difficult but important exercise for selecting the best investments. At a recent World Bank event, Professor Bent Flyvbjerg highlighted the problem of underestimation of costs and overestimation of benefits due to optimism bias and suggested some behavioral responses. Below, we spell out four reasons for adopting a zero-discount rate.
Arrow et al 2013 and Weitzman 2013).A quick review of the relative outcomes for projects 1 and 2 shown in the figure below demonstrates this. The two projects have a similar net present value at a 6% discount rate. However, project 1, which has a long tail of benefits, comes out poorly with higher discount rates. Project 2, which has a shorter period of benefits, comes out relatively worse with lower discount rates. In recent years, research has pointed out the need to move to zero or close to zero for projects with long tails of benefits or costs (see
The figure compares two projects that cost $100 million in year 1, but with project A yielding an annual return of $10 million between years 2 to 30 and for project B an annual return of $20 million between years 2-10. Discount rates impact the net present value of projects and can thereby also change the relative ordering between projects.
Here are the reasons for a discount rate of zero. Three of them general and one specific to climate projects:
Consumption/wealth might not continue to increase:
A sustained increase of consumption due to continued growth is one of the fundamental assumptions of a positive social discount rate. The past few years have shown that sustained growth cannot be assumed. Recent research suggests that about half the children born in the U.S. in the 1980s are likely to have income that is at best equal to their parents’, and this was before the dislocation of COVID. Climate change is both exacerbated by consumption and constitutes a rapidly aggravating negative supply shock. With a declining number of workers in many countries, it is hard to see output growth topping 1 percent in many places. If incomes do not increase, then marginal gains from consumption do not necessarily reduce with time and discounting the value of future consumption does not necessarily hold.
Discounting basic rights is philosophically questionable:
A social discount rate based on marginal utility of consumption should not be applied to the value of life or human suffering. When estimating the present value of losses from future extreme weather events or slow-onset disasters, let’s be mindful that these measures of loss can be a matter of life or death for vulnerable people. This type of consumption does not match well with the optimal growth model from which social discount rates are derived. The mathematician Frank Ramsey and others have argued that while discounting made sense on behalf of an individual it was ethically indefensible for society as a whole – the lives of all generations should be treated equally. While accounting for these social costs might be a necessary exercise for decision making, reducing their weight through discounting is ethically questionable.
Climate change is an existential rather than a marginal threat:
Discounted cash flow analysis is generally conducted on the premise that the project itself is marginal and doesn’t generally affect relative prices. But this might not apply when it comes to climate change massive projects in small places (like Nam Theun II in the Lao People’s Democratic Republic). Climate change is a massive change for (very finite) Planet Earth and has the potential to dramatically affect prices and growth. From that perspective, global guidance on social discount rates can shape investment decisions similarly to the way monetary policy does with interest rates. Lower social discount rates would promote projects with long-term benefits for the climate and be complementary to other methods such as setting carbon prices. Retaining coal would look worse.
A zero-discount rate limits the risks of moral hazard:
Problems of systematic bias and strong incentives for project teams and decision makers to arrive at positive present values make it necessary to set strict institutional guidelines. Cost-benefit analysis results are often driven by parameters – such as discount rates -- on which there is no scientific agreement or consensus. Swedish reviewers have noted an unsystematic range of discount rates applied for environmental projects. The review found that justifications were weak and non-transparent, which reduced accountability. If discount rates are in play, there is an incentive for project sponsors to think of planning cash flows in terms of the discounted outcome, which could incite opportunistic discretion over the timing of planned cash flows. An institutional commitment to set discount rates at a fixed number would remove one opportunity to massage the numbers.
The zero response
Zero makes for a simply communicated attention grabber. A zero-discount rate also takes a neutral stance on whether consumption growth will remain positive or dip because of climate change. Zero ensures ethical treatment between generations and of losses in human life. Finally,It has commitment value.
Zero may not be a perfect number in the strict technical sense, but it may be a very good number to improve project selection.
Join the Conversation
Interesting thoughts. However, these arguments ignore a more fundamental one -- that projects with low social rates of return will look "better" and public funds will be diverted to them, rather than spent on interventions that have a high social rate of return. Poverty reduction today is clearly better than poverty reduction in 30 years, not least because when people have higher incomes they reinvest it in their children and build societal wealth. What is the moral argument for taking resources out of income generation for the poor today to address the climate damage that the rich world has largely created? The social opportunity cost of capital in poor countries is high, so we should set a DR so that investments exceed that SOCC, regardless of the possibly lower social rate of time preference across generations. Higher returns should always be sought, including for climate-friendly interventions, and other means can be used to ensure that they are pursued than a generally lower DR.
When calculating the social rate of return, the objective is to account for the total benefits to society from a project. A lower discount rate does not mean that projects with low social rates of return will be favored over projects with higher ones. It mostly means that longer time horizons will be considered in the decision making process. In your example of poverty reduction with generational benefits on wealth accumulation, a lower discount rate will make that project look even better because the benefits on the next generation will not be discounted as much when comparing to a project with only short term benefits. In any event, the funding decision (to which you refer) is separate from the benefit cost analysis. Our task is to inform a political process with analysis – resource constraints may mean rationing between projects. The decision criteria for allocating financing may be different across jurisdiction and through time. The point is to fund only projects that are positive.