The public-sector discount rate🍋
Time for a re-think?
The public-sector discount rate plays a central role in determining which government interventions get the green light, and which stay on red.
The Treasury says it intends to review the discount rate once a year, but it has not done so since mid-2020 – almost four years ago. Since then, interest rates – which underpin the official discount rate – have risen sharply, putting Treasury’s guidance out of date. The last time interest rates were at today’s level, the Treasury’s discount rate was 8.0% – three percentage points above the current value.
In the latest NZIER Public Good Insight paper, I argue that it is time to take a more in-depth look at the prescribed rate — rather than just updating the current numbers. This post provides an overview of the main issues and approaches.
A discount rate is used to compare costs and benefits at different points in time
The New Zealand government’s preferred method for assessing public sector interventions is cost-benefit analysis (CBA). CBA involves identifying the costs and benefits, and expressing them in monetary terms so that they can be compared against each other.
Costs and benefits that occur today are thought to have greater value than costs or benefits in the future for two reasons:
Opportunity cost: Resources can be invested and produce more resources for use in the future.
Time preference: People prefer to consume today rather than tomorrow because they are impatient or expect to be better off in the future.
To compare costs and benefits in different time periods, they need to be converted into ‘present values’ through a process called discounting.
In standard approaches, the present value (PV) of future costs and benefits declines at a constant rate, known as the discount rate. With a constant discount rate, the PV of a future cost or benefit occurring in year t is given by:
The Treasury currently recommends a discount rate of 5.0% for most projects. The Treasury’s CBAx guidance also suggests an alternative rate of 2.0% but does not explain how this should be used.
Some government agencies use different discount rates. For example, NZTA uses a rate of 4.0%, and Pharmac uses 3.5%. Having different discount rates for different government organisations is at odds with the Treasury’s current approach and makes it harder to compare different types of government spending.
Two approaches underpin the debate
For decades, policy analysts and academics have been debating how the discount rate should be set. Underlying this debate are two different approaches, each tied to one of the two reasons for discounting presented above:
Social opportunity cost of capital (SOC): The discount rate should be set based on the next best alternative use of government funds measured using market interest rates.
Social rate of time preference (SRTP): The discount rate should be set by specifying society’s preferences for consumption over time.
Under certain assumptions, these approaches should yield the same result. However, these assumptions do not hold in practice, making it difficult to determine which approach is best.
The two approaches typically involve different views about:
whether the resources used in a public-sector investment would otherwise result in increased investment or consumption
whether the discount rate should reflect the risk involved in public-sector projects and how this compares to private sector projects
whether society’s time preferences can be inferred from market prices or should be specified explicitly based on other evidence or ethical reasoning.
The Treasury’s current guidance is based on the SOC approach. It argues that if the government does not undertake the project, it could invest the public funds in the share market and distribute the proceeds. The idea is that if a public investment does not yield at least as high a return as a private investment with similar risk, then society would be better off choosing that alternative instead.
This raises questions about whether investing in the share market is a realistic option for the government and whether share market returns are a good measure for increases in social welfare – which is what the government should care about.
One of the main arguments that people use for the SRTP approach is that it allows society’s preferences to differ from individuals’ preferences that are reflected in market prices. Whereas individuals’ preferences concern how consumption is distributed across different points in their lives, society’s preferences also concern how resources are distributed across different generations.
Many philosophers and economists argue that society should place more weight on the interests of future generations than individuals do. However, this can lead to some counter-intuitive results. For example, Maya Eden (2023) shows that if the market interest rate is 6%, then a 1.5% discount rate implies that it is better to increase the consumption of a 20-year-old by $6 than to increase the consumption of an 80-year-old by $100.
Non-standard approaches allow the discount rate to vary
The standard approaches assume that there is a single discount rate that is the same for all types of costs and benefits and is constant over time. This assumption makes calculating present values simple. But evidence indicates that market interest rates and individual preferences vary depending on the time horizon and the nature of the investment.
Some economists suggest using different discount rates for different costs and benefits or for distinct types of projects. This often reflects the idea that the value of many non-market benefits, such as environmental benefits, should rise relative to market benefits as the relative scarcity of non-market goods increase. In this case, it would be more transparent to model the increasing value of non-market benefits explicitly in CBA than to build it into the discount rate.
In recent years, many countries have moved toward declining discount rates for long-term projects. One reason for this is that it reflects behavioural evidence, which shows that individuals tend to apply lower discount rates to events further in the future. Another reason is that uncertainty about the discount rate results in an expected rate that declines over time. In some cases, declining discount rates can lead to time inconsistency – implying that the government should make decisions that it would reverse in the future.
Simplicity and consistency are key
As this discussion shows, determining the appropriate discount rate is a complex challenge. It is important to keep in mind that discounting is an analytical tool designed to help support political decisions by democratically elected representatives.
While many complex issues need to be considered, the discount rate needs to be as simple and easy to use as possible – the aim is to have a common approach applied widely across public investments. Decision makers and the analysts who support them need to understand what the discount rate represents and how it is determined.
Careful political and ethical judgements are required
Arguments about the discount rate are often motivated by political views about the role of government in society or ethical views about what we owe to future generations. Policy advisors face difficulties in making these judgements on behalf of society without clear evidence of New Zealanders’ views on these complex issues.
Incorporating value judgements into the discount rate that do not reflect New Zealanders’ views could potentially undermine confidence in government decisions. The credibility of public sector advice depends on taking a balanced and transparent approach.
By Adrian Katz.
Adrian is a Senior Economist at NZIER.
 | A guest post by
|
One thing that is not discussed here is whether the discount rate is a contributor to the short-termism endemic in New Zealand public policy making. Why value policies that pay off in the future when the discount rate - indeed the entire theoretical and philosophical background to the rationale for a discount rate - is that we should and will prefer current expenditure pay-off versus a pay-off that may be quite a way down the track, including for future generations? Think infrastructure, think ferries, think national super, think water quality, think climate change, think urban sprawl. All of these, and more, ask us to defer immediate gratification for future generations and a desirable future state for New Zealand. Could it be that the discount rate as constituted is an ideological bias towards presentism that is foisted onto us by the great intellects of economics past and present? Yet the young will have to live through the future we are bequesting them (and which we will not have to face)!