It Is Time to Reconsider the 3% Discount Rate

Health technology assessment (HTA) guidance often recommends a 3% real annual discount rate, the appropriateness of which has received limited attention. This article seeks to identify an appropriate rate for high-income countries because it can influence projected cost-effectiveness and hence resource allocation recommendations. The author conducted 2 Pubmed.gov searches. The first sought articles on the theory for selecting a rate. The second sought HTA guidance documents. The first search yielded 21 articles describing 2 approaches. The “Ramsey Equation” sums contributions by 4 factors: pure time preference, catastrophic risk, wealth effect, and macroeconomic risk. The first 3 factors increase the discount rate because they indicate future impacts are less important, whereas the last, suggesting greater future need, decreases the discount rate. A fifth factor—project-specific risk—increases the discount rate but does not appear in the Ramsey Equation. Market interest rates represent a second approach for identifying a discount rate because they represent competing investment returns and hence opportunity costs. The second search identified HTA guidelines for 32 high-income countries. Twenty-two provide no explicit rationale for their recommended rates, 8 appeal to market interest rates, 3 to consistency, and 3 to Ramsey Equation factors. Declining consumption growth and real interest rates imply HTA guidance should reduce recommended discount rates to 1.5 to 2+%. This change will improve projected cost-effectiveness for therapies with long-term benefits and increase the impact of accounting for long-term drug price dynamics, including reduced prices attending loss of market exclusivity. •Health technology assessment guidance often recommends an annual discount rate of 3%, but most guidance provides no supporting rationale. The literature advances 2 approaches to identify an appropriate rate: (1) bottom up (the Ramsey Equation) sums contributing factors, but economists disagree on their values; (2) real interest rates can represent returns available from other investments, but which market rate to use is unclear.•Pragmatically, we can start with the commonly recommended 3% discount rate and consider trends suggesting recent changes to this value. Projected declines in consumption growth and observed declines in real interest rates suggest health technology assessments should now recommend a discount rate of 1.5% to the low 2% range.•Using a lower discoun