Correcting the Error in Gamma Discounting

In “Gamma Discounting” Martin L. Weitzman (2001) concludes that certainty equivalent discount rates should decline significantly over time. He draws this conclusion from fitting a Gamma distribution to the responses of 2,160 economists asked to give a discount rate estimate and calculating effective discount rates from it. This paper shows that Weitzman’s model is based on an erroneous definition of expected present value. Correcting the definition changes Weitzman’s conclusions, explains the Pazner and Razin discrepancy and solves the Weitzman-Gollier puzzle under risk neutrality. The assertions of this paper are corroborated by calculations based on data found in Weitzman (2001).