John Gabrieli, Harvard College ’16


It is widely recognized that estimates of the social cost of carbon are extremely sensitive to small changes in the chosen discount rate. However, considerable disagreement remains among economists about whether to use a “risk-free” rate of return of 1-2% or the market rate of return of 6-7%. In order to resolve this question, Weitzman (2007) suggested the use of a “climate investment beta,” analogous to the investment beta used in the capital asset pricing model (CAPM), to estimate the rate of return on climate investments. In a 2013 theory paper, Weitzman established the relationship between the climate investment beta, which he terms the “real project gamma,” and the appropriate discount rate for climate investments. The real project gamma, defined as the ratio between the costs of climate change that are proportional to overall consumption levels and the costs that are independent of consumption levels, has never been empirically estimated. This paper attempts to estimate the real project gamma of carbon abatement by comparing the cost of warming in the last century across a panel of countries and identifying the proportion of consumption-related damages. Ultimately, a cross-country comparison of climate damages suggests an upper bound for gamma of roughly 0.5, significantly lower than the implied gamma of 1 that is assumed by many prominent integrated assessment models of climate damages (IAMs). The implications of this finding on gamma-adjusted discount rates and estimates of the social cost of carbon are discussed.

View Fullscreen