Working Paper: CEPR ID: DP16007
Authors: Christian Gollier
Abstract: Because of risk aversion, any sensible investment valuation system should value less projects that contribute more to the aggregate risk, i.e., that have a larger income-elasticity of net benefits. In theory, this is done by adjusting discount rates to consumption betas. But in reality, for various reasons (Arrow-Lind and WACC fallacies, market failures), most public and private institutions and people use a discount rate that is rather insensitive to the risk profile of their investment projects. I show in this paper that the economic consequences of the implied misallocation of capital are dire. To do this, I calibrate a Lucas model in which the investment opportunity set contains a myriad of projects with different expected returns and risk profiles. The welfare loss of using a single discount rate is equivalent to a permanent reduction in consumption that lies somewhere between 15% and 45%, depending upon which familiar discounting system is used. Economists should devote more energy to support a reform of public discounting systems in favor of what has been advocated by the normative interpretation of modern asset pricing theories over the last four decades.
Keywords: discounting; investment theory; asset pricing; carbon pricing; arrow-lind theorem; wacc fallacy; rare disasters; capital budgeting
JEL Codes: G12; H43; Q54
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| single discount rate (H43) | misallocation of capital (E22) |
| misallocation of capital (E22) | welfare loss (D69) |
| absence of risk adjustment in discounting (D15) | overvaluation of high beta projects (H43) |
| absence of risk adjustment in discounting (D15) | undervaluation of projects that hedge against macroeconomic risks (H43) |
| risk-free rate as discount rate (H43) | welfare loss (D69) |
| misinterpretation of Arrow-Lind theorem (D79) | misallocation of capital (E22) |