An oft-used prediction function in economics may wildly underestimate the future value of today's investments, especially those relating to the environment, says a recent Bloomberg column that mentions SFI External Professor John Geanakoplos and SFI Professor J. Doyne Farmer.
Exponential discounting is frequently used in cost/benefit analyses of policy options with future payoffs. But when policy makers invest in long-horizon projects such as carbon emissions reduction programs, exponential discounting might not provide reliable future values for investments made today.
Standard discounting says that 96 dollars invested at 4 percent annually will turn into 100 dollars a year from now, so the value of having 96 dollars today is greater than the value of having 96 dollars in the future. It would take 100 future dollars to match the value of the present-day 96. When evaluating the potential payoffs from investments, economists usually discount future resource values by a fixed percentage for each passing year. This leads to vanishingly small estimates for values in the distant future.
But Geanakoplos and Farmer argue that because real-world interest rates fluctuate constantly, the exponential discounting model is a poor indicator of long-term worth. They advocate for a "hyperbolic discounting" model, which better accounts for these fluctuations.
Their model shows that after 500 years, the value of an investment could be a million times greater than what the standard exponential model would show. This means that the people working for agencies such as the Environmental Protection Agency and the Intergovernmental Panel on Climate Change may be dramatically miscalculating the long-term impact of their decisions.
Read the Bloomberg column (July 27, 2011)
Read their paper (August 2009)