EMF Publications


EMF WP 12.19 Sorting Out Facts and Uncertainties in Economic Response

Working Paper

Author
Gary W. Yohe

Published by
Stanford University, 1993


This paper begins in Section I with a taxonomy of surprises which (1) recognizes the complication involved in understanding the scope of the natural and social processes which drive global change but (2) simplifies the portrait of the underlying natural and social systems in an effort to uncover a practical way to proceed. Focusing upon the possibility that strongly correlated, low likelihood tails of distributions of effects might feed exaggerated impacts into nonlinear damage functions provides a means with which to capture the limit of “foreseeable surprises” which (1) meets the researchers’ need for quantification and (2) recognizes the potential importance of including the damage associated with surprise into current economic analyses of various policy options designed to abate global change and thereby diminish further their relative likelihood.

Section II explores the “So what?” question. Building upon an analysis of the efficient U.S. response to the threat of greenhouse warming to her own well-being produced by William Nordhaus [1990] along a “best guess” trajectory of damage, an “uncertainty multiplier” is computed for a wide range of other, less likely scenarios. Taking the IPCC range of temperature sensitivity to a doubling of (effective) carbon concentrations as a measure of the uncertainty which surrounds current understanding of the physical phenomenon behind global warming, the Nordhaus measure of the marginal cost associated with the efficient response for the U.S. to the effects of doubling is multiplied by as little as 0.46 and as much as 14.99. The expected value of this uncertainty multiplier is 2.64, a value which increases the efficient reduction in cumulative carbon emissions for the United States through the year 2050 from 6% to nearly 15%.

Building a method of systematically recognizing the exaggerated effect of the unfortunate confidence of unlikely events and nonlinear damages clearly makes a difference. The precise numerical results recorded here are, of course, dependent upon the modeling and the interpretation of physical and economic data offered in their support. Some generalization of their significance, with particular reference to the broader issues of application is offered in Section III, but the implication of the actual multiplier computed here cannot be ignored. The effect of incorporating the potential cost of low-likelihood scenarios of what the future might hold is large enough to increase substantially the efficient response of the United States to the possible damage of greenhouse warming felt within her borders.