Uncertainty and Climate Change Geoffrey Heal∗ Columbia Business School, New York, NY 10027. Bengt Kriström SLU-Umeå, 901 83 Umeå, Sweden. November 2001, revised February 2002.
Abstract Uncertainty is pervasive in analysis of climate change. How should economists allow for this? And how have they allowed for it? This paper reviews both of these questions. Key Words: Climate change, uncertainty, risk. JEL ClassiÞcation: D 80, Q 00
1
Overview
The extent to which the earth’s climate will change as a result of human activity is still unclear. Although the basic science is well-understood and now seems conÞrmed by the data available, the precise magnitudes of the various effects that contribute to the climate regime are not yet known. The most recent assessment by the UN Intergovernmental Panel on Climate Change (IPCC), the Third Assessment Report (TAR), has for the Þrst time made an explicit attempt to give an indication of the degrees of uncertainty associated with its various predictions.1 The TAR gives an indication of the degree of conÞdence that the leading authorities in the Þeld of climate change have in their forecasts. In some cases it is clear that, while the sign of an impact is clearly known, there remains massive uncertainty about the size, with the range of possible magnitudes involving differences of several hundred percent. This is illustrated by the IPCC’s widely-quoted range for the possible change in ∗
Contact author - Heal at
[email protected], www.gsb.columbia.edu/faculty/gheal. An earlier review of the extent and nature of uncertainty in climate forecasting can be found in Roughgarden and Schneider (1999). 1
1
global mean temperature, which goes from 1.5 to 6 degrees centigrade.2 In the TAR the IPCC classiÞes its Þndings according to the degree of conÞdence in them. It recognizes Þve different conÞdence levels, ranging from “very high (>95%)” through “high (67
