Tuesday, April 9, 2013

Heal and Milner on Climate Uncertainties

Geoffrey Heal (Columbia) and Antony Millner (LSE) have produced a terrific working paper for the National Bureau of Economic Research (here), which does several important things: (1) it disaggregates the various kinds of climate-related uncertainty that impede rational policy-making and implementation; (2) it explains why a great deal of climate-related uncertainty is unlikely to be substantially reduced (or reduced enough) over the next two decades to make rational policy-making easier; and (3) they begin a long-overdue investigation of non-probabilistically-based alternatives to expected valuation-based decision-making for climate policy, including Savage's theory of Subjective Expected Utility, various maximin-based criteria, the Smooth Ambiguity Model, and Variational Preferences. Finally, they emphasize that deep uncertainty about various aspects of the climate problem (more accurately, complex of problems) are not an excuse for doing nothing because "[w]e know more than enough about climate change to know that it is a serious problem that requires immediate policy attention: uncertainty does not imply ignorance."

Here is the abstract:
Uncertainty is intrinsic to climate change: we know that the climate is changing, but not precisely how fast or in what ways. Nor do we understand fully the social and economic consequences of these changes, or the options that will be available for reducing climate change. Furthermore the uncertainty about these issues is not readily quantified and expressed in probabilistic terms: we are facing deep uncertainty or ambiguity rather than risk in the classical sense, rendering the classical expected utility framework of limited value. We review the sources of uncertainty about all aspects of climate change and resolve these into various components, commenting on their relative importance. Then we review decision-making frameworks that are appropriate in the absence of quantitative probabilistic information, including non-probabilistic approaches and those based on multiple priors, and discuss their application in climate change economics.
Heal and Millner are not the only social scientists working to improve our understanding of the uncertainties climate policy-makers confront. In his talk yesterday at the Ostrom Workshop, Scott Barrett focused on an self-evidently important distinction between "threshold uncertainty" (where tipping points may occur) and "impact uncertainty" (the consequences from passing tipping points).

No comments:

Post a Comment

I actively moderate comments for spam, advertisements, and abusive or offensive language.