Brief Summary of:
Anthony M. Barrett*,† (firstname.lastname@example.org)
†Global Catastrophic Risk Institute
Forthcoming in Decision Analysis
This version dated 4 May 2017.
Key Points Include:
· There could be great value in comprehensive, integrated assessment of global catastrophic risks and risk-reduction options.
o Ideally, assessment would address all important GCRs over key time periods (e.g. the next century) and options of stakeholders (including, but not limited to, public policy options of governments).
· For tractability, begin by taking a broad-but-shallow approach (with very approximate estimates/models), then add depth iteratively to continually improve our understanding.
· Use value-of-information considerations to guide research decisions on where to focus research as part of a GCR research agenda.
o In the context of decision analysis, the value of information derives from reducing uncertainties to help decision makers to achieve better decision outcomes.
· A risk reduction cost effectiveness based approach to assessing value of information simplifies some problems in GCR comparison.
In this paper, we develop and illustrate a framework for determining the potential value of global catastrophic risk (GCR) research in reducing uncertainties in the assessment of GCR risk levels and the effectiveness of risk-reduction options. The framework uses the decision-analysis concept of the expected value of perfect information (EVPI) in terms of the cost-effectiveness of GCR reduction. We illustrate these concepts using available information on impact risks from two types of near earth objects (asteroids or extinct comets) as well as nuclear war, and consideration of two risk reduction measures. We also discuss key challenges in extending the calculations to all GCRs and risk-reduction options, as part of an agenda for comprehensive, integrated GCR research. While real-world research would not result in perfect information, even imperfect information could have significant value in informing GCR reduction decisions. Unlike most value of information approaches, our equation for calculating value of information is based on risk reduction cost effectiveness, to avoid implicitly equating lives and dollars e.g. using a value of statistical life (VSL), which may be inappropriate given the scale of GCRs. Our equation for value of information may be useful in other domains where VSLs would not be appropriate.