Development of Probabilistic Socio-Economic Emissions Scenarios (2012)
Paper Number: EE-0574
Document Date: 11/01/2012
Author(s): Abt Associates
Subject Area(s): Economic Analysis, Climate Economics, GHG Emission Projections, Uncertainty, Avoided Damages
Keywords: Economic Analysis, Climate Economics, GHG Emission Projections, Uncertainty, Avoided Damages
Abstract:
In support of assessment for policy development, the Administration recently developed a range of values to use in regulatory analysis for quantifying the social costs of adding (or social benefits of removing) one ton of carbon dioxide from the atmosphere. This value is referred to as the ―social cost of carbon‖ (SCC). As a monetary measure of the incremental damage resulting from carbon emissions, the SCC is intended to include the global economic impacts of climate change, including but not limited to effects on agricultural productivity, human health, coastal property, and ecosystem services.
Most SCC estimates have been derived from one of three simulation or dynamic optimization models commonly referred to as integrated assessment models (IAMs): DICE (by William Nordhaus at Yale University; Nordhaus and Boyer, 2000), FUND (by Richard Tol at the Economic Social Research Institute in Dublin, Ireland; Tol, 2002), and PAGE (by Chris Hope at the University of Cambridge; Hope, 2006). These IAMs combine reduced-form representations of climate processes, economic growth, and feedbacks between the two in a single modeling framework. Ongoing work seeks to update these models by incorporating more of these complex interactions and improving the representation of physical and economic processes.
In the summer of 2009, an interagency group developed a set of interim SCC values based on existing estimates in the literature for use in Federal regulatory analysis until a more comprehensive analysis could be conducted. Subsequently, the interagency group convened to discuss key inputs and assumptions that were then used to generate SCC estimates based on DICE, PAGE, and FUND. An extensive review of the literature was conducted to select three sets of input parameters for these models: climate sensitivity, socio-economic parameters, and the discount rate. Since each IAM takes a different approach to modeling damages, all other model features were left unchanged, relying on the model developers' best estimates. The Federal government has set a preliminary goal of revisiting the SCC values within two years. In the meantime, the participating agencies, including EPA, are interested in determining how these modeling frameworks can be improved so the latest scientific and economic research are better represented in policy and regulatory analyses.
Reference socio-economic scenarios are closely tied to climate damages because, all else equal, more and wealthier people tend to emit more greenhouse gases and also have a higher (absolute) willingness to pay to avoid climate disruptions. However, there exists significant uncertainty in key parameters that underlie such projections. In the 2009-2010 U.S. Interagency Workgroup on the Social Cost of Carbon this uncertainty was not directly addressed. Instead, a scenario approach was undertaken in which four business-as-usual scenarios were considered along with a fifth "international policy" scenario represented by averaging across four model runs constrained not to exceed atmospheric concentrations of 550 ppm CO2 by 2100. In the final analysis the five scenarios were given equal weight, implicitly assuming a 20% probability for each.
Without explicit probabilities depicting the likelihood of each scenario the interagency workgroup applied equal weights, implicitly assuming each outcome to be equally likely. Schneider (2001) and Webster et al. (2003) have suggested that treating scenarios as equally likely, even when they are not in reality, can be expected if they are not attached specific probabilities by the scenario developers.
The purpose of this analysis is to help overcome these limitations through the development of a publicly available library of socio-economic-emissions projections derived from a systematic examination of uncertainty in key underlying model parameters, which may then be used in probabilistic damage assessments.
This paper is part of the Environmental Economics Research Inventory.