Prof. Baldwin and Sue Wing Forecast U.S. CO2 Emissions to Exceed Official Estimates in Recent Journal of Regional Science Paper

in 2013, 2013, Faculty, Ian Sue Wing, James Baldwin, Publications, September-13
September 25th, 2013

Recent developments in U.S. climate change policy have seen the first tentative steps toward legislating a binding aggregate emission cap and implementing curbs on GHGs at the state and regional levels.1 This state and regional level policy action has been identified as both a critical element in U.S. emissions reductions and as a force to shape national climate change mitigation policy (Byrne et al., 2007; Lutsey and Sperling, 2008; Rabe, 2008). Consequently, the resulting economic effects of these policies is the subject of intense recent interest (Grainger and Kolstad, 2009; Hassett et al., 2009; Sue Wing, 2010). The first step in making any such assessment, and one incorporated or mandated in all state climate action plans (EPA, 2012), is to forecast how states’ baseline emissions are likely to evolve. Prerequisite to such projections is the ability to characterize the geographic variations in the precursors of GHGs—particularly CO2—based on an understanding of their historical evolution.

In this paper we investigate how the driving forces behind U.S. carbon dioxide emissions have evolved over the period 1963–2008. We take an explicitly spatial approach, quantifying in detail the interregional variations in CO2 precursors that are largely absent in the literature. While several recent papers have exploited state-level databases on the prices and quantities of fuel use, their focus has been quantifying the aggregate effects of drivers such as income and prices.2 The unfortunate consequence is that the substantial interregional heterogeneity underlying these results, which is interesting in its own right, has largely been ignored. An important exception to this general trend is Metcalf’s (2008) inquiry into the drivers of the energy intensity of U.S. states, which he disaggregates into intrasectoral changes in energy efficiency and intersectoral changes in the structure of economic activity. This paper’s key feature is the use of index number decomposition analysis, which is a popular technique for apportioning the time-evolution of a composite variable into contributions associated with movements in its constituent factors.3 We build on this approach, developing an extended decomposition framework which attributes the evolution of CO2 emissions over space and time to five precursors: the emissions intensity of energy use, the energy intensity of economic activity, the composition of states’ output, per capita income and population. Click to read entire paper…