Estimating direct and indirect rebound effects for U.S. households with input-output analysis. Part 2: Simulation

This is the second part of a two-part paper that integrates economic and industrial ecology methods to estimate the indirect rebound effect from residential energy efficiency investments. We apply the model developed in part one to simulate the indirect rebound, given an estimate of the direct rebound, using a 2002 environmentally-extended input-output model and the 2004 Consumer Expenditure Survey (in 2002$) for the U.S. We find an indirect rebound of 5-15% in primary energy and CO2e emissions, assuming a 10% direct rebound, depending on the fuel saved with efficiency and household income. The indirect rebound can be as high as 30-40% in NOx or SO2 emissions for efficiency in natural gas services. The substitution effect modeled in part one is small in most cases, and we discuss appropriate applications for proportional or income elasticity spending assumptions. Large indirect rebound effects occur as the U.S. electric grid becomes less-carbon intensive, in households with large transportation demands, or as energy prices increase. Even in extreme cases, there is limited evidence for backfire, or a rebound effect greater than 100%. Enacting pollution taxes or auctioned permits that internalize the externalities of energy use would ensure that rebound effects unambiguously increase consumers' welfare.

► We apply the models analyzed in part 1 to simulate indirect rebound effects (RE). ► We find U.S. residential direct and indirect RE of 15-50%, based on efficiency type. ► RE using proportional or income elasticity assumptions are good in most cases. ► RE are sensitive to grid emission factor, energy prices, and gasoline budget share. ► Even in cases with large indirect RE, there is no evidence of backfire.