Measuring the impact of sprawl and housing stock characteristics on greenhouse gas emissions from home energy use

This paper investigates the strength of the statistical relationship between various urban form characteristics and the carbon emissions that result from household electricity and fuel use. Using data from the 100 largest metropolitan statistical areas in the United States, taken in 2000 and 2005, OLS regression methods are used to measure the correlations between per capita emissions from household energy use and independent variables describing both housing form and urban sprawl. After controlling for differences in fuel mix, energy price, and income, the research finds significantly higher emissions associated with a greater incidence of detached single-family housing when compared against high-rise buildings containing twenty or more units. Notably, it does not find the same effect for houses in smaller multi-unit buildings. It also finds significantly lower emissions associated with an increased incidence of row housing when compared against detached single-family homes. This analysis also finds a positive correlation between household energy emissions and both a history of greater rates of land conversion. This paper concludes that the literature regarding smart growth and new urbanism should explore potential impacts on household energy consumption in its discussion of urban sprawl in addition to considering impacts on VMT and auto emissions. It is likely that failure to do so results in an under-estimation of the potential emissions-cutting benefits of denser urban design.