A one-year subsidy program for energy-efficient home appliances has been implemented in China. We construct a dataset consisting of participant and non-participant households in both urban and rural areas of Rizhao city. By applying a quantile regression and counterfactual analysis, this study disentangles the impact of this subsidy program and inherent household attributes on per capita residential electricity consumption. First, contrary to the mean regression, the elasticity of electricity consumption to household's income, age, education and energy-saving awareness, varies markedly across the electricity consumption distribution and shows discrepancy between urban and rural areas. Second, while inherent household attributes are identified as a primary determinant to the changes in residential electricity consumption, the effect induced by the subsidy incentive is more significant at the middle of the electricity consumption distribution than at the tails. Third, there are significant rebound effects that lead to overall increase in household electricity consumption. Our results suggest that the disparity between urban and rural regions and targeted consumer behavior changes should be taken into account to ensure the effectiveness of a future energy-efficient subsidy program. (C) 2014 Elsevier Ltd. All rights reserved.