This was created in partnership with Environment for Development .
This paper uses the National Energy Survey Data for Kenya 2009 to investigate the main determinants of household energy conservation and savings, exploiting discrete choice and Tobit models. We estimate conservation models for five household fuels—fuel wood, charcoal, kerosene, LPG, and electricity—and generate information to predict not only the odds for household energy conservation but also the levels of possible savings. Our findings reveal that demographic variables, such as the household head’s gender and occupational and educational attainment, as well as household location and size, are key determinants of not only the propensity to conserve energy but also levels of actual energy savings. Other factors that shape these decisions and outcomes include total hours of energy utilization, perceptions about cleanliness and affordability of the energy source, and awareness of conservation measures and regulations. For successful energy conservation, policy will need to take these factors into account.