This paper uses a randomized experimental design with real-time electronic stove temperature measurements and controlled cooking tests to estimate the fuelwood and carbon dioxide savings from an improved cookstove program in the process of being implemented in rural Ethiopia. Knowing more about how households interact with improved cookstoves is important, because cooking uses a majority of the fuelwood in the country and therefore is an important determinant of greenhouse gas emissions and indoor air pollution. Creating local networks among stove users generally appears to increase fuelwood savings, and among monetary treatments the most robust positive effects come from free distribution. The paper estimates that on average one improved stove saves approximately 634 kilograms of fuelwood per year or about 0.94 tons of carbon dioxide equivalent per year, which is about half of previous estimates. Using the May 2015 California auction price of $13.39/ton, the carbon sequestration from each stove deployed is worth about $12.59. Such carbon market offset revenues would be sufficient to cover the cost of the stove within one year.