This paper uses two years of agriculture census data to build a panel dataset that consists of all the small towns in Bhutan. This dataset is used to estimate the impact of irrigation gaps and drought on the yields of paddy, maize, and other crops. The paper compares the estimated impacts from a panel fixed effects model and a spatial first differences model. The findings show that irrigation gaps reduce paddy yields and droughts reduce maize yields. Estimates from the spatial first differences model are found to be consistent relative to those from the panel fixed effects model. The paper further finds that water constraints reduce yields of vegetable crops, and other constraints, such as labor shortages, wild animals, insects, and diseases, also reduce the yields of cereal crops.