The 2011 FAO State of Food and Agriculture yearbook brought to global attention the problem of female farmers lagging in terms of agricultural productivity compared to males. The report showed that female farmers also lagged in access to assets and inputs that support productivity. Various analysts, therefore, assumed that differences in access to assets were the main drivers of gender gaps in farm productivity. This present study returns to the question of gender-based differences in farm productivity, decomposing differences in per-hectare farm yields between males and females, using the Kitagawa-Oaxaca-Blinder decomposition approach and examining the evidence from eleven developing countries. Using different definitions of the “female farm” we decompose differences in yields by gender. We show a surprising result: that yield gaps do not always favor male farms. Nevertheless, in terms of their profile, female farms are consistently at a disadvantage across countries and definitions, with older and less educated household heads, less farmland available, smaller household sizes, and fewer economically active household members. Notably, however, the decompositions show that female farms appear to have an “endowment effect” advantage. This effect, however, must be interpreted with care because it is driven by the negative correlation between female farms and farm scale. As extensively documented in the literature, in the range of scale relevant to this study, smaller farms tend to have higher yields. Therefore, the “endowment effect” advantage is likely hiding a disadvantage to women scaling up the operational size of the farm. Female farms also display a “structural” or “unobserved effect” disadvantage, similar to that which has been found in the gender wage gap literature, which offsets and many times cancels this deceptive “endowment effect” advantage