This paper examines the effect of adult literacy program participation on household consumption in Ghana. The adult literacy programs in Ghana are of special interest since they are more comprehensive than standard literacy programs and incorporate many additional topics. We use community fixed effects combined with instrumental variables to account for possible endogenous program placement and self-selection into program participation. For households where none of the adults have completed any formal education we find a substantial, positive and statistically significant effect on household consumption. Our preferred estimate of the effect of participation for households without education is equivalent to a ten percent increase in consumption per adult equivalent. The effects of participation on welfare for other households are smaller and not statistically significant, and become smaller the more educated the household is. We find positive and statistically significant effects of participation on literacy and numeracy rates, although the increases are too small to be the only explanation for the welfare effects. There is also evidence that participants are more likely to engage in market activities and to sell a variety of agricultural goods. Taking account of both direct cost and opportunity cost we argue that the social returns to adult literacy programs are substantial.