Rural households in sub-Saharan Africa face high levels of poverty and annually recurring periods of lean season food insecurity. We conduct a randomized controlled trial in rural Malawi to assess the impact of coupling unconditional cash transfers of either $17/month or $43/month with a nutrition behavior change intervention on the diets and food security of households during the lean season. We find evidence of protective effects, but only when the intensity of treatment is high. When combined with the behavior change intervention, the large monthly cash transfer of $43/month improved food security by 15 percent, increased food consumption by 16 percent, and largely enabled households to smooth energy consumption between seasons. These effects are driven by a relative increase in consumption from own production. Households receiving the large cash transfer of $43/month invested in agricultural inputs and assets, allowing them to produce and store more maize in the preceding harvest and partially insulating them against negative food price shocks. We do not find evidence of similar effects of the nutrition behavior change intervention alone or in conjunction with a smaller transfer of $17/month.