Evidence shows social protection can improve diets, but little is understood about how effects vary within a household or what factors determine how food is allocated across different household members. We use individual food intake data from two randomized control trials to estimate intrahousehold dietary impacts of cash or food transfers, with or without nutrition behavior change communication (BCC), in two regions of Bangladesh. We assess whether intrahousehold impacts 1) are consistent with different allocation "rules" hypothesized in the literature, 2) differ by transfer modality, provision of BCC, or regional context. Results indicate that households distribute food equally among their members (men, women, boys, and girls), both in absolute terms and in proportion to individual-specific requirements and deficits. Patterns are similar across regions and do not depend on transfer modality or whether BCC is provided. Findings have implications for designing nutrition-sensitive social protection with different target groups prioritized.