Title:Attention vs Choice in Welfare Take-Up: What Works for WIC?

Abstract:We decompose the causes of incomplete take-up of welfare program benefits into two latent mechanisms: inattention, where households do not consider program participation, and active choice, where households consider participation but find it not worthwhile. To capture these two mechanisms, we model households' take-up decision as a two-stage process: attention followed by choice. Our model allows covariates to influence both stages simultaneously, creating an identification challenge in quantifying covariates' impact on either stage separately. Exploiting two institutional structures of welfare programs: full attention among current participants and the recertification process, we prove that attention and choice parameters are constructively identifiable. Applied to National Longitudinal Survey of Youth 1997 (NLSY97) data on the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC), our model reveals substantial household-level heterogeneity in both attention and choice probabilities. Furthermore, counterfactual simulations predict that choice-nudging policies outperform attention-boosting policies. We test this prediction using data from a WIC pilot program in Vermont that sent choice-nudging and attention-boosting text messages to different households. Consistent with the counterfactual prediction, choice-nudging messages increased retention by eight to ten percentage points, while attention-boosting messages had statistically insignificant effects. Our results highlight the critical role of choice nudges in improving program take-up.