Working Paper: CEPR ID: DP16084
Authors: Esther Duflo; Abhijit Banerjee; John Floretta; Anna Schrimpf; Anirudh Sankar; Francine Loza; Harini Kannan; Matthew O. Jackson; Arun G. Chandrasekhar; Maheshwor Shrestha; Suresh Dalpath
Abstract: We evaluate a large-scale set of interventions to increase demand for immunizationin Haryana, India. The policies under consideration include the two mostfrequently discussed tools—reminders and incentives—as well as an intervention inspiredby the networks literature. We cross-randomize whether (a) individuals receiveSMS reminders about upcoming vaccination drives; (b) individuals receive incentives forvaccinating their children; (c) influential individuals (information hubs, trusted individuals,or both) are asked to act as “ambassadors” receiving regular reminders to spreadthe word about immunization in their community. By taking into account differentversions (or “dosages”) of each intervention, we obtain 75 unique policy combinations.We develop a new statistical technique—a smart pooling and pruning procedure—forfinding a best policy from a large set, which also determines which policies are effectiveand the effect of the best policy. We proceed in two steps. First, we use a LASSOtechnique to collapse the data: we pool dosages of the same treatment if the datacannot reject that they had the same impact, and prune policies deemed ineffective.Second, using the remaining (pooled) policies, we estimate the effect of the best policy,accounting for the winner’s curse. The key outcomes are (i) the number of measlesimmunizations and (ii) the number of immunizations per dollar spent. The policythat has the largest impact (information hubs, SMS reminders, incentives that increasewith each immunization) increases the number of immunizations by 44 % relative tothe status quo. The most cost-effective policy (information hubs, SMS reminders, noincentives) increases the number of immunizations per dollar by 9.1%.
Keywords: Development Economics; Immunization; India; Reminders; Incentives; Smart Pooling
JEL Codes: C18; C93; D83; I15; O12; O15
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| combination of information hubs, SMS reminders, and increasing incentives for vaccinations (J68) | increase in the number of immunizations (I19) |
| using information hubs and SMS reminders without incentives (L96) | increase in the number of immunizations per dollar spent (H51) |
| incentives, SMS reminders, and information hubs (L96) | immunization rates (I18) |
| presence of information hubs (D83) | magnifies the effect of other interventions (C92) |