Investing in safe deliveries: correcting misperceived social norms to provide more generous maternity grants in rural Zambia

This document describes the pre-analysis plan for a randomized experiment encouraging savings groups in rural Zambia to provide one-time maternity grants to pregnant women to encourage them to deliver at a health facility. We randomize savings groups into a Treatment and Control arm. In the Control arm, there is no intervention while in the Treatment arm we encourage savings group to give maternity grants to pregnant women. The Treatment arm is further randomized to three sub-treatments: 1) Discussion, 2) Private, and 3) Public where we experimentally vary whether or not we reveal the average preferred amount of group members, and whether we reveal this privately or publicly. At Baseline, we administer a survey to all savings group members in the Treatment arm to measure demographics, social preferences, and beliefs around pregnancy, delivery, and the role of savings groups in supporting pregnant women. Afterwards, savings group members convene to discuss and decide on the amount of the maternity grant to be provided. In both the Treatment and Control arms, we will conduct phone follow-up interviews over the 18-months period with savings group members who become pregnant and give birth to examine facility delivery and other downstream health outcomes. This document describes our research design and empirical strategy. We define our primary and secondary outcome variables and main regression specifications. However, we anticipate that we will carry out additional analyses beyond those included in the document. This document is therefore not meant to preclude additional analyses.