Overcoming Knowledge Barriers to Maternal Care: Experimental Evidence on the Calibration and Efficacy of Informational Messaging from Pakistan

Abstract We conducted a large-scale field experiment to calibrate phone messaging to its potential of overcoming behavioral barriers to maternal care uptake in the countryside of a developing country, where a significant share of women forgoes life-saving maternity-related care even when within reach. The high-arching goal of our intervention is to test if and to what extent filling out insufficient domain knowledge (i.e., childbearing-related health literacy) generates responsiveness to nudges for adopting maternal care among the rural poor of Pakistan. We find that informational nudges sent in a random order using the appropriate medium of communication, voice messages, during pregnancy significantly improve care-seeking behaviors, measured by antenatal care, postpartum checkup, and postnatal visits, through improved literacy. Importantly, we document that high-frequency voice calls timed to gestational age substantially increase the efficacy of informational nudges, including boosting facility deliveries, once domain knowledge is built and nudgeability established. Nevertheless, small financial incentives trump the productivity of informational voice calls in both improving health literacy and boosting care uptake, likely due to participants equating the intrinsic value of the intended behavioral change to the size of the monetary reward. These results are scalable to a large number of populations across developing nations. JEL codes: I12; I15; C93; D10; D91