UCLouvain Economics Seminar: Alessandro Tarozzi

Evaluation of Alternative Strategies to Increase Demand for and Responses to Information on Arsenic-contaminated Tubewell Water in Bangladesh

Alessandro Tarozzi, Univ. Pompeu Fabra

 Lack of reliable information on environmental risk is often a key constraint limiting the extent of risk-avoiding behavior in developing countries. The existence of private markets may be important if households cannot rely on the public sector to provide information, but fees may severely limit demand, especially among the poor. Such considerations are salient in Bangladesh, where naturally-occurring low-dose arsenic is frequently present in tube-well water, a primary source of drinking water for millions of households. Because arsenic contamination varies considerably across space, even within very narrow areas, the provision of information on arsenic contamination has been shown to be effective at allowing households relying on unsafe water to switch to safer and nearby sources. However, the safety status of millions of tubewell remains unknown, and there is no well-established market for tests. In this paper, we describe results from a randomized controlled trial in 123 villages in Sonargaon, Bangladesh, where tests were sold under different conditions. We evaluate both the impact on demand and the impact on risk-mitigating behavior, conditional on purchase. We find that at a relatively low price of 45 Takas (about 2 PPP USD) only 22 percent of households purchased a test, despite widespread awareness about arsenic risk and infrequent knowledge about the safety of one’s drinking water. Sales were increased neither by “nudges” in the form of visible metal placards indicating safety status, nor by offers that attempted to exploit local networks by incentivizing group purchases, but contracts requiring payment only in case of “good news” more than doubled demand. We also find that the different interventions led to heterogeneous impacts on switching away from unsafe wells, likely due to both selection and differential impacts conditional on purchase. Conditional on learning about the unsafe status of one’s tubewell water, group purchases, visible placards and fees-for-good-news (but only at lower prices) about doubled the fraction of households which stopped drinking water from contaminated wells. (Joint paper with Ricardo Maertens, Kazi Matin Ahmed and Lex van Geen)