Credit Constraints, Present Bias and Investment in Health: Evidence from Micropayments for Clean Water in Dhaka

Low rates of adoption of and low willingness to pay for preventative health technologies pose an ongoing puzzle. In the case of water-borne disease, the burden is high both in terms of poor health and cost of treatment. Inexpensive preventative technologies are available, but willingness to pay (WTP) for products such as chlorine treatment or ceramic filters has been observed to be low in a number of contexts. In this paper, we investigate whether time payments (micro-loans or dedicated micro-savings) can increase WTP for a high-quality ceramic water filter among 400 households in slums of Dhaka, Bangladesh, where water quality is poor and the burden of water-borne disease high. We use a modified Becker-DegrootMarschak mechanism to elicit WTP for the filter under a variety of payment plans. Crucially, we obtain valuations from each household across all payment plans, which (a) increases power and (b) allows us to investigate the mechanisms behind differences in WTP across plans. We find that time payments significantly increase WTP: median WTP under a lump-sum, up-front payment is USD 9.30, versus USD 17 with a simple 6-month loan and USD 20 for an up to 12-month loan. Similarly, coverage can be greatly increased: at an unsubsidized price of USD 28 (50% subsidy price of USD 14), coverage is 12% (27%) under a lump-sum but as high as 45% (71%) given time payments. Many explanations are consistent with these reduced-form results. In ongoing work, we use our rich withinhousehold WTP data, the design of the payment plans, and a simple structural model of time preference to investigate the mechanisms at work behind these large differences in WTP. In particular, we measure the relative importance of credit constraints, time-preferences and the risk associated with a new technology. JEL classifications: C93, D14, D91, G21, I15, O16, Q53, Q56 † [email protected], [email protected], [email protected], [email protected] ‡ We thank the Bill and Melinda Gates Foundation and 3ieimpact for financial support. Ariel BenYishay, Jessica Goldberg, Glenn Harrison, John Rust, Steve Stern, Ken Train, Song Yao and participants in the Columbia University Sustainable Development Seminar, Georgetown University gui2de Development Economics Seminar, UC Berkeley Development Lunch, University of Maryland Applied Micro Lunch, University of New South Wales and Yale / Citi China India Insights Conference 2013 provided helpful comments. We are grateful to Raihan Hossain and Ashraful Islam for research assistance and project management, and icddr,b for collaboration in field implementation. This work was motivated by previous collaborations with Michael Kremer and Stephen P. Luby. All errors are our own. Low rates of adoption of and low willingness to pay for preventative health technologies pose an ongoing puzzle in development economics (Dupas 2011, Jameel Poverty Action Lab 2011). In the case of water-borne disease, the burden is high both in terms of poor health and cost of treatment, and inexpensive preventative technologies are available, but willingness to pay for products such as chlorine treatment or ceramic filters has been observed to be low in a number of contexts (Ahuja, Kremer, and Zwane 2010, Ashraf, Berry, and Shapiro 2010, Berry, Fischer, and Guiteras 2012, Luoto et al. 2011). Many explanations for this puzzle have been proposed. We focus on one common characteristic of many health technologies: a relatively large up-front investment is required, while the benefits accrue over time. This is problematic for a number of interdependent reasons. First, households may find it difficult to borrow, especially for non-business purposes. Second, poor households may have high discount rates or be close to subsistence levels of consumption and therefore be unwilling to sacrifice a large amount of current consumption. Third, households may exhibit time-inconsistency in the form of present bias or hyperbolic discounting (Ashraf, Karlan, and Yin 2006). Fourth, households may be unwilling to sink a large sum into a new technology when they are unsure of its benefits. These barriers suggest a number of interventions to increase adoption and improve welfare. Consumers who face liquidity constraints or exhibit present bias may find it difficult to fund purchases even if they are willing to pay substantial amounts over time (Holla and Kremer 2009). As a result, time payments, either micro-loans or layaways (dedicated savings), may increase adoption and improve welfare (Tarozzi and Mahajan 2011, Dupas and Robinson 2013). When consumers have uncertain valuation of a new product, a free trial or money-back guarantee can allow learning at low risk (Levine and Cotterman 2012). In this paper, we examine how time payment plans (either micro-loans or layaways) and interventions to decrease the risk incurred while learning (free trial, money-back guarantee) affect willingness to pay (WTP) and attempt to understand the mechanisms at work. Both of these are empirically challenging. First, individuals with greater access to finance may have a greater taste for health relative to consumption or more resources overall. Second, even if access to finance were randomly assigned, there are many variations possible and we would typically only observe one choice per individual, so it would require an enormous sample size to determine which policies are