Health and Economic Effects of Piped Water

According to a 2006 study by the Asian Development Bank over 100 million of the 250 million Indonesians do not have access to safe drinking water and an additional 75 million rely on potentially contaminated sources. These numbers have declined however since the Indonesian economic crisis of 1998 the expansion and maintenance of water services have faltered. Estimates are that only 2 US dollars per person is being invested annually on water and sanitation in Indonesia, far less than other low and middle-income countries. (The Water Dialogues 2009). Currently, the only evidences of piped water’s impact on health and economic outcomes published in peer reviewed journals is a decrease in infant mortality (see GamperRabindran et al 2007 and Galiani et al 2005), one paper which shows increase in years of school attended for young girls (see Gould et al 2009) and another paper showing a significant negative externality of piped water operating through sanitation and resulting in an overall negative effect of piped water on health (see Bennett 2011). Previous studies have been unable to determine piped water’s impact on an individual’s health and economic prosperity because of data limitations. My research explores these individual level impacts as well as gender differential effects by exploiting the quasi-experimental variation created by the gradual expansion of piped water services in Indonesia over the past decades. I will analyze the effects of piped water on infant mortality, the long term effects on later life health and economic outcomes and how those effects differ by gender. Clean water is critical to maintaining proper hygiene during childbirth and therefore readily available safe water could result in decreased maternal and infant mortality. Furthermore, water’s influence on infant mortality is extended through diarrhea, the second leading cause of death for children under five in Indonesia and caused, mainly, by contaminated drinking water (Asian Development Bank 2006). Furthermore, health is a stock that is constantly evolving beginning in utero and early life inputs such as water can have potentially large effects on later-life economic outcomes. This is known as the fetal origins hypothesis from the seminal work by Barker (1992). Therefore, piped water is likely to affect the health of adults and consequentially their economic outcomes. And the effect of piped water on the health of adults is likely to differ by gender because of the disproportionate burden carried by women in developing countries to provide water to their home. In Indonesia, as in many other developing nations, the task of carrying water to and from its source belongs to women (World Bank Group 1994). This exhausting and physically demanding task is costly and many case studies show its negative effect on female health (see Ray (2007) for a review). The data and the quasi-experimental variation created by the gradual introduction of piped water in Indonesia allow my research to fill at least three voids in the current literature. First, the Indonesian Family Life Survey (IFLS) contains individual level data whereas the current literature uses mostly community level data without measures of individual health or economic status. Second, with individual data I am able to disentangle the differential effects of piped water on men and women. Third, because of the panel structure of IFLS I am able to exploit the quasi-experimental variation and control for household fixed effects to attain a plausibly exogenous identification strategy.