World Toilet Day presents us with the opportunity to analyse issues surrounding both toilet access and usage, particularly open defecation. India's Swacch Bharat Mission (SBM) - Grameen (The Clean India Campaign) seeks to make rural India open defecation free. There is enough evidence that supports the importance of these levers in influencing household demand. A recent trial by Mushfiq Mobarak et. al. from Yale on 'demand estimation with strategic complementarities' showcases that 'subsidies (can be) more effective than info or supply-side strategies' in influencing toilet construction. They also established that the greatest positive influence is observed when subsidies are implemented hand-in hand with awareness/behavior change campaigns. Conversely, the latter had a bigger influence on the types of products purchased and the likelihood of sustained use.1*
The current intervention strategies on open defecation and behavior change in India are at the level of a 'GP'. This serves as the unit of intervention, through village champions who are nominated by the GP leadership. While administratively, this seems like a logical choice, insights gathered from a recent pilot that Athena Infonomics implemented in Uttar Pradesh raises some questions on its efficacy. Similarly, Mushfiq Mubarak et. al.'s trial on 'estimating demand for toilets with strategic complementarities' raises critical questions on our current subsidy policy.
There are no clear patterns of ownership/use observed within a GP/village where the SBM subsidy has been implemented.2 The subsidy policy thus has not successfully leveraged the 'peer effects' or 'social influence' advantages that are associated with households building toilets through subsidy. This raises questions about the 'optimality' of the SBM subsidy policy and the need to better understand the reason for weak 'peer effects'.
One way to begin is by mapping what we know about the three most basic questions in any subsidy programme. Who is accessing the subsidy? how? and why?
Insights from our Uttar Pradesh study and learnings from other research interventions reveal that the subsidy model in India is top-down. Communities also seldom participate in terms of choice of products and it is un-targeted (those with high income can claim). This structure therefore dilutes the potential 'social network/influence' advantages. The process therefore merely creates assets, without a focus on 'use'. Also, such a subsidy structure is less likely to create/strengthen social norms that are pro-toilet construction/use.
Altering the structure of subsidy distribution (for example: cash-vouchers to households as against contractor driven asset creation) maybe worth exploring further. This approach leverages the 'social influence' effects of subsidy provision, encouraging greater participation from households.
Recent evidence3 also points to the benefits/advantages of targeting certain types of households as a priority. For example, in Bangladesh, targeting the poor/low income households (as against 'highly connected individual') demonstrated a bigger impact on the toilet construction/use behaviors of other members in the community. Here, shame was found to be being the primary driver.
Insights from the recent pilot implemented for 3ie reveal that the Swacchagrahi approach of recruiting champions who operate at the level of a GP (which could cover anywhere between 2 to 15 villages) has limited buy-in from community stakeholders. In Uttar Pradesh, GP heads usually nominate Swacchagrahis, who are often are not the most effective social nodes. This is due to relatively low levels of acceptance from members belonging to villages/castes other than the one that the Swacchagrahi is from.
Using interpersonal communication strategies to influence toilet use behaviors is a welcome improvement. Yet the absence of clear guidelines on 'selection', 'training' and deployment strategies renders this newly recruited ground-force less effective. Questions that should be asked include: Who is best equipped to be a Swacchagrahi?; How will the village champion monitor or record toilet use?; and finally, How should the incentives be structured to keep the Swacchagrahis motivated to perform?
Interventions should use low cost pyschometric tools/selection procedures, potentially extracting more value from the Swacchagrahi intervention. These should be administered directly by village administrative staff to identify more suitable candidates. Additionally, this should be coupled with a combination of specific social/monetary incentives.
* There is strong correlation observed between product quality and use, which further re-enforces the importance of sound IEC.
References:
1. R Guiteras, J Levinsohn, A Mushfiq Mubarak , 'No Sh*t: Demand Estimation with Strategic Complementarities', NCSU, Yale University, September 2017.
2. Study on Consumer Research Study to Map Drivers for Toilet Construction and Use in Uttar Pradesh supported by the World Bank.
3. R Guiteras, J Levinsohn, A Mushfiq Mubarak, 'No Sh*t: Demand Estimation with Strategic Complementarities', NCSU, Yale University, September 2017.