In this project Andrea Szabo and Gergely Ujhelyi study the causes of nonpayment and which policies are effective at addressing them.

Background

In developing countries, punishing nonpayment by denying service may not achieve much. First, the impact of denying service is limited when consumers do not have enough income left to pay their bill. Second, aggressive enforcement could go against social perceptions of fairness and erode citizens’ trust in local governments, resulting in even more nonpayment or even civil unrest. Third, in some cases nonpayment can be caused by consumer dissatisfaction with service delivery or a lack of trust in the provider. Punishing nonpayment by denying service would be highly counterproductive in these situations.

This study explores an alternative strategy to reduce nonpayment: providing information. Purchasing and using water is a complicated activity: quantities used by various appliances are not directly observed by the household, and consumption and payment occur at different times. One possible reason of nonpayment is that households are unfamiliar with the billing system or the amount of water they consume in their everyday activities and accumulate high bills which they find difficult to pay. Providing information about various aspects of water consumption could improve the households’ water management and consequently lower waste and their monthly bill.

Study design

To study this possibility, we designed, implemented, and evaluated a randomized water education campaign in a collection of low income peri-urban townships in South Africa. Education officers visited a treatment group of 500 households to give them accessible information about various aspects of the water consumption process, including the water meter, the bill, and the amount of water used by various everyday activities. We evaluate the program combining administrative billing data on the full population of consumers with in-depth survey information on the treatment group and a control group.

Findings

We find that our information campaign achieved remarkably large reductions in nonpayment. Compared to the control group, we find that treated households are more likely to pay their bill and make larger payments. On average, our treatment reduced the fraction of consumers making no payments by 4 percentage points and increased total payments by about 25% in the three months following the treatment.

We see no decrease in water use corresponding to the lower rate of nonpayment. Because water use reflects a consumer’s planned payment behavior, this finding implies that nonpayment in our setting primarily reflects unplanned circumstances (such as higher than expected bills) rather than households systematically planning to avoid payment.

The evidence shows that the education campaign did not operate solely by increasing consumers’ information, or by creating reminders to pay or a threat of enforcement. Instead, households may have increased payments to reciprocate the provider’s outreach efforts.

Overall, our findings show that strategies other than increased enforcement can lower nonpayment.

Read the full study

A. Szabó and G. Ujhelyi (2015): “Reducing Nonpayment for Public Utilities: Experimental Evidence from South Africa,” Journal of Development Economics 117, 20-31.