Case Study

Water Conservation: Evidence from a Randomized Evaluation




Project Overview

Implementing Team/Partners

ideas42, World Bank, Municipality of Belen


Costa Rica


World Bank



The availability of water is one of the most pressing challenges of our times. This is prompting policymakers to revise their water management strategies and be better positioned to respond  to both, the increasing demand of water from the public, and the need to increase the supply of water.

In an urban context, household water use constitutes the majority of water consumption and thus is a key priority to address. For the most part, this issue has been tackled through pecuniary approaches (ie. price or tax increases) as well as information campaigns, with behavioral interventions on water use relatively under explored, especially in developing countries.

In this project the team designed a set of three simple, low cost, and replicable behavioral interventions delivered by adding stickers to water bills.,  The effect of this intervention on water consumption was evaluated using a randomized control trial in the town of Belen, Costa Rica during 2014.


Methods in Action




The team conducted focus groups with residents from  a variety of socio-economic backgrounds. Insights from these focus groups informed  the development of the interventions. For example, there was a general consensus that reduction of water consumption was a good thing, but most residents did not know how much water their household used, nor could not evaluate whether their water consumption was reasonable or too high. In addition, most residents could not identify specific steps they could take to reduce their household’s water consumption.

Design of interventions

Based on the focus group findings the team designed three interventions. Two interventions that allow residents to benchmark their consumption against that of their peers:

Neighborhood Comparison intervention: residents were informed of how their household’s water consumption compared to that of the average household in their neighborhood

City Comparison intervention: residents were informed of how their household’s water consumption compared to that of average household in the city of Belen.

One planning prompt intervention:

Plan-Making intervention: Drew residents’ attention to their household’s water consumption, encouraged them to set concrete goals about how much they would reduce their water consumption, and prompted them to identify  specific steps that would help them meet this goal.


The sample was drawn from a list of active residential water consumers in the city of Belen. The households chosen were stratified by postal route in addition to neighborhood and average monthly consumption in the 12 months prior to June 2014, and were randomized into three treatment groups with 1,399 resident for each group and a control group with 1,429 residents. The project ran during a two-month period beginning when they received the water bill with the intervention

Analysis of results

The general results were compared against the control group. The Neighborhood Comparison and Plan-Making interventions led to significant reductions in water use while the City Comparison had no discernible effect on water consumption. These findings supported literature that suggests that more localized norms are typically more effective at spurring behavior change.



Decline on water consumption

Creating awareness about how much water an individual consumes and comparing the consumption level with peers can in fact change individual’s behavior regarding the use of water. Two of the three interventions significantly decreased water consumption in the months following the intervention. A descriptive social norm intervention using neighborhood comparisons reduced consumption by between 3.7 and 5.6 percent relative to a control group, while a Plan-Making intervention reduced consumption by between 3.4 and 5.6. With similar results, these interventions are effective on different sub-populations, with the Plan-Making intervention being most effective on low-consumption households and the Neighborhood Comparison intervention being most effective on high-consumption households.

Effective low-cost intervention

For the most part, behavioral interventions have utilized sophisticated software which hinders the ability of local governments in developing countries to adopt such strategies. This project is proof that it is feasible to design successful behavioral interventions even in contexts with technology and resource constraints.

Influencing policy

This project was able to contribute to the field of behavioral economics in urban governance by proving the successful application of behavioral economics to impact water use. The results encourage the use of this strategy to policy makers as a way to supplement their present pecuniary approaches. More importantly, it validates that interventions can be done in resource and technology constrained settings in the developing world.