x

NEW UPDATE: Completed trials may now upload and register supplementary documents (e.g. null results reports, populated pre-analysis plans, or post-trial results reports) in the Post Trial section under Reports, Papers, & Other Materials.
Economic and Financial Models for Pricing and Setting Sanitation Tariffs for the Benefit of the Urban Poor, Evidence from Burkina Faso
Last registered on December 19, 2016

Pre-Trial

Trial Information
General Information
Title
Economic and Financial Models for Pricing and Setting Sanitation Tariffs for the Benefit of the Urban Poor, Evidence from Burkina Faso
RCT ID
AEARCTR-0000834
Initial registration date
September 07, 2015
Last updated
December 19, 2016 3:52 PM EST
Location(s)
Region
Primary Investigator
Affiliation
University of Virginia
Other Primary Investigator(s)
PI Affiliation
University of Virginia
PI Affiliation
University of Notre Dame
Additional Trial Information
Status
On going
Start date
2014-11-01
End date
2017-09-30
Secondary IDs
Abstract
This project is a Randomized Controlled Trial designed to measure the impact of alternative pricing models on access to improved sanitation services in peri-urban West Africa.

The market for improved sanitation services (mechanized desludging in particular) is important: sanitation services have high externalities for the local communities. A government may want to intervene in the market for sanitation services but find it difficult to do so as they are unable to observe values for the consumers or costs for the service providers.

In a first round survey, we measured the underlying values of services for customers and the cost of providing services for desludging operators. We used this information to design two pricing treatment structures, one with fixed prices set based on household characteristics, and another with desludger bidding and negotiations with desludgers for lower prices. The fixed prices are set based on household characteristics; this treatment is designed to allow governments to target subsidies to poor households while financing some of the subsidies through price discrimination toward wealthier households. We will test whether these pricing structures, developed based on the underlying factors in the economy, perform better than the control group which has no pricing intervention in terms of total take-up of mechanized desludging. This project will describe and test how information on market structure can be used to help in designing interventions by governments to improve the market structure of sanitation services.
External Link(s)
Registration Citation
Citation
Johnson, Terence, Molly Lipscomb and Molly Lipscomb. 2016. "Economic and Financial Models for Pricing and Setting Sanitation Tariffs for the Benefit of the Urban Poor, Evidence from Burkina Faso." AEA RCT Registry. December 19. https://doi.org/10.1257/rct.834-2.0.
Former Citation
Johnson, Terence et al. 2016. "Economic and Financial Models for Pricing and Setting Sanitation Tariffs for the Benefit of the Urban Poor, Evidence from Burkina Faso." AEA RCT Registry. December 19. http://www.socialscienceregistry.org/trials/834/history/12652.
Sponsors & Partners

There are documents in this trial unavailable to the public. Use the button below to request access to this information.

Request Information
Experimental Details
Interventions
Intervention(s)

In a first round survey, we measured the underlying values of desludging services for customers and the cost of providing services for desludging operators. We used this information to design two pricing treatment structures, one with fixed prices set based on household characteristics, and another with desludger bidding and negotiations with desludgers for lower prices. The fixed prices are set based on household characteristics; this treatment is designed to allow governments to target subsidies to poor households while financing some of the subsidies through price discrimination toward wealthier households. We will test whether these pricing structures, developed based on the underlying factors in the economy, perform better than the control group which has no pricing intervention in terms of total take-up of mechanized desludging. This project will describe and test how information on market structure can be used to help in designing interventions by governments to improve the market structure of sanitation services.
Intervention Start Date
2015-07-25
Intervention End Date
2016-12-31
Primary Outcomes
Primary Outcomes (end points)
takeup of mechanical desludging, takeup of mechanical desludging among heterogeneous consumer groups, mechanical desludging prices offered by suppliers
Primary Outcomes (explanation)
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
In many peri-urban communities in developing countries (like Ouagadougou), the market for sanitation services (especially mechanical desludging) is under-developed, and households often resort to using inferior (manual desludging) methods.

In Burkina Faso, we use measures of the underlying values for improved desludging services from a survey run in December, 2014, and second-price auctions for the procurement of desludging services run from December 2014 through July 2015 to develop a model of the optimal tariffs to charge each household for desludging services in order to increase take-up as much as possible given a constrained budget for subsidies.

We test the modeled “ideal” prices against two options: first, a pure control group of households which are offered no pricing intervention, and second, a group of households that are asked to call into the call center for a desludging when they need one, and are given the lowest cost desludger that the call center is able to find through a bid-undercutting mechanism.

The endline survey will be rolled out after the randomized price structures have been in effect for at least 12 months. At this point, we expect the median household to have used 2 desludgings since the roll out of the price structures, and we will be able to observe the relative impact of the different price structures, estimate demand for sanitation services, and estimate the impact of the monitoring interventions.
Experimental Design Details
Pricing Model Intervention: The characteristics of the model for the optimal pricing structure of desludging services are as shown in the attached Market Design Appendix. Undercutting-bid intervention: The undercutting-bid mechanism is meant to instigate competition among desludgers while maintaining a group of desludgers who are willing to provide desludging services to the household at short notice. We ran second-price auctions for the procurement of desludging services from December 2014-July 2015. We are continuing these auctions for the procurement of desludging services to be used for our pricing model treatment. In order to encourage increased competition while offering services to households that call in as quickly as possible, we generate a spot-price setting mechanism. The spot price will be derived from calling desludgers and offering them the job for a slightly lower price than those bid in the second price auctions.
Randomization Method
Randomization done by computer
Randomization Unit
Neighborhood clusters of 30-40 households
Was the treatment clustered?
Yes
Experiment Characteristics
Sample size: planned number of clusters
126
Sample size: planned number of observations
3,750
Sample size (or number of clusters) by treatment arms
1,550 in the price structure intervention, 1000 in the bid-undercutting treatment, and 1,200 control
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Please see page 13 of the Market Design Appendix for power size calculations and expected effect size of the price structure treatment. For the under-cutting bid treatment, we expect prices to go down by approximately 1,000 CFA. This is based on the bidding in the second price auctions, together with the fact that the bidders will be encouraged to reduce their prices through the price-undercutting mechanism. With an effect size of 1,000 CFA, a standard deviation of 5964, and a sample mean of 16,836, this means that the necessary sample size is 282. We multiply the sample size by 1.9 since the intra-cluster correlation coefficient is 0.03 and clusters have approximately 30 households in order to be conservative (decisions are made at the household, not cluster level). This gives us a sample size requirement of 535 households. We expect this price reduction of 1,000 CFA on average to increase take-up of mechanical desludging by 0.085 from 0.455 to .555 (.455 was the mean in our data from an earlier survey sampled from the same areas). Our data from the prior survey shows that the standard deviation for take-up is 0.49. This means that for power of .8 on take-up, we need a sample size of 197 (To be conservative, we increase the sample by a multiple of 3 because sampling is by neighborhood-cluster of 25-35 households. However, these clusters are used for logistical ease. Decisions are made at the household level, and we do not expect high intra-cluster correlation—intracluster correlation in an earlier survey was calculated as .07. With clusters of size 30, this generates an inflation factor of 3). On average about 50% of households get a desludging over the course of a year, with 50% of those who get at least one desludging getting 2 or more desludgings in a year. We sample 1000 households with the offer of the negotiation treatment in order to get an expected 500 households offered the treatment and actually needing a desludging over the course of the year (with over 1000 desludgings total expected among those households over the year).
Supporting Documents and Materials

There are documents in this trial unavailable to the public. Use the button below to request access to this information.

Request Information
IRB
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Name
University of Virginia
IRB Approval Date
2013-05-14
IRB Approval Number
2013019100
Post-Trial
Post Trial Information
Study Withdrawal
Intervention
Is the intervention completed?
No
Is data collection complete?
Data Publication
Data Publication
Is public data available?
No
Program Files
Program Files
Reports, Papers & Other Materials
Relevant Paper(s)
REPORTS & OTHER MATERIALS