x

We are happy to announce that all trial registrations will now be issued DOIs (digital object identifiers). For more information, see here.
Market Structuring of Sludge Management for the Benefit of Vulnerable Households in Dakar (Demand-side Trial)
Last registered on August 12, 2014

Pre-Trial

Trial Information
General Information
Title
Market Structuring of Sludge Management for the Benefit of Vulnerable Households in Dakar (Demand-side Trial)
RCT ID
AEARCTR-0000344
Initial registration date
August 12, 2014
Last updated
August 12, 2014 6:05 PM EDT
Location(s)
Region
Primary Investigator
Affiliation
University of Virginia
Other Primary Investigator(s)
PI Affiliation
University of Pennsylvania
PI Affiliation
University of Virginia
PI Affiliation
University of Wisconsin-Madison
Additional Trial Information
Status
On going
Start date
2013-08-28
End date
2015-06-30
Secondary IDs
Abstract
Poor sanitation is an important cause of childhood diarrhea, which often leads to child mortality. We will study how to increase demand and decrease prices for an improved sanitation technology, mechanical desludging. On the demand side, we will measure the effects of social and behavioral factors (social pressure, learning, and procrastination) on household demand. In a related trial focused on the supply side, we will study the effect of different auction mechanisms on collusion and prices paid by consumers.
Registration Citation
Citation
Houde, Jean et al. 2014. "Market Structuring of Sludge Management for the Benefit of Vulnerable Households in Dakar (Demand-side Trial)." AEA RCT Registry. August 12. https://doi.org/10.1257/rct.344-3.0.
Former Citation
Houde, Jean et al. 2014. "Market Structuring of Sludge Management for the Benefit of Vulnerable Households in Dakar (Demand-side Trial)." AEA RCT Registry. August 12. https://www.socialscienceregistry.org/trials/344/history/2395.
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)
This project seeks to identify methods of increasing the use of mechanized desludging by increasing household willingness to pay for improved sanitation services. We investigate two broad mechanisms through which willingness to pay may change: social network impacts and liquidity. We test the impact of several social network interventions that encourage take-up through learning and social pressure, and the impact of payment and deposit interventions which affect the liquidity of the households.

The treatment group is comprised of four thousand households, grouped into 400 neighborhoods of 10 households each. A further 800 households (two per neighborhood) will be surveyed to evaluate the effects on those nearby who were not offered the subsidized mechanized desludgings.
Four thousand households will be randomly assigned to receive a subsidized price of either $34 or $48 for mechanized desludging services, which normally cost around $50. The subsidy is valid for a maximum of 2 desludgings over a nine month period. To measure learning from doing, researchers will see whether those who received subsidized services continue using mechanical desludging after the discount ends.

To understand the extent to which social pressure influences use of mechanized desludging, the attribution of the discounts will be made public for half of the clusters through the distribution of discount lists, and for the other half it will be offered privately. The impacts of learning from others and coordination will be measured when 1000 randomly chosen households in 200 clusters are first told either how many or specifically which of their neighbors have signed up, and are then given the opportunity to sign up.

A number of payment structures will also be tested. Eighty-seven percent of households (3500) will be asked to leave a deposit at the time of the survey if they would like to sign up for the subsidized mechanized desludging. The remaining 500 treatment households will receive a subsidy for mechanical desludging, but will not be asked to leave a deposit. One third of households will be asked to pay the remainder at the time of service, one third will be given a savings account earmarked for desludging and billed monthly, and one third will be given the same earmarked savings account, but allowed to contribute whenever they wish. Half of households will also be offered a general, non-earmarked savings account. Varying the frequency of payments and savings options will test the relative importance of commitments and mental accounting to encourage payment and usage.
Intervention Start Date
2014-02-10
Intervention End Date
2015-06-10
Primary Outcomes
Primary Outcomes (end points)
prices paid for mechanical desludging, takeup of mechanical desludging, household health, willingness to pay for desludging services, savings account balances, follow-through on plans to use mechanical desludging
Primary Outcomes (explanation)
Willingness to pay is measured as the takeup rate (or follow-through rate) of mechanical desludging at different prices (subsidy levels). Our project will also have access to data from a related study that collects auction data from a call center that matches clients and desludgers. We will observe if callers accepted or rejected the auction winner's price for different bids.

Health outcomes will be measured using diarrhea rates through baseline and endline survey data, as compared to incidence of cold or cough symptoms. Sanitation will affect diarrhea rates, while respiratory problems (cold/cough) should see no effect.

Follow-through rate is important as realized take up is key to understanding the true effect of the treatment on willingness to pay. We will compare the commitment to take up the improved sanitation service and the realized take up at the end of the study.
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
We use a randomized controlled trial (RCT) to investigate social effects on the adoption of mechanized desludging. First, we will offer randomly selected neighborhoods subsidized desludging services coupled with various social pressure treatments to measure direct social effects.

We will construct 400 groups of twelve neighboring households each. These neighborhoods will be far enough apart that, in general, their sanitation decisions will not affect one another and the households will not know one another. In our 400 selected neighborhoods, each neighborhood will include 10 randomly chosen treated households and 2 untreated households; these 2 untreated households will allow us to measure indirect spillovers.

We will first conduct a baseline survey of demographic information, including household composition, education, health, membership and participation in associations and cooperatives, and savings habits. We will also collect GPS data on the locations of the households.

We then conduct a second survey on willingness to pay for improved sanitation with the household member who is in charge of making decisions regarding desludging ("the decider"). The decider survey will cover savings and loans, wealth and durable assets, brief questions on income and spending, sanitation practices, and social networks. The social network component of the survey will include questions asking who in their neighborhood they talk with about waste disposal, who they would choose to lead a neighborhood sensitization on health, who is a member of the same association or cooperative as them, from whom they would borrow or to whom they would lend money, who they did borrow from or lend to within the past year, with whom they are related, which households use mechanized desludging, and where each household dumps its sludge.

At the time of the second survey, in the 10 treatment households, we will offer the decider his randomly assigned treatment. There are several treatment arms (please see experimental design) but one main treatment involves randomizing discounts of different sizes to households that sign up for a subscription of two desludgings, and randomizing whether this discount is private or public information.

At the end of the decider survey, households in the "deposit" group which would like to sign up for the subscription will be asked to pay a deposit of roughly US$6 – an amount equal to the respondent's participation gift. The deposit will be credited towards the second desludging, and will be unavailable to them until the end of the nine-month subsidy period. After the subsidy period ends, they will have the option to continue using the subscription for an unsubsidized third desludging depending on the interest of the Senegalese Ministry of Sanitation in continuing the program following the main research period.

After the year in which treated households have access to subsidized desludgings, we will re-interview both the 4000 treated households and the 800 untreated households, allowing us to measure their sanitation practices and relationships with neighbors.
Experimental Design Details
We will work in the Pikine and Guediawaye departments of peri-urban Dakar, Senegal. This area is home to approximately 1.2 million individuals, and roughly 80% of households in the area are not connected to the city's sewage network; each household is responsible for deciding how to dispose of its own waste. This area is also poorer than the rest of Dakar. Many areas in Pikine and Guediawaye are far from the waste treatment plants or less directly accessible, so mechanized desludging is expensive. Collusion among desludging service providers also seems to contribute to high prices. These facts mean that adoption of mechanized desludging is particularly low relative to the rest of Dakar. We will measure the following direct social effects on the treated households. Social pressure and incoming reciprocity: We have randomly assigned discounts of 4% or 32% to each household, sharing the distribution of these discounts with 'public' clusters and not sharing this information in 'private' clusters. Households that have been assigned the high discount in 'public' clusters will face stronger social pressure to purchase the sanitation service than households that have been assigned the same discount in the ‘private’ clusters. We offer only two levels of subsidies in order to increase the social pressure felt by households drawing the high subsidy and in order to avoid small sample problems when estimating the impact of the subsidies. The public-private treatment is randomized at the cluster level. There are 400 clusters, each with 10 households. In this treatment we will compare the overall use of mechanized desludging versus manual desludging. We estimate the necessary sample size assuming a standardized effect size of 0.2 and an intra-cluster correlation coefficient of 0.2, based on desludging data from a related study in Dakar. With 10 households per cluster, the cluster-corrected power calculation suggests that we need at least 364 clusters. This corresponds to a power level of 95% and confidence level of 5%. We have selected 400 project clusters. We will also evaluate the effect of this treatment on the follow-through rate for those who commit to mechanized desludging. Because this outcome can only be measured for participants who sign up for a subsidized desludging through the project, there will be fewer participants per cluster. We adjust the power calculation assuming only 4 households per cluster. We expect a greater standardized effect size for this group of 0.25. Keeping the same power and confidence levels, we need at least 333 clusters. Learning-by-Doing: People may be more likely to continue mechanized desludging after they experience it. Households are given discounts for the first two desludgings during the 9 months following their subscription, while the price for the third desludging, or any desludging after the 9-month period, will no longer be subsidized. Learning-from-others, coordination, and outgoing reciprocity: People may be more likely to sign up for mechanized desludging after they have seen their neighbors do it. Some of the neighborhoods (200 clusters of 10 households) will be randomly divided into two halves when surveying. In the second half, we will provide information about the first five households' desludging decisions to the last five participants. We will notify the last five of how many of their first five neighbors have signed up for a desludging subscription, with half of the treated group learning only the number of neighbors and half also learning the names of subscribing neighbors. We will then compare the takeup and follow-through rates of the first fives with the last fives, and also of the number-only treatment group with the names treatment group. Within a cluster, the assignment of households to the first five or last five is randomized at the individual level. In order to detect a standardized effect size of 0.2 for the learning from others treatment, we will need a sample size of at least 1300 households for a 5% confidence level with 95% power. Conducting this experiment with 2000 households is therefore sufficient. We will be able to compare the 1000 last five households with control groups from the 1000 first five households, as well as with the 2000 households not included in this experiment, giving a total of 1000 treated and 3000 control. We will also compare the two treated groups – whether the households are told how many or who among their neighbors sign up for a mechanized desludging. These treatments are randomized at the cluster level, with 5 treated households per cluster. Because this intervention is logistically difficult and costly, we are not able to increase the sample beyond 100 clusters (500 households) per treatment arm. Still, this sample size will be sufficient to detect a standardized effect of 0.25 with 82% power, or a standardized effect of 0.3 with 94% power. Incoming reciprocity: People may sign up for desludging in order to reap future benefits from their reciprocal neighbors. We will carry out baseline and endline social network surveys to test this. In addition to the RCT, we will also play an economic game. During the survey, we will run a variant of a dictator game measuring incoming reciprocity. Each individual is given 500 CFA. He will choose how much to keep and how much to distribute anonymously to each of the other households. Any money sent to others will be doubled. This game will be played again during the endline survey. We will compare transfers to households that purchased the desludging service with those who did not. Payment mechanisms: We will also separately measure the effects of payment mechanisms on the willingness to pay for desludging services. There are several different payment types that we are testing: frequency of payments, the requirement of a deposit in addition to payment at the time of service, and the availability of savings accounts for general purposes in addition to the savings account earmarked for desludging. Frequency of payments: Households receive reminders to save for their desludgings, and the frequency with which they are asked to pay in the reminders varies between at will, monthly, and payment at the time of service. Earmarking: The earmarking treatment will compare those who received a single savings account earmarked for desludging with those who received both a desludging account and a normal savings account. Because not all participants will opt into the payment plan, we have adjusted our sample size to account for an estimated 40% expected takeup (hence 1600 participants in the earmarking experiment). For the earmarking experiment, our outcomes of interest are both total levels of savings and final use of the desludging services. We will compare the proportion of households who follow through with mechanical desludging after signing up for a subscription plan. We estimate that use of the desludging services (follow-through rate) will be about 70% for those with one account. In order to detect an increase to 80% or higher, we require a sample size of at least 962. We therefore believe our sample size of 1600 should be conservative for the earmarking treatment arms. In terms of the savings level outcome for the earmarking experiment, we expect a standardized effect size of at least 0.2, which suggests that we need a sample size of at least 1300. Deposit: About 3500 households will be asked to pay a deposit toward a subsidized desludging at the time of the decider survey, and 500 will be offered a subsidized desludging with no deposit. We will compare the proportion of households who follow through with mechanical desludging among those who left a deposit and those who did not. This sample size is sufficient to detect a 0.2 standardized effect size at 5% confidence with over 98% power. Note that this treatment is cross-randomized with the payment plans. Half of each deposit treatment arm will receive a subscription payment plan. We expect the effect of the commitment deposit treatment to be similar across the different subsidy levels. Spillovers: Two households (out of 12) within each cluster will not receive any treatment in order to measure spillover effects. We estimate the sample size necessary in order to have 95% power to calculate an effect that is statistically different from 0 at a 5% level of confidence, and still adjust for likely intra-cluster correlation of the two households. Using a similar framework in estimating the effect of learning on adoption of mosquito nets, Dupas (2010) finds a standardized effect size of having all neighbors receive the maximum subsidy of 0.44. (The coefficient on the share of households with the maximum subsidy within 500 meters is 0.215, and the standard deviation of adoption is 0.489.) We estimate the necessary sample size assuming a more conservative standardized effect size of 0.2 and an intra-cluster correlation coefficient of 0.2. With 2 households per cluster, the cluster-corrected power calculation suggests that we need 780 spillover households to compare with 3900 participant households. We will therefore use all 400 of our project clusters, for a total of 800 spillover households.
Randomization Method
We use cluster randomization across neighborhoods in Dakar, with each cluster and household assigned to treatment through randomization in Matlab.
Randomization Unit
There are various randomized treatments. Some of these are randomized at the neighborhood cluster level, and some are individually randomized. Each neighborhood cluster contains 12 households: 10 treated and 2 spillover.


Cluster-level randomized treatments:
Public / private treatment: the values of neighbors' subsidy prices are shared with others in their cluster in half of the clusters in order to observe the effect of pressure from the neighbors on the take up of desludging services.
Learning from others treatment: half of neighbors are told who (half are told how many) of their neighbors chose to take up the desludging subscription.

Individual-level randomized treatments:
Subsidy levels: subsidy levels are randomized at the household level, and the number of high versus low subsidies varies across the different clusters.
Payment frequency: households receive reminders to save for their desludgings, and the frequency with which they are asked to pay in the reminders varies between at will, monthly, and payment at the time of service.
Earmarking treatment: some households are offered one earmarked desludging account while others are offered two accounts (one earmarked and one general savings account)
Commitment deposit treatment: 87% of households are asked to pay a deposit toward a subsidized desludging, a randomly selected 13% are not asked to pay a deposit.
Spillover estimation: 2 households per cluster are selected to be surveyed but not receive the treatment or the subsidies. We observe whether these households are more likely to take up when there are more households with high discounts in their neighborhood.
Was the treatment clustered?
Yes
Experiment Characteristics
Sample size: planned number of clusters
400 neighborhood clusters
Sample size: planned number of observations
4800 households
Sample size (or number of clusters) by treatment arms
Cluster-level randomized treatments:
Public / private treatment: 200 public clusters (2000 households) vs. 200 private clusters (2000 households)
Learning from others treatment: 100 number clusters (1000 households) vs. 100 name clusters (1000 households) vs. 200 no info clusters (2000 households)

Individual-level randomized treatments:
Subsidy levels: 2000 high vs. 2000 low subsidies
Commitment deposit treatment: 3500 deposit households vs. 500 no-deposit households
Earmarking treatment: 1000 single account households vs. 1000 two accounts households vs. 2000 no account households
Households receiving at least one account are enrolled in one of three payment plan options: Payment frequency: 666 at will vs. 667 monthly vs. 667 time of service
Spillover estimation: 4000 households receiving subsidy vs. 800 spillover households (2 per cluster)
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Cluster-level randomized treatments (the public / private treatment and the learning from others treatment), as well as the individually randomized earmarking treatment, have an MDES of 0.17. The commitment deposit treatment has an MDES of 0.14, and the spillover estimation has an MDES of 0.13.
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
Innovations for Poverty Action
IRB Approval Date
2012-09-14
IRB Approval Number
574
IRB Name
Comite National d'Ethique pour la Recherche en Sante (Senegal)
IRB Approval Date
2012-02-01
IRB Approval Number
000227
IRB Name
University of Virginia
IRB Approval Date
2013-08-23
IRB Approval Number
2012-0272-00
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 and Papers
Preliminary Reports
Relevant Papers