x

We are happy to announce that all trial registrations will now be issued DOIs (digital object identifiers). For more information, see here.
Rural Institutional Innovation: Can Village Courts in Bangladesh Accelerate Access to Justice and Improve Socio-Economic Outcomes?
Last registered on August 05, 2019

Pre-Trial

Trial Information
General Information
Title
Rural Institutional Innovation: Can Village Courts in Bangladesh Accelerate Access to Justice and Improve Socio-Economic Outcomes?
RCT ID
AEARCTR-0001563
Initial registration date
March 07, 2017
Last updated
August 05, 2019 11:10 AM EDT
Location(s)

This section is unavailable to the public. Use the button below to request access to this information.

Request Information
Primary Investigator
Affiliation
Yale University
Other Primary Investigator(s)
PI Affiliation
Yale University
Additional Trial Information
Status
On going
Start date
2017-04-01
End date
2020-12-31
Secondary IDs
3ie application number: DPW1/1100
Abstract
Lack of access to the formal justice system is large problem for rural populations in most low and middle income countries. In Bangladesh informal dispute resolution mechanisms are common, but often lack the power to enforce decisions. There is also a perception of bias stemming from local power structures, which may discourage marginalized groups from using such mechanisms.

In 2006 the Government of Bangladesh passed into law a system of Village Courts (VCs) which are given the power to resolve smaller disputes (up to BDT 75,000, or roughly US$ 1,000). The VCs are anchored in the lowest tier of government, the Union Parishad (UP), with the UP Chair acting as the Chair of the court. While this is a great idea in theory to resolve small disputes cheaply with lower administrative complications, qualitative reports indicate that frequency of formation and use of the VCs have thus far been low. To counter this deficit, the Government of Bangladesh has launched a program called Activating Village Courts in Bangladesh (AVCB).

The AVCB program helps UPs to set up VC by educating local government representatives, providing material and human resources, and running awareness campaigns to encourage the local population to seek help from the VC. We will conduct an impact evaluation of the AVCB program using a Randomized Controlled Trial (RCT).

The impact evaluation would compare a treatment group of UPs where the AVCB program is implemented to a “control group” of UPs that do receive the program during the two-year study period. The impact evaluation will answer questions about the causal effect of the AVCB program on a range of outcome variables designed to test each step in our theory of how the VC affects village life. We will examine three broad categories of outcome variables. First, an “Access to justice” component will measure the cost and time it takes to resolve disputes, and the mechanisms households choose to use to resolve any dispute that they are involved in. Second, “quality of the justice process” will measure the nature and duration of continued conflict after the dispute was “resolved”, and self-reported satisfaction measures. Third, we will track longer term downstream effects of improved dispute resolution such as changes in economic activities where cooperation and contract enforcement are valuable, crime rates, trust and communal harmony, as well as political outcomes such as the satisfaction with the UP leaders. The impact evaluation will also be able to assess the spillover effect of the AVCB program on the Bangladeshi local justice system. Does, for example, the VCs decrease the burden on the district courts?
Registration Citation
Citation
Mattsson, Martin and Mushfiq Mobarak. 2019. "Rural Institutional Innovation: Can Village Courts in Bangladesh Accelerate Access to Justice and Improve Socio-Economic Outcomes?." AEA RCT Registry. August 05. https://doi.org/10.1257/rct.1563-3.0.
Former Citation
Mattsson, Martin and Mushfiq Mobarak. 2019. "Rural Institutional Innovation: Can Village Courts in Bangladesh Accelerate Access to Justice and Improve Socio-Economic Outcomes?." AEA RCT Registry. August 05. https://www.socialscienceregistry.org/trials/1563/history/51236.
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)
Our intervention is the Activating Village Courts Bangladesh (AVCB) program.

The Village Court Act of 2006 requires that a Village Court (VC) be composed of an UP chair (who acts as the Chair of the VC) and four jury members nominated by the plaintiff and the defendant. The plaintiff and the defendant each nominate two jury members, of which two have to be from the twelve-person UP council. These five-person courts take on both criminal and civil cases and have the authority to adjudicate disputes up to BDT 75,000 or approximately USD 1,000. The VCs do not have the ability to adjudicate more serious criminal cases, such as abduction, rape, or murder. VCs can only impose financial punishments and cannot send defendants to prison. In disputes that involve women or children, at least one woman has to be appointed to the court.

The AVCB program was initiated because the VC structure had been implemented by almost no Ups by 2009—three years after the passing of the Village Courts Act. The AVCB program trains relevant local government officials, provides physical infrastructure in the form of court room furnishings and stationary, and provides a Village Court assistant (human capital) who helps with the daily administrative tasks of the VCs. The AVCB program also publicizes the VC system through television commercials, flyers, and even open-air theatre, informing local citizens of their right to seek justice from the VCs.

Intervention Start Date
2017-05-01
Intervention End Date
2020-12-31
Primary Outcomes
Primary Outcomes (end points)
At this stage of the research project we have not yet determined the outcome variables for our final analysis. We will provide a full pre-analysis plan specifying all of the final outcomes after we have analysed the baseline data. The variables that we want to specify before the baseline data collection are the variables that will be used in our balance of randomisation analysis. We have selected these variables into keeping two rationales in mind. The first is that the main outcome variables should be included since if these are unbalanced that may raise questions about the main results. The second is to include variables that are important to give the reader a overview of our study population such as age, household size and income.
Primary Outcomes (explanation)
Here is a detailed list of outcome variables that we will use in our balance of randomisation analysis:

Household level variables:
• Average age in household (in years)
• Number of individuals in household
• Natural logarithm of per capita household income (winzorised at the 99th percentile)
• Yes to the question if there is a functional VC in the UP (binary variable)
• Putting VC as the institution where to resolve hypothetical disputes within the VC jurisdiction (sum of 3 binary variables)
• Total number of disputes (ongoing or resolved within the last year)
• Number of disputes resolved using Shalish (within the last year)
• Number of disputes resolved using VC (within the last year)
• Number of disputes resolved using DC (within the last year)
• Number of disputes resolved without an institution (within the last year)
• Number of unresolved disputes
• Total amount of resources spent on dispute resolution within the last year (winzorised at the 99th percentile)
• Percentage of households that have been the victim of a crime (within the last year)
• Total amount of investment within the last year (winzorised at the 99th percentile)
– of which: was made jointly with a business partner
• Total outstanding amount of money lent out to friends and family
• Total outstanding amount of money borrowed from friends and family
• Satisfaction with overall justice system (Scale 1 to 5)
• Satisfaction with the UP chair (Scale 1 to 5)

UP level variables:
• UP population (number of households)
• Fraction of randomly selected households with a dispute (ongoing or resolved within last year)
• What fraction of the VC records did the UP keep updated
• What fraction of the knowledge question on how to conduct a VC could the UP chair respond to correctly?
• What fraction of the knowledge question on how to conduct a VC could the UP members respond to correctly?
• Number of cases from UP that has gone to the DC (Number collected from DCs administrative records)
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
The experiment is based around the random selection of 187 UPs that will receive the AVCB program from a pool of 276 eligible UPs, 89 UPs are kept as a control group.
Experimental Design Details
Not available
Randomization Method
Randomization done in office by a computer using Stata.
Randomization Unit
Union Parishad (UP), the lowest tier of Government in Bangladesh.
Was the treatment clustered?
Yes
Experiment Characteristics
Sample size: planned number of clusters
The total number of clusters in the experiment is 276 UPs. 187 UPs in the treatment group and 89 UPs in the control group.

However, in the first round of data collection we only collected data from 107 UPs, 53 in the treatment group and 54 in the control group. For the follow up data collection we are planning to collect data from these UPs as well as an additional 70 UPs (35 in the control group and 35 in the treatment group).
Sample size: planned number of observations
We plan to collect data from 30 households in each UP making the total sample size in the first round of data collection 3,210 while in the follow-up data collection it will be 5,310. Furthermore we will collect data from the UP chair, two UP members and the UP secretary in each UP we survey as well as administrative data from the district courts for all 267 UPs.
Sample size (or number of clusters) by treatment arms
In the first two rounds of data collection: 53 UPs in the treatment group and 54 UPs in the control group. In the follow-up data collection 86 UPs in the control group and 91 UPs in the treatment group. In each surveyed UP 30 households as well as the UP chair, two UP members and the UP secretary will be surveyed.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Power calculations will be done after the baseline survey data been collected. This way we can use the correct standard deviations and intra-cluster correlations.
IRB
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Name
Yale University Institutional Review Board
IRB Approval Date
2016-11-03
IRB Approval Number
1609018380
Analysis Plan

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

Request Information