Building Legal Capacity: Evidence from the D.R. Congo

How to cite

Ngoma, Marina Mavungu; Weigel, Jonathan; Bergeron, Augustin; Davoine, Eva; Granato, Gabriel; Robinson, James. 2023. "Building Legal Capacity: Evidence from the D.R. Congo." AEA RCT Registry. AEARCTR-0011953. https://doi.org/10.1257/rct.11953-1.1

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Marina Mavungu Ngoma* , Jonathan Weigel , Augustin Bergeron , Eva Davoine , Gabriel Granato , James Robinson

* Primary investigator

Registry ID
AEARCTR-0011953
DOI
https://doi.org/10.1257/rct.11953-1.1
First registered
Last updated
Status
On going
Trial dates
2023-02-01 – 2026-06-02
Countries
Private
JEL codes
D63, H20, H71, P48, O17
Registry record
http://www.socialscienceregistry.org/trials/11953

Abstract

Resolving disputes is integral to the accumulation of state capacity. Yet often policymakers privilege fiscal capacity building in fragile states over legal capacity building. In this project, we study a low-capacity state—the D.R. Congo—seeking to establish legal authority and how its efforts to do so shape citizens’ demand for the state. Specifically, we examine the randomized rollout of a legal capacity building program implemented at scale in the city of Kananga (DRC) by the Ministry of Justice and a local NGO. In this program, citizens with local disputes are randomly assigned to receive subsidized mediation overseen by state lawyers, customary chiefs, or to remain in the status quo. We compare the impartiality of the state and customary justice system and examine effects of the program on property rights security and citizens’ views of and willingness to pay for the formal state.

Intervention

We evaluate the randomized implementation of a large-scale legal capacity building program in the city of Kananga (DRC). This program, overseen by the Ministry of Justice, has the goal of boosting the resolution of local disputes over property, debt, and petty theft. Given the coexistence of the state and customary sectors in DRC, the program has two main arms: a state legal capacity building arm, where state lawyers serve as mediation authorities, and a customary capacity building arm where customary chiefs instead take on this role. After creating a systematic register of all disputes in the city, cases are randomly assigned to mediation overseen by a state lawyer, a customary chief, or to remain in the status quo in which citizens could pursue either of these options (unsubsidized) or seek mediation by the neighborhood chief, though most often disputes remain unresolved.

In addition, we cross-randomize three sub-treatments:
1. Escalation to higher court: in both treatments, a share of cases is randomly assigned to receive the option of escalation to a higher state or customary court. If the first stage of mediation fails, parties have the opportunity to escalate the case to the Public Prosecutor in the state legal arm or to the customary Royal Court in the customary arm, at a subsidized cost.
2. Formal Document: in both treatments, a share of cases is randomly assigned to have the adjudicators issue a formal letter outlining the mediation outcome with signatures from all relevant parties.
3. Leopard Skin: in the customary arm, a share of cases is randomly assigned to grant the option to customary mediators to use a leopard skin in the mediation process. Leopard skins, in customary tradition, represent a high-stake tool to coerce conflict parties into telling the truth.

We compare the impartiality of the state and customary justice system and examine effects of the program on property rights security and citizens’ views of and willingness to pay for the formal state.

Experimental design

In collaboration with the Ministry of Justice, we study a randomized controlled trial. The randomization is stratified on conflict type, co-ethnicity of the mediator and plaintiff, the gender of the plaintiff, and the perceived difficulty of the case.

Design details

Randomization method
The randomization is done by a computer.
Randomization unit
The randomization unit is the case.
Clusters
N/A. We are studying a case-level randomization.
Observations
There are approximately 60,614 compounds in Kananga. We anticipate an average of 8% conflict reporting, which corresponds to approximately 4,849 conflict cases in the city. In addition, we randomly sample roughly 4,100 households across 432 neighborhoods and invite them to participate in a baseline and endline survey. At endline, we will collect a very short survey with all compounds in Kananga.
Arms
See PAP
Primary outcomes
The perceived and observed impartiality, capacity, and effectiveness of the state and customary justice systems; property rights security and social cohesion; citizens’ conceptions of justice; and citizens’ views of and willingness to pay for the formal state.
Secondary outcomes
Detailed dispute resolution outcomes: costs (official and unofficial), time to resolve, durability of outcomes, characteristics of the mediation process.

Cited by

No citation data available for this registration yet.

Keywords