Home Effects in Tax Collection: Evidence from the DR Congo

Last registered on August 04, 2026

Pre-Trial

Trial Information

General Information

Title
Home Effects in Tax Collection: Evidence from the DR Congo
RCT ID
AEARCTR-0019265
Initial registration date
July 29, 2026

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
August 04, 2026, 9:27 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Primary Investigator

Affiliation
UC Berkeley

Other Primary Investigator(s)

PI Affiliation
University of Zurich
PI Affiliation
UC Berkeley
PI Affiliation
World Bank
PI Affiliation
UC Davis

Additional Trial Information

Status
On going
Start date
2025-08-01
End date
2026-12-31
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
A key question for governments is whether to assign officials to work in the communities they come from. Local postings can raise government effectiveness because officials have local information. But local postings also create possible conflicts of interest because officials share social ties and may thus internalize costs or benefits of government policy. We study this tradeoff by exploiting variation in whether tax collectors are assigned to their home neighborhoods or not in the context of property tax collection in Kananga, DRC. We study how postings shape interactions between tax collectors and taxpayers, tax compliance, revenue, and bribes in a low-capacity environment.
External Link(s)

Registration Citation

Citation
Danner, Sarah et al. 2026. "Home Effects in Tax Collection: Evidence from the DR Congo." AEA RCT Registry. August 04. https://doi.org/10.1257/rct.19265-1.0
Experimental Details

Interventions

Intervention(s)
Intervention Start Date
2025-08-01
Intervention End Date
2026-12-31

Primary Outcomes

Primary Outcomes (end points)
Primary outcomes, measured in administrative records, surveys and recordings, thus include:
1. Tax compliance (any payment and complete payment)
2. Tax revenue
3. Cost-effectiveness
4. Tax collector effort and messaging
5. Leakage (bribes, etc.)
Administrative records will provide official data on tax compliance and revenue for individual properties. Cost-effectiveness is assessed based on cost data detailed in the plan for a companion study (Danner et al., 2025). Tax collector effort will be measured in terms of number of visits, time spent at households, and persuasion techniques as captured in the recordings. Leakage will be estimated through the endline survey of property owners, in which reports of bribe requests by collectors or
collusive agreements between owners and collectors will be elicited, along with self-reported compliance (to compare against official records). Past studies in this setting confirm the viability of eliciting citizen-reported bribe measures as indicators for the presence of corruption (Balan et al., 2022). We plan to study heterogeneity by collector and delivery zone characteristics that we describe in the full PAP.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
A recurring dilemma in public administration is whether frontline officials should be posted "at home" or rotated away from the communities where they are socially embedded. Posting officials locally can be attractive because it may reduce frictions in implementation: local officials are easier to place, they speak the local social "language", they often know where to find people and how to navigate the area, they may have private information about local dynamics or particular citizens' characteristics and they may have additional leverage over local citizens through their personal ties. Yet the relationships that make local implementation easier may also reshape incentives. When an official’s decisions directly impose costs or confer benefits on friends, relatives, and neighbors, the official may internalize these consequences -- creating pressure to bend rules, soften enforcement, or redirect effort, especially where monitoring is weak and discretion is high. Such internalization may result from intrinsic preferences of local administrators, or from the ability of citizens to exert pressure on the local administrator as a result of their local ties. Whether local embeddedness strengthens or weakens state capacity is therefore an empirical question that depends on both the task and the institutional environment.

We study this tradeoff in the context of tax collection in the Democratic Republic of the Congo. In our context, tax collectors choose their enforcement effort and are paid based on performance. Local information may increase the productivity of effort by raising the probability that enforcement leads to payment. Social ties, however, can also create a countervailing motive: inducing payment imposes real costs on the collector's social network, reducing the collector's net incentive to push for compliance even when they have better information. Home postings may increase the capacity to collect (via information) while simultaneously reducing the willingness to enforce (because of internalized costs), making the overall effect on revenue and corruption ambiguous ex ante.

In this project, we compare tax collectors who are assigned to their "home zones" (where they live), "quasi-home zones" (where they frequent), or "past zones" (where they were assigned randomly as collectors last year) -- compared to neighborhoods where they have no personal ties. Does proximity to one's social network improve tax compliance through better information and targeting or stronger influence over taxpayers? Or does proximity undermine compliance because collectors internalize the burden imposed on citizens in their network and adjust enforcement accordingly? To examine these questions, we leverage variation generated by the assignment protocols used in 2025 and 2026 property tax campaigns conducted in the city of Kananga, DRC. Each year, collectors rotate across multiple collection zones over the course of the campaign. These assignments generate plausibly exogenous differences in whether a collector works in their home, quasi-home, or past zones in a given wave of the tax campaign. This structure allows us to estimate how home postings shape tax compliance, revenue, and bribe extraction, and administrator-citizen interactions in a low-capacity environment. We will also explore how dimensions of the tax collector and property owner relationship may affect tax compliance decisions by owners and tax enforcement by collectors, including age, co-ethnicity, co-partinsanship, and religious affiliation.
Experimental Design Details
Not available
Randomization Method
Randomization is done by a computer.
Randomization Unit
Treatment is randomized at the delivery-zone level.
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
1,359 delivery zones
Sample size: planned number of observations
45,687 tax-assessable and non-exempted properties
Sample size (or number of clusters) by treatment arms
1. HDZ: N = 42 [among: N = 39 in treatment and N = 3 in control]
2. QHZ: N = 138 [among: N = 68 in treatment and N = 70 in control]
3. PDZ: N = 990 [among: N = 493 in treatment and N = 497 in control]
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
Committee for Protection of Human Subjects (CPHS) at UC Berkeley
IRB Approval Date
2024-02-14
IRB Approval Number
#2023-10-16862
Analysis Plan

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

Request Information