Local Authorities and Property tax collection: Experimental evidence from Tanzania

Last registered on May 21, 2021
View Trial History

Pre-Trial

Trial Information

General Information

Title
Local Authorities and Property tax collection: Experimental evidence from Tanzania
RCT ID
AEARCTR-0004960
Initial registration date
May 20, 2021

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
May 21, 2021, 9:33 AM EDT

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

Locations

Country
Tanzania, United Republic of
Region

Primary Investigator

Name
Francis Mwaijande
Affiliation
Mzumbe University
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Salama Ally
PI Affiliation
Simiyu Administrative Secretariat
Contact Investigator
PI Name
Jasmin Sereri
PI Affiliation
Tanzania Revenue Authority
Contact Investigator
PI Name
Eva Lyimo
PI Affiliation
Simiyu Administrative Secretariat
Contact Investigator
PI Name
Miriam Mmbaga
PI Affiliation
Simiyu Regional administrative Secretariat
Contact Investigator
PI Name
Diego Shirima
PI Affiliation
MASA Institute of Social Sciences Research
Contact Investigator
PI Name
Muhajir Kachwamba
PI Affiliation
MASA Institute of Social Sciences Research
Contact Investigator
PI Name
Kija Badala
PI Affiliation
Simiyu Administrative Secretariat
Contact Investigator
PI Name
Joshua Mwakalikamo
PI Affiliation
MASA Institute of Social Sciences Research
Contact Investigator
PI Name
Mwanahamisi Kawega
PI Affiliation
Simiyu Administrative Secretariat
Contact Investigator

Additional Trial Information

Status
On going
Start date
2020-11-20
End date
2021-06-30
Keywords
Other
Additional Keywords
Local, Authorities, Tax, Compliance
JEL code(s)
Secondary IDs
Abstract
This study aims to analyze the extent to which local community leaders in local government authorities can be leveraged to increase property tax collection in a developing country context. The property tax collection system is one of the main sources of revenue for many countries. Unfortunately, in Tanzania, property tax collection is as low compared to other types of tax such as VAT and Pay as You Earn (PAYE) tax. The Property tax collection in Tanzania is as low as 30% of the total expected revenue from property tax. The government has been changing strategies on collecting property tax, these include both centralized and decentralized approaches, but with less significant impact on compliance.
The study design employs a randomized control trial to assess the effect of messaging strategies and the involvement of local leaders/authorities to increase tax collection in Tanzania. Participants were randomly allocated to three treatment arms and a control group. The first treatment arm involved SMS reminder with a threat of fine, which is standard in tax compliance literature. The second treatment arm combined treatment for the first arm with an additional threat of involvement of local leaders in the tax collection process. The third arm involved the treatment in the first arm with an actual request for help from local leaders to increase tax compliance. Therefore, the innovation and value addition of this study to the existing body of literature stems from two additional experimental treatments in the second and third treatment arms stated above. While we acknowledge that there are many studies using SMS reminders/threats of fine to influence tax compliance; we leverage this strategy to study the additional effect of the involvement of local authorities in tax compliance.
External Link(s)

Registration Citation

Citation
Ally, Salama et al. 2021. "Local Authorities and Property tax collection: Experimental evidence from Tanzania." AEA RCT Registry. May 21. https://doi.org/10.1257/rct.4960-1.0
Sponsors & Partners

Sponsors

Experimental Details

Interventions

Intervention(s)
The study involves four experimental arms. These include control group (T0) which involves list of participants who neither receive SMS reminder nor local leaders are involved. The second treatment arm involved SMS reminder with both a threat of fine and a threat of involvement of local leaders. The third arm involved SMS reminder with both a threat of fine and an actual request for local leaders’ involvement in the tax collection process. Here below are the SMS scripts used in the intervention.

T0: T0-Neither SMS reminder nor Mjumbe will be involved in tax collection
T1: BASIC SMS for TaxPayers in Treatment 1 and Treatment 3 (control group receives nothing). Here below is the SMS content for T1.
Dear [NAME], Pay property tax for your building No [BRN], located at [STREET NAME] before June 30, 2021 amount of TZS [AMOUNT].
NON-PAYMENT IMPLIES FINES AND SANCTIONS FROM TANZANIA REVENUE AUTHORITY

T2: SMS for TaxPayers in Treatment 2 (basic T1 SMS+threat of Mjumbe action). Here below is the SMS content for T2.
Dear [NAME], Pay property tax for your building No [BRN], located at [STREET NAME] before June 30, 2021 amount of TZS [AMOUNT].
NON-PAYMENT IMPLIES FINES AND SANCTIONS FROM TANZANIA REVENUE AUTHORITY.
YOUR AREA’S MJUMBE, [MJUMBE NAME], WILL HELP TRA IN COLLECTING THIS TAX

T3: SMS for Mjumbe (only Treatment 3 – mjumbes in control group and Treatments 1 and 2 receive nothing). Here below is the SMS content for T3
Dear X, please inform property owners in your area to pay property taxes before June 30 2021. NON-PAYMENT IMPLIES FINES AND SANCTIONS FROM TANZANIA REVENUE AUTHORITY.
Intervention Start Date
2021-02-25
Intervention End Date
2021-06-30

Primary Outcomes

Primary Outcomes (end points)
Our primary outcome variable of interest is compliance measured in terms of compliance rate
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
The study consists of a randomized control trial that involved 670 Mjumbes (local leaders) and 21,327 taxpayers from 4 Local Government Authorities: Busega District. Kahama Town Council, Bariadi Town Council and Shinyanga Municipal. The experiment has been rolled out as follows:

1. Baseline data collection: Data was extracted from the database of property taxpayers from the Tanzania Revenue Authority (TRA) in the financial year 2018/19-2020/201 covering 4 LGAs: BARIADI Town Council, BUSEGA District, KAHAMA Town Council and SHINYANGA Municipal.

2. Randomizationand Intervention: cluster-level randomization was carried out to assign subjects to one of 4 groups described above (3 treatment groups and one control group). A Big stick rerandomization approach was used to allocate 670 local leaders (Mjumbes, under which property taxpayers are organized), with the goal of achieving a balanced sample among main covariates. 4 waves of SMS reminders will be sent to treated subjects (2 per month) and 2 waves of phone calls will be made to leaders (one at the end of each intervention month).

(3) Endline data collection: Endline data will be collected between May and the end of June 2021.
Experimental Design Details
Randomization Method
Randomization Method: Big stick rerandomization was employed to ensure a balanced sample between treatment and control group across most relevant covariates.
Randomization Unit
Property taxpayers were randomized at the Mjumbe level.
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
Sample
The study will include 670 clusters (at Mjumbe level, using Mjumbe IDs)

Sample size: planned number of observations
Sample size for Property taxpayers = 21,328 Sample size for Mjumbe = 670 In line with the existing literature on similar interventions (Antinyan et al. 2020), we expect Minimum Detectable Effects of +5% between Treatment 1 and control group, and of +2% between Treatment 2 and Treatment 1 and Treatment 3 and Treatment 1 respectively. Based on these conservative MDEs, simulation of cluster-level random assignment of the existing 670 clusters result in more than 80% power for the detection of the aforementioned differences. Simulation code available upon request.
Sample size (or number of clusters) by treatment arms
Control Group = 5273
Treatment (T1), Reminder+Threat of fine = 5696
Treatment (T2), "Reminder+Threat of fine + *Threat* of Mjumbe involvement = 5012
Treatment (T3), Reminder+Threat of fine+ *Actual* request for Mjumbe involvement = 5347

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
In line with the existing literature on similar interventions (Antinyan et al. 2020), we expect Minimum Detectable Effects of +5% between Treatment 1 and control group, and of +2% between Treatment 2 and Treatment 1 and Treatment 3 and Treatment 1 respectively. Based on these conservative MDEs, simulation of cluster-level random assignment of the existing 670 clusters result in more than 80% power for the detection of the aforementioned differences. Simulation code available upon request.
Supporting Documents and Materials

Documents

Document Name
Working Paper
Document Type
other
Document Description
Antinyan, Armenak and Asatryan, Zareh, Nudging for Tax Compliance: A Meta-Analysis (2020). CESifo Working Paper No. 8500, Available at SSRN: https://ssrn.com/abstract=3680357
File
Working Paper

MD5: 3eb06a5bdfbb896a6c7c3dd11204368d

SHA1: 6156479c865a861609d2b9c8708e97a700f2a1f8

Uploaded At: May 20, 2021

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

Request Information
IRB

Institutional Review Boards (IRBs)

IRB Name
IRB Approval Date
IRB Approval Number
Analysis Plan

Post-Trial

Post Trial Information

Study Withdrawal

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

Request Information

Intervention

Is the intervention completed?
No
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