Impact of Networks on Non-Compliance in VAT

Last registered on October 27, 2021

Pre-Trial

Trial Information

General Information

Title
Impact of Networks on Non-Compliance in VAT
RCT ID
AEARCTR-0008357
Initial registration date
October 11, 2021
Last updated
October 27, 2021, 4:01 AM EDT

Locations

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

Request Information

Primary Investigator

Affiliation
Ashoka University

Other Primary Investigator(s)

PI Affiliation
University of California - Irvine

Additional Trial Information

Status
On going
Start date
2021-06-01
End date
2022-06-30
Secondary IDs
Prior work
This trial is based on or builds upon one or more prior RCTs.
Abstract
Although different forms of Value Added Taxes have been implemented in several countries, effective implementation is constrained by state capacity. One form of deviation from effective implementation is large-scale non-compliance of filing requirements by the eligible tax payers. According to the official Indian government data, the cumulative (on time + delayed) filing of various tax forms in Delhi--the context of this study--is around 75% in February 2021. The on-time filing rate is around 55 percent.

This study nudges the taxpayers to file their pending returns and file their future returns on-time in a developing country context. The nudges take the form of electronic calls. During the intervention, we randomize the content of the messages to determine their relative effectiveness. Relatedly, we also analyse the effect of the non-compliance of a firm's business partners on its own compliance behaviour. The project has policy relevance since frequent communications from the tax department are costly in terms in monetary costs and impose a cognitive burden on the taxpayers. By measuring the persistence of the treatment effects, we can potentially comment on the optimal frequency of sending nudges to the taxpayers.
External Link(s)

Registration Citation

Citation
Gupta, Bhanu and Tejaswi Velayudhan. 2021. "Impact of Networks on Non-Compliance in VAT." AEA RCT Registry. October 27. https://doi.org/10.1257/rct.8357-1.1
Experimental Details

Interventions

Intervention(s)
We are sending pre-recorded messages to target population among Delhi's taxpayers to increase compliance of Goods and Services Tax (GST) tax laws. The target populations consists of late filers and non-filer of two different GST forms. The intervention takes the form of pre-recorded calls which are made to the taxpayers around 4 times in a month. Different treatment arms vary in terms of the content of the message.

The primary objective of the intervention is to increase the rate of overall filing and reduce the rate of late-filing. Since past late filing is indicative of future late-filing, nudging the past late-filers can potentially reduce the rate of future late-filing.

The second objective of this study is to analyze the spillover effects of nudging the taxpayer on the their commercial network. Specifically, non-compliance of upstream taxpayer can cause non-compliance of their downstream taxpayers. If the upstream partner hasn't filed the return, then the downstream partners can't claim input tax credits which can significantly increase their short-term tax liability. Thus, nudging the upstream partner can significantly increase the overall compliance.

The final objective is to assess the net benefit of government communication as frequent communications from the tax department are costly in terms in monetary costs and impose a cognitive burden on the taxpayers. By measuring the persistence of the treatment effects, we can potentially comment on the optimal frequency of sending nudges to the taxpayers.
Intervention Start Date
2021-09-09
Intervention End Date
2021-12-31

Primary Outcomes

Primary Outcomes (end points)
Our primary outcomes will be fall in the category of measuring filing behaviour. We would measure on-time filing rate, as well as, filing rate of previous delinquents. Filing behaviour will be observed for both the taxpayer and the network partners. Additionally, we would measure the revenue impact of the nudges and if the nudges cause some firms to deregister.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Our target population is drawn from the administrative records of all the GST taxpayers in Delhi. Because the intervention is to send automated messages to taxpayers by phone, we have randomized unique phone numbers into treatment groups, and not unique taxpayers. We exclude any taxpayers who share a phone number from our target population. Among taxpayers who have a unique phone number, the number is classified as a “non-filer” or “late-filer”. After these restrictions, we are left with a sampling frame of 283,794 taxpayers who we then stratify according to nature of business, type of non-compliance and imputed turnover. Our experimental design has 3 treatment arms. All taxpayers registered with the central government, and 25% of taxpayers registered with the state government are assigned to the control group. The three type of messages are as follows: 1. The reminder messages informs the taxpayers of impending deadlines; 2. The deterrence messages, besides reminding the taxpayers, also inform them about the fees/penalties associated with non-compliance; and 3. The custom deterrence message also includes some information of the past non-compliant behaviour.
Experimental Design Details
Not available
Randomization Method
Randomization done in office by a computer
Randomization Unit
Unique phone numbers of a firm
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
283,710 taxpayers
Sample size: planned number of observations
283,710 taxpayers
Sample size (or number of clusters) by treatment arms
70,928 taxpayers control; 42,606 taxpayers reminder; 85,088 taxpayers deterrence; 85,088 taxpayers custom deterrence
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
Ashoka Institutional Review Board
IRB Approval Date
2021-09-27
IRB Approval Number
18_21_Gupta
Analysis Plan

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

Request Information