Tax Evasion Experiment

Last registered on August 24, 2022

View Trial History

Pre-Trial

Trial Information

General Information

Title

Tax Evasion Experiment

RCT ID

AEARCTR-0008468

Initial registration date

November 02, 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

November 05, 2021, 6:18 PM EDT

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

Last updated

August 24, 2022, 5:42 AM EDT

Last updated is the most recent time when changes to the trial's registration were published.

Locations

Country

India

Region

Delhi

Primary Investigator

Name

Narges Hajimoladarvish

Affiliation

Centre for Social and Behaviour Change, Ashoka University

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Sanjit Dhami

PI Affiliation

University of Leicester

Contact Investigator

PI Name

Pavan Mamidi

PI Affiliation

Centre for Social and Behaviour Change, Ashoka University

Contact Investigator

Additional Trial Information

Status

On going

Start date

2021-10-18

End date

2022-11-30

Keywords

Behavior, Crime, Violence, & Conflict, Governance, Welfare

Additional Keywords

JEL code(s)

C91, C92, D82, D91, G21

Secondary IDs

Prior work

This trial does not extend or rely on any prior RCTs.

Abstract

This experiment measures loss and risk aversion in the context of entitled and unentitled income in order to understand tax evasion with relation to mental accounting. Participants will be asked to make decisions regarding paying taxes while taking into account various factors like tax, audit, and fine rates. Subject-specific loss aversions and risk aversions are elicited through binary lotteries.

External Link(s)

Registration Citation

Citation

Dhami, Sanjit, Narges Hajimoladarvish and Pavan Mamidi. 2022. "Tax Evasion Experiment." AEA RCT Registry. August 24. https://doi.org/10.1257/rct.8468-1.2

Sponsors & Partners

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

Request Information

Experimental Details

Interventions

Intervention(s)

Intervention Start Date

2021-11-10

Intervention End Date

2021-12-31

Primary Outcomes

Primary Outcomes (end points)

Tax evasion, Loss aversion, Risk aversion

Primary Outcomes (explanation)

Tax evasion is calculated as income minus declared income in the experiment (we have tax evasion for varying policy parameters. However our primary variable of interest is when the probability of audit is decreasing in declared income).
Risk aversion is calculated through a 6 bisection procedure to elicit certainty equivalent of a lottery. This indifference helps us to estimate the parameter of utility function which is required for the calculation of loss aversion.
Loss aversion is calculated through an indifference between subjects' income (y) and a lottery that pays y-z with 50 per cent chance and y + x otherwise.

Secondary Outcomes

Secondary Outcomes (end points)

Our control variables include Age, Gender, Marital Status, Education, Expenses, Employment, Religion, Number of earning members in the family, Household annual income and time spent in the experiment.

Secondary Outcomes (explanation)

Experimental Design

We are interested to investigate if tax evasion is related to the source of income (Labour and non-labour income) and behavioural parameters such as loss aversion as predicted by our model. Subjects will either earn some income (labour) in the experiment or they are randomly given some income and have to declare it for tax purposes. We also elicit subject-specific loss aversion and risk aversion. We test if the source of income, loss aversion and risk aversion affects tax evasion. We also randomly prime subjects to tax evasion data in India and see if has any effect on subjects tax evasion.

Experimental Design Details

Randomization Method

Randomization is through a random draw by a computer built within the experiment online.

Randomization Unit

Our experiment design has multiple layers of randomization that take into account the various effects we are interested in exploring. We first randomly assign subjects into the primed group and the unprimed group, wherein the primed group will be exposed to data on tax evasion in India and the unprimed group will not be exposed to that data. We are also interested in distinguishing between the effects of the type of income source, hence subjects are further assigned between the entitled/earned income arm and the unentitled/random income arm. Finally, we take into account task order randomization between the lottery (loss aversion) task and the tax payment task to control for any possible order effects.

Was the treatment clustered?

Yes

Experiment Characteristics

Sample size: planned number of clusters

We will have 8 groups of people.

Sample size: planned number of observations

We will have a maximum of 60 individuals in each group.

Sample size (or number of clusters) by treatment arms

60 primed entitled income (lottery-tax), 60 primed entitled income (tax-lottery), 60 primed unentitled income (lottery-tax), 60 primed unentitled income (tax-lottery), 60 unprimed entitled income (lottery-tax), 60 unprimed entitled income (tax-lottery), 60 unprimed unentitled income (lottery-tax), 60 unprimed unentitled income (tax-lottery). Hence, in total we will have 480 individuals.

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

Supporting Documents and Materials

IRB