Back to History Current Version

Automatic Payments and Tax Evasion: an experimental study

Last registered on March 31, 2025
View Trial History

Pre-Trial

Trial Information

General Information

Title
Automatic Payments and Tax Evasion: an experimental study
RCT ID
AEARCTR-0013243
Initial registration date
March 26, 2024

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
April 02, 2024, 10:57 AM EDT

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

Last updated
March 31, 2025, 5:48 AM EDT

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

Locations

Country
Italy
Region
Toscana

Primary Investigator

Name
Damiano Paoli
Affiliation
Alma Mater Studiorum - Università di Bologna
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Vincenzo Valori
PI Affiliation
Università degli studi di Firenze
Contact Investigator
PI Name
Maria Grazia Pazienza
PI Affiliation
Università degli studi di Firenze
Contact Investigator
PI Name
Domenico Colucci
PI Affiliation
Università degli studi di Firenze
Contact Investigator

Additional Trial Information

Status
Completed
Start date
2024-03-26
End date
2024-09-01
Keywords
Behavior, Crime, Violence, & Conflict, Governance
Additional Keywords
Tax Evasion, Tax Evasion Game
JEL code(s)
C91, H26, H31
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
This study aims to investigate the impact of automatic payment systems on tax evasion. We design a laboratory experiment where subjects play a Tax Evasion Game. Participants receive a fixed endowment and engage in an effort task under time pressure where they have the opportunity to earn income but are also simultaneously required to pay a tax, spending a fraction of the available time to successfully pay it. While evading the tax offers a higher expected payoff, it also carries the risk of a significant fine if detected. In the intervention treatment, we introduce an automatic payment option, allowing participants to commit to the tax payment before beginning the effort task. The findings will examine the impact of automatic payments on tax compliance and their potential to effectively deter tax evasion. This research contributes to our understanding of how individuals react to tax systems and the potential for automated tax collection and tax payment subscriptions to enhance revenue and tax compliance.
External Link(s)

Registration Citation

Citation
Colucci, Domenico et al. 2025. "Automatic Payments and Tax Evasion: an experimental study." AEA RCT Registry. March 31. https://doi.org/10.1257/rct.13243-2.0
Sponsors & Partners
Experimental Details

Interventions

Intervention(s)
We designed a Tax Evasion Game where subjects have to complete a real effort task under time pressure to obtain a payoff and simultaneously have to decide whether to pay or not a tax by inserting a code. The intervention introduced in one of the two treatments is an automatic payment option that allows participants to pay the tax automatically without inserting the code during the real effort task.
Intervention (Hidden)
Intervention Start Date
2024-03-27
Intervention End Date
2024-04-30

Primary Outcomes

Primary Outcomes (end points)
The main outcome variable of the study is compliance at the individual level.
With our design, we aim to measure the Average Treatment Effect of the automatic payment system on individual compliance.

Furthermore, we can also estimate the ATE on tax revenue and individual earnings, to obtain valuable policy implications of the intervention.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Performance in the real effort task.
Risk aversion, measured with the Bomb Risk Elicitation Task (Crosetto and Filippin, 2013), and its influence on the adhesion to the automatic payment system.
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Participants receive an endowment m. They will have to carry out an activity that will require constant commitment (effort task) and that can generate a maximum income of 2m. In addition, at the same time as the effort task, they will also be required to pay a tax equal to the entire endowment m. If they don't pay it, they will be audited with a given probability and will pay a fine equal to m which is added to the entire tax, resulting in an outflow equal to 2m. The expected payoff of not paying the tax is positive.
The effort task is an activity that lasts 45 seconds and consists of counting the number of 1s in several strings composed only of 0s and 1s. Simultaneously, there will be another box in the desktop window, in which subjects have to copy a given code if they want to pay the tax.
In the Automatic Payment Treatment, participants have the option of adhering to the automatic payment of the tax before starting the effort task. If they adhere to this system, during the effort task they will not have to enter the code because the tax will be automatically subtracted from their endowment. They will have to choose whether to keep the adhesion or leave the automatic system in each round, before starting the effort task.
Experimental Design Details
Subjects will play 3 practice rounds and 12 payoff-relevant rounds.
In each round participants receive an endowment m (100 ECUs). They will have to carry out an activity that will require constant commitment (effort task) and that can generate a maximum income of 2m (200 ECUs). In addition, at the same time as the effort task, they will also be required to pay a tax equal to the entire endowment m. If they don't pay it, they will be audited with a 30% probability and will pay a fine equal to m which is added to the entire tax, resulting in an outflow equal to 2m. Thus, the expected payoff of not paying the tax is positive.
The effort task is an activity that lasts 45 seconds and consists of counting the number of 1s in a string composed only of 0s and 1s. Each string will contain a variable number of digits and a random number of 1s. Each correct answer will earn 10 ECUs to the subject, and there will be 20 strings. Simultaneously, there will be another box in the desktop window, in which subjects have to copy an alphanumeric code if they want to pay the tax. This side task will therefore take a fraction of the available time.
In the Automatic Payment Treatment, participants have the option of adhering to the automatic payment of the tax before starting the effort task. If they adhere to this system, during the effort task they will not have to enter the alphanumeric code because the tax will be automatically subtracted from their endowment. They will have to choose whether to keep the adhesion or leave the automatic system in each round, before starting the effort task.
After the effort task, subjects are informed of the overall earnings of the round. These earnings include the income from the effort task and the net result from the choice made on taxation:
• If they pay the tax, the initial endowment will be used in full to pay the tax. Therefore, the total gain will be equal to the income from the effort task.
• If they do not pay the tax, there are two possible outcomes:
- They are not audited: the endowment is kept, and the total gain is equal to the sum of it and the income from the effort task.
- They are audited: the endowment is used to pay the tax and the fine is imposed. The total earnings are equal to the difference between the income from the effort task and the fine received.
Randomization Method
Randomization will happen during recruitment through ORSEE.
We will identify a sample of eligible subjects and randomize them in the two treatments.
Randomization Unit
Randomization at the individual level (participants).
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
134 subjects.
Sample size: planned number of observations
134 subjects x 12 rounds = 1608 observations.
Sample size (or number of clusters) by treatment arms
67 subjects per treatment.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
a two-tailed Wilcoxon-Mann-Whitney test on the average individual compliance in the two treatments (Control vs Automatic Payment) with alpha equal to 0.05 and power equal to 0.80 gives a minimum detectable effect of 0.5 SD.
Supporting Documents and Materials

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

Analysis Plan Documents

Pre-analysis plan

MD5: 1eabcc83d1b073dcc1a498e20d192428

SHA1: dec30fac80e9aff7840dfb91905d890c470cae92

Uploaded At: March 26, 2024

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?
Yes
Intervention Completion Date
September 01, 2024, 12:00 +00:00
Data Collection Complete
Yes
Data Collection Completion Date
September 01, 2024, 12:00 +00:00
Final Sample Size: Number of Clusters (Unit of Randomization)
133 subjects.
Was attrition correlated with treatment status?
No
Final Sample Size: Total Number of Observations
1596 observations, equal to 133 subjects times 12 periods.
Final Sample Size (or Number of Clusters) by Treatment Arms
768 (64x12) in control arm, 828 (69x12) in treatment arm
Data Publication

Data Publication

Is public data available?
No

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

Request Information

Program Files

Program Files
No
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials