Mind the gap. Who evades VAT and by how much?

Last registered on January 27, 2023


Trial Information

General Information

Mind the gap. Who evades VAT and by how much?
Initial registration date
January 17, 2023

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
January 27, 2023, 1:07 PM EST

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



Primary Investigator

University of Mannheim

Other Primary Investigator(s)

PI Affiliation
ZEW Mannheim
PI Affiliation
University of Mannheim
PI Affiliation
University of Mannheim

Additional Trial Information

In development
Start date
End date
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
VAT tax evasion erodes government budgets available for public goods, is a source of unfair competition, and exposes individuals to exploitative work conditions. The prevalence of VAT evasion and its determinants are difficult to study, because of its illegal nature. Our objective in this study is to identify the share of firms in Germany that have ever evaded VAT tax. We implement the randomized response technique in a sample of firm decision makers active in Germany. Additional information on the position in the value chain and the firm's industry allows us to compare the occurrence of VAT evasion across parts of the firm population that differ in their firm characteristics and scope of activity. We supplement the results on the prevalence of tax evasion with estimates of its quantitative importance using the indirect questioning technique. The results will provide important insights into the mechanics of tax evasion and valuable starting points for a refined discussion of policy measures to combat this phenomenon.
External Link(s)

Registration Citation

Dörrenberg, Philipp et al. 2023. "Mind the gap. Who evades VAT and by how much?." AEA RCT Registry. January 27. https://doi.org/10.1257/rct.10699-1.0
Sponsors & Partners


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

Request Information
Experimental Details


Intervention Start Date
Intervention End Date

Primary Outcomes

Primary Outcomes (end points)
- share of respondents who have evaded VAT
- magnitude VAT evasion
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
The experiment is designed as a combination of a between and within subjects treatment.

The randomization assigns 12.5% of respondents to a control group, that forces a response. 1-2(1/2)^4=87.5% of respondents may answer truthfully.

The between subjects component compares the proportion of respondents that indicate having evaded VAT beyond the control group. That is, the causal effect is the difference between the number of respondents who answer yes in the control treatment arm compared to the number of respondents who answer yes in the free responses treatment arm. The treatment works via the anonymity given by the randomization device. This effectively blinds the treatment allocations to avoid underreporting bias. A well-blinded RCT is considered the gold standard for clinical trials but is less known in economic applications.

The within subjects component then investigates how those who answered with yes in the first question respond to a subsequent question that uses the indirect questioning technique. The indirect questioning technique asks about the behavior of a competitor.
Experimental Design Details
Not available
Randomization Method
We implement a randomized response technique (RRT) forced response design in a Computer Assisted Telephone Interview (CATI) survey in Germany. The survey experiment is carried out by the partner company uzbonn independently. As a randomization device, we ask respondents to use three coins (or if not available an online coin toss on https://zufallsgenerator.org) but not tell the outcome to the interviewer. The coin tosses determine whether the participant answeres the sensitive question or is forced to answer ‘Yes’ Or ’No’.

The respondent is requested to
- always answer ‘Yes’, irrespective of the truthfulness of the response, if she has received three tails,
- always answer ‘No’, irrespective of the truthfulness of the response, if she received three heads,
- answering truthfully ‘Yes’ or ‘No’, if she received a combination of heads and tails.

The probability of getting three times head or tails is 0.5x0.5x0.5=12.5%. The probability of getting a combination of head and tails is 1-0.125=87.5%. That is, the treatment arm with a forced answer is assigned to 12.5% of respondents, the treatment arm with the potential to answer truthfully is assigned to 87.5% of respondents.
Randomization Unit
individual firm representative
Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters
no clustering
Sample size: planned number of observations
1500 firm representatives
Sample size (or number of clusters) by treatment arms
The randomization assigns respondents to two treatment arms: 12.5% of respondents are assigned to a forced response, 87.5% are assigned to a treatment arm that allows a truthful response.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

Institutional Review Boards (IRBs)

IRB Name
IRB Approval Date
IRB Approval Number
Analysis Plan

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

Request Information