Embezzlement and taxation: Experimental evidence

Last registered on February 19, 2020

View Trial History

Pre-Trial

Trial Information

General Information

Title

Embezzlement and taxation: Experimental evidence

RCT ID

AEARCTR-0004646

Initial registration date

January 27, 2020

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 28, 2020, 10:19 AM EST

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

Last updated

February 19, 2020, 8:42 AM EST

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

Locations

Country

United States of America

Region

Primary Investigator

Name

Ingrid Hoem Sjursen

Affiliation

Centre for Applied Research at NHH (SNF) / Stockholm School of Economics

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Jean-Robert Tyran

PI Affiliation

University of Vienna

Contact Investigator

PI Name

Louis Putterman

PI Affiliation

Brown University

Contact Investigator

Additional Trial Information

Status

In development

Start date

2020-02-19

End date

2020-03-30

Keywords

Governance

Additional Keywords

Tax, embezzlement, punishment

JEL code(s)

Secondary IDs

Abstract

We provide causal evidence on how two institutional features, source of revenue (tax vs. windfall) and whether the affected party can punish, affect embezzlement.

External Link(s)

Registration Citation

Citation

Putterman, Louis, Ingrid Hoem Sjursen and Jean-Robert Tyran. 2020. "Embezzlement and taxation: Experimental evidence." AEA RCT Registry. February 19. https://doi.org/10.1257/rct.4646-1.1

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

In the experimental treatments, we exogenously vary whether the public revenue is a windfall or is collected as a tax on earnings from a real effort task. Additionally, we vary whether the parties affected by the embezzlement can demand accountability from the person embezzling public revenue (by implementing a costly punishment).

Intervention Start Date

2020-02-19

Intervention End Date

2020-03-30

Primary Outcomes

Primary Outcomes (end points)

1. Embezzlement: Share of public revenue the public official does not invest in the public project.

2. Punishment: Minimum acceptable investment: highest share of public revenue allocated to the citizens for which the citizen does not punish the public official.

Primary Outcomes (explanation)

1. Embezzlement: (2 - amount of public revenue the public official invests in the public project)/2.

2. Punishment: Highest amount of public revenue allocated to the citizens for which the citizen does not punish the public official/2.

Secondary Outcomes

Secondary Outcomes (end points)

1. Beliefs about punishment: Highest share of public revenue allocated to the citizens for which public officials believe that the (randomly selected) citizens are willing to punish them.

2. Social norm for embezzlement: Guess about other participant's judgement of five possible investment decisions the public official could make.

3. Beliefs about embezzlement: Citizen's guess about the public official's embezzlement.

Secondary Outcomes (explanation)

1. Beliefs about punishment: Highest amount of public revenue allocated to the citizens for which public officials believe that the (randomly selected) citizens are willing to punish them/2.

2. Social norm for embezzlement: For each of the five social appropriateness judgements, we construct a discrete variable coded in the following way: "very socially inappropriate" = 1, "socially inappropriate" = 2, "somewhat socially inappropriate" = 3, "somewhat socially appropriate" = 4, "socially appropriate" =5, "very socially appropriate" = 6.

3. Beliefs about embezzlement: (2 - citizen's guess about the public official's investment)/2.

Experimental Design

Participants interact in groups of three, where one participant has the role of public official and two participants have the role of citizens. In the interaction, there is some public revenue, the purpose of which is to benefi t the two citizens. The public official is in charge of investing the public revenue in a public project that benefi ts the citizens, but has the power to keep some for him or herself.

In the experimental treatments, we exogenously vary whether the public revenue is a windfall or is collected as a tax on earnings from a real effort task. Additionally, we vary whether the parties affected by the embezzlement can demand accountability from the person embezzling public revenue (by implementing a costly punishment).

Experimental Design Details

Randomization Method

Computer (Qualtrics software)

Randomization Unit

Individual

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

1600 groups (consisting of three individuals)

Sample size: planned number of observations

4800 individuals, 1600 groups

Sample size (or number of clusters) by treatment arms

1600 individuals (400 groups) in each of the four treatments

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

1. Embezzlement: 0.2 standard deviation (5.5 percentage points, assuming a standard deviation of 0.28) 2. Punishment: 0.14 standard deviation (4.35 percentage points, assuming a standard deviation of 0.31)

Supporting Documents and Materials

IRB