Tax Compliance of SMEs

Last registered on August 21, 2017

Pre-Trial

Trial Information

General Information

Title
Tax Compliance of SMEs
RCT ID
AEARCTR-0002390
Initial registration date
August 20, 2017

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
August 21, 2017, 10:58 AM EDT

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

Locations

Region

Primary Investigator

Affiliation
ETH Zürich

Other Primary Investigator(s)

PI Affiliation
ZEW Mannheim

Additional Trial Information

Status
On going
Start date
2017-07-18
End date
2018-03-01
Secondary IDs
Abstract
We present the design and pre-analysis plan for a randomized control trial evaluating the effects of randomized audit threats and moral appeals on tax compliance of small and medium sized firms in Bulgaria. True audit probabilities are randomly varied between treatments and communicated to the firms. In addition, we employ treatments in which firms receive mailings highlighting the importance of taxes and specifying the purpose of tax money. Similar to the treatments with varying audit probabilities, we also manipulate the strength of the moral appeals to taxpayers. We benchmark our findings in the different treatments against a control group which receives a neutral mail from the tax authorities.
External Link(s)

Registration Citation

Citation
Doerrenberg, Philipp and Jan Schmitz. 2017. "Tax Compliance of SMEs." AEA RCT Registry. August 21. https://doi.org/10.1257/rct.2390-1.0
Former Citation
Doerrenberg, Philipp and Jan Schmitz. 2017. "Tax Compliance of SMEs." AEA RCT Registry. August 21. https://www.socialscienceregistry.org/trials/2390/history/20649
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)
The study will be conducted between July and December 2017 (or January/February 2018--depending on the capabilities of the NRA) as a randomized control trial (RCT) in collaboration with the Bulgarian tax authorities. 172.172 SMEs and self-employed taxpayers have been randomly divided into an untreated control group and three main treatment groups. In the first treatment group - the treated control group -, taxpayers receive a neutral mail informing about the website of the Bulgarian tax authorities which contains valuable instruction on how to file taxes correctly. In the second treatment group - the audit group -, the probability with which a taxpayer will be subject to an audit in the upcoming month is communicated to the taxpayer by mail. In the third treatment group - the moral appeals group -, taxpayers receive a mail informing them about the moral obligation to pay taxes as well as what taxes are used for. In all treatments, taxpayers receive the same information as in the treated control group but the content of the mailings is extended to either include information about the true audit probability or information about the moral obligation to pay taxes and what taxes are used for (content invoking reciprocity). The two main treatment groups - audit group and moral appeals group - consist of several subgroups in which the audit probability and the strength of the morale appeal are varied; see below for detailed descriptions of all treatment groups.

Because tax evasion is a major problem -- not only in Bulgaria -- , the tax authorities and we are interested in identifying the optimal and most cost effective audit probability (from the whole possible set between 0\% and 100\%). Specifically, in six sub-treatment groups we introduce individual firm specific audit probabilities of 0\%, 1\%, 10\%, 40\%, 60\%, 100\% and inform taxpayers about their individual probability to be audited. We further introduce one treatment with an ambiguous positive audit probability (of 1\%) where taxpayers are not informed about the exact probability and only that the probability is positive. Additionally, since audits are costly, the tax authorities are also interested in how effective moral appeals to pay taxes work in comparison to an officially announced probability of receiving an audit. Therefore, we want to use tools from behavioral economics that have proven to effectively increase cooperation such as, e.g., inducing an increased feeling of reciprocity through moral messages. Precisely, we have four sub-treatments using moral appeals. In one treatment, we invoke weak reciprocity. In another treatment, we invoke strong reciprocity indicating what taxes are used for. In the third moral appeal treatment we directly address the taxpayer and mention that the taxpayer benefits from services and infrastructure paid for by tax money. In the fourth treatment using moral appeals the mail is accompanied by a graphical example (i.e., a picture of a playground for children) of tax spending. We then aim at testing how well these moral messages work in increasing tax compliance compared to an announced audit.



All interventions have been prepared in collaboration with the tax authorities. The interventions take into account the capabilities and interests of the NRA. The NRA, however, will communicate and interact with the taxpayers. The researchers will not have any contact with the taxpayers. All mailings that taxpayers receive will be sent by the tax authorities.

Taxpayers in the untreated control group will not be subject to any intervention. However, some taxpayers in the this group will be invited to voluntarily fill in a questionnaire and state their perceived probability of receiving an audit and how immoral they think tax evasion is. This survey will - depending on the response rate - inform us about the underlying belief about the probability of receiving an audit and the underlying immorality of cheating on tax returns in the population of SMEs in Bulgaria. However, the survey is mainly independent from the experiment presented here. Therefore, we will not specify the details of the survey in more detail here. Firms in the survey, however, will be excluded from the sample as they maybe behave differently because of filling in a tax (and honesty) related questionnaire.

Intervention Start Date
2017-07-18
Intervention End Date
2017-12-31

Primary Outcomes

Primary Outcomes (end points)
The aim of the RCT is to compare the tax-compliance response of firms across treatment groups. For this purpose, we particularly study (monthly reported) VAT payments and social security payments of SMEs for their employees across firms in the treatment groups.
Primary Outcomes (explanation)
See PAP.

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
The study will be conducted between July and December 2017 (or January/February 2018--depending on the capabilities of the NRA) as a randomized control trial (RCT) in collaboration with the Bulgarian tax authorities. 172.172 SMEs and self-employed taxpayers have been randomly divided into an untreated control group and three main treatment groups. In the first treatment group - the treated control group -, taxpayers receive a neutral mail informing about the website of the Bulgarian tax authorities which contains valuable instruction on how to file taxes correctly. In the second treatment group - the audit group -, the probability with which a taxpayer will be subject to an audit in the upcoming month is communicated to the taxpayer by mail. In the third treatment group - the moral appeals group -, taxpayers receive a mail informing them about the moral obligation to pay taxes as well as what taxes are used for. In all treatments, taxpayers receive the same information as in the treated control group but the content of the mailings is extended to either include information about the true audit probability or information about the moral obligation to pay taxes and what taxes are used for (content invoking reciprocity). The two main treatment groups - audit group and moral appeals group - consist of several subgroups in which the audit probability and the strength of the morale appeal are varied; see below for detailed descriptions of all treatment groups.

Because tax evasion is a major problem -- not only in Bulgaria -- , the tax authorities and we are interested in identifying the optimal and most cost effective audit probability (from the whole possible set between 0\% and 100\%). Specifically, in six sub-treatment groups we introduce individual firm specific audit probabilities of 0\%, 1\%, 10\%, 40\%, 60\%, 100\% and inform taxpayers about their individual probability to be audited. We further introduce one treatment with an ambiguous positive audit probability (of 1\%) where taxpayers are not informed about the exact probability and only that the probability is positive. Additionally, since audits are costly, the tax authorities are also interested in how effective moral appeals to pay taxes work in comparison to an officially announced probability of receiving an audit. Therefore, we want to use tools from behavioral economics that have proven to effectively increase cooperation such as, e.g., inducing an increased feeling of reciprocity through moral messages. Precisely, we have four sub-treatments using moral appeals. In one treatment, we invoke weak reciprocity. In another treatment, we invoke strong reciprocity indicating what taxes are used for. In the third moral appeal treatment we directly address the taxpayer and mention that the taxpayer benefits from services and infrastructure paid for by tax money. In the fourth treatment using moral appeals the mail is accompanied by a graphical example (i.e., a picture of a playground for children) of tax spending. We then aim at testing how well these moral messages work in increasing tax compliance compared to an announced audit.



All interventions have been prepared in collaboration with the tax authorities. The interventions take into account the capabilities and interests of the NRA. The NRA, however, will communicate and interact with the taxpayers. The researchers will not have any contact with the taxpayers. All mailings that taxpayers receive will be sent by the tax authorities.

Taxpayers in the untreated control group will not be subject to any intervention. However, some taxpayers in the this group will be invited to voluntarily fill in a questionnaire and state their perceived probability of receiving an audit and how immoral they think tax evasion is. This survey will - depending on the response rate - inform us about the underlying belief about the probability of receiving an audit and the underlying immorality of cheating on tax returns in the population of SMEs in Bulgaria. However, the survey is mainly independent from the experiment presented here. Therefore, we will not specify the details of the survey in more detail here. Firms in the survey, however, will be excluded from the sample as they maybe behave differently because of filling in a tax (and honesty) related questionnaire.

Experimental Design Details
Randomization Method
To randomly allocate SMEs and self employed taxpayers into the different treatments, the tax authorities provided us with data from 172.172 SMEs and self employed for the year 2016 (i.e., pre-experiment data). Specifically, we received annonymized information about taxpayers' monthly VAT payments, monthly social security payments for workers and firm size. Thus, while the tax authority interacts with the taxpayers, i.e., sends out mailing, performs audits and collects the data, we, the researchers, performed the randomization process taking into account the capabilities and interests of the tax authorities, i.e., number of audits that can be performed and number of mailings that can be sent out.

Precisely, the tax authorities allow us to send out 68.200 mailings and are willing to perform 2210 audits. Importantly, these audits will be performed anyway and therefore do not increase nor decrease the overall likelihood of a firm in the sample to be audited in Bulgaria compared to 2016.

Moreover, to cover the range of audit probabilities the tax authorities are interested in, we are able to assign 5000 SMEs and self employed to the ambiguity treatment, 5200 firms to the 1% audit probability treatment, 5000 firms to the 10% audit probability treatment, 2000 firms to the 40% audit probability treatment, 1180 firms to the 60% audit probability treatment and 10.000 firms to the control group, and 10.000 firms to each of the moral appeals treatments and 10.000 firms to the survey group. We are limited by strategical considerations and manpower restrictions of the tax authorities to only include 100 taxpayers in the 0% audit treatment and 100 firms in the 100% audit treatment (treatment 4 and 9).

Given these considerations and restrictions and in light of our goal to have comparable firms in each treatment group, we randomized as follows. First, we ranked taxpayers according to mean VAT payments in the year 2016 in ascending order. Second, we divided the taxpayers in deciles; that is, we created 10 groups where the first group consisted of the 10% taxpayers with the largest VAT payments and the 10th group consisted of the 10% taxpayers with the lowest VAT payments. Third, within each decile we randomly assigned a number to each firm and ranked firms within each decile by this random number. Fourth, based on this random ranking in each decile, we assigned the taxpayers to treatment groups by assigning n/10 firms to treatment group X, where X is one of the treatment groups and $n$ is the total number of firms we intended to assign to group X. Since we do this within each decile, we in total assigned 10 * n/10 = n firms to group X.


This procedure ensures that an equal number of taxpayers from each decile is assigned to the respective treatments or control group. Since we want the survey to be comparable to the taxpayers in the treatments and representative for the population of SMEs in Bulgaria, we also included the taxpayers who will be invited to fill in the survey into the randomization process.
Randomization Unit
firm tax base (VAT)
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
172.172 (individual firm), depending on subgroup for subgroup analysis e.g., ZiP code (see attached PAP)
Sample size: planned number of observations
172.172
Sample size (or number of clusters) by treatment arms
see attached PAP
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
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

Experimental design and Pre Analysis Plan - Doerrenberg and Schmitz

MD5: 0568ad365e420597d7510c02f6eb5e4c

SHA1: dfdcd8eff9f474e33a88efb0e1f5586006bf8de9

Uploaded At: August 20, 2017

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?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials