Improving tax compliance: deductions for work related expenses in income tax returns filed through tax agents
Last registered on March 26, 2017


Trial Information
General Information
Improving tax compliance: deductions for work related expenses in income tax returns filed through tax agents
Initial registration date
March 24, 2017
Last updated
March 26, 2017 12:40 PM EDT
Primary Investigator
Harvard University
Other Primary Investigator(s)
Additional Trial Information
In development
Start date
End date
Secondary IDs
Australians can claim income tax deductions when they incur costs in the course of earning income (e.g., vehicle and other work-related travel costs). Work related expenses are the most common deduction claimed by Australians. In 2014-15 total claims amounted to a total of $21 billion.

The majority of Australian taxpayers that claim work-related expenses in their income tax returns lodge these forms through a registered tax agent. This trial, delivered in collaboration with the Australian Tax Office (ATO), will test a new method, informed by behavioural insights, aimed at supporting Australians to lodge accurate work related expense claims. The trial will test the effectiveness of a letter sent to tax agents reminding them the ATO has a data analytics system that identifies larger-than-expected claims and encouraging them to review their clients’ claims.
External Link(s)
Registration Citation
Hiscox, Michael. 2017. "Improving tax compliance: deductions for work related expenses in income tax returns filed through tax agents." AEA RCT Registry. March 26.
Experimental Details
Tax agents in the treatment group will receive a letter aimed at improving tax compliance amongst their individual clients.
Intervention Start Date
Intervention End Date
Primary Outcomes
Primary Outcomes (end points)
Primary: (1) Average work-related expense claim (2) Amendments to past returns (3) Amendments to past returns of clients on attached list
Secondary: (1) Total deductions (2) Net tax paid (3) Proportion of clients on attached list who amend past returns (4) Average WRE claim for subsequent tax year
Outcomes will be evaluated at the agent level and the individual level.
Primary Outcomes (explanation)
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
Tax agents will be randomly allocated to receive the letter or not.
Experimental Design Details
The sample will be selected from a group of tax agents with higher claiming clients who are not already involved in compliance action with the ATO. Agents in scope will be randomly allocated into a treatment group that will receive a letter from the ATO, and a control group that will not receive the letter. Outcome measures will be recorded in several months’ time to allow for any late lodgements made by tax agents.
Randomization Method
Randomization will be done by a computer algorithm. Agents will be stratified by type, and then matched using non-bipartite pair matching.
Randomization Unit
Tax agents
Was the treatment clustered?
Experiment Characteristics
Sample size: planned number of clusters
2287 tax agents
Sample size: planned number of observations
Approximately 20,000 individual clients of tax agents
Sample size (or number of clusters) by treatment arms
1,144 tax agent clusters in treatment and 1,143 tax agent clusters in control
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
For the outcome measure of average WRE claims by agent, assuming alpha = 0.05 and power = 0.8 the study is powered to detect a 9.0% effect size.
IRB Name
BETA Ethics Committee
IRB Approval Date
IRB Approval Number
BETA ETH 2017 - 004
Post Trial Information
Study Withdrawal
Is the intervention completed?
Is data collection complete?
Data Publication
Data Publication
Is public data available?
Program Files
Program Files
Reports and Papers
Preliminary Reports
Relevant Papers