Irish Revenue Randomised Trials

This month the Irish Revenue Commissioners (responsible for tax administration) published the results of 20 randomised trials they have conducted in the area of behavioural design. This is a significant report in terms of Irish public policy and also contributes to the growing international literature in this area. A summary of the report is below. 

While audit and other risk management interventions are effective compliance tools, they can be expensive and time consuming for both Revenue and taxpayers. Targeted treatments using behavioural science can be a complementary and cost-effective tool to improve compliance. A summary of key findings across four behavioural insights is below. 

Deterrence: Deterrence strategies (e.g., highlighting possible sanctions) dissuade taxpayers from non-compliant behaviour. The research confirms that deterrent effects significantly improve taxpayer compliance, particularly when combined with other insights. They impact different taxpayer segments differently. 

Simplification and Salience: Compliance or other behaviours can be enhanced through simpler presentation of information and by drawing attention to key details. For tax administrations, this may involve highlighting the third-party information held, for example through correspondence or the pre-filling of tax returns. Information in any communications should be presented in the most clear and simple way possible, including bolding and centred text. 

Personalisation: International research has shown the potential of more personalised correspondence, which is increasingly becoming a possibility given technological advancements. Revenue trials confirm that personalisation leads to greater and quicker engagement, especially when multiple elements of personalisation are applied. 

Social Norms: The behaviour of others can influence an individual’s choices. Revenue research finds that social norms are generally not effective at influencing behaviour. However, there is limited evidence indicating that these may improve taxpayer compliance when combined with other insights. 

According to a meta-analysis, which weights the result of 20 trials by sample size, the most effective insights tend to be deterrence (+8.0% improvement in targeted behaviour), personalisation (+4.0%) simplification and salience (+3.3%) and social norms (-1.6%). The wording and design of communications can affect taxpayer compliance. Even seemingly insignificant changes to correspondence can significantly change behaviour. While these lessons have mainly been learned from letters, they should be considered in any Revenue communication with taxpayers.