Close button
The Guardian
Email YouTube Facebook Instagram Twitter WhatsApp

If you want to improve tax compliance, it is a bad idea to keep telling taxpayers that tax compliance is low


Paul is serious about changing jobs. He has interviewed with 7 potential employers in the last 5 months. None of them has called him back. Today, he is sitting across the table from Anne; potential employer number 8. Paul spends about 5 minutes of his 25 minute interview slot telling Anne how 6 out of his previous 7 interviewers did not call him back. You are Anne. How likely are you to recommend Paul for the job?

Even if his resume and experience are impressive, you will have your doubts. You will keep thinking: what am I missing? What did the other 6 people see? If you are like most people, the fact that majority of the previous interviewers did not think Paul was good enough could be enough reason for you to reject him.

When public officials are trying to get taxpayers to pay their taxes, they often make exactly the same mistake as Paul. They focus on low compliance rates and the fact that majority of the people are not paying taxes. This inadvertently leaves the potential taxpayers thinking: if no one is paying, should I really be the fool that decides to pay? Many times, this feeling is enough to prevent voluntary compliance.


When people are not sure, they do what everybody else is doing If Anne rejects Paul because of his track record with other interviewers, she has just made a call based on what social scientists call social proof. Social proof is our tendency to see an action as more appropriate when others are doing it. Social proof is observed frequently under conditions of uncertainty. When people are not sure of what to do, they will look to what other people are doing and copy it.

Take a common example. You are driving to work in the morning. You observe traffic building up ahead of you and make up your mind to join in as you usually do. As you approach, many of the cars in front of you start to make a u-turn. What do you do? Do you forge ahead, determined to figure out what has informed this strange behaviour before deciding on the appropriate course of action? Or do you simply turn around like everyone else convinced that all those people must have seen something.

Even when people are very sure, they can still succumb to social proof In his now famous conformity experiments, Solomon Asch set out to show that social proof would only happen under conditions of uncertainty. He ended up proving the opposite.

In the experiments, subjects were shown several sets of two cards. For each set, the first card had a single straight line and the second card had 3 straight lines. One of the lines on the second card was the same length as the line on the first card but each one of the other two lines was either longer or shorter. Each person was asked to judge which of the 3 lines on the second card was the same length as the line on the first card. In order to see what would happen without social proof, the cards were shown to a control group of subjects individually. All the subjects chose the right match nearly 100% of the time. The same cards were then shown to the test group. For this group however, there were six other people in the room (who were actually actors) in addition to the subject. Each person was asked to state his assessment out aloud with the only genuine subject of the experiment speaking last. The six actors had been primed to give an incorrect answer every now and again. Solomon Asch wanted to know if the subjects would also follow the crowd against their better judgement.

Under this condition, 75% of the participants went with the crowd on at least one occasion. When they were interviewed after the experiment, many of them confessed to following the crowd in spite of knowing the right answer.

What would you have done if you were one of the subjects? Are you sure? How many times have you listened to a recommendation during a meeting, concluded that the sensible thing to do would be to object to it only to keep quiet and go along with everyone else after the first four people that spoke before you spoke in favour of the proposal? Even when people are sure about what to do, the behaviour of others can still get them moving in a different direction.

Some countries are using this knowledge to improve tax compliance
In a study performed by the UK’s Behavioural Insights Team (BIT), different types of letters were sent to about 100,000 individual taxpayers who were yet to pay their taxes. The purpose was to determine which type of message would have the best impact on tax compliance. Compared to the standard letter, the letters which indicated that majority (90%) of taxpayers in the recipient’s category had already paid their taxes and that the recipient was in the minority led to an increase of about 16% in tax compliance. These letters also outperformed “public goods messages” which emphasised the gains from paying taxes such as roads, schools etc.

A similar study carried out in Guatemala by Kettle et al (2016) also reported consistent findings. Tax payments resulting from messages that are based on social proof increased tax payments significantly compared to the control message.

These studies provide scientific evidence of something many taxpayers already know. Whenever taxpayers have to make a decision on an issue (this includes both issues where the law is clear as well as issues where the law is vague), many of them never do so until they have asked the question: “what are other people doing?” A lot of times, you can predict the final decision based on the answer to this question.

The evidence suggests that focusing on how the majority are non-compliant will likely increase non compliance
Messaging about what everyone else is doing can either increase or reduce tax compliance. When messages such as “only 13% of taxpayers are compliant” go out to the public consistently, taxpayers interpret this to mean “nobody is paying taxes”, I had better not pay taxes too. These messages use the power of social proof against their senders.

So what should tax administrators do? If non-compliance is high, are they to lie about it?
The answer is no. Robert Cialdini et al have a useful suggestion based on several studies they conducted. When the way most people behave is contrary to the behaviour that you want to encourage, then you are better off not drawing the attention of your audience to this fact. Instead, you should focus your audience on what the appropriate behaviours are. Instead of emphasising the fact that most of their constituents do not pay tax, tax administrators should be telling their constituents the behaviours they should desist from without drawing their focus to size of the defaulters.

It could also be better to ignore social proof all together and focus the message on other behavioural attributes that can drive compliance given the circumstances. For example deterrent messages which frame non-declaration as an intentional and deliberate act have proved to be quite effective.


Is this all that is needed to take tax compliance to the desired levels?
Of course not. There are lot of things that have to work together to get tax compliance to the desired levels. For example, Governments must become more accountable so that taxpayers can see concrete evidence of their tax money being put to use. Tax rules also need to be made clearer and compliance easier.

The point I make here is that the framing of messages can have a big impact on compliance and many government officials end up creating messages that work against them. Framing the message right can therefore move the tax authority closer to, rather than farther away from, its goals. It is also cheap.

Seun Adu is an Associate Director and Transfer Pricing Leader at PwC Nigeria. He is a regular writer and public speaker on tax and transfer pricing matters.Visit our tax blog for in-depth analyses, unique insight and superlative perspective on tax matters: Subscription is free!

In this article:
Tax compliance
Receive News Alerts on Whatsapp: +2348136370421

No comments yet