Nudge Economics and Tax Compliance

It would seem the latest bright shiny ball in the policy world is Nudge Economics. On the one hand, I suppose it is good that governments are finally getting around to reading some economics books. On the other hand, governments don’t seem to read the books too carefully since they can’t seem to understand the difference between Nudge economics and, well, economics.

If you take Ec101 you will learn something about the rationality assumption. This assumption means that people make decisions that make them happy. Those decisions may appear to be weird, nonsensical, and ‘irrational’ to some observers, but in economics (well Ec101 that is) we have this condition that one should not judge the preferences underlying someone’s decision. That fact that Mike Moffat hops on one leg throwing erasers at his students to teach this concept does not mean he is irrational. It means he is rational because doing so makes him happy (his students learn the concept).

But wait, you say. People often do things that are irrational and that makes economics bunk. Well, you might want to first read this, but you might also want to take a course beyond Ec101. Sometimes we make mistakes in making decisions because we don’t have the right information, we misunderstand something like information or risks, we don’t pay attention, the costs and benefits of the choice are too closely matched, herd mentality, and so on. So we want to make decision A, but for whatever reason we make decision B.

A good example is organ donation. Almost everyone says they want to donate their organs when they die, but very few of these people actually register to be an organ donor. To become a registered organ donor you have to incur the cost of registration and simple inertia fails to make that happen. So the rational decision is to register, but we observe the irrational decision instead.

This is where Nudge comes in. Nudge is about a cost-effective ways to ensure voluntary compliance usually through a change to the choice architecture to ensure the rational decision is made. So instead of having to register to become an organ donor, we change the choice to having to register to not become an organ donor. That is, moving from an opt-in to an opt-out system is simple Nudge economics. Similar techniques have been used for retirement savings.

This is not the same as incentives. Incentives are used to change a rational choice. So a pigouvian tax, like a carbon tax is pure economics based on rationality. What I find in going to these sessions where government is talking about Nudge economics is that they do not understand the difference between Nudge economics and economics. While they talk Nudge, the actions, suggestions, phrasing are all about pure and simple incentives.

One of the areas where governments want to apply Nudge economics is to tax compliance. After all, Nudge is all about cost-effective voluntary compliance, which is pretty much how we administer our tax system. To do so, we have to understand that Nudge will not work on people who are typical tax evaders. It will not work on those for whom the tax evasion decision is a rational decision. These are people who have strong negative attitudes about paying taxes, have low risk aversion, do not care about getting caught, understand the consequences, and have a strong social norm towards tax evasion. For these people, we have incentives. Nudge will work on people who made a mistake, don’t understand the rule, lack information, follow the herd, misunderstand the risks, and so on.

There have been a number of tests of Nudge theory on tax compliance. Prepopulated forms, which opts someone into filing their taxes, increased the number of people filing their taxes. The reason is simple, the application process for taxes is a significant compliance factor for many non-filers. The down side is this UK report shows there is little evidence that pre-populated forms decrease tax non-compliance by those already filing. By sending the pre-populated form, the revenue authority shows its hand to the taxpayer. There is little incentive for a taxpayer to disclose additional income that the tax authority is not aware of.

The UK used tax letters that stated the majority of individuals pay their taxes on time, which resulted in a 15% increase in on time filings. However, when Switzerland, Minnesota, and France used the same method, they found no effect on compliance behavior. This tells us that, as always, we have to be careful about generalizing findings obtained from methodologies that don’t allow for generalizations (causality 101).

There was also interesting work that examined whether telling taxpayers where their money went helped with tax compliance. It did not. Instead, it made it worse. What did help compliance a lot (16%) was to let the taxpayer allocate some of their tax payment to a particular program.

The difficulty though with some of these Nudge interventions is that you do risk future non-compliance if these interventions are targeted to the wrong people or if taxpayers feel they are not treated respectfully. Meaning that all the Nudge work on compliance can be undone by one wrong move by CRA (its communication with taxpayers is historically dreadful) or messages coming politicians (like those in Ottawa that are already suggesting that tax is a dirty word).

So what are things CRA could do to increase tax compliance?

I’ve talked about tax lotteries before, which is an excellent example of Nudge Economics and tax compliance. And is a great idea for CRA to consider in those pesky cash industries like the renovation industry.

But the simplest of all Nudge techniques in the field of tax compliance is third party reporting. Yup, that can work as a Nudge. Consider tip income. By simply subjecting it to third-party reporting, which is already required for controlled tips, you change the choice architecture related to tip reporting. Another area is rental income, where CRA can use tenants to report their rental income paid and match it to the landlords receipts.

I would like to see CRA more active in this area, but I also it hope it takes the time to understand what Nudge is (and what it is not), who to target, and what will work in Canada. There are more than a few academics who would be thrilled to be involved in such initiatives, in exchange for publications of course.

Advertisements

7 thoughts on “Nudge Economics and Tax Compliance

  1. Is the difference between nudging and economic incentives really that black and white?

    You can argue that third-party reporting of tip income, for example, is not a nudge but a change to economic incentives. Third-party reporting increases the chances of getting caught if you attempt tax evasion, and may swing a rational actor’s cost-benefit analysis in favour of compliance rather than evasion.

    Similarly, with your organ donation example, it may be that people are extremely rational, but just don’t have very strong preferences about organ donation. It may be that, on average, people have extremely mild preferences toward not donating their organs, such that if you were to force people to choose, more people would opt-out (survey responses notwithstanding; people may just be trying to impress the pollster). But making the choice is not costless — I have to think about it, discuss with my family, fill out the paperwork, buy a stamp, walk it to the mailbox, etc. — so when you switch to an opt-out system, rational actors may make a rational choice not to opt out, since the benefit they’d get from opting out exceeds the cost of doing the paperwork.

    I think nudging is often about policymakers deciding what they think is best for people (e.g. having them donate organs) and then setting default options that make it more costly for people to choose an alternative option. Or sometimes we do it to ourselves (e.g. freezing our credit cards in a block of ice, or setting up automatic bank transfers to a savings account, even though we’d arguably prefer buying some new electronics now if it weren’t so difficult to unfreeze our card or transfer funds back to our bank account).

    In these cases, it’s not really true that nudging doesn’t change economic incentives. It does. What makes a nudge a nudge, rather than your run-of-the-mill incentive change, is that you do the nudge by changing a default that has to be set somehow. Absent clear information from an individual, we need to decide whether to use their organs or not after their death, so policymakers basically have to choose an opt-in or an opt-out one. They might as well pick the system that gives them a more desirable policy outcome. Similarly, we have to have tax forms if we want to collect taxes, but we could make them pre-filled, or unfilled. Policymakers can choose the pre-filled ones if they think it improves compliance. But it’s not fair to argue that there’s no incentive change going on at the same time.

    • All great questions. There is a raging debate going on about what exactly is Nudge vs. behavioural economics vs incentives and there is no real consensus. It depends on who you talk to/read and how you think about the underlying behaviour and choice architecture.

  2. Really, nudge isn’t different from other economics at all. The simple Nudge examples, like RRSP enrolment, are about eliminating the cost of enrolment and increasing the cost of not enrolling. When people say they’d like to save more for retirement, but aren’t enrolled in an RRSP matching program at work, they’re not making the “irrational decision,” just saying that I’d like to enroll, but the cost of doing so is higher than my preference for it.

    Same thing with 3rd party reporting. You’re changing the expected cost of evasion by increasing the likelihood of an audit, penalties.

    • I’ll just point you to my response above. There is a raging debate going on about what exactly is Nudge vs. behavioural economics vs incentives and there is no real consensus. It depends on who you talk to/read and how you think about the underlying behaviour and choice architecture.

  3. […] Rather, the technique “will work on people who made a mistake, don’t understand the rule, lack information, follow the herd, misunderstand the risks, and so on,” Tedds has written. […]

  4. […] Rather, the technique “will work on people who made a mistake, don’t understand the rule, lack information, follow the herd, misunderstand the risks, and so on,” Tedds has written. […]

  5. […] Rather, the technique “will work on people who made a mistake, don’t understand the rule, lack information, follow the herd, misunderstand the risks, and so on,” Tedds has written. […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s