In the previous post I outlined the basic premise of nudging, and examplified it. The examples are well-known throughout academic research and have gathered a lot of popularity as such. Simple changes lead to big outcomes. Nudging works.
However successful nudge has proven to be, it has proven to be successful in very specific situations. This is something the authors, both Thaler and Sunstein, warn about as well. Nudging is not the “be all end all” of behavioural change interventions. They have dubbed nudging gone wrong as “sludge.” And the concept of sludging is what I’ll dive into this post.
To nudge or not to nudge? That is the question. As said before, nudging is the changing of choice architecture. We change the architecture to make it easier for people to make the most desirable (most beneficial) choice. We remove obstacles, we change defaults, we emphasize, we inform, we compare etc. etc. All of these interventions have to be based on insights we already have from the fields of the behavioural sciences. If our intervention is not based on anything grounded in (behavioural) science, would it be that surprising if it backfired?
What does a nudge gone wrong look like? In the United States, there is a program called the earned income tax credit. This program is supposed to encourage work and transfer income to the working poor. The Internal Revenue Service (IRS) has all the information necessary to make adjustments for credit claims by any eligible taxpayer who files a tax return. So far so good. The issue is, however, that the rules require people to fill out a form that many eligible taxpayers do not manage to complete. So rather than using information that the IRS actually has, the decision and the actions are handed back to the people. Now this is not just a case of laziness. Many of these forms are hardly comprehensible. You’d need a degree in Tax Law or Economics to understand what exactly is being asked. So due to these two reasons many taxpayers end up depriving themselves of the subsidy that Congress intended they receive. This is a pretty good example of a pretty bad nudge. Especially as the information necessary is available to the IRS.
Another example focusses on voting. As voter participation has been on the decline for years, this is an important topic to focus on. Keeping behavioural insights in mind, you can quite easily nudge participation by automatically registering anyone who applies for a driver’s license. They have the right age, they are in the system, and as such eligible to vote. This sounds great. This is not being done. Again, when looking at the US, we see that a massive sludge has occurred in the state of Ohio. This state has removed people who have not voted recently and those who have not responded to a postcard prompt (who does physical mail anymore these days anyway...?) from the eligible voters list. Apparently, this sludge is being defended on the grounds of voter fraud. But I’ll bet you five quid a cow can see through that argument, especially as it is a fact that people who intentionally vote illegally are rare.
So the sludges above are examples from the public sector not really understanding what damage they are doing. Ironically, the public sector does not necessarily benefit too much from this (maybe lobbyists and the IRS do, to some extent, but still). Sludges get much more insidious when they move into the domain of the private sector, in which we are talking money, sales and profit. An example of this is rebates, or “buying baits.” Firms often offer a rebate to customers if they buy a certain product. In line with behavioural science, we know that most consumers are lazy, or at least suffer inertia. So, what do these companies do? They require customers to take action to obtain the rebate. Such as mailing in a form or a copy of the receipt, the bar code on the packaging, etc. Unsurprisingly, redemption rates for rebates tend to be low. For the company this is a successful nudge: increased sales with hardly any rebates having to be given out. However, the lure of the rebate has tricked the customer into buying a product they otherwise might not even have considered buying, without ever getting the benefit of the rebate. As such, this was the worst possible decision outcome for them. It’s a total sludge.
Sludges can take different forms. It can discourage behavior that is in a person’s best interest such as claiming tax refunds, but it can also encourage self-defeating behavior such as buying a product and never claiming the rebate. Nudging is supposed to be done for good, in the words of Richard Thaler. However, there are plenty of nudges done for bad. Holding on to that thought, a criticism that we haven't even considered, which I won't do in this article is the simple distinction between a good or a bad nudge. Or rather, to nudge someone to choose whatever action/option is in your best interest. Who determines what is in my best interest? This is a question of what (political) ideology you support. Nudging has dubbed itself as libertarian paternalism. You might be a fan of that. You might not be... Regardless of political or governance preference, if it seems that you are being pushed into a very specific decision outcome, by its choice architecture (again, not by force), ask yourself whether this is truly the best decision for you. Are you going to claim that rebate? If not, maybe just leave the product in the store. You don’t need to be sludged.
References Thaler, R. H. (2018). Nudge, not sludge. Science (New York, NY), 361(6401), 431.