Richard H. Thaler: "HR departments are the ones that should be reading 'Nudge'"

In a Zoom interview, Thaler spoke about nudge versus sludge, why he doesn’t favour an opt-out system for organ donations and why the slippery slope argument that nudges can become shoves falls flat.

Nudge versus sludge. How would you categorise these examples: One, a stock broking company called Zerodha recently announced a fitness incentive for employees working from home. The idea is that we’ve all picked up some unhealthy habits (many of us have put on weight) during the last 18 months. So, employees set a 12-month fitness goal for themselves. They report their progress with respect to this goal each month. At the end, employees who meet or beat the goal get a month’s salary as bonus. Plus, they stand to win an additional Rs10 lakh in a lucky draw. Is this an example of nudging?

How is the reporting monitored?

This is self-reported.

Well, cheating aside. I would say this is nudging. And certainly well-intended. And it could work, on several grounds, there is research to suggest it could work. First of all, giving somebody a goal can help. And then keeping track can help. And then giving a reward can help. So all of those things can help.

On the other hand, these are among the most difficult habits to alter. So if you are asking me to predict how well it worked, I would not expect big gains. I don’t know anything about this (specific example). But I would guess a lot of people would just drop out. But the people who stay in, will improve.

What suggestion would you give to the company to improve adherence/success rate?

I think you could improve on the self-reports. For example, suppose you just have to take a photograph of the weighing scale. Admittedly that’s not foolproof. But at least it requires a little bit of work to make up the numbers. And the same with whatever exercise. If you have a smartphone, it would be very easy to add a monitoring feature and you could even have people opt into that. For example, my iPhone, if I go out for a walk, asks me, and it is always keeping track of my steps. So you can easily imagine linking that. And that would have the dual goal of sort of keeping people honest and allowing people an easy way of checking other goals.

What about the Rs 10 lakh lucky draw?

I don’t know. How many people are involved?

The company has over 1,000 employees. I don’t know how many will sign up to the programme or reach the end line.

It could help. People like lotteries. But I’ve been involved in a recent example using lotteries to try to encourage vaccine take-up, and it didn’t help. Now what we don’t know is to what extent that was because of the stage we are in. In the US, roughly two-thirds of the eligible people have been vaccinated. Which means that the remaining third are the most hesitant. And they may not be as nudgable. Roughly, the first third didn’t need any nudging because there was more demand than supply. In the middle period, the key was to make it easy - so, no appointments; go to the people in rural areas, and so forth. For the last third, sterner measures may be needed.

Second example: Vaccine registration in India requires people to sign up to the COWIN website, which requires a smart phone or computer, an Internet connection, and the ability to navigate the website. Is this sludge, especially for people in remote corners and the elderly?

That’s kind of an empirical question. Even the elderly may have children and grandchildren who are tech-savvy. You can easily imagine creating places where people can get help doing that. Even in a rural place, a village centre where they have a computer where people can go and have somebody enter the information. To the extent that it won’t be sludge for most of the people. And to the extent that there are some people for whom it is, then take some steps to reduce the sludge.

Is sludge also a way of nudging people towards certain behaviours?

Well. My preferred usage is that it (sludge) is always a pejorative. It’s like unnecessary friction. It is the case that friction can be helpful. So, for example, sometimes on the highway, if there’s going to be a toll that induces you to slow down, that’s helpful. You can think of it as helpful sludge.

No, of course not. No publisher wanted it. No, we were hoping that maybe our wives and children would read it. We never expected what happened next.

Why final edition? What prompted it?

One of the things I have been interested in for a long time - going back to graduate school - is what are the situations in which people eliminate options in order to prevent themselves from doing something stupid. Ulysses famously ties himself to the mast so that he wouldn’t drive the ship against the rocks; people who are trying to quit smoking don’t keep cigarettes in the house. 

Rewriting this book was a pandemic exercise. It’s an accident. Somebody at the publisher hadn’t noticed that the contract with the paperback publisher had expired 18 months earlier and they got around to telling us in April of 2020. And there wasn’t that much to do. (Once we started writing) Then we got carried away, and it turned into a lot of work. And I decided, OK, we’re not going to do this again. And we kept trying to find what title would convey that the book is substantially new: Nudge 2.0, Nudge remastered. We settled on the final edition to signal that it’s new and it’s done. 

What is the “while we are at it” bias you talk about in this book?

This is familiar to anyone with a home improvement project. You start with “I’m just going to remodel this one thing”. And then (you add something else) while you’re at it. And there was a lot of while-you’re-at-it (in this book).

The while-you’re-at-it and the low opportunity cost - there wasn’t anything else to do. But I am glad we did it. There were a bunch of things that were out of date. And there were things that readers kept misunderstanding. Eventually, if enough people misunderstand it, you have to stop thinking that the readers are dumb and conclude that the writers are bad. 

What kind of misunderstanding?

For example, our chapter on organ donation is completely rewritten to make sure that absolutely everyone understands what our view on the subject is, which is (that) changing the default isn’t enough. And if anything, the evidence suggests that it may hurt.

When countries announced that they had switched to so-called presumed consent, which is an opt-out system for organ donation, on Twitter I would get congratulatory notes: “Hey, your idea has been adopted.” In the book we say no, we are against this. It’s not a surprising mistake. When we started the book, we thought that is what we would say. But then we ended up doing the research and realised that it doesn’t work. This one (edition), I don’t think anyone can miss what our view is.

What are smart disclosures?

This may be the most important and most boring topic in the book, which is a little unfortunate. But the idea is that governments have thousands of rules and regulations around disclosures - what you have to tell consumers or investors. And if we think about the way in which those disclosures are made, they are not 21st century. They’re not even 20th century. They might as well be 1st century because they are just a document.

(For example,) the ingredients that are just listed on the side of a package of food. We urge that wherever possible, those disclosures be made available in a machine-readable form and the benefits of that are that it would create a new industry of what we call choice engines.

A good example of a choice engine is a travel website. If I wanted to fly from Chicago to Mumbai, 20 years ago, I would have had to call a travel agent. Now I go online and I can book it myself in 5 minutes. And I can be pretty sure I am getting the best price. So why shouldn’t that be true for other more complicated things like choosing a mortgage? Or finding a food (item) that doesn’t contain an ingredient to which one of your children is allergic? 

My mantra is make it easy. That’s my answer to all of life’s problems: Make doing the right thing easier.

It’s very easy to buy a book on Amazon. It’s easy to get around if you have GPS on your phone; you don’t get lost. We want people to get lost less often in their other endeavours.

If you want people to do something, make it fun. For example, I think mathematics education is quite ineffective. And it’s because most kids don’t see how it relates to anything in their life. Geometry or algebra just seems like some abstract thing. And there are so many ways that you could make it real to them in some way they would enjoy. Say, keeping track of their favourite cricket star or what have you.

I wrote a little oped a few years ago about make it fun. Somebody read it, and the office where he worked, you could get from one floor to the other only by going into an internal stairwell. And people find those off-putting. So they (the office) put music in the stairwell, and blackboards where people could write things. You had to swipe a card to get into the stairwell, so they could monitor who is going in, and the number of people using the stairs went up. 

So we could add some fun to that weight loss and exercise programme that you were talking about and that would also make it probably more effective.

Example 3: Next Episode button on Netflix that automatically starts the new episode, is that a nudge?

Well, for sure Netflix thinks so. It’s certainly (makes you) more likely to binge. Especially on shows where there’s a cliffhanger at the end of every episode. I think it would be nice if Netflix let you turn it off. That would be my advice to Reed Hastings - to let people opt out of that.

Could companies transpose this idea of automatically starting one thing after another, to nudge employees towards certain behaviours?

I think, yes. The part of business that I think is most ready for a revolution is human resource (HR) management. I think HR departments are the ones that should be reading Nudge. It’s more likely that the CFO is reading it, but I think it is at least as useful, maybe more useful, for the head of HR. Both in terms of benefits management and thinking about how you want to relate to your workforce, and how to make their lives better and make them productive.

What are the 2-3 things that you would recommend for readers to nudge themselves towards better behaviours?

It’s very individual-specific. And it kind of depends on the job and what your weaknesses are. So, some introspection can be helpful. 

I think turning off all notifications except from your calendar is a good thing. Especially when most of the news in the world is bad, notifications just make you unhappy and are distracting. Especially email notifications - I don’t need those.

But people have been making lists for as long as there have been places to write things down. And how you do that in a way that’s good for you again is very individual-specific. 

The last thing is that the things you dread doing, maybe figure out a way to make them more fun or to give yourself some reward if you get it done. 

What are the best and worst feedbacks you’ve got for the book in the 13 years since it was published?

I would say the best feedback is that 400 governments around the world, including India, have created so-called nudge units, which is quite remarkable.

The feedback I find the most annoying is something we talk about in the book: the slippery slope argument, which is when people say, “Yeah, when you start nudging then the next thing you are going to do is start shoving.” And I find that argument unconvincing. There’s never any evidence of any slope in slippery slope arguments. Show me why automatically enrolling people into a pension plan has led to something bad, and then I’ll start worrying about a slope. I think those slopes are perfectly flat.

This is the first in a series of interviews with authors of books that have changed our world and worldview.