Video: A behavioural lens on debt-repayment

Dilip Soman: Wow. So you've given us the difficult task of ending the day. And we know that things that end well are good things. Right? So it's up to me and my panel to make sure we end today well. I've heard the saying that we save the best for last, so thank you for doing that, Jane, and everybody responsible for this.

Look, I've got three jobs to do. Job number one is I want to talk about an academic concept called the academic graveyard. So those of you who are academics in the – in the room, you might or might not know what a graveyard is. So I'm going to tell you what the graveyard is. I'm going to actually pull out some things from my graveyard and share some ideas from there. I want to set up sort of the three speakers. So I want to sort of serve as a preface to what they're going to say, talk a little bit about the problem of debt and what a behavioural lens on debt repayment actually means. And then I want to get all of you to not forget the lovely online voting poll thing that we did this morning, Slido.com.

So can we go to Slido? Every time I have been asked to speak about a problem or to analyze a problem, I guess I have three responses or three sets of questions. Firstly, how big is the problem? Why is it a problem at all? Can it be avoided? And then how is it being solved and how should it be solved? So here's one question that I'm curious to hear your responses to. For every dollar of household disposable income, how much debt does the average Canadian household owe? Is it about 50 cents? About a buck?More than $1.50? Or I have no clue? If you can go to Slido.com, let's take a look at what our numbers are saying at this point in time. More than $1.50. Ninety-six percent of you think it is more than $1.50. Guess what: you're correct.

We talked about the Bank of Canada study that came on line in August. The Bank of Canada study estimated a total debt of about two trillion, which works out to about $1.70 per dollar of disposable income. That's a lot of debt. Right? It's the highest it's ever been. It is pretty high compared to many other countries, though it's not as high as some other countries, but it is still significantly – and that's the average. Right? That's the average Canadian family, $1.70, which means there's about 20 percent of Canadian families that do not have any debt. You can only imagine how many debts some families have. So for example, eight percent of families owe 350 percent of their income as debt. Right? So that's the kind of numbers we're looking at.

The average amount of credit card debt in Canada per household is about $2,700, give or take. The average consumer debt, non-mortgage, is about $20,000 – or $21,000. Right? So is it a big problem? Yes, it is a big problem. Right? So let's go away from Slido for a minute and – and talk a little bit more about the big picture here.

So what I'd like to do is to do the following. So it started with the claim– and somebody had a question about this as well. Rather than focusing on people who are already in the hole, why don't we think about developing interventions to prevent people from falling into the hole in the first place? Why don't we just work towards preventing debt? Right? We obviously will help people who are already in debt, and I'm going to make the proposition that that's difficult. It's not impossible; it's difficult. And it's difficult for two reasons. The first has to do with this lovely panel we had right after lunch on scarcity. There is a rich literature now developing in that area. Jiaying, you're certainly one of the big proponents there. But if you haven't read an amazing book called Scarcity by Sendhil Mullainathan and Eldar Shafir, you should. Alright? And – and in the book, Sendhil and Eldar talk about the psychological burden of scarcity.

So think about a family who spends as much as they earn. You would think they don't need debt. Right? Wrong. Right? And the answer is wrong because there are often constraints that the real world poses in terms of when you need to make payments. So abstract case: if I earn all of my dollars on the first of each month, but my bills are due on the 27th, I'm going to need debt. It's just got to happen, right? And so the question really isn't should I educate these people about debt. I mean, a lot of people who are at the lower rungs of the income pyramid know that debt isn't a good thing. They know they shouldn't get into it. But they just can't– they need it. Right? And so the question really isn't should we teach them about avoiding debt; it is how do we help, either by changing the context in which they live, or by other institutional arrangements. That's one friction. Right?

But then, as you look at people – and Caroline talked about the fact that even people high up in the income pyramid sometimes feel scarcity, there's a lot of psychological reasons for why people are in debt. So impulsive spending. You know, as payments become more and more frictionless, it is becoming easier to spend. Those of us who are in the behavioural science, behavioural economics community obviously know about the concept of nudge. But there's also something called sludge, which is anti-nudge, the opposite of nudge. Right? And in an effort to make payments sludge-free, I'd make the argument we are making it too easy for people to spend. People spend without thinking. Right? And so at some level there is a non-zero optimal for sludge. We must kind of think about what we can do to get people to think before they spend. So there's – there's a lot of psychological reasons for why people get into it. So can it be avoided as a problem? Well, possibly, but I really don't see that happening. I think debt will stay with us. The concept of taking debt is going to stay with us, so we do need to think a little bit about how we help people get out of debt.

The behavioural lens. So what do we mean by a behavioural lens? My fellow panellists are all amazing behavioural scientists. But what does it mean to be behavioural when we think about solutions? Well, let's keep aside normative behaviour. Let's focus on what people actually do and develop solutions that are consistent with actual behaviour. That's the thrust of what these three folks are going to talk about. So for example, in – this is the most complex equation I'll show you. Alright? The claim I make is that the utility that people get from psychological factors is oftentimes greater than the utility that they get from money. Right? They will do things that are easier or that are psychologically more pleasing, that make them feel good, over things that save them money. And I think that's an important thing to keep in mind.

So those of you who've heard me talk about this – “econs”: rational people; “humans”: irrational people – the room is full of humans, the econs aren't here. They live on the pages of economics textbooks. They have no place in society, but yet we design financial systems for econs, knowing that it's humans that are going to use them. And so at some level we need to think about bigger questions about how we design financial products and services.

OK, so the academic graveyard. Right? Projects that should become published papers but don't. Alright? And the reason you know that you are a senior academic is when your graveyard is bigger than your front lawn. Right? So I now have a prosperous graveyard. I have lots of projects that should be published but aren't– oftentimes for good reasons, some not for good reasons. And I want to pull out a project from my graveyard to kind of highlight the difference between humans and econs.

So think about a credit card statement. You're going to get a credit card statement. Today most credit card statements will say the following: you owe 3000, minimum payment ten bucks. Alright? And under those circumstances, we find the distribution of payments looks something like this. Most people either pay ten or 3000. Nobody pays anything in the middle. Right? And the reason for that is, when you think about people who have credit card debt, well, they're obviously struggling with other things in life as well. There's cognitive deficiencies along the lines that Jiaying talked about. They don't go about thinking about, you know, how much do you think I can afford. They're going to say can I afford ten? Yeah. Can I afford 3000? No. Right? And so they'll either go to ten or to 3000, right? So this is research by Keys and Wang. This was done by the CFPB [Consumer Financial Protection Bureau] a couple of years back.

And so I and a colleague developed a couple of hypotheses. I'm just going to show you the hypotheses, but I will simply say, because I only have 36 seconds left, that we do have the data to support that. We tried to do a field study, but unfortunately the partner we worked with in the United States failed to randomize the intervention, therefore we have a paper that belongs in the graveyard and not in an academic journal.

But here's the hypotheses, right? Suppose, instead of offering people zero, or a minimum of ten, but you owe 3000, we give them four options: minimum ten, but you could pay 500; you could pay a thousand, but you owe 3000. Right? And they actually find that simply giving people more options increases the likelihood that they pay back more. Why? Because if I've got a human with limited attention, who's looking at the statement and saying can I afford 3000, no, I now give them a second target, which is higher than the minimum payment. Right? And so we tried a bunch of interventions in the lab. They worked really well. We took this in the field, but, like I say, we had problems with randomization, and therefore that didn't quite work out.

So that's a classic illustration of the human versus econ story. Econs wouldn't do this. They wouldn't show this pattern of behaviour. They'd calculate exactly how much they can afford and pay that. Humans aren't doing that. They're just answering a simpler question. Right? So that's the kind of dichotomy that we're going to talk about: humans versus econs.

And so the panel that I have, starting from my right, Nicole Robitaille, who's a professor at Queen's. Welcome back, Nicole, because Nicole got her PhD here. Rick Scott at the University of Michigan, Keri Kettle at the University of Manitoba. I'm not going to give you their bios because they're in your binders. As the last session, we'll do something different, which is we'll start bottom up. And so we'll start with Keri, then Rick, and Nicole. They'll each take ten minutes, and then we will open up for questions and discussion. So Keri Kettle.