Week 6: Introduction to Unit 5

Week 6: Introduction to Unit 5

“Introduction to Unit 5 … Nudging 2.0 … The Nudgers Toolkit … A Practitioner’s Guide to Nudging … Debate … Skills and Knowledge”
(Source URL)

Summaries

  • Unit 5 > 5.0 Introduction to Unit 5 > 5.0.2 Unit 4 Debate Debrief
  • Unit 5 > 5.1 Nudging 2.0 > 5.1.1 Nudge: Five Years Later
  • Unit 5 > 5.1 Nudging 2.0 > 5.1.5 Nudging: Getting Started
  • Unit 5 > 5.2 The Nudgers Toolkit > 5.2.3 Debiasing and Rebiasing
  • 5.3.1 Nudging: An Organizing Framework
  • Unit 5 > 5.3 A Practitioner's Guide to Nudging > 5.3.7 A Guide: Design and Iterate
  • Unit 5 > 5.3 A Practitioner's Guide to Nudging > 5.3.7 A Guide: Design and Iterate
  • Unit 5 > 5.4 Debate > 5.5.1 Debate 5

Unit 5 > 5.0 Introduction to Unit 5 > 5.0.2 Unit 4 Debate Debrief

  • DILIP SOMAN: Should we focus as a field on developing a Grand Unified Theory, or should we just focus on developing what I call a culture of experimentation.
  • John Lynch, professor at the University of Colorado, basically sort of acknowledged the difficulty in coming up with a Grand Theory.
  • “If we don’t have a theory and we’re just experimenting, then we don’t want to do what we are supposed to do as academics. The goal of academia is to try and understand, theoretically, what drives behavior.” The other interesting point he made was having a Theory guides us towards practice.
  • Rory Sutherland from Ogilvy in the United Kingdom again sort of presented arguments for both to having a theory and not.
  • He said, “Now, just because we will never have a perfect theory doesn’t mean that we should stop trying.” So the desire to push for a good theory is still something that echoed loudly through his comments, but he did sort of seem to make the point that you, at this point in time, need to think through a culture of experimentation and of empiricism, because that’s what’s going to give us the data that will eventually help us with building a Grand Theory.
  • Everything matters, as we’ve said so many times, and so to have a theory of everything is a challenge.
  • The best way we prepare ourselves to develop that theory is the idea of experimentation.
  • Why do we need a Grand Theory? And not in his words, but one of the points he was getting to is that if our goal is to create, not just to help make the world a better place, and those not just are going to affect decision-making that is tied to a given context, should we just be contextual and study empirically what’s happening in that particular context.
  • “Maybe there isn’t a Grand Theory. Maybe we don’t need a Grand Theory, and maybe it’s just an illusion that there a Grand Theory.” And maybe Chris is right.
  • We had my colleague Professor Chen-Bo Zhong from the Rotman School at the University of Toronto, who talked about this [INAUDIBLE] theory and empiricism, or theory and experimentation, as a topic that’s being discussed and debated hotly not just in this forum here, but in the field of psychology more generally.
  • Because unlike the previous debates where we saw our panelists taking sides, here most people were sort of somewhere in the middle, saying that yeah, we want a theory, but it’s not clear that we’re going to get that theory any soon.
  • There are obviously going to be universities and labs that specialize on theory development that will focus on the theory aspect.
  • We don’t need to be in a world where everybody’s developing theory or everybody’s developing empirical skills.
  • I think that’s really the way we think about knowledge creation and knowledge dissemination, I think it’s the role of the university to develop the insights.
  • I agree with John Lynch who said that if we didn’t do that, if we didn’t develop theory, we wouldn’t be scientists.
  • We also need to then translate that theory into practical, actionable insights and make sure we build that culture of experimentation at the crossroads, where we are actually doing evidence-based policy.
  • We now have a menu out there where we know what other people have tried, what’s worked, what’s not, and having that evidence at hand, I think, would make a big difference.
  • Lastly, if that data is made available to people in academia, people whose business it is to create the theory, I think that makes things easier for us, in terms of theory development.
  • Remember last time, we talked about the notion of discovery, was this theory testing.
  • If we had access to more data on the discovery phase, I think it becomes easier to let academics focus on theory testing.

Unit 5 > 5.1 Nudging 2.0 > 5.1.1 Nudge: Five Years Later

  • DILIP SOMAN: Now that you know a lot of the basic principles of behavioral economics and you’ve studied a little bit about how to design experiments, you might want to think about topics with which you could actually apply some of these skills.
  • There’s a story on why people don’t save as much as we think they should.
  • There are stories on why people don’t donate to charities.
  • There are stories about the fact that people generate too much trash, including items that they could put in the recycling bin.
  • A lot of these organizations had an economic advisor.
  • Perhaps the one book that changed the fact that companies and organizations like governments started looking towards the behavioral sciences was, of course, the book that you’ve talked a lot about, Nudge by Dick Thaler and Cass Sunstein.
  • Let’s start off by reminding ourselves about the four different strategies of behavioral change that we’ve talked about in week one.
  • We said initially that one common strategy for governments in which to influence behavior change is simply to impose restrictions or constraints on what people can choose.
  • The fourth strategy, of course, that we talked about, was the nudging or the choice architecture strategy where you simply make it easy for people to accomplish the desired choice.
  • Now, we’ve got to be careful about making sure that we do not fall into the trap of thinking that these four strategies are mutually exclusive, that we should simply be nudging while ignoring the other strategies.
  • This is not a story about nudging versus the other three strategies.
  • It is a story about how we can best utilize the right combination of these strategies in behavior change.
  • Plastic bags, starting in 2009, were going to cost $0.05 per bag, with the idea that they would discourage people from asking for plastic bags, and as a result, build of culture of using reusable cloth bags when they went ahead and did their grocery shopping.
  • Is this an economic strategy to change behavior, or is there more to it? On the face of it, $0.05 seems like a punishment, a tax for following the undesirable behavior.
  • We’ve done a number of experiments in India we were trying to get people in the rural parts of the country to consume financial advice.
  • Our initial response based on economic theory was if you give advice for free, it is more likely to be consumed.
  • We actually found that if you had people who make small payments for advice, they were much more likely to follow through on that advice.
  • The consistent finding we found was that people who had paid 5 or 10 rupees were much more likely to follow through on the advice as compared to people who had paid nothing.
  • The point I want to make here is that while an element of financial incentives, or economic incentives have been used to get people to change behavior, we could strengthen that particular intervention by thinking through a behavioral contribution to that.
  • Now, let’s think about a few other examples where there are differences between an economic approach to changing behavior versus a behavioral approach to changing behavior.
  • Why does that happen? It happens in situations where households typically pay their energy bills by direct debit from their bank accounts, so a lot of people actually don’t see the increased price that they pay.
  • On the other hand, think about a simple strategy where you look at people who make a payment for certain consumption and fail to make the consumption.
  • Let’s replace ski tickets with something that you want people to consume.
  • Most people who are entitled to a check up do not get it.
  • Here’s an example of a situation where you have an economic component, the fact that people pay a price, but the strength of that economic component has been increased by simply changing a behavioral factor, i.e. the way in which the tickets are presented.
  • So in other words, you could think about the effectiveness or one specific kind of strategy, be it a nudging strategy, be it an economic strategy, be it a persuasion strategy, that is affected or changed by the presence of a second strategy.
  • The presence of an economic strategy can be strengthened by simply crossing it with a second behavior, or a nudging approach.

Unit 5 > 5.1 Nudging 2.0 > 5.1.5 Nudging: Getting Started

  • How do we help people who clearly have a deficiency in making optimal decisions to make better choices.
  • When I thought about how to best help myself get around from place to place, I realize that there are two kinds of strategies that we typically use to help people with physical deficiencies.
  • We do that by providing people with a crutch, with an assistive device, with a wheelchair, or a walker.
  • The second thing we need to do is to make sure that the context, or the environment in which they move around, is a safe environment, and it is easy for people to walk around or move around in that environment.
  • If you’ve got that same analogy to cognitive deficiency, we’ll again see that there are two kinds of ways in which we can help people make better decisions.
  • Now, before we actually get into the design of nudges, there are a few things that we need to keep in mind.
  • Just like any plumber does before he or she starts a new job, or any carpenter, or any mason, you want to make sure you understand exactly how many hammers, and spanners, and wrenches you have in your disposal.
  • What are these tools? These tools are twofold- an understanding of the common mistakes that people make in their decision making process and an understanding of things we could do to correct those mistakes.
  • In the next module, we’ll actually review a lot of these tools with a view to understanding exactly what are the different kinds of strategies we could use to help people make better choices.
  • Second, before you start designing you nudge intervention, you want to make sure you know exactly what behavior you want to change.
  • I want people to eat more healthy food.
  • Or you could say something like, I want for people’s savings rates to go up.
  • How do I get people to save more money.
  • If I want people to increase their recycling behaviors, there are a few specific things that I need to do.
  • So when do you think about a nudge intervention, you want to make sure you’re very precise in terms of what, exactly, you want to happen as a result of the nudge.
  • Do I want people that currently have a bank account to save more? Or do I want more people to open bank accounts? These are two very, very different outcomes.
  • Let’s imagine I have a problem where I run a call center where people call me for help on how to save money.
  • There are dropped calls because people call in, they wait for a period of time.
  • I could think about a number of interesting behavioral strategies to keep those people in the line.
  • I could give them information about how many people are ahead of them in the call.
  • So what do you really need to do is to make sure that your processes for delivering the service are as good as they can be before you nudge people into doing what you want them to do.
  • Once you come up with a nudge, once you come up with an intervention, you need to be able to assign people to a control condition, some other people to the treatment condition, and then to be able to test for the effectiveness of the nudge.
  • Fifth, ask yourself the question, is nudging indeed the best option here? Are there other alternatives? Is it the case that, perhaps, an economic incentive is better than a nudging incentive.
  • We need to think about having then collaborate to make sure that we end up with the kinds of outcomes that we want.

Unit 5 > 5.2 The Nudgers Toolkit > 5.2.3 Debiasing and Rebiasing

  • DILIP SOMAN: We’ve all heard the term “Bias” before.
  • What exactly is a bias? A bias is any systematic deviation from a response or a decision that you would expect a rational decision-maker to perform.
  • Think about a simple bias called overconfidence, the idea that people believe that they know more than they actually do.
  • This is a bias, because people on an average always believe that they know more than they do.
  • The finding that most people believe to they know more than they do is a bias.
  • What I’ve done is I’ve identified the cause of the bias, and I have fixed the cause so that the bias no longer happens.
  • Again, keep in mind the bias here is the fact that the door that is supposed to be in a closed position stays open.
  • What I have done here- the bias still stays because there is still an error in the spring.
  • A de-bias is a situation where you identify the cause of the bias, you fix it, and the bias goes away.
  • A re-bias is a situation where you have a bias, you impose a second bias that acts in the opposite direction, and now the door stays shut, people behave as if they were not biased.
  • How do we make sure that people are not overconfident? It turns out that there are two common strategies for reducing overconfidence.
  • One strategy is simply getting people to write down two or three reasons why they think that they would be wrong with whatever judgement or prediction that they have made.
  • Simply getting people to think about the what if scenario reduces overconfidence.
  • Why? Because the mechanism for overconfidence is the fact that people only look at information or reasons that support their judgment, not for reasons that oppose it.
  • Now, let’s look at a second example, a self-control failure, the fact that people want to save more for the future, but don’t.
  • It turns out people who participate in the Smart Save More Tomorrow program actually save more, but they would become any better at self-control.
  • You pick two other biases, the fact that people forget, the fact that they have loss aversion, to cancel out the effects of the first bias.
  • If you motivate people to come up with the right answer, they will.
  • How do we motivate people? You could give them financial incentives.
  • Once there is some external stimulus that gets people to be more motivated to be accurate, they are more likely to process information more effortfully and to draw more information into their decision.
  • The second set of strategies that we’ve used to reduce biases are cognitive strategies, getting people to think differently or training people or generating templates that they could use which provide them with the right kind of data, or forcing them to think through model-based approaches for making decisions.
  • Finally, technological strategies include the use of decision supports or aids or online databases to inform people with the right data to make the best choices.

5.3.1 Nudging: An Organizing Framework

  • DILIP SOMAN: Now that we know what choice architecture is and what a Nudge is, let’s think for a moment about a framework to organize all of the different kinds of Nudges that we have come across.
  • The first dimension is one in which you can categorize Nudges just whether they boost self-control or they activate a desired behavior.
  • In the boosting self-control type of Nudge, we’re typically faced with a situation where the consumer or the individual knows what they need to do, but simply need a Nudge to stay committed to that particular behavior.
  • Whereas, in the activation kind of Nudge, the individual is perhaps not even aware of the right outcome, or the right behavior, and so they need to be steered towards that desired behavior.
  • These are the kinds individuals who might want to impose a Nudge on themselves to make sure that they follow through on the chosen action.
  • Externally-imposed Nudges, on the other hand, might be Nudges imposed by some other party.
  • We think about a mindful Nudge as one that increases cognition and gets people to think more a little bit about their decision at hand.
  • Mindless Nudges, on the other hand, go the opposite way.
  • Mindless Nudges are effective because the individual typically currently faces too much of a cognitive overload. And so we simply try and make it easy for them to move towards the final outcome by making it easy and getting them to think less about the decision.
  • Finally, another way of classifying Nudges is by what we call encouraging Nudges versus discouraging Nudges.
  • In which case, we want to come up with a Nudge that discourages that particular competing activity.
  • We have Nudges that activate versus those that boost self-control.
  • A simple Nudge where an individual can channel money into a separate bank account to reduce the likelihood that it is spent is a mindless Nudge because it makes the decision easy by moving the money into a separate part.
  • By putting together all of these different dimensions into a unifying framework, it helps us develop a ground for a process for designing Nudges.
  • Keep in mind that different programs could employ multiple Nudges.

Unit 5 > 5.3 A Practitioner’s Guide to Nudging > 5.3.7 A Guide: Design and Iterate

  • DILIP SOMAN: With the view to helping people understand the process and develop a framework for how to develop nudges, my group here at the Rotman School of Management and I have developed a guide that we call the Practitioner’s Guide to Nudging.
  • Mapping exactly what the sequence of decisions looks like and what there key considerations are when they make those decisions is perhaps one of the most important things to do when we start working on a nudging strategy.
  • As part of the PDF version of the guide I just talked about, an appendix in that guide gives you a simple form with a checklist for the kinds of things you need to make notes on when you ordered that decision making process.
  • Or are there multiple stages in that decision? Are there many options from which the person needs to choose? Or are there simply two? Are there simple choices? Or do people need to make judgments on a continuous case? For example, a simple choice would be do I choose Fund A versus Fund B? A more complex judgment would be how many dollars should I put into my retirement account each month, and should that number change as a function of time? The other set of factors to look at include the information that people need and use in making that decision.
  • What is the mindset of the individual when they actually make that decision? Are they implementation focused? Or are they in a planning focus? Are they myopic? Or are they employing a long term window? Where do they make the choice? Is it made in the store where they have time constraints? Or are they making the choice at home, sitting alone in the comfort of their living room? Is it a choice that is made in the presence of somebody else? Or was it a choice that they make independently? And finally, again, environmental and social factors need to be audited and documented.
  • Is the decision going to be publicly visible? Or is it something that the person can make on their own and nobody else will know about their choice? Is it a product category, which is conspicuously consumed? Or is it something that is silent and hidden? Is the choice made in a group setting? Is it made in an individual setting? These and other factors are part of this checklist and are always a great place to start when you design a nudge, because it gives you all the information that you need to know when you start thinking through what the right intervention should be.
  • Without people realizing that this is something they need to attend to, the rest of the stages simply don’t matter.
  • Now, once you’ve got the chart, and once you start thinking about what the right kind of nudge should be to affect these different stages in the decision making process, there are four kinds of questions that you need to ask yourself.
  • Is the decision maker aware of the correct decision and are they simply unable to execute it? Or do they not even know what the right choice is? Think of a simple example.
  • We’ve talked about the fact that everybody knows they need to save more, but they just don’t get down to doing it.
  • People know what they need to do, but they simply fail on the execution.
  • Is the individual aware, or do you need to make them aware? Do you need to steer them towards a particular outcome? Second question- is the individual motivated to impose a nudge on themselves? Think about the Save More Tomorrow program.
  • If people have a self control problem, they’re aware of it, and they’re motivated to solve it, then in fact, you can apply certain kinds of nudges.
  • Another important question- are people making the wrong choice because of inertia, laziness? Or is something else getting in the way? In other words, do I need to provoke and excite people, or nudge them, into doing something? Or do I want to nudge them into not doing the opposite thing? And again, on the face of it, these seem like complementary processes.
  • Which is the box at which there is the greatest attrition or drop off that happens? In other words, how many people that are in stage one actually go to stage two? How many people that express interest or motivation or recognition that they need to save more actually go out and open a bank account? How many of those actually collect the information needed to figure out exactly what they need to invest in and so on and so forth? So if you actually map that out in your chart, you can identify the big bottlenecks.
  • Now, what actually causes these bottlenecks? You need to think through the underlying psychological principles that cause bottlenecks.
  • What’s the biggest bottleneck at the need recognition stage? Why do so few you, as a percent, actually begin to feel the need to plan for their retirement? What part of it is simple psychological bias? The distance bias- it’s too far away.

Unit 5 > 5.3 A Practitioner’s Guide to Nudging > 5.3.7 A Guide: Design and Iterate

  • DILIP SOMAN: Remember again that there are four steps to the process of designing a nudge.
  • Mapping the context, selecting a nudge, identifying a specific lever to execute the nudge, and finally, the process of designing and iterating through experimentation.
  • We’ve talked about the first two, and at the end of those two stages, what you’ve now got is a diagram that represents decision making over time, as well as a set of nudges that you could potentially implement at each stage.
  • It’s now time to think about specific ways in which we can implement those nudging strategies.
  • What you see here is a list of common levers that we use in implementing some of those nudging strategies.
  • So that’s a very powerful, simple nudging strategy.
  • The default for the nudge book is to not read it, but you’ve got to do something to go out and get it.
  • Could we somehow tinker around with the defaults to steer people along that decision path? Framing of information is another big potential nudging strategy.
  • We’ve spoken about that a lot in this course, and then finally, another common nudging implementation strategy is the way in which you change the salience of the payment that people often need to make.
  • So what we have here is a list of several strategies that are commonly used to implement some of the nudges that we talked about.
  • Now again, at the end of this process we’ve got perhaps a list of three, four, five different nudging interventions, different nudges we could use to smoothen the flow of this decision making process.
  • Principle number one, nudges that attack a bottleneck in the upstream part of any decision making process is generally preferred to one that works downstream.
  • Principle number two, think about nudges that have the highest possible reach.
  • Think about interventions that deliver the nudge to the most maximum number of people that you possibly can, and choose the one that reaches the highest.
  • The simpler you make the nudge and the quicker you make it, the more likely are you to see results sooner.
  • Finally, think through the long term effectiveness of the nudge.
  • Some nudges work really well in the short term, but once you un-nudge people, and once you change the context back, then in fact, the effectiveness drops off.
  • Pick nudges that have higher, longer term effectiveness than ones that have lower long term effectiveness.
  • If it’s a nudge that involves the manipulation of information, you need to think through how best to communicate that, what’s your vision strategy, what is your media strategy? You then need to design an experiment where you can effectively identify the cause and the effect, the fact that your selected nudge actually results in the outcome that you want it to.
  • Are you going to use a simple two condition design? Are you going to use a fully crossed design, or are you going to use a before and after design? And then finally, once you get data back from your experiment, it potentially gives you room to iterate and update your nudge.
  • Perhaps you’ve done an experiment in which you’ve tried to pit three nudges against each other to see which one of them creates the biggest impact on the outcome.
  • Let me end this module by talking about three mantras for designing nudges and working with a nudging initiative.
  • There is no central database where nudges all over the world can go and look for interventions that other people have done that have worked.
  • If only there were a common database that in fact it would make it easier for us, and that we would not need to reinvent the wheel every single time we wanted to come up with a new nudging intervention.

Unit 5 > 5.4 Debate > 5.5.1 Debate 5

  • ANNAMARIA LUSARDI: I am Anna Lusardi, and I am the academic director of the global center for financial literacy in Washington, DC. I believe that financial literacy is an essential skill that is necessary to participate in today’s economy and society.
  • As in the past, one would not be able to participate in society without being literate, meaning without being able to read and write, today, people need to be financially literate to participate in today’s society.
  • I’m very happy and proud to be the chair of the financial literacy expert for the OECD Program for International Student Assessment.
  • In 2012, a new module measuring financial literacy has been added to PISA.
  • So we will have, soon, data and evidence measuring financial literacy among 15-year-olds across countries.
  • I really hope that this will provide evidence and motivation for adding financial literacy in high school.
  • DILIP SOMAN: So actually, there’s a very good case to be made for why nudging is perhaps the most critical component of helping people help themselves.
  • In the world of financial literacy, for example, we know that people who don’t save enough know they need to be saving more.
  • I think people, in all these domains, know what the right thing.
  • I think that’s where nudging can be tremendously helpful.
  • We’ve seen similar other work by people like Eldar Shafir and Sendhil Mullainathan where, again, it’s not so much the education or giving people be the knowledge that they need to open accounts, or need to set up retirement funds, it’s also nudging them to actually go ahead and do it.
  • I think, can help, but without a nudge, there’s very little that that education can do for you.
  • Hey, do you know a lot of people who complain about fast food, and kind of think it’s “Evil”? Yeah, I think we all know a lot of people like that.
  • What if we were to design a restaurant that could profitably nudge people to eat a little bit better, and eat a little bit less, at the same time without chasing away the people who want a bacon cheeseburger with American cheese.
  • Now the key to behavioral economics and nudging is actually giving companies, and giving ourselves, the suggestions about what can profitably help them help us eat better.
  • That’s what the challenge, for us, is to do- is to come up with ways and say, here’s how you can help me eat better.
  • There’s not a lot of things that restaurants won’t do, and companies, won’t do, and grocery stores won’t do if they think they can make a lot of money.

Return to Summaries List.

(image source)

Leave a Reply

Your email address will not be published. Required fields are marked *