Unit 1 > 1.2 Behavioural Economics and Choice Architecture > 1.2.1 What is Behavioural Economics
- Human beings are people like you and me- very different.
- Human beings sometimes do things because other people would like them to do things.
- Human beings typically form judgments or make choices based on the context around them, as opposed to Econs, who don’t care about context.
- Perhaps the best way to illustrate this point is to talk about a product category that’s one of my favorites- coffee.
- Most coffee shops all over the world will sell you coffee in one of three sizes- small, medium, or large.
- That’s probably going to come as no surprise to you if I tell you that the most popular size of coffee anywhere in the world is the medium size of coffee.
- It doesn’t matter how much coffee is in that medium cup of coffee.
- In this scenario, this is probably going to be the most popular cup of coffee.
- What happens if I remove the largest size, and replaced that with an even smaller size? It turns out now the new medium cup of coffee will be the most popular cup of coffee.
- A number of years back when I lived in Hong Kong, I and a bunch of my students did a study in a coffee shop which was in an office building.
- We asked people why they picked the medium cup of coffee, and you typically heard something like, the larger one has too much, the little one has too little, the one in the middle has the right amount of coffee.
- What we then did was we actually increased the size of every cup of coffee by two ounces- just what I showed you before- and we found that the new medium now became the most popular size.
- Now it turns out that this works not just for coffee, but for a whole number of other product categories.
- As we said before, humans are extremely prone to the role of context in decision-making.
- A second example of the role of context in decision-making comes from a completely different domain- the domain of organ donations.
- Simple question- why are organ donation consent rates really low in countries like Canada and the United States? Why are they extremely high in other countries, like France and Austria? It turns out there are a number of differences between countries like Canada on the one hand, and Austria on the other hand, but they don’t seem to explain the differences in organ donation rates.
- In Canada, organ donation rates are about 2.5%. In Austria, they’re close to 99%, and that’s a big difference.
- The answer as to why there is a big difference in donation rates comes from the process that people need to engage in order to donate their organs.
- In Canada, for example, if you want to donate organs, you have to go to the Department of Motor Vehicles, ask for a form, fill it out, send it in.
- You then get a web code that you go online, access, and register, and then you are an organ donor.
- In Austria, it is assumed you will donate your organs.
- If you want to not donate your organs, you would go to the same Department of Motor Vehicles, ask for a form, fill it out, send it in, get a web code, and then de-register from being an organ donor.
- If everybody else is donating organs- if that’s the default- perhaps I should donate organs because that’s the right thing to do.
- So what we’ve seen across these two examples of the compromise effect- Itamar Simonson’s work- and the default effect is the role that context plays in shaping human behavior.
Unit 1 > 1.2 Behavioural Economics and Choice Architecture > 1.2.3 Nudging and Behaviour Change
- The third point I want to touch upon is the fact that we can use our knowledge about behavioral economics to help people help themselves.
- We can design tools and technologies and products and gadgets that would actually help people make better choices.
- A, how can I change the context to help people make better choices? B, how can I actually get them more information, more insight, and more analytics to help them make those better choices? Now, before I end this segment, I want to talk about the other big element in this course, which is the notion of changing the context to influence choice.
- You’ve got people that currently chose option A, and you want to move them towards option B. What are the different tools you have at your disposal? One tool that is popular with policymakers and governments is the notion of restriction.
- When I do this, I leave people with no option but to choose product B. That’s a simple example.
- What you can do is you can create positive incentives, like a subsidy or surplus or some kind of payment, for people that move from point A to point B. Or you could create a negative incentive for people that choose to stay at point A. You could tax them.
- We’ll talk about the notion of choice architecture.
- How can I influence choice without changing economic incentives, without imposing restrictions, and without advertising? Simple example, defaults.
- If I know that people stick with the default, simply changing the default in a given choice is going to change the percent of people choosing that option.
- If I know that items displayed in the middle are more likely to get chosen, can I take a preferred option and surround it by an option that is weaker and cheaper, another one that is stronger and more expensive? And the research would show that, in that case, choices are more likely to gravitate to the one in the middle.
- We talked about choice architecture, we talked about influence, we talked about stealing people’s choices, so what’s the right option? Who decides what is right? And again, there are multiple ways of thinking through that.
- We can think through a parental body, a government, a policy board, a cooperative unit, a dieting society, someone that we trust that tells us what the right choice is.
- Can we actually think about what the majority wants? Can we think about options that are socially more desirable and steer people towards those choices? That’s a second norm, a second way, of thinking about who decides.
- The third and most important point is you, the individual, decide what is right for you, and then you compose a choice architecture.
- You compose a context in which your own choices are now influenced by what you decide is right for you.
- Remember, we said that there’s a big gap between what people want to do and what they end up doing, and choice architecture presents a fantastic opportunity to lock yourself into the kind of behavior that you want to do.
Unit 1 > 1.3 Rational Choice > 1.3.1 Rational Choice
- DILIP SOMAN: What does it mean to be rational? Perhaps the best way of answering that question is to go through a simple example.
- Suppose a friend of yours comes to you and says, Dilip, how much should I save for a diamond? How would you advise that person to proceed? Well, it turns out that economics has a very handy model to describe how this person should go about making that decision.
- Without getting into too much detail, what the hypothesis essentially says is, you want to look at the net present value of your future income stream.
- You want to look at the net present value of your future consumption stream.
- Then set your savings decision such that the present value of those two numbers is the same.
- I need to know how many children I’m going to have.
- Whether we’re going to enjoy the movies, we’re going to enjoy living in an apartment, a condo building, a house.
- I need to know exactly what my income streams are going to look like.
- I’m going to need to think about what the economy is going to look like.
- So what does it mean to be rational? Well, there’s a lot of things that go into the definition of rationality, but I’m going to be very simple in this video, and I’m going to focus on what I call the four C’s of rationality.
- Cognition, the ability to think through problems, computational ability, which is the ability to process information and make fairly complex calculations, and finally, consistency in decision making, both internal and external.
- There might be some decisions where you have completeness, but in most decisions, you don’t.
- The second point that we talked about was cognition, the ability of human beings to make completely unemotional decisions.
- That’s important from a rational perspective.
- If you’re a rational agent, what you want to do is to look at only elements of the decision-making that adds something called utility to your choice, and look at those, and only those.
- It turns out human beings look at all kinds of other things.
- I had a friend, curiously, an economist, who was looking at a membership to a health club.
- The number of people that should have bought a stock that’s going up for $10 but don’t, because they could have bought it for $7 is stunningly high.
- So what human beings tend to do is, they tend to try and find short cuts so that they can actually make decisions, given their limited cognitive abilities.
- They could have been rational if their heads were replaced by a supercomputer, but now that they only have that little laptop, they had to be boundedly rational.
- So in the context of the limited cognitive ability that they have, they’re still rational, but it’s just that they can’t go beyond a certain limit.
- We’re looking at the products that they’re looking to buy.
- When you think about a world as complex as that, we can only devote so much attention to the stimulus, and that’s why we never pick up all of the information that’s necessary to make a good decision.
- Now, when you think about the mathematical modeling of rationality, there’s a fairly complex set of algorithms and theorems that actually describe what rational choice looks like.
- In order to understand those, we want to first look at this, which is a simple mathematical expression of the utility that a product offers to an individual.
- If it comes heads, then I give you $100. If it comes tails, you give me $50. The way I’m going to capture that using this equation is, I’m going to say that the weight is here, the probability of each of these outcomes, 0.5, multiplied by the utility of the value that I assign to the $100, or the minus $50. Think about applying the same equation in a different setting, where you’re choosing a product from a multiple set.
- In this case, the x’s are going to stand for the different attributes of the product, the quality, the price, the size, the shipping costs, and the w’s are going to stand for how important each of these attributes are to you, the decision maker.
Unit 1 > 1.3 Rational Choice > 1.3.3 The Axioms of Rational Choice
- We’re going to focus on the first three, because the last two are really there for mathematical tractability off the model, and they don’t have interesting behavioural insights coming out of those.
- Completeness says that when I give a decision maker two options, let’s say x and y, the decision maker should be able to articulate a preference between those two options.
- In other words, they should either be able to prefer x over y, or y over x, or they should be able to say that they’re totally indifferent between those two options.
- Let’s go onto to the second axiom, and that’s called transitivity.
- Let’s imagine a decision maker prefers x over y, and then chooses y over z. What this axiom says is, in that case, x should be preferred over z. Now this sounds fairly obvious.
- If you prefer receiving $100 over $50, and you prefer receiving $50 over $20, it makes perfect sense that you want to receive $100 instead of $20. But there are many different domains in which this particular axiom is violated.
- You’re looking to hire someone, and you have three applicants, Mr. A, Mr. B, and Mr. C. And for each one of them, you’ve got data on two attributes.
- You want to choose the most intelligent person, and if in fact, you have two people of equal intelligence, then you will choose the one with higher work experience.
- I would prefer C over B. However, if I now look at the whole picture, I will end up preferring A over C, because now, the difference between 120 and 100 is significantly large, that by my rule, I end up picking the more intelligent person.
- Here’s the third and more important axiom, and this axiom is called the axiom of substitution.
- If I have two options, x and y, and I’m indifferent between x and y, then I should be indifferent between two lotteries that have x and y as their price.
- Then I should be indifferent between those two lotteries because I was indifferent between the two prizes to begin with.
- Cancellation says that if I’ve got two options, and I remove something identical from those two options, I should still now have the same preference between those two options as I had before.
- Option A, I give you a million dollars, no questions asked.
- In this circumstance, what Maurice Allais found was that most people prefer Option A. It’s a no brainer.
- Again, if you’re choosing here between two options, we’re going to call it A and B. In Option A, we have a lottery in which there’s an 11% chance of winning one million and an 89% chance of winning nothing.
- Now, as you might expect a lot of people prefer Option B. If you think about, the difference between 10%, 11% isn’t very large.
- People tend to pick the Option B, which gives them a chance of winning the larger price.
- If I put these two side by side, what you’ll notice is that these two options are essentially identical.
- What I’ve done is I’ve taken away an 89% probability of winning a million dollars away from both options.
- Essentially, by doing that, I have now created a situation in which I can reverse preferences between A and B. And that, again, is a violation of the axiom.
- Now, as we go through this course we’ll see a lot of different ways in which consumers and people in general tend to violate, not only these axioms of choice, but tend to make inconsistent preferences.
- People routinely do not behave in consistence with those four C’s, but they’re just being people.
Unit 1 > 1.4 Decision Points > 1.4.1 A Theory of Decision Points
- What’s interesting is that a lot of people who get popcorn of a very large size don’t actually want to eat all of it.
- A lot of times when people decide to eat popcorn or potato chips or drink soda or eat chocolates, they tend to make what we call a meta-decision.
- We found that when cookies are partitioned by putting glazed paper in the middle of different cookies again people slow down the consumption of cookies.
- We invited a number of people to participate in a gambling study.
- They were each given 100 gambling coupons that looked like that.
- Each coupon had a cash value of $0.50, which meant that every individual in our experiment was getting about $50 worth of assets, which they could exchange for cash at any point in time.
- So people could actually choose to simply leave the room, cash in their $50, and be done.
- A lot of people in our experiment told us that they knew they shouldn’t gamble, but they had fun doing it.
- Our idea was to try and see if partitioning these coupons into different set sizes actually changes how much people gamble.
- In one condition, there were 100 coupons in an envelope sealed and given to participants.
- In a second condition, there were 10 coupons in an envelope, and 10 of these envelopes were given to the participant.
- We were interested in seeing how much people gambled as a function of this particular manipulation.
- People had to actually go online, click on a button.
- With some probability, they won five coupons for every coupon that they had gambled.
- Rule number one was that they could not gamble away their winnings, so they could only gamble up to 100 coupons.
- The panel on the top shows you the number of people who actually gambled a given number of coupons when they had 100 coupons in one envelope.
- So the panel on the top is for people that had all 100 in the same envelope.
- What you see is a lot of people chose not to gamble.
- Once they started gambling, the bars keep increasing in height.
- What you’ll see is that there was one person that actually gambled away all 100 of their coupons.
- The panel at the bottom, on the other hand, shows you conditions where people had 10 envelopes, each with 10 coupons.
- What you see there is that almost everybody gambled.
- 40 was the highest number that was gambled.
- What’s interesting is that these amounts gambled represent multiples of 10, which essentially means people opened an envelope and then stopped, or opened two and stopped or three and stopped.
- The coupons were like popcorn in bags that are different sizes.
- When people opened one envelope, the coupons kind of became free.
- When people had to multiply open a number of envelopes, that was a decision point which forced them to stop.
Unit 1 > 1.4 Decision Points > 1.4.3 Using Decision Points as a Nudge
- DILIP SOMAN: Now that we found that the decision point idea worked with food and with gambling tokens, our next question was, does it work with cash? We did a bunch of experiments in rural India with people that were in a cash economy.
- The banker went with us to different households and would suggest something like, we think you should save 40 rupees over the next two weeks.
- Of course, adding pictures of children increased that savings rate even further because again, it was a decision point.
- So we found across a number of experiments that the idea of creating decision point seems to work because it converts decision-making from an automatic process to a fairly deliberative, rational process.
- The impulsive guy doesn’t think too much and eats until it’s all gone.
- What’s happening with the theory of decision points is that the part of us that actually is the visionary, the planner, is imposing a constraint on the impulsive person by creating these multiple opportunities to go back and think.
- So what is a decision point? It is any intervention that is designed to get people to stop, pause consumption, and think a little bit.
- There are different ways in which you can actually implement a decision point.
- The act of that small transaction cost actually made people stop and think.
- The second way in which you can create a decision point is to send people a reminder.
- So think about a study where you send people a reminder about saving or about taking their medication on a text message or on email.
- The third way in which you can actually create a decision point is to simply create some sort of interruption to a consumption’s routine.
- If you wanted people to say no, make it easy to say no.
- If at any point in time, there is something you want to control, impose decision points on yourself.
Unit 1 > 1.5 The Allais Paradox Revisited > 1.5.1 The Allais Paradox Revisited
- In the first problem, the first choice that participants were presented with, Option A was 100% percent chance of getting $1 million.
- In the second problem- the one that had a yellow background- we had two gambles- 11% chance of $1 million, 89% chance of nothing- that was the first option.
- The second option was 10% chance of $2.5 million and a 90% chance of zero.
- If I’m faced with this choice, I’m going to say option A is a certain $1 million.
- Option B, there is a small, but a clear chance, of getting nothing.
- So one line of reasoning would say that people are going to go for Option A, which is a safe option.
- So the prediction would be that a lot of people might pick Option A. In the second problem, the yellow box, now we are comparing two gambles.
- So the prediction might be that people would pick Option B in the second problem set.
- Most people pick Option A in the left-hand side, in the blue box.
- In the right-hand side, in the yellow box, people pick Option B which, by itself, is not a problem, except for the following argument- that the problems in the blue box and the yellow box are essentially identical.
- Option A, 100% chance of $1 million in the blue box, and one of the outcomes of Option B in the yellow box- a 90% chance of 0.
- I’m going to take the 100% chance of $1 million, and I’m going to break that down into an 11% chance of $1 million plus an 89% chance of the same $1 million.
- It’s there in Option A. It’s also there in Option B. And then, I’m going to take that $1 million in each of these two features, and replace that by nothing.
- It’s a paradox because you have two options, or two choice sets, two problems, which are functionally identical.
- In one case, people choose Option A. In the other case, people choose Option B. That is Allais’ Paradox.
Unit 1 > 1.6 Debate > 1.6.1 Debate
- Second, irrational behaviors result in negative emotional feelings, such as regret or general unhappiness.
- From this angle, I think irrationality can indeed damage welfare and well-being.
- If people can be a little more rational, use reason more adequately in decision making, their welfare can often be improved a lot.
- I think it’s very obvious this is very little money.
- I want to argue this is because we are irrational.
- SENDHIL MULLAINATHAN: I think in behavioral economics, we often use that word, “Irrationality.” And when we use it, we’re making a very big mistake.
- Irrationality suggests, at its very root, that people are doing something stupid.
- If you called a friend irrational, they’re probably not going to be your friend for very long.
- Actually, we use that word for technical historical reasons, because somebody wrote down some axioms that they called “Rational.” And as a result, when people violated those axioms, we called them irrational.
- “Irrational,” doesn’t in any way, shape, or form mean that the person is being stupid.
- If somebody underweights base rates in forming probability estimates, we would call them irrational in the technical sense.
- It’s foolish to call them irrational in the colloquial sense, because what does it even mean to get base rates correct? The mind is configured the way the mind is configured.
- So once we think of it that way, then we can ask the question, are the quirks of the human mind- can they ever be damaging to our welfare? The answer is, of course.
- Look at your own behavior, think about it, and ask yourself, what are the times I’m going to make that mistake? Obviously, when I’m rushed, I’m going to make that mistake.
- I think a lot of social policy that tries to fix, quote, “Irrationality” is nothing more than that.
- I think all the heat in this debate is just generated because we use the word “Irrational.” CHRISTOPHER HSEE: Well, is irrationality really damaging? Well, it depends on how we define rationality.
- I think we should focus more on the second type of irrationality.
- I don’t really think that that’s how well well-being or welfare should be defined, anyway.
- If people are being irrational but that makes them more happy, I don’t see any real problem with that.
- I’m going to be making the case that nonrational or irrational thinking is not at all damaging for human welfare.
- In the last 10 or 20 years, there have been literally thousands of studies published that suggest in ways big and small human beings act in irrational or nonrational ways.
- I might pay somebody $20 to mow my lawn because that’s how much I think mowing a lawn is worth.
- Let me give you one final concrete example of nonrational thinking at work before I make the case on why such thinking might actually be very good for us.
- Let me give you an example of nonrational thinking at work.
- Now my mind starts to think, how far down is it to the Earth? I look out the window.
- The irrational fears that many of us have about airline travel have led to policies and procedures that have made aviation as safe as it is today.
- When you think about nonrational thinking across the board, there’s two ways to think about it.
- You can think about the thinking and say, that’s crazy.
- Or you can ask yourself, why is it that our brains have such a great propensity to think in nonrational or irrational ways? Why do we have nonrational thinking in our brains? Now, unless you’re somebody who believes in intelligent design, the reason our brains work the way they do is because they have been designed by evolution to work the way that they do.
- There’s a reason we have nonrational thinking in our heads.
- That’s because over the long term, thousands of years, millions of years, nonrational thinking, the biases that are built into the brain have protected us.
- So when you think about these biases, you can think about them in two ways.
- If you want to think about them like an engineer, you would say, that’s a terribly designed system.
- Why is this system so dumb? Or if you think about them like an economist, you would say, look, it makes no sense that you would pay somebody $20 to mow your lawn, but you wouldn’t take $100 from your neighbor to mow his lawn.
- So if you think about those things in a very narrow fashion, and you want to mock those kinds of behaviors, you should probably become an engineer or an economist.
- Well, I say disproportionately, but if you think about it in evolutionary terms, it’s much better for our brain to actually be somewhere on the side of, err on the side of caution.
- So it’s not unreasonable, even if it’s irrational, for our brain to contain this bias.
- If you like, is irrational in economic terms.
- Thinking that a stick is a snake is almost certainly ecologically rational.
- Effectively, to some degree, copying other people’s behavior without really thinking about it sometimes gives you the benefit of having 200 pairs of eyes rather than just one.
