PRESIDENT ZIMMER: Good afternoon, and
welcome to the 2018 Ryerson Lecture. The Ryerson Lectures are named
in honor of Nora and Edward Ryerson. Edward L. Ryerson Jr. was
a long-standing member of the Board of Trustees,
serving 48 years from 1923 to 1971. And he served as chairman
of the Board of Trustees from 1953 to 1956. The Ryerson family,
now the Ranney family, has an association
with the University that dates back to
our earliest days when Martin A. Ryerson served
as a member of our first board of trustees and presented the
gift for the Ryerson Physical Laboratory named in
honor of his father. We have members of the
Ranney family here tonight. And I especially
want to acknowledge the presence of George Ranney. We are delighted
to see you, George. Thank you very much
for being here. The annual Ryerson
Lecture is an opportunity to hear from an
esteemed faculty member of the University of Chicago. Each year a committee of faculty
from around the University chooses one faculty member to
deliver the Ryerson Lecture, an individual whose
research contributions are believed to be of
lasting significance, and who the committee
believes is someone that the community
would like to hear from. The first Ryerson
Lecture was delivered by John Hope Franklin in 1974. The collected Ryerson
Lectures represent 45 years of the University’s
scholarly leadership across the disciplines,
and also provide an intellectual history of
our broader academic society over that period. This year, we are
delighted to have Richard Thaler as our speaker. Richard is the Charles R.
Walgreen Distinguished Service Professor of Behavioral
Science and Economics, and Director of the
Center for Decision Research at the University
of Chicago Booth School of Business. He is also co-director of the
Behavioral Economics Project at the National Bureau
of Economic Research. His paradigm-shifting
work, which bridges psychology
and economics, exemplifies the
spirit of scholarship at the University of
Chicago and has earned him a well-deserved place among
the University’s long list of influential
economic scholars. Richard’s contributions
to the field were recognized last year
when he was awarded the Nobel Prize in Economic Sciences. Richard joined the University
of Chicago faculty in 1995 as the Robert P. Gwinn
Distinguished Service Professor of Behavioral
Science and Economics at the then Graduate
School of Business, now, of course,
the Booth School. Before coming to
Chicago, he taught at the University of Rochester
and Cornell University. He earned his bachelor’s degree
from Case Western Reserve University and his PhD from
the University of Rochester. He’s a member of the American
Academy of Arts and Sciences, a fellow of the American Finance
Association, the Econometric Society. And in 2015, he served as
President of the American Economics Association. And in recently fast-breaking
news, this morning it was announced
that Richard was elected as a member of the
National Academy of Sciences. He’s the author of several
books including Nudge: Improving Decisions about
Health, Wealth, and Happiness, published in 2008
with Cass Sunstein, and Misbehaving: The Making of
Behavioral Economics, published in 2015. Richard’s talk today is
titled “Behavioral Economics: Past, Present and Future.” Richard? RICHARD H. THALER: OK, so
about some time last June, I think, I got a call from
President Zimmer’s office. His assistant
called me up, said, the president wants
to speak to you. And when he got on
the phone, I said, no! Which is my– the dean knows
that’s my standard operating procedure. But then he said he wanted
me to give this lecture. And I decided I
couldn’t say no to that. So thank you all for coming. And thanks to all the donors
and Ryerson mishpucha. And here’s what we’ll see. So let me start by saying
what behavioral economics is, and, I hope
fittingly, use a quote from a distinguished
University of Chicago alum, namely Herb Simon. He says, “The phrase
‘behavioral economics’ appears to be a
pleonasm,” which is a word even University of
Chicago students may not know. It means a redundant phrase. And he explains, well, what
would non-behavioral economics be? If you think about
it, economics is about the behavior of firms
and people and consumers and employees. And they’re all people. And so what would this
non-behavioral economics be? And Herb answers the question. He says the answer
to this puzzle of why we need that
adjective is found in the assumptions of
neoclassical economic theory. So the core
assumption, the thing that drives everything
in economics, is the assumption that
agents choose by optimizing. And so if you open
any economics paper, you’ll see a symbol
max something. And that’s the way
the theory starts. So it wasn’t always that way. Economics started
out behavioral. Let’s go to the founder of
economics, Adam Smith, who was the first behavioral economist. Let me just give you three
quotes from Adam Smith on three topics that behavioral
economists have talked about endlessly– the first, “overconfidence.” Here’s Smith, “The
overweening conceit which the greater part of men
have of their own abilities.” What about loss aversion,
the fact that we seem to– losses hurt about twice as much
as gains make you feel good. Kahneman, Tversky thought
they discovered that. No. 1759, “Pain is in
almost all cases a more pungent sensation than
the opposite and corespondent pleasure.” So then last, “self-control,”
a subject that I’ve studied. “The pleasure which we are to
enjoy 10 years hence interests us so little in comparison with
that which we may enjoy today.” Now, the language may
be slightly different if you read modern
papers, but the idea– these are the core ideas
of behavioral economics. And they get
attributed to people like me and Kahneman
and Tversky, but Adam Smith had
them all first. Then fast forward to
Keynes, who I claim is the inventor of
behavioral finance. Here’s a passage from
the general theory– day to day fluctuations
in the profits of existing investments, which are
obviously of an ephemeral and non-significant
character, tend to have an altogether excessive and even
absurd influence on the market. So this is Keynes’
writing in 1936. My friend Bob Shiller,
who shared the Nobel Prize with Eugene Fama and Lars
Hansen five years ago, won in large part for
work documenting the fact that stock prices seem
to move too much compared to the movement of fundamentals. And I would say economics was
behavioral up until Keynes. I’ll come back to that
though in a minute. Another early
behavioral economist, Pareto, he says, “The
foundation of political economy, and, in general of
every social science, is evidently psychology.” My psychology colleagues
are always telling me this. “A day may come when we
shall be able to decide the laws of social science from
the principles of psychology.” I’m not sure we’re
quite there yet. But then let me give you the
University of Chicago view. Now, University
of Chicago is not often thought of as a hotbed
of behavioral economics. In fact, there was the Journal
of Political Economy recently published a 125th
Anniversary issue, where they asked various
Chicago economists to write short papers. They asked me to write one
about behavioral economics. And I will save the page limit
was not a binding constraint. But I began with this quote
from exactly 100 years ago by a professor in the economics
department, John Maurice Clark. He was the son of a more famous
economist, John Bates Clark, a famous award is named after. So here’s what John
Maurice Clark says. “The economist may attempt
to ignore psychology, but it is sheer impossibility
for him to ignore human nature. If the economist borrows
his conception of man from the psychologist,
his constructive work may have some chance of
remaining purely economic in character.” But if he does not, he will
not thereby avoid psychology. Rather, he will force
himself to make his own, and it will be bad psychology.” Applause is fine. You know, I– So I’ve been saying for years– my former student, Werner
De Bondt is here today. And it’s thanks to him
that I have that quote. I owe him many other things,
but especially that quote. Werner reads these
kinds of old papers. I don’t know where
the hell he found it. But it’s a wonderful passage. And so since Werner
showed me that, I’ve been saying that
behavioral economics is simply borrowing good psychology rather
than inventing bad psychology. So let’s go back to
those assumptions that Simon was talking about. And there are really four. So optimization I mentioned. People are assumed to choose
the best option of those they can afford. Second, consumer sovereignty. What that means is
that people know what’s best for themselves. And in particular,
they know better than anyone else could know. And especially, they know better
than the government would know. And what that means– you never will read
in an economics paper, “I assume there are no
self-control problems.” That sentence has
never been written. But it’s implicit
in every paper. Because the idea of
consumer sovereignty, that we choose
what’s best for us, means that we never choose
what’s wrong for us, like eating that dessert
i having that extra drink or what have you. Unbiased beliefs. There’s a large
literature in economics called rational
expectations literature, that just formalizes
something that was always assumed informally
in economics, which is that people’s expectations
about the future are unbiased. Meaning that you couldn’t
improve on them even if you were as good an
econometrician as Lars Hansen. And then finally, self-interest. Economists assume that people
are pretty much jerks and only care for themselves. Possibly their family, depending
if the family has reciprocated. So this set of assumptions
defines homo economicus. And I’d like to
eliminate the Latin, and I just call these people– well, these fictional
characters– “econs.” So how do econs
differ from humans? The people we deal with. And how big are the disparities? That is really what the field of
behavioral economics is about. It’s about that. So this is as close to an econ
as we have in the literature. He was half human. And the people I study are– more like that. Homer has none of those four
character– well, selfish. OK, so let me pose an
obnoxious question. Is the idea of optimization
even plausible? I mean, is it a starter, as
a model of human behavior? And I would argue no. That Adam Smith had it right,
and for the 30 years or so following World War II when
economists got busy making their field rigorous,
led by people like Paul Samuelson
and Ken Arrow and Robert Solow, that
whole generation who redid all of economics, but right. Was that a plausible model? And I’d say no for two reasons. One is some things are
harder than others. And some people are
smarter than others. And economics assumes that we’re
all equally good at everything. So let me give you an example. Here are some tasks that are
of increasing difficulty. Now, the life-cycle theory
of saving, if you don’t know, is a theory that says
that we figure out– at birth, I guess, but
let’s say upon graduation, how much we’re going to
make over our lifetime, how we want to allocate that
across the lifetime. Then what investments we
need to make to do that. And then smooth optimally
updating presumably daily, or, if you’re really
sophisticated, in continuous time. So anyway– so let’s take
somebody really smart, and I figured I
would take a local. So Fermi, who is, I am
told, a pretty smart guy, he was probably capable of
solving the lifecycle model. Unlike me. Now, let’s take somebody
not as smart as Fermi. OK, so now, what about
self-control problems? Again, we have an array
of self-control problems. I must say that one
of my former students thinks that my PowerPoint
skills are inadequate. And she’s responsible for these. And you’ll see some of
her editorial comments in the following. So let’s take somebody who
had mastered self-control. And again, the economic
model would work perfectly. He would find resisting
golf and sharing a ’45 Mouton equally easy. Here’s her view of– So the model was a nonstarter. It couldn’t possibly be right. But in 1976 or ’75
or whenever I started having these deviant thoughts,
me getting up and saying, look, this model is ridiculous,
was not a winning strategy. And the attempts I made
were met with resistance. In my book, Misbehaving, I
refer to it as the gauntlet. And my friend Matthew
Rabin has come up with a nice phrase for the
one line put-downs that were used for three
decades to just say, we don’t need to
bother with you. So what are they? One, as a university– actually
both, both of these I’m going to mention have
University of Chicago origins. And in particular,
Milton Friedman. And the first one
has just two words– “as if.” And I would hear
this in workshops. In the early years, I
hardly ever gave a workshop without somebody just raising
their hand and saying, “as if,” and, like, I should stop
talking at that point. So there is a passage in a
paper that Friedman actually wrote with Savage, where he
describes an expert billiards player. And he says that the
billiards player behaves as if he knew math and physics. And that model,
assuming all of that, would predict very
well how he plays. So what do I say to that? Well, first of all,
economics is not supposed to be a
theory about experts. It’s a theory about everybody. The life-cycle
theory of saving is a theory about how everybody
saves for retirement, not just Frank Modigliani
who invented the theory. Well, the problem
economists have is that they assume all agents
are as smart as they are. When they’ve spent the last
year solving some problem, then they say, uh, now assume– right? So I don’t play chess
like a world champion. I don’t play billiards
like a world champion. And neither do most
of you, I assume. And the second key thing,
and my biggest discovery, my biggest scientific
discovery, was discovering two
Israeli psychologists named Kahneman and Tversky. Now, you may think they existed
before I discovered them sometime in the ’70s. But economists didn’t
know about them, so my discovery
counted as a just– kind of like Columbus, you know? So the big insight from
Kahneman and Tversky is that people make errors,
and that those errors are predictable, or systematic. So if errors are predictable,
“as if” goes out the window. Because people are
not behaving “as if.” They’re optimizing. They’re deviating from
that in a predictable way. So the “as if” line was
a verbal sleight of hand. Friedman was almost certainly
the most brilliant debater in the history of economics. And the two words
worked for a long time. Now, another argument that I
would often hear, especially in the early years when
the data we had was mostly from laboratory
experiments, people would say, well, yeah,
they get that wrong– “that” being whatever
you just studied– but if you raised
the stakes enough, they would get it right. That was claim one. Then another claim was,
well, in your experiments, you just gave them
one chance at this. And in the real world,
we get to learn. And so in the real world,
people behave like econs. Now, the first
thing to notice is I would often hear those two
remarks from the same person, often in the same workshop. And they’re self-contradictory. Why? Because the higher the stakes,
the less often we get to do it. We buy milk twice a week
maybe, soups less often, cars less often, houses
very infrequently. So as we raise the stakes,
the amount of practice we get goes down. So you can have one of these
arguments, but not both. And then there’s
the annoying fact that there’s no evidence to
support the claim that people improve as the stakes go up. None. So then we get to
another rude line I have, called the invisible hand wave. I suppose I’m referring
again to Milton Friedman. And here’s the way it goes. It’s– and this is never written
down but was heard often. And I think it’s still heard
in some economics workshops, right here on this campus. And the line is, “Well,
if people behaved in markets the way they
do in your experiments or your studies, then–“. And now I– the reason I
call this a hand wave is I claim no one has ever finished
this sentence with both hands in their pockets. And that’s because it’s
not possible to do. Try it. I mean, there has to
be a lot of hand-waving because markets have no way of
transforming humans into econs. So how would this
work if it worked? What is supposed to happen
if you engage in a market? So suppose you choose
the wrong career, or marry the wrong
spouse as my wife did, or fail to save for retirement. If you failed to
save for retirement, barring reincarnation,
you’re screwed. If you take out a
mortgage, you won’t be able to repay unless
house prices keep going up or interest rates go down. You know, you get poor. That’s all. You don’t disappear. Stupidity is not
fatal, unfortunately. And here’s maybe the
most important point, so I put it in bold. It’s much easier to make money
exploiting irrationalities than fixing them. So here’s an example. And this is a dated number. The actual number is
almost certainly bigger. This is the amount of extended
warranties sold annually in the US. Now, almost always
you should just turn down extended warranties. That’s a good rule of thumb. Just say no, like
Nancy Reagan said. Now, but people are
making billions of dollars selling extended warranties. I have been trying to
convince people not to buy extended warranties
for a long time. So far, I’ve made
exactly $950 before tax in this effort, which is what
I got paid for writing one New York Times column. And you know, $950 compared
to $27 billion, you know, they’re winning. And as a general principle,
it’s just really hard to make money convincing people
that they’re making mistakes. They just think that there’s
something wrong with you. And, you know, go away. So where are we so far? The one-liners didn’t work. So we have to just
get down to work, and we have to look at the
data and say, all right, how differently do humans
behave compared to econs? And if that’s a lot, then what
do we do with economic models to stick in some new
thing instead of the econ? So let me emphasize that
nowhere have I ever said– and in Misbehaving I disavow
on numerous occasions the idea that we
should just throw away all of standard economics. In fact, behavioral
economics would not exist without the
neoclassical economics, because it was the
starting point. It’s the benchmark
for everything. And just like in
high school you learn physics, the simple
physics in a vacuum, and that’s a good way to start
learning physics I guess. But if you want to
fire up a rocket ship, you probably ought to pay
attention to atmosphere. So and for the role of stakes,
that’s an empirical question, and I’ll show you a
little bit of data. So let’s take the argument
about self-interest. There is lots of evidence
from behavioral economists that people have a
preference for outcomes that they perceive to be fair. Now, take the
prisoner’s dilemma game, which I don’t think there are
any courses at the University of Chicago that don’t cover
the prisoner’s dilemma, so I won’t explain it. But the theory is that the
dominant strategy is to defect, and so everyone should defect. But in laboratory studies,
a little less than half of the people cooperate. So my friends Steve Levitt
and John List wrote a paper a decade or so ago
saying, well, they’re not so sure about
these experiments because the stakes are low. So how do we address that? It’s hard to get a budget to
run really high [INAUDIBLE] prisoner’s dilemma games,
unless you use a game show. So somebody invented
a game show that was only shown in the UK that
ends with a prisoner’s dilemma. And here’s how it works. There would be two finalists. They will have earned a
certain amount of money. And then they play
what we’ll technically call a weak prisoner’s dilemma. They called it “split or steal.” The two players were
given these golden balls, which was where the name
of the show comes from. And they could choose
split or steal. If they both choose
split, they each get half. If one splits and the other
steals, the one who steals gets everything. And if they both steal,
they both get nothing. OK, so if we had
run this study, it would have cost us 2.8 billion– sorry, well, it’s pounds
so it’s almost billion. But 2.8 million
pounds, which, so far, none of the presidents or
deans sitting in the front row have offered me that kind of
money to run an experiment. So we studied this one. And the stakes vary a lot. So from 100 pounds
to 100,000 pounds. And how does cooperation vary? That’s a pretty
enormous range to study. Does cooperation fall? Sort of. So here’s the game if
you want to get a– this is the high
stakes, highest stakes. Some of you may have
seen this on YouTube. If you haven’t, check it out. This is the expression the
two players’ faces when they discover they’re– the pot that they can share
or not is 100,000 pounds. They’re pretty amazed. Now, the guy on the left
is pleading and promising that he’s going to split. And she is saying if
she steals, no one would ever talk to her again. And here you can see they’re
bonding and making promises. And you can guess,
you can decide which one you think is nicer. I think it’s fair to say
she had some misgivings about her choice, but
not 100,000 pounds worth. So all right, so what
happened over this range? Well, John and
Steve were kind of right, sort of, in
that cooperation rates fall as stakes go up. But they fall to the level
we see in labs, right? Does this– anyway, on the right
side here where the money gets big, you can see the cooperation
rates are sort of 40% to 50%, which is what we’ve seen
in games for $10 or $20. The only reason it falls
is that the stake– that the cooperation
was really high when the stakes were relatively low. This is something my colleagues
and I have observed in– I’ve now written three
papers about game shows, so how low I’ve gone. And one was on the
dreadful show Deal or No Deal, which
was almost designed to be an economics experiment. But what we see in
this game and in that one is something that I’ve
called the Big Peanuts Phenomenon. And what I mean by that
is, suppose that you’re playing for only 500 pounds. Well, you were expecting to
play for tens of thousands of pounds. So that seems like peanuts. Now, if we ever tried to run
an experiment with 500 pounds– I don’t know what
that is now, $750– that would be a lot
of money, right? And in any other
time of their life, they would think that
was quite a bit of money. But in this particular
case, they thought, well, for $750 it doesn’t– it’s not
worth it to be a jerk on TV. So we got very high
cooperation rates, but there’s no evidence
that cooperation disappears at very, very high stakes. All right, let’s
raise the stakes further from hundreds of
thousands to trillions, and talk about the
Efficient Market Hypothesis, a phrase that was
coined by my friend and golf buddy Eugene Fama. And I like to describe this
as having two components. The first is what I call the
“no free lunch” component, which is, you can’t beat the market. The second component I
call “the price is right.” You can see I have an
affinity with game shows. And this component says that
asset prices are correct. That is, they’re equal to
their intrinsic value, whatever that is. Now, in my judgment, the
“no free lunch” component is approximately true. If I were grading it, I would
give it maybe an A-minus. And what do I mean by that? It may be possible to beat the
market, but most people don’t. So it’s hard. But I will say the following
three things cannot all be true. Investors are rational,
markets are efficient, and the financial sector takes
up almost 10% of the economy. You can have two, but
not three, of those. You pick which ones you want. I’m not saying the financial
sector has to be 0, but 9% is a lot. So all right, what about
“the price is right”? The defenders of the
Efficient Market Hypothesis long slept well at
night about this part because they thought
it was untestable, and there’s nothing
better in theory than it being untestable. So and the reason why it
was thought to be untestable is we can’t measure
intrinsic value. So if I think Apple is cheap,
and I think it’s undervalued, and Bob thinks Apple is
great and is going to go up, who’s right? How would anybody know? There’s no place you can look
up the intrinsic value of Apple shares? So how can we test this? We have to get tricky and
look for special situations where we can test it. And there are several of
them that I’ve studied. But I’m going to show you
the one that’s the most fun. So there’s a
closed-end mutual fund. I guess I have to
explain what that is. Unlike the prisoner’s dilemma,
it’s not taught in every class. A closed-end mutual fund is
just like a regular mutual fund except that the shares
are traded on exchanges. So if you want to buy
shares in the CUBA fund, you would call your
broker in the old days, or go to your account
and say, I want to buy $1,000 worth of
shares in the CUBA fund. Now, the interesting thing
about these closed-end funds is that they don’t always sell
for the value of the assets that they own, which
is a little odd. And in fact, I
wrote a early paper about closed-end funds
talking about that. It infuriated Merton Miller
for reasons I never learned, but it did. So but let’s go back to CUBA. The CUBA fund has the
ticker symbol, C-U-B-A, but cannot and could
not ever invest in Cuba. For one thing, there are
no securities in Cuba. For second, it would
be against the law. So we can stop with those. So the fund has never
owned any interest in Cuba. They have mostly US stocks,
some Mexican stocks, some cruise ship lines. I’m not sure why they
named it the CUBA fund. Maybe they appeal to the mojito
set or something like that. So OK, so that’s the CUBA fund. And like most closed-end
funds, historically, it traded at a discount,
meaning the shares traded at about 10% to
15% less than the assets that they owned. So here’s a plot. The orange-ish line
is net asset value, which means the value of the
assets that the fund owns. The green line is the price. And you can see over
there at the beginning it’s trading at a discount. And then something
strange happens. And all of a sudden, it starts
trading at a surplus of 70% So a week earlier, you could buy
$100 worth of Caribbean stocks for $85. Then the next week $170. Anybody have any idea? So this was the day
that President Obama announced his intention to
relax relations with Cuba. Now, remember this fund doesn’t
have any ownership in Cuba. And you might think, well,
yeah, but if he goes through with this, it could boost the
economy throughout the region. But if that were the
case, then the orange line would go up, right? Because these cruise
ship lines that are going to be frantically
going back and forth are going to make more money. If anything, the
market goes down a little bit over this period. If there’s a rational
explanation for this, I don’t know it. So Fischer Black, co-inventor
of the Black-Scholes formula, said maybe we should define
an efficient market as one where prices are right
within a factor of 2. Now, economists, when they
are grading themselves, tend to be easy graders. And this is an example. The Dow is about 25,000 now? So that means the right
price is somewhere between 12 and 50,000, you know? I don’t know whether you
want to call that efficient. But some policy makers have
had the belief that prices have to be equal to intrinsic value. And in fact, Chairman
Greenspan admits to having made that mistake. In a famous “mea
culpa” speech sometime after the financial
crisis, he admitted possibly putting too
much faith in the idea that there couldn’t be bubbles. Now, I think this,
as an assumption, is innocuous, as long
as you don’t believe it. If you start to think
it’s true, then you’re going to think that there
are all sorts of things that we shouldn’t worry about,
like real estate prices in Las Vegas. And I’m not going to
talk about this today. The same argument
applies to labor markets. So if you think some
people get paid too much and some too
little, there’s even more reason to doubt
the idea that it must be the case
everybody is getting paid exactly what they’re worth. LeBron James is, but– Now, here’s another rude
phrase that I’ve introduced. The economic model usually
makes vague predictions. For example, if price goes up,
quantity demanded goes down. There’s no magnitude there. But in some specific
situations, economics makes precise
predictions, which is that there is a variable that
will have exactly zero effect. And I call these supposedly
irrelevant factors. These are things that economists
are sure don’t matter at all, but they matter. So sunk cost, there’s– if you eat some dessert because
you paid a lot of money for it, you’re committing the
sunk cost fallacy. Framing the idea, my
friends Kahneman and Tversky invented, where they showed
that showing physicians data on the percentage of
people who survive an operation versus the
percent who die matters a lot. Although obviously one
is 1 minus the other. Which options are
named the default. I’ll show you some data
on that in a minute. Something I’ve
spent a lot of time in my career talking
about, mental accounting– people put labels on money,
and they’re not supposed to. So you’ll have to read my book
or something, which should really be mandatory I think. But let me just show
you data on one of– a couple of these, for
a topic that I think is important which
is retirement saving. Now, earlier in the talk
I made fun of the idea that people would be able to
solve this problem themselves. And so I’ve devoted
a lot of attention to trying to help people
save for retirement. And notice saving for
retirement is hard in two ways– hard conceptually,
the math problem is hard, and hard self-control. You have to delay
gratification and go back to Adam Smith’s quote. So about 20 years
ago, 1994 I think, I wrote a little paper saying
the only thing we could do to increase
retirement saving is just a tiny little change
in 401(k) plans where the default
used to be, when you were first eligible there
were a big pile of forms to fill out. If you didn’t fill them out,
you [? weren’t ?] [? in. ?] The switch was, you
get that pile of forms, and on the top one it
says, if you don’t fill out these forms we’re going to
enroll you at this saving rate and in this fund. Now, that’s a SIF, right? That’s a Supposedly
Irrelevant Factor because there’s a lot
of money at stake, and the cost of
filling out the form can’t really be that great. So here’s what happens. Let’s call those maroon lines. The maroon lines
are the percentage of people who enroll, and
the blue lines are the ones– no, sorry. All of those are the
percent who enrolled. The maroon lines are for people
with automatic enrollment. The blue lines are for
plans that don’t use it. You can see, for
any income group, if you just switch the default
you get 90% of the people in. And you did nothing. Here’s where we’ve just
grouped people by age. Same thing. Now, there’s a problem with
automatic enrollment, which is that most plans default
people into too low of a saving rate. This is not true
at universities. Universities have historically
had very high, generous retirement plans, either
because they love their faculty or because they want to
get rid of old people. You can decide which. But anyway, so in
the private sector, they typically start
people out at a 3% saving rate, which is not great. So how can we fix that? Former student of mine
and I created something we call Save More Tomorrow. And the idea is we give
people the opportunity to increase their saving rate
later, because we all have more self-control in the future. Right? Many of us are planning
diets in June, or July maybe. So I was going around
the country kind of like Johnny Appleseed trying
to get some firm to try this. Finally one firm in
Chicago agreed to do it. And they did it by hiring
a financial adviser and having the adviser
go offer to speak one on one with each employee. There were only
about 300 employees. And the Save More Tomorrow
option was his second choice. His first choice was
to suggest raising saving by 5 percentage points. For the ones who said,
no, they won’t do that, he offered them
Save More Tomorrow. So here’s what happens. This is a plot of the saving
rates of people who say, I don’t want to
talk to that guy. This is a plot of
the saving rates for the people who
accepted his advice to go up 5 percentage points. And you can see they go
up and then flatline. Here are our guys. Notice they were the worst
savers to begin with. And we almost quadrupled
their saving rates in less than four years. So these two ideas,
automatic enrollment and Save More Tomorrow, it’s now
called Automatic Escalation, are now used in a majority
of large companies in the US. And we guesstimate that we’ve
increased people’s saving by $30 billion, but
that’s a wild guess. OK, last topic. And I’ve saved this one for
last because it sounds horrible, as this cartoon suggests. How do people do choosing
health-care options? So a large employer– I’m almost done. A large employer tried
what an economist would think of as the
ideal system, namely they gave people
every possible option. So there were 48
options altogether. This is like you go
into Alinia, and Grant sends out a list of
4,000 ingredients and says, tell me what you want. Right? So now, this is a great
strategy for econs. They can just really fine
tune the plan perfectly. How did it work for humans? Not so much. More than half picked
a plan for which there was another that dominated it. What do I mean by dominated? I mean that in any state
of the world, no matter how much health care
you consume, there was a plan that was better. Think of this like there are
two Starbucks located next to each other. One is 20% cheaper. Half the people are going
to the expensive one. Right? They serve the same coffee, the
same network, same everything. OK, well, you know, this was a
mere private-sector employee. What was the mistake
they were making? Choosing too low
of a deductible. Similar to my point about
extended warranties. So let’s see. Would this apply to a
place where the employees have really high IQ? OK, so here at the
University of Chicago we have four health-care plans. Two are health
maintenance organizations. I’m not going to talk
about them at all. The other two are Blue
Cross Blue Shield. They’re called PPOs. I’m going to only
talk about those. And there is lots of
complicated details. And I’m going to
gloss over them. The prices depend on
how much you make. The less you make, the
more the university subsidizes the health insurance. But the math of this is
all pretty much the same. So I’m going to
quote the numbers for high-income employees,
which is defined here as making more than $175,000. And for a family plan. And there are two deductibles–
$1,000 and $4,000. The low-deductible plan comes
with an option of the dreaded Flexible Spending Account. Which everyone hates. Is there anyone here who likes
their Flexible Spending– no. So you have to file
forms, and sometimes they get rejected for reasons
you don’t understand. It’s awful. The high-deductible plan comes
with a wonderful Health Savings Account that doesn’t
require any paperwork. You just swipe a debit
card or pay for it– it works just like
a credit card. And there are no
forms to fill in. OK, which plan is better? Now, I’m going to show
you a chart that’s a little out of date. And I am not giving any advice. If there are any
lawyers in the room, I am not giving any advice. These lines reflect how
much it would cost you– that’s the vertical axis– as a function of how much
spending you do on health care. You can see one is below
the other one everywhere. That’s good. Right? So low is good here. It means you’re saving money. You’re really saving money
if you’re young and healthy. Then you’re over in
this left-hand corner. And there could be a
few thousand at stake. For everybody else,
it’s merely 800 or 1,000 or something like that. OK, well, this is the
University of Chicago. Everybody is going to be
in the lower plan, right? No. 80% are in the wrong one. Now, guess which plan is newer? So there’s something
called Status Quo Bias, which is a fancy
term for sticking with what you’ve always done. And the high-deductible
plan was an innovation about three years ago. And the default– remember
how important defaults are– the default in open enrollment,
if you don’t fill it out, if you don’t go log in,
is same as last year. So the upper line I’m giving
people the benefit of the doubt and saying they’re all asleep. And the lower line are
the ones who are awake. What’s the basic problem? Economists are using one
theory for two different tasks. Which is to say what’s the
rational, smart thing to do and what people do. And we need two theories. OK, that’s point one. Point two, there’s no reason
why economics necessarily has to be based on rational choice. Here’s a quote
from Kenneth Arrow, possibly the smartest
economist of the 20th century, who just says it’s
obvious that we don’t have to build theories that way. And he gives us an example. You could have a theory where
people take the first one. So what instead, if we’re not– if we’re going to get rid of
econs, what are we going to do? We could try to borrow
some good psychology. Now, I falsely
advertised forecasts about the future, which,
as Yogi Berra said, forecasts are hard,
especially about the future. So I have exactly one
slide, and it’s my last one, you’ll be happy to hear. So if every economist
were convinced by this, then behavioral
economics will disappear. I’ve been predicting its
disappearance for 20 years. So far it hasn’t happened. It will disappear because
economics will be as behavioral as it needs to be. So for easy things we can have
Fermis, and for hard things we can have Thalers. And that’ll be more
work for economists. Behavioral economics is harder
because almost anybody who has had high school calculus can
solve an optimization problem. And figuring out what
people like me do is harder. The result will be economics
with more explanatory power and opportunities to help
or to nudge for good. Like this. Thank you. AUDIENCE: Thank you for
your presentation today. When an individual reaches their
maximum 401(k) contribution, what should a
consumer do to force themselves to save more money? RICHARD THALER: Piggy bank? I mean, look, there
are lots of ways of– so commitment
strategies are useful for self-control problems. So the reason why 401(k)s are
so successful is the money gets taken out before you have
a chance to spend it. And people can recreate
that on their own. And another thing people can
do is if they have a mortgage and it’s a 30-year mortgage and
they can afford it, refinance and get a 15-year mortgage. And then 15 years later,
you won’t have a mortgage. AUDIENCE: Thank you
for the presentation. So there is plenty of
evidence for mistakes people– sorry, that people
make individually. Is there evidence
for mistakes doing in groups or socially as a– yeah, as a mass? RICHARD THALER:
Well, I’m staring at two social psychologists
who could but won’t volunteer to answer this question for me. Nick, correct me
if I’m wrong, but I think the way most groups
operate, if anything, they make things worse. Now, they could
make things better. Because there is the
wisdom of crowds. So if you have 10
people in a room, then collectively they know more. But the way many groups
operate, what will happen is that the group
will get polarized, and the most extreme
members of the group will have
disproportionate weight. The loudest members
of the group will have disproportionate weight. So it’s certainly not
the case that groups are a solution to the problem. Groups that are run well, where
the opinions are collected independently so people don’t
influence each other before they’ve had a chance to think,
it could help but it’s– the way groups
normally work, no. OK, we had somebody
in the front. Yeah? AUDIENCE: I wonder if you have
a wish list for public policy areas where you would like
behavioral economics to have more influence. Are there some
out there that are on your list for future work? RICHARD THALER:
Well, in my book, Misbehaving, I talk
about how I would like to see behavioral macro. Macroeconomics, you know,
the study of the big stuff. There has been some
work but not very much. I think there are huge
opportunities for that. There are lots of people doing
work in development and poverty where there are
big opportunities. Health care, the combination
of behavioral insights plus technology I
think is the best hope for reducing health-care
expenditures. And the reason is that a lot
of our health care expenditures are because people don’t
take their medicines. So take diabetes. If people just take their
medicine, it can be handled. If they don’t, a
lot of bad stuff can happen, like losing
your vision and your limbs. And so how to fix that? Well, the technology now
exists, or close to exists, to stick something in your body
that will measure your blood sugar in real time and fix it. Now, if we skip the fix-it part,
just the continual monitoring, if your cell phone
beeped every time that you should do something,
that would also help. So I think medicine is– obviously we have to– we can’t spend a sixth of our
GDP on health care and more and more forever. We have to figure out something. And part of it is just helping
people take care of themselves better. It’s not the only solution. The last I would
mention is corruption. The biggest problem in many
poor countries is corruption. And getting people to be more
honest is a behavioral problem. Look, let me just say,
all the problems we face, when you boil it down,
are behavioral problems. Climate change is a
behavioral problem. You know, dealing with North
Korea is a behavioral problem. AUDIENCE: Would you care to
comment on the income disparity that we’ve seen in
this country, which seems to have increased
enormously, certainly in my lifetime? Is it amenable? If so, in what ways? RICHARD THALER: Well, I mean,
look, we know how to change it. For example, just this
year we’ve made it worse. So you know, we could
repeal that bill. That would at least get us back
to where we were a year ago. This relates to– I had this one glancing
point about salaries. My colleague Steve
Kaplan has never met a CEO who didn’t
deserve a higher pay. And that’s one point of view. There are others who
make comments like yours that it’s certainly the
case that the top 1% of the population, or
in the top tenth of 1%, have owned an
increasingly large share. But the argument
against doing this, of course, the argument
made on the right, is that we will not have any
innovation and entrepreneurship if we do that. The counter to
that is Bill Gates who seems to have
made a lot of money and wants to give it all away. Same with his buddy
Warren Buffett. So this is a problem– this is not a
structural problem. Now, obviously,
at the bottom end there’s a lot we
could do to increase the skills of the people
in the bottom quintile or what have you. But that’s much easier
to say than to implement. But this is eminently fixable. AUDIENCE: Thank you
for speaking today. I was wondering
if you can comment on how you see
neoclassical economics and behavioral
economics eventually merging into one discipline. Do you see it as a
qualitative overlay on the mathematical
models that people study in schools like this one? Or do you see it more of
as an explicit buildout into those mathematical models? RICHARD THALER:
Well, again, it’s hard to forecast the future. What’s happened so
far is we can divide empirical and theoretical. There’s been a lot
of work on trying to build more
behaviorally-realistic theoretical models. Probably the leading
practitioner of that field is my friend
Matthew Rabin, who I was giving the credit or blame
for the term “explainoations.” And his strategy has been
to study things piecemeal. So he’ll take one bit of
behavior like self-control and build a model of that. And there are still
is going to be some maximizing going on in there. And he’s leaving
everything else aside. And then he’ll move over
to expectation formation and build a model of that, and
again, ignore everything else. There will not–
here’s a prediction. There will not be a
new overarching theory of economic behavior. We have one. It’s the best possible one. It just doesn’t
work at predicting. And that comes out meaner
than I intend it to. We don’t want to throw those
away, because we still want– look, the theory
of the firm tells firms what they should do if
they want to maximize profits. It’s good to know that. Black-Scholes
formula tells you how option prices should be set,
and it works amazingly well. So that will continue. But at least for
the near future, like the next decade, what
we’ll see on the theory side is piecemeal. On the empirical side,
and economics has– the biggest trend in
economics over my career has not been it becoming
more behavioral. It’s become more empirical. And partly that’s because
of the existence of data. The field of financial
economics was created here at the University of Chicago
because some people created a data set, now
called the Chicago– CRSP, Chicago– I don’t
remember what it stands for. But all of a sudden
you had monthly data on stock prices
going back to 1926. And the field of financial
economics existed. Gene Fama would have
studied something else if CRSP hadn’t come along. So there are data sets like
that popping up everywhere. One of the most exciting
young economists in the world, Raj
Chetty, has become a master of exploiting
large administrative data sets that include tax
returns, anonymized of course. And to go back to this
inequality question, he and his team have done some
of the most exciting work just documenting that. So and the thing about
Raj Chetty’s work and the work of
many economists is that it’s kind of atheoretical. And I mean that in a good way. That if you’re
trying to figure out what parts of the country
people are more likely to escape the bottom to the top, that’s
an empirical question mostly. And the more people
just get their hands dirty with data, they– first of all, when they
do that, they find things like the health-care
stuff I showed you, and they’re kind of
confronted with the fact that people aren’t
choosing the best one. And so that’s the future
I see for the next decade, is continued exponential
growth in empirical economics with better and
better data sets. And I would say most
economists under 40 don’t think anything I said
tonight is very revolutionary. This was an argument
I was having with people my age and older. And the younger generation will
lead us to the promised land.