The psychology of scarcity: Why having too little means so much

♪ [upbeat violin and piano] ♪

♪ ♪

>[ ELDAR SHAFIR]
It's a pleasure to be here.

I'm going to talk to you about

how scarcity imposes demand
on our cognitive capacity.

In other words, on how the psychology
that comes from not having enough

demands attention
and cognitive capacity

and leaves us with
less mind for other things.

I want to include— I want to quickly
mention scarcity of money, certainly,

poverty, scarcity of time,

scarcity of calories
if you're dieting, water, etc.

And the argument is going to be then,

when you don't have
enough of something,

it captures your attention,
it captures your mental capacity,

and leaves you with less.

You'll see that how all this
connects actually very nicely

with the talks we had before.

We all know that attention is limited.

One of the things that's interesting

is that we fail to appreciate
the extent to which it's limited.

Many of you may have seen recent
research on using cell phone in cars.

We all know it's not so good.

It's a disaster, as it compares to
being legally drunk by U.S. standards.

This is a classroom in New Haven.

It's a lovely study that was done,

where they noticed that the fifth graders
in this class, randomly assigned,

happened to find themselves
either on the quiet side of the school

or on the other side of the school
where trains were going by.

And so, when you sit in a class
and this train goes by,

there should be a sound.

If you could hear it,

that has an enormous distraction
on kids' attentional capacity,

and what they found is that
the kids near the train tracks

were one year in academic
performance behind their friends

on the other side of
the school at fifth grade.

They then installed soundproofing,

and a couple of years later,
they were matched again.

So, this is just a passing train
with an enormous impact,

and the point is that these
are trains in the external world.

You don't need them
to be in the external world.

Imagine yourself in an office,
very quiet, no distractions,

except you're worried about paying rent
or a child who is sick or anything else.

Those internal trains are going to
have an impact that could compare,

in terms of the attentional distraction
and the cognitive capacity,

to having distractions on the outside.

An amazing study
that was done in the ’40s

had to do with when the allied forces
realized they were about to inherit

a lot of hungry people in Europe,

and they had no idea
how to feed them.

Feeding hungry people
is a non-trivial task.

A famous researcher
into nutrition in Minnesota

conducted a series of studies.

These are conscientious objectors who,
since they're objecting to “the Good War,”

were very eager to volunteer
to these studies.

These are capable young men
who volunteered

to starve, not to death,
but to immense discomfort.

And when you look at
the descriptions of what happens,

you know, there is
the obvious physical stuff.

They're too tired to keep
their hands up to wash their hair.

They need pillows to sit down
because their butts are too naked.

But the amazing stuff
also is the psychological.

These are young talented men

who spend their entire time

planning to open restaurants,
reading recipes,

comparing prices of foods
in different stores in the newspapers.

That's all they're thinking about.
It's taking all their cognitive capacity.

It's about food.

At some point, the researchers
decide to distract them

by showing them some films,
and these guys describe the films,

and they couldn't care less
about the love scenes.

They want to see the meals.

And what's important here is,

this is not something
they're choosing to do;

in a sense, they'd rather
do something else.

It just imposes itself on their minds
and it's very hard to avoid.

So the notion is that cognitive capacity,
which is a very limited resource,

is captured by when you
don't have enough of something,

and you basically find yourself
automatically paying a lot of mind

to that thing you don't have.

Now, a footnote, if this works well,

you should have these
annoying blinks on the screen.

If it was darker in here,
it would be very annoying.

These blinks that you're seeing
last 350 milliseconds.

That's how long you get
darkness every time you blink.

You blink approximately
15 times a minute.

During the 12 hours
that you're awake every day,

you get one hour of nothing
but dark input into your eyes,

which you have never seen.

That's to say that your brain
is doing amazing stuff

to which you have no introspective access.

And that's sort of important

because we walk around
with this amazing machine

and we think we know what it's doing.

You know, we know roughly,
like we know what our liver is doing,

unless you're a physician.

And so, basically, a lot
of the behavioral research,

and I'm saying this because it's very
important when you get behavioral research

not to do that,
“Yeah, does it feel right to me or not?”

The argument is a lot of what
we discover in behavioral research

is about a mechanism
that's magnificently sophisticated,

that's behind the eyes
and between the ears

to which you have very little access.

I want to talk about particularly about
poverty and a little bit about time,

but again, the notion is
that scarcity in food, friends, water

will capture your attention in this way
and be very profound.

Nice studies have been shown,
for example, that it's top of mind.

So, you get subjects into a laboratory

who have been asked
not to drink for 4 hours.

They're now very thirsty.

Half of them are randomly
assigned to get water,

the other half get pretzels.

Not a good idea when you're thirsty.

Then they sit in front of a computer
and have to identify words.

These are words that are flashing

roughly as long as you had the [blink]
(it's 400 milliseconds),

and you have to decide:
yes, it's a word or not it's a word.

And what you find is, when you
get words that are related to thirst,

those who have had pretzels

identify these words much faster
than those who've had water.

If you compare the performance
to neutral words, they're identical.

Again, this is not at the level--
this is pre-decisional.

It's way too fast.

It's literally showing that the words
that are related to thirst

have higher semantic activation.

They're literally top of mind.

They're there to be tickled the minute
anything happens that's relevant,

as opposed to other words.

And what the notion is,
that when it's top of mind,

loneliness, for example,
there's nice research showing

(it's kind of a sad research,
but it's very compelling),

when you're very lonely and
loneliness is mentioned or brought up,

you function less well.

It's distracting.

So, the sort of tragic impact
that loneliness has

is that people are socially
as capable as anybody else,

[but] when they think about loneliness,

such as, for example,
when you're interacting with somebody,

what really matters to you to do well,
that's when you do the least well

because you have cognitive capacity taken
by the concerns related to loneliness.

Dieting and [Cinnabons?], very similar.

Here was a study within California.

This was a study done
with dieters and non-dieters.

We have them play fun word searches.

You see these are words

where the odd-number words are
food-related (cake, doughnuts, sweets)

and the even-number words are neutral.

There's two conditions: they get either
this one or they get another condition

where now the food-related
words have been replaced

by other irrelevant, neutral words.

We have the dieters look
and the non-dieters look

to find all these words.

And what I'm going to do now
is look at how long it takes them

to find the neutral words,
the words that are common to both.

So basically, I'm asking: How long
does it take you to find the word “cloud”

when it's preceded by a doughnut
or when it's preceded by a picture?

And what you see is
that [in] the non-dieters

it makes no significant difference.

The dieters take twice as long to find
“cloud,” having just seen a doughnut,

than having just seen a neutral word.

[light laughter from audience]

It literally interferes
with what you do next.

So, it's top of mind, and when it's
triggered, it sits there, and it lingers,

and it takes cognitive capacity
for what else you want to do.

Financial poverty is
the ultimate of all scarcities.

We've talked about
a lot of other scarcities.

It's important to keep in mind,
this is one that's nondiscretionary.

I'll talk briefly about the busy.

You know, when you're very busy,
you don't have time.

We all talk about how busy we are.

Don't tell anybody, but between us,
if you stop doing what you're doing,

nothing would happen [chuckles]
except for a few physicians in the room.

[light laughter from audience]

But basically, it's discretionary.
We choose to be busy.

When you're poor –

and we can get into
interesting discussions about

what is the minimum that you need
and how that changes culturally,

but when you don't have enough,

when you can't pay rent,
when you can't put food on the table,

that's not discretionary,
and you can't take a break.

When you're a dieter
(there's really nice research on this)

and you have a very important project
this week, you say, “You know what?

This week, I'll just do what I need to do.
I'll go back to dieting next week.”

Notice you can't say,
“I'll be rich this week.

I'll go back to being poor next.”

Not available.

So, it's a much more chronic,
permanent, and imposing condition.

And what we did in this case is,

we ran some studies
in a mall in New Jersey.

We go to people in the mall,
ask them to participate.

They sit in front of a computer,
we give them classic cognitive tests

that have been used in cognitive
science for the last 30 or 40 years.

One is a test of cognitive control,
divided attention of sorts.

So, you sit in front of a computer.

When you see heart,
you have to quickly press the same side.

When you see flower,
you have to press the opposite side.

You're getting a headache right away.
It's confusing. It takes a lot of attention.

The other-- All of you have seen if you've
done an SAT or GRE, any kind of test,

it's a classic test that's supposed
to capture fluid intelligence.

It's a major component of the IQ test.

Basically, what shape fits best
in the missing space?

So, they sit and do these tests.

And while they're doing these tests,

we give them
financial scenarios to contemplate.

Your car breaks down, you have to think:
Are you going to take care of it?

In one condition,
the non-menacing condition,

the car is going to cost $300 to fix.

In the other condition,
the challenging condition,

it's going to cost $1,500.

As you're thinking about:
Are you going to take care of the car?

You do these, and then
when you finish them, you tell us:

how you’re going to solve
your financial problem, okay?

Finally, we'll get people's annual
household income and divide them

by basically splitting in half
into rich and poor, okay?

So, we know you're either rich or poor,
we'll give you these problems,

either a cheap car to take care of
or an expensive car to take care of,

and you do these cognitive
control and intelligence tests,

what do you get?

Let's look first at the rich.

The rich, when they contemplate
the easy or the difficult car,

perform equally well
in the cognitive control.

It's like a driving test.

The poor, when they
contemplate the easy car,

the one that they can easily afford,

look exactly like the rich.

But when they're thinking about
a car that's hard to manage,

that's hard to afford,

they're now driving
significantly less well.

The cognitive control has diminished.

Let's move to IQ.

The rich are not impacted
by what kind of car they're fixing

while they're taking
these cognitive tests.

The poor, when they're thinking about
the car that's easy to take care of

look indistinguishable from the rich.

But when the poor are worried

about a car that's presenting
a real financial challenge,

they're performing here--

We've done this now
four different ways,

replicated four times
with financial incentives,

they're losing the equivalent
of 13 IQ points.

So, these are the same people
who, minutes ago,

when the car was manageable,
performed just like the rich.

Now that the car presents a serious
imposition (takes a lot of mind),

perform significantly less well
on these IQ tests,

on these fluid intelligence tests.

We did all the controls we could
in New Jersey,

but these are, at the end of the day,
different people.

They're richer and less rich.

They have had different education

and they have different heart rates
and everything else.

The dream was, could we do this
within subjects? The same person.

It's not so easy to get people,

you know, hundreds of them
who are both rich and poor,

but the third world does
present cases like these,

and this is the best one we found.

These are sugarcane harvests
outside Chennai in the fields in India,

and the sugarcane is
a particularly good case

because you harvest only once a year.

These are all people
whose bulk of their income

is received only once a year.

They're rich after the harvest;

and because they're living
relatively tight lives

and have a hard time
smoothing their consumption,

they find themselves poor
before the harvest.

So, we run the same people
now four months apart,

two months before harvest
and two months after harvest,

A lot of nuances that make this
very nice because the mills,

because they're over capacity,
they tell the farmer when to harvest.

So, the harvests are spread
over many months.

We get rid of all kinds
of months effects, etc.

What you find, essentially,
like in New Jersey,

the same people score
significantly less well

on cognitive control
and fluid intelligence tests

before harvest when they're poor

as opposed to after harvest
when they're rich.

And now, of course, we've kept health,
education, everything else intact.

It's literally a function of
whether you have enough or not

and how much cognitive capacity
that takes from you

as you're trying to do these stupid tests.

Time-- I'll spend less time today
on time, but many of us,

many of you are time-poor in ways
that are not completely different

from the way that our subjects in India
and New Jersey are money-poor.

You have to think trade-offs.

You have to borrow from tomorrow

to do things that you haven't had
time to do today at high interest, etc.

What we did here is run—
These are Princeton students.

Nobody would blame them for being myopic.

They're highly educated,
they're very sophisticated.

We have them play a game, a classic
many of you know, Family Feud.

It's a game where they're
trying to get the answers.

They're very eager to do well.

The performance gets
translated into payoff.

And we make them randomly
either time-rich or time-poor.

So, you either have 50 seconds
per round or 15 seconds per round,

which is not quite enough.

And in some conditions,
you cannot borrow.

So, when you're out of time,
you have to move to the next round.

In other cases, we allow you to borrow.

We offer you payday loans,
basically predatory lending rates.

You can borrow at high rates.

So, every second you take now,

you have 2 seconds less
left at the end of the game.

We look at how they do.

What happens? There's two measures.

There's rounds completed
and points earned.

I'm not going into the details.

This is what you see
when there's no ability to borrow.

So, when you cannot borrow,

notice the rich obviously play
more rounds and get more points

because they're rich,
they're playing more games.

Now I'm going to let you borrow.

Now, if you run out of time,

if you want to,
you can take more seconds.

The rich notice—

(This should come on at a time.
Maybe I don't have it.)

[For] the rich, notice,
there's no impact at all.

There is high-interest
borrowing available.

Do I want to borrow?
I look and I say,

“No, it's not really useful.
It's not really worth it.

I'm not doing it.”

The poor (same Princeton students
just with less money),

are dying to get this right,
are running out of time,

are focused on what
they're doing right now.

The periphery gets less mind,
and they borrow.

And they run out of time too quickly,

and they score and they
leave the experiment

with substantially less money.

So, we're basically seeing
the sophisticated Princeton students

taking payday loans of the kind
that we typically claim

the poor are doing out of
lack of insight and intelligence.

It looks very much
not a function of who the person is,

but the context you put them in
that makes them act poor

in ways that minutes earlier

in the mall, in the fields of Chennai,
in the Princeton lab,

you avoided if you were
just a bit richer.

Implications, we all need to take
bandwidth a lot more seriously.

We have a very limited mind
when it's devoted to some areas --

and in poverty, it's a massive one
that never leaves us alone --

there is just less mind
to pay attention to other things.

And policymakers
typically don't think this way.

So, you saw Phil's FAFSA application.

Think about a poor person
who comes for some benefits program.

Imagine I propose—
Let's charge them $300 to join.

You'd say to me,
“Wait, they have no money.

They're coming for help.

Why would you charge them money?”

Well, they also don't have bandwidth.

And when we charge them massive bandwidth

to attend all kinds of events
at the right hour

and fill out complicated forms

and do all the things that exactly
are what they don't have enough of,

namely bandwidth
and attentional resources,

we're kind of doing the wrong policies.

So, Phil's FAFSA, beautiful study
that you saw, is one example

where you basically could
alleviate some bandwidth limitations.

A lot of work has been done on defaulting
workers into retirement savings

as supposed to expecting them
to do it on their own.

That has a big effect among
people who are too busy.

And in healthcare,
there's a massive effect.

So, this is a glow cap.

A glow cap is a $12 plastic bottle
that delivers you medication,

and it's structured so that
if it's not open at the right time,

it blips and blinks and
sends you an email saying,

“I haven't been opened in time.“

And it turns out there are some
estimates in the third world

where HIV is one case where
you can't just take 60% of the time.

You have to take all of it all the time.

Some estimates that $12 block cap

is having a two-decade
life expectancy impact.

And clearly, it's a case where people
want to take it, they intend to take it,

but bandwidth is limited, and they forget

And so, if you take that more seriously,
there's an enormous amount you can do.

Last thought, I'll leave you just
a picture that might stay in your mind.

In aviation, it's become
very clear very early on

that you can't train pilots any better.

That's all they can do.

As avionics get more complicated,

it's up to you to design
more sophisticated cockpits.

And the argument here is, if you look
at people managing their poverty,

there is scarcity in different resources,

it's up to us, policymakers and others,
to design basically a cockpit a life

that's more manageable
given severely limited resources.

Okay, I'll stop here.

[audience applauds]