00:00:00
Katy Milkman: A lot of the psychology
in this space really relates to
00:00:02
sort of how I relate to my past
self and my future self.
00:00:05
Thinking about that carefully
led us to want to use the fresh
00:00:11
start effect, or this
motivation, to try to propel
00:00:14
people to save more for
retirement, to think more about
00:00:17
future me, to pursue those
goals. And so we thought fresh
00:00:21
start moments would be an ideal
time, because people have that
00:00:24
extra motivation to pursue goals
and to think about the future
00:00:27
and to feel disconnected from
past failings.
00:00:30
Welcome to <i>The Ripple Effect</i>, the
podcast that takes you on a
00:00:34
journey through the minds of
Wharton faculty. I'm your host,
00:00:37
Dan Loney, and in each episode,
we'll be diving deep into the
00:00:40
inspiration behind the
groundbreaking research that
00:00:43
Wharton professors have
conducted and exploring how
00:00:46
their findings resonate with the
world today.
00:00:49
- The decision to
head into retirement
00:00:51
not always an easy one.
People consider
00:00:54
this option to have to bring in
a variety of factors, including,
00:00:58
will they have enough savings?
Especially since people are
00:01:01
living longer right now. This is
an area of research focused on
00:01:04
by our guest here today, Katy
Milkman, who's a Professor of
00:01:07
Operations, Information and
Decisions here at the Wharton
00:01:10
School. Katy, great to talk to
you again, as always. Thanks
00:01:13
very much for your time.
00:01:14
Thanks for having me.
00:01:16
I think I'll start out with like—
when you— when people think about
00:01:19
the decision to make retirement,
obviously, all these factors
00:01:22
come in, but from the research
you've done and people you've
00:01:26
talked with, is it a harder
process to make some of these
00:01:29
decisions now than maybe it was
a few decades ago?
00:01:33
Yeah, things have changed a lot. So we
now live in a society, in the
00:01:37
US, where it's common to work
for an employer who offers you
00:01:42
what's called a defined
contribution plan instead of a
00:01:44
defined benefit plan. So it used
to be, basically, that you would
00:01:48
be guaranteed some amount of
income for the rest of your life
00:01:51
if you worked for an employer.
That's a defined benefit, even
00:01:54
after retirement. And there's
been a major shift over the last
00:01:57
30 or 40 years towards giving
people the opportunity to
00:02:02
contribute a defined amount and
maybe have an employer match to
00:02:05
a retirement fund, but then the
income they'll have in
00:02:08
retirement is a function of how
much they choose to save,
00:02:11
whether they potentially dip
into those funds prior to
00:02:14
retiring. And there's no
guarantee, because it depends,
00:02:18
really, on the performance of
those assets. So it's a really
00:02:21
different world, and it has led
to not great results, honestly,
00:02:26
in terms of the retirement
income people have to live on
00:02:30
who've worked their whole
careers. A lot of people are
00:02:32
ending up working longer or
living less comfortably than
00:02:36
they would have liked because of
this new era. That's my— my
00:02:41
takeaway. Though, I should say
I'm not an expert on either
00:02:43
defined contribution or defined
benefit plans, but that's my
00:02:46
rough understanding of the
change. And then the work I do
00:02:48
with my collaborators really
focuses on, okay, in this current
00:02:51
era where defined contribution
plans have become so common,
00:02:56
it's a big challenge to convince
people to save. We have to make
00:02:59
sure people are saving an
adequate amount, and we face a
00:03:03
lot of biases when we're trying
to make that pitch. And one of
00:03:07
them is present bias, the fact
that we tend to be more attuned
00:03:10
to the instant gratification we
get from, say, spending a
00:03:14
paycheck now, than saving it for
later. We focus more on the here
00:03:18
and now. That's what present
bias means, and undervalue
00:03:21
future everything. Future me.
Whatever money I'll have in the
00:03:26
future, it's worth less to me. I
think, you know, that's forever
00:03:29
away, and I don't value it as
much. And so that's a major
00:03:32
bias. It's something I teach my
MBA students about. And given
00:03:35
that we're fighting that uphill
battle when we're trying to
00:03:37
convince people to save for
future— essentially future me,
00:03:42
we try to use a lot of
psychological insights to— to
00:03:47
propel people to save what
they'll need.
00:03:49
Well, part of this discussion, as you
just mentioned, is around
00:03:52
savings being one of the big
factors. And you research this
00:03:56
through this idea of a fresh
start.
00:03:58
Give us an idea of what
it is.
00:04:00
Yeah. So this is a topic I've studied
with my former Wharton PhD
00:04:03
student, Hengchen Dai of UCLA.
And Hengchen and I look at this
00:04:08
idea that not only at the
beginning of a new year, but
00:04:10
there are lots of moments in our
lives that we feel motivated by
00:04:14
the sense that we have a clean
slate or a fresh start. So we're
00:04:17
most familiar with New Year's
resolutions as part of the fresh
00:04:20
start effect. At the start of
every new year, we think, "Oh,
00:04:22
it's a clean, it's a new
year. It's a new me.
00:04:25
I can achieve more." And we set
these resolutions. But we have
00:04:28
actually documented that this
phenomenon of feeling like we
00:04:31
have a fresh start and and are
more motivated to make change
00:04:34
arises at lots of other moments
too. At every birthday, on
00:04:38
Mondays, at the start of a new
month, following the celebrations
00:04:44
of any kind of sort of major
event that feels like the start
00:04:47
of a new cycle. So all sorts of
new beginnings give us that
00:04:50
fresh start feeling, which
increases our motivation to
00:04:53
pursue our goals and makes us
feel more disconnected from past
00:04:56
failures. Because we can say,
"That was the old me, and this is
00:04:59
the new me." So a lot of— a lot
of the psychology in this space
00:05:02
really relates to sort of how I
relate to my past self and my
00:05:04
future self. Thinking about that
carefully led us to want to use
00:05:11
the fresh start effect, or this
motivation, to try to propel
00:05:14
people to save more for
retirement, to think more about
00:05:17
future me, to pursue those
goals. And so we thought fresh
00:05:21
start moments would be an ideal
time, because people have that
00:05:24
extra motivation to pursue goals
and to think about the future
00:05:28
and to feel disconnected from
past failings.
00:05:31
So when you say fresh start
moments, you mean what exactly?
00:05:34
Yeah. So a fresh start moment is a
moment that's the beginning of a new
00:05:37
cycle in our life or on our
calendar, and they can be
00:05:40
personal, right? I might feel
that I have a fresh start on my
00:05:43
birthday, or maybe right after a
promotion at work, or after the
00:05:47
birth of a child. Those are all
major turning points in a
00:05:49
person's life. We— we tend to
think about our lives like we're
00:05:52
characters in a book, and we're
living through these chapters.
00:05:55
And the chapter breaks are not
linear, necessarily, right? When
00:05:58
you move to a new community, you
take a new job, those are
00:06:02
chapter breaks too. But these
chapter breaks at the start of a
00:06:05
new year or a new season or a
new month or a Monday or a
00:06:08
birthday, those are also small
fresh starts, or sometimes big
00:06:12
fresh starts, that make us more
motivated to pursue our goals.
00:06:15
And so employees were receptive
to— to this, and looking at this kind of
00:06:19
moment in their life as a fresh
start and a great way to kind of
00:06:23
maybe head down a different,
different path in their lives?
00:06:26
So what we did to explore that
is we ran an experiment with
00:06:31
thousands of employees who were not
yet saving for retirement, or
00:06:34
who were saving but at a very
low rate, a rate that was well
00:06:37
below what they would need to
save to have a comfortable
00:06:39
retirement. We partnered with
four different organizations and
00:06:43
sent mailings to employees who
were in these categories. Non-
00:06:47
savers, or a small number of
very low savers. And we tried to
00:06:51
use this insight about fresh
starts to increase the
00:06:54
likelihood that people would
save. What we did is we sent
00:06:56
mailings that invited people,
certainly if they were up for
00:06:59
it, to start saving right away.
But we know that people like to
00:07:02
procrastinate on anything that
sounds difficult, like starting
00:07:06
towards a savings goal. So we
also invited them an opportunity
00:07:09
to delay. We said, "You could
also, if you don't want to save
00:07:11
now, start saving on this future
date." And what we randomized in
00:07:15
our experiment is, some people
the future date was a fresh
00:07:17
start date, and for some people
it wasn't. So for instance, if
00:07:21
you had a birthday, Dan, coming up
in two months, we might flip a
00:07:24
coin and decide, are you going
to be invited to start saving
00:07:27
now or after an upcoming
birthday, or after your next
00:07:29
birthday? That would be what
your mailing might say in one
00:07:31
condition. Or— and this is what's
kind of really tightly
00:07:35
controlled about it— we would
say, "Would you like to start
00:07:37
saving now or in two months?" So
actually, in both cases, it's
00:07:41
exactly the same offer. But in
one case, we've tied it to your
00:07:44
birthday, which makes it more
clear that this is an
00:07:48
opportunity we think you might
want to align with that fresh
00:07:51
start moment. And we tried this
with several fresh start dates.
00:07:53
We tried aligning New Year's, so
inviting you to start saving
00:07:58
after a new year. We invited you
to start saving after the start
00:08:00
of spring, and we invited you to
start saving after a birthday.
00:08:04
So you'd just get one of these
offers. It was random assignment
00:08:07
in this trial, but we tested all
of those different Fresh Start
00:08:10
opportunities. And we tested it
against just inviting people to
00:08:13
save at an equivalent time
delay. Or we also tried some
00:08:16
dates that don't feel so much
like fresh starts, just as sort
00:08:19
of— we think of them as placebos.
So Valentine's Day. It's a— it's
00:08:22
also a day that is sort of
notable on the calendar that we
00:08:25
can label. But nobody typically
thinks of Valentine's Day as a
00:08:29
fresh start moment, unless— you
know, if you met the love of
00:08:31
your life, then maybe for you,
it's meaningful. - Right. - But for most
00:08:35
people, that's not a day that
feels like the beginning of a
00:08:37
new cycle in life. So we ran
this experiment, and what we
00:08:40
found is that when we invited
people to begin saving after a
00:08:44
fresh start date, and we compare
what happened to people's
00:08:47
savings rates who were invited
to save now at an equivalent
00:08:51
time, but without that fresh
start date call out, we see
00:08:54
more savings, significantly more
savings, over the next eight to
00:08:57
nine months. We look at savings
rates, and we see, depending
00:09:01
on how you model it, a 20 to 30%
increase in the next eight
00:09:05
months savings among the
population that is invited to
00:09:09
start saving after a fresh start
date. So this does seem to
00:09:12
motivate more people. It doesn't
lead them to decline saving now,
00:09:16
and it increases the total
number of people who are saving,
00:09:19
leading to this higher savings
rate over the subsequent months.
00:09:23
And I think it's also interesting,
because you mentioned in the
00:09:27
paper that we tend to make
decisions, in many cases, that
00:09:31
don't help us in the long term.
So I guess by doing this, and
00:09:35
maybe there's a little bit more
immediacy to it, that maybe to a
00:09:38
degree, we're also changing
behaviors as well.
00:09:42
That's right, we are— we're— we
are changing behavior by getting
00:09:45
people to recognize this
opportunity is one that's
00:09:48
aligned with their goals and
sort of— and seeing, yeah,
00:09:52
actually, I do want to start
saving following my birthday.
00:09:55
That— that sounds exactly like
the right moment to do this. Or
00:09:58
yeah, at the start of spring
really does feel like a moment
00:10:00
when I should be upping my
savings contributions. But if
00:10:03
I'd asked you, "Do you want to
start saving next month?" which
00:10:06
happens to be the start of
spring, but I haven't called
00:10:08
your attention to it, you don't
have that same resonance, and
00:10:11
you don't make the same
decision. So it's changing
00:10:13
behaviors. It's changing long-
term outcomes for people by
00:10:16
increasing their savings rates.
And we think that's a really
00:10:19
important insight. And the more
we can leverage these kinds of
00:10:22
moments that people see as fresh
starts to increase savings or
00:10:26
any other goal-directed
behavior, the better.
00:10:30
You've also done research that
looks at the component of peer
00:10:33
information in retirement savings
decisions. Tell us about that.
00:10:37
Yeah. So this was another
randomized, controlled trial, also inviting
00:10:41
people who were either non-
savers or low savers to save. So
00:10:44
there's a pattern here. These
are projects— and there's a— I
00:10:47
should also say there's a common
co-author on both of these. John
00:10:51
Beshears of Harvard Business
School is a fantastic
00:10:53
researcher, does a lot of work
on retirement savings, and was
00:10:56
involved in both of these
projects. And so that's— that's
00:11:00
the other common variable. We,
in this case, partnered with one
00:11:05
big company that had a lot of
employees who had been part of a
00:11:10
union, and as a result of their
union bargaining, they had not
00:11:13
been automatically enrolled in
the— the savings program, the— the
00:11:19
contribution program that they
had on offer. So it's a 401(k)
00:11:23
plan, where you put a portion of
every paycheck into this plan,
00:11:26
and you get a tax benefit when
you do so. That money is not
00:11:30
taxed, it's put in before taxes.
So lots of people were not
00:11:35
saving, because if you're not
automatically enrolled, they had
00:11:38
to take steps to start saving,
and a lot of people don't bother
00:11:41
to do that. So we tested this
both with those union employees
00:11:45
whose union had not negotiated
for them to be automatically
00:11:47
enrolled, as well as a non-union
population. But the non-union,
00:11:52
non-savers, were different
because they had— they had
00:11:55
intentionally opted out of
savings. So they're— there's
00:11:58
sort of selection bias, if you
will. That's our nerd term for
00:12:01
saying they're slightly
different populations.
00:12:03
So that effect that you saw play
out, was it similar for pretty
00:12:06
much everyone across the
company?
00:12:09
So, it wasn't. It actually turned
out to matter what— which
00:12:13
population they were in. And what
we were testing was whether or
00:12:16
not telling them about how many
of their peers were already
00:12:19
saving, or were already saving
at a higher rate, whether or not
00:12:22
that might increase their
savings likelihood. So we tried
00:12:26
actually two things. One, we
varied whether or not they got a
00:12:29
mailing that told them about the
high number of their peers who
00:12:32
are already enrolled and
encouraged them to follow suit.
00:12:35
And then the other thing we did
is that we varied what number
00:12:40
they saw. So we were never
deceptive, but we randomly
00:12:43
assigned people to either find
out about the savings rate of
00:12:46
peers in their five-year age
cohort— so you know, if you're
00:12:50
in, you know, age 40 to 45, you
might find out about others age
00:12:54
40 to 45, or in their 10-year age
cohort. So if they were age 44
00:13:01
they could also see the 40 to 50-
year age bucket instead of the
00:13:05
40 to 45 age bucket. What that
did is it meant we had some
00:13:09
variation in the number people
saw, so we could test not only
00:13:13
what's the impact of finding out
how many of your peers are
00:13:15
saving, but also, what is the
impact on finding out a slightly
00:13:20
different number when you hear—
when you learn that your peers
00:13:23
are saving, right? So when
you're in the 40 to 45 age
00:13:26
bucket, you might find out 75%
of other 40 to 45-year-olds are
00:13:30
saving. But if you saw the 40 to
50 age bucket, you might find
00:13:33
out that, you know, 80% of folks
in that group are saving. So you
00:13:38
see a different number. So we
have these two ways we can look
00:13:41
at what's the impact of peer
influence. One, finding out that
00:13:44
lots of peers are saving and
two, what's the actual number
00:13:47
you see? And how does that
matter? And what was really, I
00:13:50
will say, surprising, given what
we know about how influenced we
00:13:53
are by our peers, is that in
this study, what we found is a
00:13:56
backfire effect. Meaning, for— and
specifically among union
00:14:00
employees. So this was the group
we expected to be most malleable
00:14:04
because they hadn't previously
made an active decision about
00:14:07
retirement. They just passively
not signed up. This group, when
00:14:11
they saw a peer comparison and
learned, hey, you know, 75% of
00:14:16
your peers are saving. That
reduced the likelihood that they
00:14:20
chose to save. In addition, when
we look at the specific number
00:14:25
they saw, we see the higher the
number they observe, randomly
00:14:29
assigned, the less likely they
are to save. And again, this is
00:14:32
two things that go against what
we normally expect to see in
00:14:34
terms of peer effects. Because
one, when I find out everybody
00:14:37
else is doing something and the
majority of other people are
00:14:39
doing it, I normally decide to
do it too, at a higher rate.
00:14:41
- Right. Right. - And two, the higher the
number of my peers doing
00:14:44
something, the more likely I am
to want to join. But we see the
00:14:48
opposite in both cases. So this
was really puzzling. We don't
00:14:52
see this among people who
weren't members of the union, so
00:14:55
they hadn't actively opted out.
There, we don't see any effect.
00:14:58
But we wanted to dig into this
backfire effect. And I want to say,
00:15:01
first of all, I still feel that
we don't know for sure what
00:15:05
happened. But our best
explanation at this point, based
00:15:08
on additional analyzes we ran,
is that it seems to be sort of
00:15:11
an upward social comparison
reaction, where people are
00:15:16
feeling like they could never
possibly catch up, because it's
00:15:19
driven by lower-income members
of the population. So when we do
00:15:23
a median split on earnings and
look at people who are below
00:15:27
median earners in all the
different states around the
00:15:30
country where this company has
employees who received our
00:15:33
mailings, we see that the effect
is really driven by the lower-
00:15:36
income folks. And that leads us
to conclude, potentially, this
00:15:42
is driven by that sense that I
can never catch up. So you know,
00:15:45
the idea of social norms is, I
want to keep up with the
00:15:48
Joneses, so I'm going to try to
do what the Joneses are doing.
00:15:50
But if you feel you can't
possibly keep up with the
00:15:53
Joneses because their income is
so much higher and they're
00:15:56
already way ahead of you on so
many dimensions, it may just be
00:15:58
demotivating to hear, yeah, they
bought another luxury car, and
00:16:01
they saved more for retirement,
and so on. And it may make you
00:16:03
feel that it's hopeless. And so
that's our guess as to what
00:16:06
happened in this particular
study. And it was very
00:16:08
disappointing, but also useful
to know.
00:16:11
So is there kind of a next logical
step that you would like to
00:16:15
take, having learned all this
and done this research about,
00:16:19
you know, deeper understanding
about retirement decisions?
00:16:22
Well, the answer to everything
right now, Dan, is AI. You know,
00:16:24
that's the answer to everything
we want to do next. So a lot of
00:16:28
the work that my collaborators
and I are talking about in the
00:16:30
space of behavior change is
related to, how can we use these
00:16:34
large language models to
incorporate some of the best
00:16:38
behavioral insights we know into
dialogs to help people make
00:16:42
better decisions? So that's the
next natural step. And certainly
00:16:45
the LLMs that we train to try to
help support people's retirement
00:16:48
savings ambitions will be armed
with the knowledge from this
00:16:52
research that, you know, it may
not be as effective as we
00:16:55
thought to use social norming,
particularly on low-income
00:16:59
consumers, when we're trying to
encourage them to save more for
00:17:03
retirement, and that it can be
effective to leverage fresh
00:17:06
start dates as moments when
people feel it's appropriate and
00:17:10
optimal to begin saving.
00:17:11
Katy, always great to talk to you
and discuss your research. Thanks
00:17:15
very much.
- Always great to be here.
00:17:16
Thank you so much for the great
00:17:17
questions and the opportunity.
00:17:19
Thank you. Katy Milkman, who's a
Professor of Operations,
00:17:22
Information and Decisions here
at the Wharton School.
00:17:24
Thank you for listening to <i>The
Ripple Effect</i>. We hope you found this
00:17:27
episode informative and
engaging. Don't forget to
00:17:30
subscribe and leave us a review
so that we can continue to bring
00:17:34
you the best insight from the
Wharton School.
