0:00

Cologent isn't just for people who want nice

0:02

hair, skin, and nails. It's also

0:04

for people who want nice joints, and needs,

0:07

and bones, and limbs, and

0:09

wrists, and arms, and index fingers,

0:12

and shoulders, and tipulas, and

0:15

ears, and hands, and fingertips.

0:18

Toes and hips and power vital

0:20

proteins for everybody with a body.

0:23

These statements have not been evaluated by the Food and

0:25

Drug Administration. This product is not intended

0:27

to diagnose, treat cure or prevent any disease.

0:31

As if the Macrispy couldn't get any better,

0:34

bacon and ranch just entered

0:36

the chat. The bacon ranch

0:39

with crispy, available and participate

0:41

in McDonald's for a limited time.

0:51

Hey there, it's Steven Dubner. Do you

0:53

ever wonder how your boss became

0:55

a boss in the first place? Yeah.

0:58

We do too, especially when

1:00

things go sideways. In retrospect,

1:03

it's easy to say that Sam Bankman Freed

1:05

was not the best person to be

1:07

running FTX, the bankrupt

1:09

cryptocurrency exchange. But retrospect,

1:12

is always easy. What about right

1:14

now? Do you have

1:17

a bad boss? Have you ever had one?

1:19

Chances are Yes. So

1:22

I hope you enjoy today's episode, which

1:24

we first published last year.

1:26

It's called why are there so many

1:29

bad bosses? My

1:36

name is Katie Johnson, and I'm

1:38

a data scientist. Johnson

1:40

is thirty two years old and lives in London.

1:42

She grew up near Bristol, went to university

1:45

in Birmingham, and since then has held

1:47

series of increasingly impressive jobs

1:49

at a series of companies. These were

1:51

all what are known as I c jobs,

1:54

I c standing for individual

1:55

contributor. Which means what?

1:58

It is someone who makes

2:00

as opposed to managing people who

2:03

make Johnson loved

2:05

being an IC. She loved analyzing

2:07

data, and she was really good

2:09

at her job. But after a while,

2:12

she thought it might be nice to come

2:14

a boss. Yeah. I wanted to manage

2:16

more and more people. And you wanted to

2:18

manage more people because why? You were

2:21

just power hungry like the rest of us?

2:23

I think there's a couple of reasons. So the

2:25

first is that I wanted to start

2:27

getting more autonomy over what I was

2:29

working on. I would be working on staff

2:32

in my IC role and I think this isn't the most

2:34

important thing. And I thought that if

2:36

I became the leader of the

2:37

team, then I would get to pick. What

2:39

I worked on.

2:40

Okay. That seems sensible.

2:42

The other reason was to have

2:44

more impact at the companies I was working

2:47

at, so you could describe this as having a

2:49

seat at the table. Also sensible.

2:52

I guess the final reason is that we

2:54

all kind of, not everyone, I guess, but

2:56

I was included in this, have a concept that

2:58

being more successful means being more senior.

3:01

And so in order to not

3:03

necessarily show others, but definitely myself

3:06

that I had achieved and become successful,

3:08

I needed to keep moving upwards within

3:10

a

3:10

company. Johnson's father

3:12

in his own career had seen things differently.

3:15

So my dad has been a network

3:17

engineer. He recently retired, but he's been that

3:19

for his whole career, and he had absolutely no

3:21

aspiration to become the

3:23

manager. He's like, why would I want to do that?

3:24

But Katie Johnson did want

3:27

to become a manager and several

3:29

firms were willing make her one. She

3:31

took the most appealing offer at

3:34

a software firm that helps companies acquire

3:36

new customers.

3:38

And I was sent on some management training

3:40

and had to do what can only be described

3:42

as a very long personality

3:44

test. And the idea was to tell

3:46

me what I was good at being good at.

3:49

And what was she particularly good at?

3:51

Critical thinking, attention

3:53

to detail, courage, all these

3:55

internal thinking type characteristics.

3:58

You can see why Katie Johnson would seem to

4:00

be a great boss. Her

4:02

new job title was head of data

4:04

and analytics. She had roughly ten

4:06

people reporting to her. The

4:09

promotion came with more money, more

4:11

prestige, more leverage to set

4:13

the agenda. Also however,

4:15

more meetings,

4:17

Oh, so many meetings,

4:19

like compared to being a data scientist, I'd

4:21

maybe have a half hour meeting in the morning

4:23

and then I'd just be free. To do

4:25

coding and thinking and making

4:28

stuff. But I was in

4:29

meetings. I think Tuesdays, I

4:31

used

4:32

to be in meetings for, like, seven hours

4:34

No offense, but did you not see that coming?

4:37

No. I really didn't. I thought

4:39

it would just be like my normal data scientist

4:41

job of a few one to ones on the side.

4:44

That was okay because it's quite interesting. You're talking

4:46

about the work you get into quite depth and problems

4:48

with my team. It's more like the meetings

4:50

like an hours coffee someone to try

4:52

and set up a better working relationship with their

4:55

team times that by like five or

4:57

ten other teams. It's just draining.

5:02

Keep in mind, this was happening during the pandemic

5:05

shutdown, so all these meetings

5:07

were virtual. And as drained

5:09

as Johnson was from all those

5:11

meetings, she was getting good reviews as

5:13

manager.

5:14

Yes. People would tell me what a great job I was

5:16

doing. I was coming across well.

5:19

But she found that being a boss

5:22

made her miserable. I

5:24

would finish my day in my study walking

5:26

to the lip room, put a blanket over

5:28

my head and cry because I was

5:31

in so much pain

5:33

at high board I was. In

5:35

retrospect, Katie Johnson had

5:37

plainly aired in wanting to become

5:39

a boss. But she'd also felt that

5:41

management was the only sensible way to

5:43

advance her career and If

5:46

you look at how most firms and institutions around

5:48

the world operate, you'd have to agree with

5:50

her. The question is, does

5:53

this standard operating procedure produce

5:55

good bosses or bad bosses or

5:57

even horrible ones. The horrible

6:00

boss is a familiar caricature. We

6:02

all know the stereotypes, the screamer,

6:05

the sadist, the idea stealer,

6:07

the passive aggressiveness, These

6:10

are some of our most enduring characters

6:12

in film. You remember Blake

6:14

from Glen Gary Glen Ross played by

6:16

Alec

6:17

Baldwin? Put that coffee

6:19

down. Coffee's

6:21

for clothes is only.

6:24

Do call yourself a salesman who's son of a

6:27

Or in the film office space when Peter

6:29

is trying to escape the office on Friday

6:32

afternoon and he gets snagged by the

6:34

boss. Hello, Peter.

6:36

What's happening? I'm

6:40

gonna need you to go ahead and come

6:42

in tomorrow. So

6:44

if you could be here around

6:48

nine would be great. Okay.

6:51

Oh, oh, and I almost forgot. I'm

6:55

also gonna need you to go ahead and come in

6:58

on Sunday too. Then

7:00

there's Miranda Priestley played by

7:02

Meryl Streep in the devil wears

7:04

Prada.

7:05

You have no style or sense of

7:07

fashion?

7:09

Well, I

7:12

think that depends on what's your No. No.

7:15

That was a new question. The horrible

7:17

boss Motif is so attractive that

7:19

the director Seth Gordon made a film

7:22

called horrible

7:23

bosses.

7:24

Yeah. We gotta trim some of the fat around here.

7:25

Trimmer. What do you mean by trim the fat?

7:28

I want you to fire the fat people. Truly

7:33

horror irrible bosses do occasionally turn

7:35

up in real life, especially in Hollywood

7:37

itself. The producer Scott Ruedin,

7:39

for instance, has been accused of years

7:42

worth of alleged abuses, like

7:44

smashing an assistant's hand with

7:46

a computer monitor. But

7:48

even in Hollywood these days, and especially

7:51

in more normal industries. This sort

7:53

of grotesquery is harder to get

7:55

away with. Bosses who are

7:57

outright monsters are more likely

8:00

to lose their jobs. But how much

8:02

attention are we paying to the more common

8:04

type a bad boss. Someone who's simply

8:07

incompetent or overstretched

8:09

or even just miserable being a

8:12

boss like Katie Johnson was.

8:14

Do you even know how many bad bosses

8:16

are out there? The more

8:18

you dig, the more you learn, the the

8:20

science of boss behavior is not very

8:22

scientific. One Gallip poll

8:24

shows that roughly fifty percent of American

8:26

employees have, at some point in their career,

8:29

left a job because of

8:31

a bad boss. But an employee

8:33

might have ten or twenty bosses

8:35

over a career, so Maybe that

8:37

number isn't so bad. A survey

8:40

of European employees found that

8:42

only thirteen percent rated their

8:44

current boss as bad.

8:47

So maybe the Hollywood caricature is

8:49

way off. Still considering

8:52

that nearly all of us will at some point

8:54

in our lives have a boss or b

8:56

one. We thought there might

8:58

be some boss questions worth asking.

9:00

And so today on Freakonomics

9:02

Radio, when a boss is

9:04

a bad boss, Have you ever wondered

9:06

why? There's no reason to

9:08

believe that a great salesperson who'll be

9:10

a great

9:11

manager. And yet, this kind of promotion

9:13

happens all the time. Why is

9:15

that?

9:16

So there are two ways to motivate people. We can

9:18

pay them a whole lot more or we can

9:20

give them an opportunity for

9:21

promotion. Today on the show

9:24

why good employees become

9:26

bad bosses and whether

9:28

that will ever change. Spoiler

9:30

alert, probably not.

9:46

This is Freakonomics radio. The

9:48

pod cast that explores the hidden side

9:50

of everything with your host,

9:53

Steven Dubner.

10:04

One of the reasons I became a writer

10:06

years ago is because I didn't

10:08

particularly like having a boss.

10:11

Like Katie Johnson, I prefer

10:13

to set my own agenda, my own pace,

10:15

I also really like working alone,

10:18

also like Katie Johnson, I am

10:20

not particularly fond of meetings, so I

10:22

wouldn't be a very good boss either. Fortunately,

10:25

at Freakonomics Radio, There are

10:28

a couple other people who do all the bossy

10:30

stuff, leaving me pretty much free

10:32

to do this. What

10:35

we're doing right now, asking questions

10:37

trying to find answers. So here's

10:39

a question I've always been curious about. How

10:42

important are bosses anyway? I

10:44

don't mean CEOs, the ultimate

10:46

boss. If you're interested in that, we once

10:48

did a series called The Secret Life

10:51

of CEOs. Today,

10:53

we are just talking about your standard issue

10:55

middle manager. Do they really

10:57

matter? Yes.

11:01

Broadly speaking, managers matter.

11:04

Bosses matter for outcomes. That

11:06

is Steve Tedellis. He is an professor

11:09

at UC Berkeley's Hass school

11:11

of business, a training ground for

11:13

future bosses. Management

11:15

is not something that Tadellus himself

11:18

aspires to. Telling me how close

11:20

you are to administration so I know how far

11:22

away to be from you. But he has

11:24

spent time while on Sabbatical working

11:27

as a boss at some well known firms.

11:30

When I was at eBay and

11:32

Amazon, I managed Teams and I enjoyed

11:34

it very

11:34

much. How do you assess yourself

11:37

as a manager in that realm. I'm blushing.

11:39

So Because

11:42

you're the best ever? No.

11:44

But I'm pretty good, so I'm feeling a little

11:46

uncomfortable. Your positive self assessment

11:49

is based on direct feedback

11:51

or just a general warm glow

11:54

feeling. At eBay and Amazon, the

11:56

feedback was actually formal food surveys.

11:59

Surveys, that

12:00

is, with questions like, on

12:02

a scale of one to five, how much do you

12:04

agree with the following statement? My

12:06

boss generates a positive attitude

12:08

in the team. Or my

12:10

boss is someone I can trust, or

12:13

My boss provides continuous coaching

12:15

and guidance on how I can improve my

12:18

performance. These surveys

12:20

led Steve Tedellis to ask his own

12:22

bigger questions about

12:23

bosses. For instance, does

12:25

it really matter? Do these measures of

12:28

manager skills or characteristics

12:31

do they really have any value for the firm?

12:34

Is there some way in which managers

12:36

who score higher on these surveys are

12:39

actually contributing more.

12:43

These are eternal questions in

12:45

the field known as personnel economics.

12:48

You could ask the same questions about any

12:50

manager, the head coach of a football

12:52

team, the chairperson of

12:54

your homeowner's association, the

12:56

president of the United States. But

12:59

as I mentioned earlier, the academic literature

13:01

on the impact of bosses is not

13:03

particularly advanced. can

13:06

see why if you think about it There

13:08

are so many variables in the relationship

13:10

between a boss and their employees that it can

13:12

be hard to pinpoint the effects of

13:14

the

13:15

boss. This is why most research

13:17

focuses on one single

13:19

quantifiable metric, productivity.

13:22

For example, there is a

13:24

paper by the late

13:27

wonderful economist Eddie Lasir, Katherine

13:30

Shaw, and Chris Stanton, where

13:32

they show that there is variation

13:35

in output

13:36

of employees based on the

13:38

managers that are in charge of them

13:41

That paper from twenty fifteen

13:43

analyzed data from a single firm

13:46

that the researchers were not allowed

13:48

to identify, but it appears to be something

13:50

a call center. The analysis

13:52

looked at what happened when a worker moved

13:54

from what the researchers identified as an

13:56

average boss to a high quality

13:59

boss. Such a move, they

14:01

found increased productivity by

14:03

as much as fifty percent. So

14:06

if this were a call center and

14:08

a given worker, handled a hundred

14:10

calls per shift under an average boss,

14:12

an excellent boss could boost that

14:15

to a hundred and fifty calls. So

14:17

at least in this type of setting,

14:19

a quote good boss is doing

14:22

something right, but the data couldn't say what.

14:25

Steve Tedellis wanted to learn

14:26

more, so he teamed up with Mitchell

14:28

Hoffman, an economist at the

14:30

University of Toronto's Rockman School

14:32

of Management, to write a research

14:34

paper. I have access

14:37

to interesting data and people

14:39

in this company that will

14:41

have to be unnamed because when it comes

14:44

to personnel data, companies are very

14:46

hesitant. Ted Ellis would only

14:48

say that this firm did high-tech

14:51

knowledge based work. Maybe

14:53

given his history, you might picture

14:55

a firm like an eBay or

14:58

an Amazon. In any case, he

15:00

is looking at a very different type

15:02

of work than the earlier research with

15:04

its narrow measure of

15:05

productivity. What we're doing

15:07

is opening the hood

15:09

up a little bit and what sort of data

15:11

did they have access

15:12

to? We have data that

15:14

allows us to measure the impact of a

15:16

particular manager skill

15:18

that we're calling people management skills

15:21

as opposed to just do managers

15:23

matter. People management skills,

15:25

meaning the sort of things you find on

15:27

those employee feedback surveys.

15:30

How well the manager coaches and communicates?

15:33

How trustworthy they are. So

15:35

that's the boss data on the

15:37

employee side. Ted Ellis and Hoffman

15:39

had a lot of concrete data. Subjective

15:42

performance scores, as well as how

15:44

often a given employee was promoted or given

15:46

a raise, the number of patents

15:49

they

15:49

filed, and whether they stayed

15:51

at the firm or left. In

15:53

these high-tech knowledge based companies,

15:56

retention is a very, very

15:59

important focus because

16:02

getting these high skilled workers is

16:04

not easy, and there's a lot of

16:06

competition. And when you

16:08

lose an employee, especially an

16:11

employee that's very valuable, then

16:13

it could take

16:15

months to replace them.

16:17

So Tedllis and Hoffman set about

16:19

to sort through all this data to look for any

16:21

causal relationships between the rating

16:24

of a given manager and the various

16:26

outcomes of the employees working

16:28

under

16:28

them. Would they find? For

16:31

the most part, it was a big bag of

16:33

nothing. We didn't find

16:35

that the ratings of

16:37

the managers seemed to impact

16:40

the subject performance of their employees,

16:42

their income, their promotions,

16:44

or patent applications in a

16:46

meaningful way. That's right. On all

16:48

those employee outcomes, performance,

16:52

earnings, patents, it

16:54

just didn't seem to matter whether

16:56

the manager was highly rated

16:58

were poorly rated. But there

17:00

was one other outcome to look

17:02

at employee

17:04

retention, bingo, To

17:06

Dallas and Hoffman look at employees at

17:08

this one firm who moved from a manager

17:11

with a poor rating to one with a high rating.

17:13

That's associated with an attrition drop

17:15

of about sixty percent That is

17:18

huge and within that huge

17:20

effect was an important

17:21

nuance. What we see then

17:24

is that managers

17:27

help retain better employees more

17:30

than worse employees, which shows that the

17:32

impact of being a better manager is

17:34

strongest where it matters the

17:36

most. So

17:38

a good boss seems capable of

17:40

keeping the best employees happy

17:43

and presumably productive. Conversely,

17:46

a bad boss might drive away

17:48

the best employees. The

17:50

Fidelis Hoffman paper was published

17:52

in twenty twenty one in the journal of

17:54

political economy, one of the best econ

17:57

journals. So okay.

17:59

The economics literature on bosses and management

18:01

just got a little bit deeper But

18:04

remember, employee retention was

18:06

the only outcome where it seemed to matter

18:08

whether a boss was good or

18:10

bad. And if you ask Steve

18:12

Tedellis a more fundamental question,

18:14

like, what does a good boss

18:16

actually do to instill this

18:19

loyalty. This is where

18:21

I have to take a step back and say

18:24

that there are certain things that may

18:26

be outside the scope of

18:27

what economists should be dealing

18:29

with?

18:30

If you were to make a list of things that

18:32

you would like to measure, were it possible given

18:34

the data? What would some of those things be? Really

18:36

good question. Something that's

18:39

very hard to measure

18:41

that I believe is

18:43

important is compassion.

18:46

I guess if this is gonna be on the radio or

18:48

might lose my Economist card. Steve

18:50

Didellis is not the only Economist who's

18:52

been frustrated by the lack of evidence

18:55

for what makes a good boss

18:57

good. Maybe compassion

18:59

is as important as he suspects,

19:01

but We just don't have any

19:04

large scale empirical evidence yet.

19:06

The Stanford Nicholas Blum,

19:09

has been studying leadership and management

19:11

for years. And yet,

19:14

no one could really give us a straight answer

19:16

on what to find a good or a bad leader you

19:18

look at the data and there's ten different recipes

19:20

for success. Maybe they each work for a particular

19:23

case study, but I've still twenty years

19:25

later struggled to find anything

19:27

that's the secret recipe beyond

19:29

saying, sure there are some people that are better than others,

19:31

but it's damn hard to tell what it is.

19:33

This has not stopped leadership gurus

19:36

from promoting their pet theories.

19:39

As Bloom puts it, there is a

19:41

ton of BS around this from

19:43

airport bookstore pulp fiction.

19:46

And here's another reason to question the

19:48

literature on management and bosses.

19:51

As we've been hearing, most of the

19:53

BOSS data comes from employee

19:56

surveys. Have you ever

19:58

taken a survey that rates your manager?

20:01

If so, were you told it was

20:04

anonymous? Did you

20:06

believe it was anonymous? Were

20:08

your answers objective? Or

20:12

did you maybe think, well, my

20:14

boss thinks I'm good at my job, so

20:16

I'm gonna say they're good at theirs. Or

20:18

vice versa, I don't think my boss

20:21

likes me, so I'm sure not gonna give them

20:23

a good rating. As we have

20:25

said before on this show, survey data

20:27

can be the lowest form

20:29

of

20:30

data. Here again is Steve

20:32

Tedellis. I'm sure you know

20:34

that he economists are very wary about

20:36

using surveys. And

20:38

economists believe in what we call a

20:40

revealed preference approach. Meaning

20:43

how you behave

20:44

is telling me a lot more about you than

20:47

what you say about yourself. Just

20:49

how big is the gap between what people

20:51

say and how they behave? Over

20:54

the years, I have heard many economists give

20:57

many examples of this gap

20:59

Steve Tedellis' example is my

21:01

all time favorite. There is a lot

21:03

of discussion about privacy and privacy

21:06

regulation these days.

21:07

And you hear a lot of people saying how

21:09

their privacy is important to them, and

21:12

then you turn them and say, here's the snickers

21:14

bar. Could I have your mother's maiden name? And they

21:16

say yes. So it's a little

21:18

bit confusing when you tell me that you

21:20

really care about privacy and then you just

21:23

scroll down on every app you

21:25

download and click yes, yes, yes,

21:27

that doesn't tell me that you really care

21:29

about privacy.

21:31

So the same is true for many other

21:34

types of behavior. So

21:37

let's keep in mind that much of what we have

21:39

been told in the past about good

21:41

bosses and bad bosses is not

21:43

exactly evidence based. Researchers

21:46

like Fidelis and Hoffman and

21:49

Bloom have been shipping away

21:51

at the black box of boss behavior, but

21:53

we've got a long way to go. This

21:56

means we need to keep looking for

21:58

good data and asking Good

22:00

questions. So coming up after the break,

22:02

who becomes a boss and why?

22:05

If so many people think the boss selection

22:07

process is stupid, Why

22:10

do firms keep doing it? And whatever

22:13

happened to Katie

22:14

Johnson.

22:15

I get to the end of the day and the lack thing I

22:17

wanna do is talk to someone else.

22:19

I'm Steven Dubner. This is Freakonomics

22:21

Radio, and remember, you can get our series

22:24

the Secret Life of CEOs and

22:26

all our past episodes on any podcast

22:29

app will be right back.

22:32

Freakonomics radio sponsored

22:35

by Staples. Did you know that with

22:37

Staples' business advantage, you've got access

22:39

to a team two thousand real life

22:41

purchasing experts to help your business

22:44

run better and your dollar go

22:46

further. They take data to

22:48

the next level and add human insight to

22:50

create custom solutions for you that

22:52

data alone could never provide. They

22:54

do the leg work for you, curating a

22:56

selection of products so you can be sure

22:58

you are getting just what your business needs.

23:00

That is the Staples Business Advantage

23:03

sign up at staplesbusinessadvantage dot

23:06

com Freakonomics

23:11

radio sponsored by Satya. This is

23:13

story of one company whose products fly

23:15

in the face of everything we've learned as consumers.

23:18

The company is Satva, and the luxury

23:20

mattresses they make are unique. Unique

23:22

in that they are so luxury various, so

23:24

plush, so comfortable, you'd think they'd

23:26

cost a small fortune. But in reality,

23:29

they cost half the price of the top retail

23:31

brands. But that's not possible.

23:33

Our consumer experience tells us.

23:35

Right? Wrong. SatBA has been

23:37

in the mattress business for a long time.

23:40

They will also set it up for you and take your old

23:42

mattress no extra charge, all of

23:44

which proves that luxury and affordability can

23:47

peacefully coexist, you just have

23:49

to know where to look. Right now,

23:51

save two hundred dollars on a thousand dollars

23:54

or more at sottva dot com

23:56

slash freakonomics. That's s

23:58

double ATBA dot

24:00

com slash freakonomix. Have

24:10

you ever thought about where a boss

24:12

comes from? What I mean is,

24:14

why a given employee will

24:17

rise from the ranks to become

24:19

a

24:19

manager? Here's someone

24:21

who's been thinking about that a lot.

24:23

My name is Kelly Xu. I'm a professor

24:26

of finance at the Yale School of Management.

24:29

Kelly Xu, along with Allen Benson

24:31

and Danielle Lee, published a paper

24:33

in the Quarterly Journal of

24:35

Economics, another top journal

24:37

called Promotions and the

24:39

Peter Principle. So the

24:41

Peter Principle is a very funny

24:44

and popular management book written

24:46

by Lawrence j, Peter, and his

24:48

book offers an explanation for why

24:50

we might see incompetent bosses

24:52

everywhere.

24:53

Incompetent bosses everywhere. Okay.

24:56

I'm listening. What is this

24:59

explanation? Let's go back to

25:01

Lawrence j Peter himself. This

25:03

is from a nineteen seventy three documentary.

25:07

The Peter principal states very simply that

25:09

in any hierarchy, an employee

25:12

tends to rise to his level of

25:14

incompetence. I'm

25:17

sorry. As many times as I've heard that, phrase,

25:20

I still laugh at it just because it

25:22

sounds like it's gonna be not irreverent

25:25

and then it turns immediately irreverent which

25:27

makes me

25:27

chuckle. Well, except I think it's a funny

25:30

idea, but it also rings true

25:32

and it's funny in a kind of unpleasant

25:34

way because it reminds people how much

25:36

they dislike their bosses.

25:38

Peter was a Canadian education scholar.

25:41

He used his daily observations

25:43

to form a theory about job

25:45

promotions. I saw that very often

25:47

a competent individual was

25:50

promoted to something he

25:52

couldn't do. I saw

25:54

a competent mechanic where

25:56

he stayed in my car. He

25:58

was terrific. He

26:01

was very responsible, very precise,

26:03

knew exactly what he was doing. So they made

26:05

an informant. Now he's no longer

26:07

fixing

26:07

cars, and he's trying to manage other mechanics,

26:11

and he's very incompetent. The

26:13

more Peter looked around. The more

26:15

he saw people who were good at

26:17

their jobs routinely stumbling into

26:20

bigger jobs they weren't good

26:21

at. In any organization where

26:25

competence is essentially

26:27

eligibility for

26:29

promotion, and incompetence is

26:31

a bar to promotion. People

26:33

who arise to the level of incompetence intend

26:36

to stay there. The book he wrote

26:38

with Raymond Hall was called

26:40

the Peter Principle, why things

26:43

always go wrong. It wound up

26:45

selling millions of copies. The

26:47

book was meant to saturize corporate

26:49

strategy. Nevertheless, a

26:52

variety of big firms tried to recruit

26:54

Peter to become their management

26:56

guru. He declined saying

26:58

that he didn't wish to rise

27:00

to his own level of

27:02

incompetence. Kelly Shoe again.

27:04

His idea is that

27:06

firms and organizations tend to promote

27:08

people based upon their performance so

27:10

far. What that means is a worker

27:12

who is good at her job will be quickly promoted

27:15

to a new job role, which might require

27:17

different set of skills. If she's

27:19

good at that new role, she's gonna be promoted

27:21

again, until she reaches a position

27:23

where she's actually a bad match for that

27:26

new job role. And then she will no longer

27:28

be promoted. On the one hand, it

27:30

would seem to make perfect sense that you promote

27:32

someone who's good at their

27:33

job. You don't want to promote the bad workers.

27:36

On the other hand, managing is

27:38

not the same as

27:39

doing. There's no reason to

27:41

believe that a great salesperson who knows

27:43

how to negotiate deals will be a great manager.

27:46

That again is the Berkeley economist Steve

27:48

Tedellis. I look here in my

27:50

company, Berkeley great researchers often make

27:53

for lousy department chairs. Great engineers

27:55

often

27:55

make for lousy engineering managers. But

27:59

here's the thing about the Peter principle. Even

28:01

though the theory had been around, for

28:03

half a century. No one had ever

28:06

checked with real data from

28:08

real companies whether Lawrence

28:10

Peter was right. A

28:12

few observations about car

28:15

mechanic or an academic researcher

28:17

turned department chair, those

28:19

do not constitute empirical proof,

28:22

especially in the realm of management

28:24

in all that airport bookstore pulp

28:27

fiction. This is where Kelly Shoe

28:29

and her coauthors come in. They wanted

28:31

to see if the Peter Principle actually

28:34

exists. And if so, what

28:36

should be done about it? First

28:38

step, get hold of some

28:40

data.

28:41

We got our data from a company

28:43

that offers sales performance management

28:46

software and services

28:48

Shoe can't tell us the name of the company,

28:50

but pictures something like Salesforce.

28:52

A typical client of our data provider

28:55

is a firm that employs business

28:57

to business sales workers. And

29:00

that client firm would input the

29:02

sales numbers and the whole organizational structure

29:05

into a software program,

29:07

and what we're doing is we're studying the data

29:10

that these client firms uploaded into

29:12

the software

29:12

program. How

29:13

many firms and how many workers? We

29:15

see data for about forty

29:18

thousand business to business sales workers

29:20

at over a hundred and thirty different US

29:22

based firms. And how many of

29:24

those were in managerial roles?

29:27

Five thousand managers and about

29:29

fifteen hundred promotion events.

29:32

So in terms of empirical

29:34

studies in your realm, this is

29:36

considered a pretty large and robust data

29:38

set or would you have liked it to be even

29:41

bigger than

29:41

that? I would always prefer a bigger data

29:43

set, but for this type of question, a

29:45

very large and comprehensive data set.

29:48

So these are sales

29:50

workers and sales managers, what

29:53

makes sales a good business

29:55

function to study?

29:56

One is important to study sales workers

29:58

because almost ten percent of the

30:00

US labor force or somehow involved

30:02

in the sales function. The other

30:04

benefit is that we have

30:07

a very good measure of their

30:08

performance, so we can test

30:11

are the stronger performers more likely

30:13

to be promoted. So that makes

30:15

a lot of sense from your perspective as

30:17

the scholar from my perspective as

30:20

someone who's not in sales, I would think,

30:22

well, your findings may not translate

30:24

very well that in a field like journalism

30:26

or in healthcare or in many

30:28

other fields, the measurables

30:31

aren't nearly as measurable as

30:33

they are in sales. So how generalizable

30:35

do you think your findings

30:36

are? I believe it's likely to

30:39

apply to other settings where

30:41

the skills required to succeed at one

30:43

level differ from skills required

30:46

to succeed in the next level. So

30:48

some examples are science,

30:51

manufacturing,

30:52

academia, entrepreneurship.

30:55

Can you think of industries or

30:57

sectors where this problem

30:59

wouldn't apply? It's actually hard

31:02

for me to think of a setting which this problem

31:04

wouldn't apply at all. I've also

31:06

seen it in the context of government structures.

31:09

A good example is actually the ancient

31:11

Chinese Imperialexamination system.

31:15

It's famous for being a meritocracy even

31:17

thousands of years ago. So you would take

31:19

a test and the top scores on the

31:21

test would become administrators

31:24

within the government bureaucracy. But

31:26

their problem was they would make the test based

31:28

upon familiarity with classical

31:31

poetry. And the people who are

31:33

vest at that test would then become

31:35

tax

31:36

collectors, which is a different skill set.

31:38

But ancient Chinese poetry was

31:41

an incredibly rich and diverse

31:43

body of literature. Yes. So I could imagine

31:45

how a mastery or even a

31:47

deep appreciation of that could theoretically

31:50

apply across a number

31:52

of skills? Theoretically, yes.

31:55

Ethan, I'm convinced. And to be

31:58

fair, I do not have the historical data

32:00

to prove that being

32:02

the best at classical poetry means

32:05

you are not the best at tax collection. Since

32:08

you don't have that data, let's look at, say,

32:10

modern US politics, how

32:12

would you assess the relationship

32:14

between a person who's electable

32:17

and a person who will govern well. Oh,

32:20

that is a very good point. So someone

32:22

who is electable might be very charismatic,

32:25

very good at public speaking, whereas

32:28

the actual function once someone

32:31

has been elected might involve being

32:33

good at deal

32:34

making, back office politics,

32:37

or understanding the actual details

32:39

of the policies that they're passed

32:41

Do you know anything about that question

32:43

empirically? I'm drawing a blank, but

32:45

you really did raise a very good research

32:48

idea. Maybe I will look into this. We've

32:50

been thinking about settings where this type

32:52

of problem might apply for a long

32:54

time, but somehow it never thought about

32:56

the government or elected official example

32:59

you just raised, but it seems like spot

33:01

on for having potential as a problem.

33:04

Okay. Before I hijacked this conversation

33:06

with Kelly Xu to talk about politics in

33:08

ancient Chinese poetry. We were talking

33:10

about her research paper that tried to

33:13

identify the Peter principle in

33:15

the wild. As Shu

33:17

told us, she had performance data

33:20

on roughly forty thousand sales workers

33:22

and around a hundred thirty companies. The

33:25

next step was to confirm that

33:27

companies indeed use an

33:29

employee's job performance as a

33:31

trigger for

33:32

promotion. The answer? Yes.

33:36

We find that doubling in

33:38

worker sales corresponds to

33:41

a thirty percent increase in their probability

33:43

of being promoted. Another

33:45

way to look at it is if someone

33:48

is the top sales worker, within

33:50

their team of five or six people, then

33:53

that top sales worker has about

33:55

tripled the probability

33:57

of being promoted relative

33:59

to the average sales worker.

34:00

Now is that alone evidence

34:03

of the Peter principle? No.

34:06

Just to promote based upon past performance

34:08

isn't necessarily a Peter principal problem

34:11

because it could be that the best salespeople

34:13

really

34:14

are. The best managers of salespeople.

34:16

In that case, you wanna promote the best salespeople.

34:19

Okay. So the next step, I guess,

34:21

is seeing whether the best salespeople

34:24

indeed do become the best

34:26

managers. How do you do that? So first,

34:28

we're going to measure the quality

34:30

of each manager. Managers

34:32

in our data are no longer directly

34:35

involved in sales. Their job as

34:37

a manager is to coordinate and

34:39

facilitate the sales of their subordinate.

34:42

And

34:42

presumably those subordinates are people

34:44

they worked with side by side and maybe

34:46

competed against just the week before they were

34:48

promoted. Is that the case often? We actually

34:50

see for the most part, people when they're

34:53

promoted, they're rotated to a different

34:55

team. Possibly because the

34:57

firm overall is exactly afraid

35:00

of those internal team dynamics

35:02

that you've just described. So we don't

35:04

want to call someone a good manager because

35:07

her team sells a lot. Because we're

35:09

worried that maybe she was lucky and she

35:11

was assigned to great sales workers.

35:13

And those sales workers could have been great regardless

35:16

of her manager input. To

35:18

get around that problem, we're gonna measure

35:20

manager quality as the manager's

35:23

value added to her subordinate

35:25

sales. If my subordinate sell

35:27

more when they work under me, than

35:29

when they worked under other managers,

35:32

then I would be considered a high

35:34

quality manager.

35:40

So here's the key question Kelly

35:42

Xu is asking. Does being

35:44

a good salesperson make you a good

35:46

manager of other

35:47

salespeople? Here's what she found.

35:49

The manager with double

35:52

the pre promotion sales as another

35:54

manager

35:55

leads to about a six percent decline

35:58

in subordinate sales.

35:59

Oh my goodness. Yes. What we

36:01

find is that among promoted managers,

36:05

those with low sales prior

36:07

to their promotion

36:08

they are actually better at managing

36:11

their supportness. Let

36:13

me say that again. Oh, my

36:15

goodness. When these firms

36:18

select people to be managers based on their current

36:20

job performance, they are actively making

36:22

themselves worse off. In

36:25

other words, the Peter principle is

36:27

as real as Lawrence Peter said

36:29

it was. And I'm

36:31

editorializing here, it would also seem

36:33

to be incredibly stupid.

36:37

If the firm's only goal were

36:40

to have the best possible managers, then

36:43

the firm could by putting more weight

36:45

on collaboration experience and

36:47

less weight on sales numbers. The

36:50

firm could promote better managers

36:52

and raise overall firm sales numbers

36:54

by about thirty percent.

36:56

That's assuming that collaboration experience

36:59

is in fact more important for a

37:01

manager than just high sales numbers.

37:04

Still, a thirty percent increase in

37:06

revenue simply by killing off the

37:08

Peter principle, that would seem to

37:10

be a no brainer. So

37:12

does this mean that modern firms simply

37:14

aren't aware of the age

37:16

old Peter principle?

37:18

Most firms are aware of the

37:20

Peter Principle problem, and it's

37:22

a problem that they purposely choose

37:25

to live with. Some evidence

37:27

we have indicating that in

37:29

situations when the firm

37:32

is trying to select a new manager who is going

37:34

to be in charge of a very large team.

37:36

So that's situation which manager quality

37:39

matters a lot. In

37:41

those situations, firms put less

37:43

weight on a worker sales numbers.

37:46

Probably because they know they're gonna end up

37:48

with a bad manager.

37:50

So she was arguing that firms know

37:52

they will get worse managers by simply

37:54

promoting people who've been good at their previous jobs

37:57

rather than people who might actually be good

37:59

managers. And yet for

38:01

the most part, they continue to

38:03

do it even though it hurts their profits.

38:06

Why would they do that?

38:08

Economists are always telling us that companies

38:11

are by definition profit

38:13

maximizing machines. Knowingly

38:16

promoting a bad manager does not

38:18

sound very profit maximizing So

38:21

are companies just making a mistake?

38:23

Affirm having a Peter principal

38:26

problem doesn't necessarily mean that

38:28

the firm doesn't understand what it's doing

38:30

or it's making a mistake.

38:34

So what is going on? After

38:37

the break, Kelly Xu gives us some answers.

38:39

You

38:39

don't wanna brag about your pay on your resume.

38:42

I'm Steven Dubner. This is Freakonomics

38:44

Radio. We'll be right back

38:49

It's time to reboot your credit card with

38:51

Apple card. Apple card is a different

38:53

kind of credit card. It gives you up to three

38:55

percent unlimited cash back on

38:57

everything you buy. It's real cash

38:59

that never expires or loses value.

39:01

And you can use it on anything. Grab a morning

39:04

coffee, pick up a tab, or pay back a

39:06

friend. Apply now in the wallet app on

39:08

iPhone and start using it right away.

39:10

Apple Card issued by Goldman Sachs Bank USA,

39:12

Salt Lake City branch, subject to credit

39:14

approval. Daily Cash is available via Apple

39:17

Cash Card issued by Green Dot Bank member

39:19

FDIC, as a statement credit, terms

39:21

Freakonomics

39:21

radio

39:24

is sponsored by JustWorks. If you are

39:26

small business leader looking for easier way

39:28

to onboard and manage remote employees

39:30

than you need to learn about JustWorks. JustWorks

39:33

makes it easier for you to start, run,

39:35

and grow a business. Justworks

39:37

gives you the peace of mind in knowing the payroll

39:39

will run smoothly. Your team has

39:42

access to high quality benefits and

39:44

your business is getting the support it

39:46

needs. JustWorks makes it simple

39:48

to hire and manage remote employees across

39:50

all fifty states. JustWorks help

39:53

to comply with state payroll tax

39:55

requirements, keep up with state labor

39:57

laws, and access health insurance plans

39:59

in any state. You have access

40:01

to a cloud based platform that enables

40:03

managers and employees alike to

40:05

quickly and securely access benefits,

40:08

payroll, and other HR functionality from

40:10

anywhere anytime. Plus, there

40:12

are no hidden costs. Take

40:14

a look at Justworks' transparent pricing

40:17

by visiting justworks dot com

40:19

slash pricing. Before

40:26

the break, the Yale Finance Professor

40:29

Kelly Xu was telling us about a study

40:31

she and her colleagues published about the Peter

40:33

principle. That's the idea that a good

40:35

employee will be promoted to

40:37

bigger and bigger jobs until

40:39

they get to a job, they're not gonna have,

40:41

and then they tend to stay there. But

40:44

for years, the Peter Principle was just

40:46

a fury. Kelly Xu wanted to see if it's

40:49

real. Using data on

40:51

thousands of promotions, she did

40:53

find that when top performing sales people

40:55

were promoted into management, the

40:57

sales performance on the teams they managed

41:00

declined. In other words, just because

41:02

somebody's good at their job, doesn't mean

41:04

they'll be good at managing people doing

41:06

that same job. Shu also

41:08

found evidence that firms know

41:10

that the best salespeople make bad

41:13

managers and they choose to promote

41:15

them

41:15

anyway. So what is

41:17

happening? What we believe is happening

41:19

is the firm is doing its best to

41:22

motivate workers, and they face

41:24

a trade off. Okay. This is

41:26

where it gets really interesting. Promoting

41:28

based upon past performance is

41:31

very motivating to workers.

41:33

So it's a very strong incentive system.

41:36

We can also work out this in some ways cheaper

41:39

than offering really strong pay for

41:41

performance. So there are two ways to

41:43

motivate people. We can pay them a whole lot

41:45

more or we can give them an opportunity

41:47

for promotion, which they might

41:49

value a whole lot because that's

41:51

something that they can put on their resume. And

41:53

it increases their status in society.

41:56

You don't wanna brag about your pay on your

41:58

resume. I mean, the minute you say that, it

42:00

makes me think wait. Maybe we should

42:02

make it more acceptable for people to brag about

42:04

their pay because wouldn't that be more efficient

42:06

in the end and encourage

42:08

less promotion of people who are going

42:11

to be bad managers? That's a

42:13

fantastic idea. I don't know of

42:15

research testing that directly. But

42:17

I do know in other

42:18

cultures, there's differences in

42:21

it being more socially acceptable to

42:23

talk about your compensation. I mean,

42:25

I was kind of half joking, but

42:28

it would be interesting if there was some metric

42:30

or badge you get saying,

42:33

I'm really good at what I do. And

42:35

I'm so good that I've been rewarded

42:37

a lot of raises. And plainly,

42:40

I'm very valuable to the firm and I could

42:42

be a manager if I wanted, but

42:45

I'm better than that. That's an

42:47

incredibly hand handed naive way

42:49

of putting it, but Is there any mechanism

42:52

in managerial science for

42:54

that kind of

42:56

delineation between success on

42:58

financial level and success on managerial

43:00

level? There have been some interesting

43:03

attempts in that direction, so I've

43:05

heard of many technology focused

43:07

firms, especially those in Silicon Valley.

43:10

They face this problem that they have

43:12

a pool of very talented and

43:14

skilled engineers and those engineers

43:17

may not be the best managers of engineers.

43:19

Many of those firms offer something

43:21

called a dual career track where

43:24

someone can rise in the ranks of

43:26

being an engineer, basically having a higher

43:28

and higher title. So you can start

43:30

as engineer, then distinguish

43:32

engineer, then lifetime distinguish engineer.

43:36

And that's a way for the

43:38

firm to recognize someone's contributions

43:41

in a public way without moving them

43:43

over to management.

43:47

The Berkeley economist Steve Tedellis

43:49

has also noticed this movement in

43:51

companies like eBay, Google,

43:53

Amazon, Facebook, there's the term

43:55

of IC. There are independent contributor.

43:59

And you will have people who are

44:01

at the level of VP not

44:03

managing a single person. Because they

44:05

are just gods in their realm

44:07

of engineering or coding

44:09

or architecture and so on.

44:12

By distinguishing between ICs

44:15

and the so called management talent,

44:18

the firm is saying, look, we are going to

44:20

promote people in ways

44:22

that reward them for what they're great at.

44:24

You're not a great manager. You're not

44:26

going to get incentivized by becoming

44:29

a manager. Has that model

44:32

trickled out at all of that high-tech

44:34

realm? One area

44:36

where I have seen it is

44:39

in consulting companies where

44:42

you have the kind of deep technical talent,

44:45

think of PhDs, etcetera. That

44:48

will remain and be very heavily rewarded

44:50

for the work they do and they will not

44:53

manage people.

44:56

The fact that Kelly Xu and Steve Tedellis

44:59

can identify handful of cases

45:01

where career success is not tied

45:04

to promotion into management Well,

45:06

those are exceptions that prove

45:08

the rule. As she found

45:10

in her research, the Peter principal is

45:13

alive and well, as absurd as

45:15

that may seem. It is yet

45:17

another confirmation that management science

45:20

as lovely phrase as that may seem

45:22

is not yet very scientific. Most

45:25

firms stick with what they've always done.

45:28

When an employee is good at what they

45:30

do, You turn them into a manager

45:33

to oversee other people who do what they

45:35

used to do, even if they are not

45:37

cut out to be a manager, like

45:40

our friend, Katie

45:41

Johnson, the data scientists

45:44

we met earlier. I didn't see that

45:46

there was another path, whereby I could

45:48

be director level but

45:50

not have direct

45:51

reports. I just didn't ever see that.

45:52

Looking back, there were

45:54

some clues that Johnson wasn't

45:56

quite manager material. You

45:59

remember, during management training,

46:01

she took that personality test, and

46:03

she told us the areas where she got high scores?

46:06

Critical thinking, attention to

46:08

detail, courage, all these kind of

46:10

internal thinking type characteristics.

46:13

Well, those were not the only

46:15

results of this test. Things

46:18

that I can do that I struggle with

46:20

was compassion, empathy

46:23

relationship building. I saw this

46:25

output and I was like, why didn't anyone

46:27

do this to me before I got this

46:29

job? Because this just screams great

46:31

data scientist, not so great manager.

46:34

But it was too late. She'd

46:36

already been made a manager. And

46:38

as you'll recall, it was not going

46:40

well. I would

46:41

finish my day and my study, walk into

46:43

the living room, put a blanket over my

46:45

head and cry.

46:50

So let's say we're talking a scale

46:52

of zero to ten. Where would

46:54

you put your median

46:56

satisfaction and when you were an

46:58

IC or a maker? When I

47:00

was a maker, I put myself as an eight

47:02

and a half that I actually loved

47:05

what I did. I at absolutely loved it. The

47:07

only reason I would even deduct one point

47:09

five points is because there were some

47:11

frustrations, as I mentioned, about not

47:13

being heard and not being

47:15

autonomous. And

47:15

then where would you put it at zero to ten when you'd

47:18

become a full on manager? I would

47:20

say I'd put myself more at like a

47:22

four or A5A6. Would

47:24

be a great day.

47:25

Okay. That's your personal

47:28

satisfaction. I do see, however,

47:30

on LinkedIn a review

47:33

from your manager. He writes,

47:35

Katie is a rounded and passionate

47:37

data leader with all the qualities

47:40

required to inspire manage

47:42

and lead a team. Plus, she has got

47:44

brilliant IC skills to boot.

47:47

And he notes that you are a real unicorn

47:49

in the data analytics

47:51

field. So that sounds like

47:53

you were the best manager ever.

47:55

Yeah. Is

47:58

there any nicest in it?

47:59

It's Really nice. Did he write that before

48:01

or after you decided to quit? He wrote that after.

48:07

You heard that right. Katie Johnson

48:10

quit that management job.

48:12

She quit being a boss entirely.

48:15

She went back to working as a data

48:17

scientist. At a different firm.

48:20

I don't know if you can ever

48:22

be successful at something you don't like.

48:24

I wanna do something that I love and

48:26

I'm really passionate about it because that's the only

48:29

way. Maybe other people are different, but I

48:31

have to love it. I have to be, like, on a Sunday

48:33

night, I can't wait to stop my work tomorrow and get back

48:35

what I was

48:35

doing. And I was never ever ever

48:37

gonna have that in my management job. So

48:40

before you ever became a manager, as a maker,

48:42

you said you're average satisfaction

48:45

or happiness was around eight and a half. When

48:47

you became a manager, dropped to let's call it

48:49

five six on a great

48:52

day. What is it now?

48:54

I'd say it's a nine and a half night. I'm super

48:56

happy. Are you getting paid less

48:58

now as a data scientist then you

49:00

were as a manager? I'm getting paid

49:02

more. How did that happen? I think there

49:05

are more individual contributors now

49:07

that paid good money. I think that this

49:09

technical specialist group is becoming

49:12

more prominent and more rewarded, and

49:14

people do realize that there are gonna

49:16

be a lot of people who don't want become the

49:18

manager. And how do you motivate

49:19

them?

49:20

I believe you looked at the Peter principal

49:22

paper. Is that right?

49:23

Yeah. I did. The way the Peter principal is

49:25

usually described is, to me, almost comical

49:27

to that people rise to the level of their incompetence,

49:30

which I find is a bit cruel

49:32

sounding because one could also

49:34

say that people rise to their

49:36

ceiling of competence. Right?

49:38

And then maybe they're not as good of that. It's not like

49:40

they suddenly turn into idiots. But

49:43

I am curious just your thoughts on

49:45

the notion of promotion into management

49:47

as a reward for being good at what

49:49

you've been doing all

49:50

along. For me, this is where the

49:53

idea of splitting out those levels

49:55

of seniority. So maybe

49:57

you don't become the manager, but you can become a

49:59

technical expert and you are paid and rewarded

50:01

for that. Is something that helps with

50:03

the incentives. What I would say on that though is often

50:06

we have this dual career track of, okay,

50:08

you can be a manager or you can be a technical

50:10

specialist. But even though you might get

50:13

a quote unquote promotion and be paid more,

50:15

the technical specialists still might be excluded

50:17

from high level conversations. So being

50:19

a manager just has this connotation of

50:22

seniority that a technical specialist

50:24

doesn't necessarily and you still might be

50:26

overlooked in terms of just the respect.

50:29

And I think that is motivating more than just hey,

50:31

here's a promotion, here's a new job title. I think

50:33

people want that autonomy and that having

50:35

a seat at the

50:35

table, people caring what you think,

50:38

it has to come with that. I

50:42

would think that many people who are

50:44

promoted from some sort of maker

50:46

to some sort of manager that it would be

50:48

hard to step

50:49

back. If for no other reason, then it seems like

50:52

a loss of status. Yes? It definitely

50:54

feels like a loss of status. I guess, for

50:56

me, I'm lucky that I don't

50:58

care what people think as much as other

51:00

people or at

51:01

all.

51:02

I'm sure that was identified in your personality

51:04

test as well. Yeah, complete rogue

51:06

doesn't care what others think. People

51:08

judge you, which I find interesting because

51:10

I didn't know anyone who likes their job as much

51:13

as I do. So for people to

51:15

look upon me and feel sorry for me in

51:17

a sense that I have chosen to go

51:20

backwards in terms of career hierarchy

51:23

it's kind of telling in terms of what we value

51:25

out for career.

51:26

And you can tell them that if you hadn't done this,

51:29

you wouldn't be on radio.

51:30

Well, it's not play. I got what I

51:32

wanted. Was

51:34

this the plan all along? Yeah. It's an

51:36

update long game. Thanks

51:42

to Katie Johnson for sharing her Boston

51:44

and backstory and to Kelly

51:46

Xu, Steve Tedellis, Nick Blum,

51:48

and all their collaborators for trying to make this

51:51

thing we call management science a

51:53

bit more

51:54

scientific. Coming up next time

51:56

on radio. Insurance

51:58

markets offer an incredibly sexy

52:01

prospect of providing

52:03

a measure of certainty in a

52:05

dangerous and uncertain world. Seriously,

52:09

insurance, sexy? Apparently,

52:12

yes. That's next time on the

52:14

show. Until then, take care of

52:16

yourself. And if you can, someone

52:18

else too. Freakonomics

52:23

radio is produced by Stitcher and Renbud

52:25

Radio. This episode is produced

52:27

by Ryan Kelly and mixed by Greg

52:29

Ripon. We had help from Jeremy Johnston

52:31

and Jared Holt. Our staff also

52:34

includes Zach Lipinski, Morgan Levy,

52:36

Catherine Cure, Elena Coleman, Rebecca

52:38

Lee Douglas, Julie Canford, Eleanor

52:40

Osborn, Jasmine Klinger, Daria Kleenert,

52:43

Ematorel, Lyric Boudic, and Elsa

52:45

Hernandez. The executive team of the

52:47

Freakonomics Radio Network is Neil Carruth,

52:49

Gabriel Roth, and me, Stephen Dubner,

52:52

our theme song is mister Fortune by

52:54

the hitchhikers, all the other music was composed

52:56

by Luis Guerra. As always,

52:59

thanks for listening. Do

53:09

you ever want to tell

53:11

your dean, thanks so

53:13

much for giving me all this managerial

53:16

responsibility, but I don't want it.

53:18

I don't want my reward

53:20

for being a good scholar to be that I have to

53:22

do a bunch of management. I

53:25

haven't thought about that. The

53:33

Freakonomics Radio Network, the

53:35

hidden side of everything. Stitcher.

53:46

This show is brought to you by better help.

53:48

Getting to know yourself can be a lifelong process,

53:51

especially since you're always growing and changing.

53:53

Therapy is all about deepening that self awareness

53:56

because sometimes you don't know what you

53:58

really want until you talk things through.

54:00

Better help connects you with a licensed therapist

54:02

online. Who can take you on that journey

54:04

of self discovery from wherever you are.

54:07

Visit better help dot com slash stitcher

54:09

today to get ten percent off your first month.

54:11

That's better

54:12

help. HELP dot com

54:14

slash stitcher. As

54:17

if the Macrispy couldn't get any better,

54:19

bacon and ranch just entered the

54:21

chat. The bacon ranch

54:23

with crispy, available and participating

54:26

indonald's for a limited time.