0:00

graduates from a small number of elite

0:03

private colleges account for a

0:04

disproportionate share of America's

0:07

business and political leaders in this

0:10

episode we discuss a recent study that

0:13

looks at how admissions criteria at

0:15

these institutions privilege students

0:17

from high income

0:24

families thanks for joining us for t for

0:27

teaching an informal discussion of

0:29

innovative and effective practices in

0:31

teaching and learning this podcast

0:33

series is hosted by John Kae an

0:35

economist and Rebecca mush the graphic

0:38

designer and features guests doing

0:40

important research and advocacy work to

0:42

make higher education more inclusive and

0:45

supportive of all

0:54

Learners Our Guest today is John fredman

0:57

John is the briger family distinguished

0:59

professor of economics and international

1:01

public affairs at Brown University where

1:04

he is the chair of the economics

1:05

Department he is a research associate at

1:08

the National Bureau of economic research

1:10

and has served in the white house as

1:11

special assistant to the president for

1:13

economic policy at the National Economic

1:15

Council John is also a member of the US

1:18

Treasury Council on racial Equity a

1:21

co-editor of the American Economic

1:23

Review and a founding co-director of

1:25

opportunity insights welcome thank you

1:28

so much for having me it's a pleasure to

1:29

be with you today's teas are John are

1:32

you drinking any tea with us today so

1:34

I'm a big tea drinker but I drink tea in

1:37

the morning and so I had a delightful

1:41

yunan tea this morning which I will draw

1:44

on the reserves of that energy

1:46

throughout this conversation well

1:49

played and I am drinking a ginger peach

1:52

black tea from the Republican tea not so

1:54

fancy but I enjoy it I have an awake tea

1:57

because I also need some energy

2:01

we've invited you here today to discuss

2:03

your 2023 working paper with Raj shett

2:06

and David Deming diversifying society's

2:09

leaders the determinance and causal

2:11

effects of admission to highly selective

2:13

private colleges this paper created a

2:15

big stir in higher ed and other circles

2:18

as well you note in the study that less

2:20

than one half of 1% of college students

2:22

attend IV plus institutions While most

2:25

of our listeners will be familiar with

2:27

IV League colleges what are the other

2:29

colleges that are included in the IV

2:31

plus designation thanks then it's

2:33

helpful to clarify up front the colleges

2:37

that were directly studying are the

2:39

eight Ivy League schools Harvard Yale

2:42

Princeton Dartmouth Brown Columbia

2:44

Princeton pen and four close peers which

2:48

are Stanford MIT Duke and Chicago the

2:51

important thing to know here you're

2:52

right that there's a pretty small share

2:54

of students it's not that something

2:56

changes discreetly when you move out of

2:59

that group of 12 schools and you go to

3:01

another outstanding private institution

3:04

like Northwestern or Johns Hopkins or

3:06

something like that we have some data it

3:09

seems like there are some pretty similar

3:10

things going on across a lot of these

3:13

very highly selected private

3:15

institutions where you do see things

3:17

being quite different where we have some

3:19

data as well is at the most selective

3:21

public institutions places like UC

3:22

Berkeley University of Michigan UT

3:25

Austin places like that you still have

3:27

to draw the line somewhere when you have

3:29

prestigious institutions that's right so

3:32

you noted that these institutions enroll

3:34

a small share of our students why are

3:36

they so important why do we need to

3:37

study them that's right less than 1% of

3:41

college students in the country go to

3:43

one of these schools and of course

3:45

college students themselves are just a

3:46

small share of students born in any

3:49

given cohort what we found though was

3:52

that students from these institutions

3:54

are really highly over represented in

3:58

leadership positions in society you see

4:01

that if you look at who's at the top of

4:03

the income distribution or who's a CEO

4:05

of a Fortune 500 company more than 10%

4:08

of those individuals are from these IV

4:11

plus institutions but it even gets

4:13

higher when you look at who's in the US

4:15

Senate about three quarters of the

4:17

Supreme Court Justices over the past 50

4:20

years have come from these schools and

4:23

so for sure these schools themselves are

4:26

not going to be making broadscale

4:29

changes in upward mobility in our

4:31

society they're just too small but in

4:33

terms of creating both a diverse group

4:36

of leaders and a broads set of Pathways

4:38

where children from any background have

4:41

the chance to be a senator Supreme Court

4:43

Justice whatever these schools are

4:45

incredibly important one of the things

4:46

that your study did is it investigated

4:49

questions that couldn't be investigated

4:51

before because of the data that you were

4:52

able to assemble could you tell us a

4:54

little bit about the data set that you

4:56

used sure our study like so many others

5:00

has been the beneficiary of the Big Data

5:03

Revolution it's affected so many aspects

5:05

of society and this is the academic part

5:07

of it we're merging together data sets

5:10

from three different places the starting

5:12

point for this paper and for many of my

5:15

other research is the universe of us tax

5:19

and Census records which have been

5:20

merged together at the US Census Bureau

5:22

and what that allows us to do is to

5:26

identify individuals when they're kids

5:28

and then actually follow them through to

5:31

not just project what we think their

5:33

outcomes might be but really actually

5:34

observe them after they get out of

5:36

college and they've entered the labor

5:37

force those data are incredibly

5:39

important in terms of measuring upward

5:42

Mobility directly then on top of that in

5:45

order to study really in depth what's

5:47

going on at these institutions we have

5:51

internal admissions data from several IV

5:55

plus colleges as well as a bunch of

5:58

these most most selective public

6:01

universities and University Systems and

6:04

we see where children are coming from or

6:07

where they grew up in the tax data we

6:08

see where they end up in the tax data

6:10

the college data are really filling in

6:12

this in between how do they go through

6:14

the college application process we both

6:17

learn a lot of other information about

6:19

them like where they applied there's a

6:21

lot of detail about the evaluations of

6:24

their applications as well as of course

6:25

whether they eventually get in and

6:26

matriculate the final data that we're

6:28

using is a set of standardized test

6:31

scores from the two main testing

6:34

companies college board that runs the

6:36

SAT and then act which runs the

6:39

theonomous test and the way we use those

6:42

data are to start from a baseline of

6:47

academic achievement at the point when

6:51

these students are applying to

6:53

University and we can talk about how

6:56

that works and of course it's not a

6:57

perfect proxy for where students are but

7:00

when we think about the role that

7:02

universities are going to be able to

7:04

play we just have to be realistic about

7:06

the fact that they are starting to

7:08

interact with students when they are 17

7:10

or 18 and there's a whole lot of

7:12

inequality in our country that's going

7:14

to affect students long before then and

7:16

so we talk of course as a policy matter

7:18

about how to deal with all that

7:20

inequality but it's a reality of the

7:22

situation especially at this highly

7:24

selective level they're going to be some

7:26

students that just aren't academically

7:28

prepared so that's going to shape the

7:30

set of students that these colleges can

7:32

recruit or admit one of the main

7:34

questions that you address in the paper

7:36

is do highly selective colleges amplify

7:38

the Persistence of privilege across

7:40

Generations by taking students from high

7:42

income families and helping them obtain

7:44

High status high-paying leadership

7:46

positions what do your results suggest

7:49

so that's exactly the kind of broad goal

7:52

of our paper to answer that question and

7:55

I think unfortunately the answer is that

7:57

on average yes do amplify the

8:00

Persistence of privilege that comes from

8:03

two different parts so first of all the

8:06

students who attend these colleges we

8:09

measure a pretty large causal effect on

8:13

their outcomes specifically in these

8:15

leadership positions as adults of course

8:18

the students are very highly selected

8:20

when they come in and so even if the

8:22

colleges weren't doing anything you'd

8:24

expect these set of students to be doing

8:27

some impressive things afterwards but

8:29

what we find and we can talk about more

8:31

of the details of how we do this later

8:33

there's a very large causal effect and

8:35

so these universities it's not just that

8:39

a large share of senators come from them

8:42

they do seem to be very important

8:44

pipeline effect where it's really

8:46

propelling students up into these

8:47

leadership positions now on the

8:49

admission side who are the students that

8:52

are coming into this setad of

8:53

Institutions that are benefiting from

8:55

this really positive effect the problem

8:58

here is that even relative to the

9:01

distribution of test scores for high

9:04

school graduating seniors which as we

9:06

talked about before exhibit a whole

9:10

amount of inequality due to differences

9:12

in education and neighborhoods that

9:14

different students from different

9:15

backgrounds have been exposed to before

9:18

they're applying to college even just

9:20

looking at students that have the very

9:22

same test scores High income students

9:25

are substantially more likely to be

9:28

admitted to an attend these institutions

9:31

relative to lower inome students and

9:33

especially middle inome students the

9:35

gaps are largest When comparing students

9:37

from very high income families to

9:39

students from middle class upper middle

9:41

class families and in your study you tie

9:44

some of the selection process to

9:46

athletic scholarships to Legacy students

9:48

as well as attendance at private high

9:50

schools could you talk a little bit

9:52

about how those factors influence the

9:54

decisions sure so the approach that we

9:58

take is a decomposition of this pipeline

10:02

we see that students are coming in with

10:05

let's just say everybody has the same

10:07

test score group of students at the

10:08

beginning we see that the students from

10:11

high-income families are more likely to

10:13

end up attending this set of schools at

10:15

the end of the day we're going to try to

10:17

decompose where in the pipeline these

10:21

disparities are emerging and so the way

10:23

we first start is actually at a somewhat

10:25

higher level than you asked the question

10:28

which is just to decompose these

10:30

differences between the application

10:32

Phase which of the set of students with

10:34

a given test score applies to these

10:36

institutions the admissions phase of

10:38

those students with that given set of

10:40

test scores that applied which are

10:42

admitted and then the matriculation or

10:44

the yield phase of those that are

10:46

admitted who's actually going to choose

10:47

to come at the end of the day and what I

10:50

found interesting coming into this

10:51

project is that there are many different

10:55

analyses or ideas about how each of

10:57

those three phases could be affecting

11:00

there's concerns about who has the

11:02

information or the resources to apply

11:05

there's concerns about potential biases

11:07

in the admissions process from some of

11:09

the factors that you mentioned Legacy

11:11

preferences or private schools and

11:13

there's a concern that maybe schools

11:15

aren't offering financial aid that's

11:17

sufficient in order for students from

11:19

less affluent families to attend in our

11:22

data we see that about 2third of that

11:26

entire disparity is coming from the

11:28

admission part of that alone so that's

11:31

not all of it but I mean just to give

11:33

some numbers there are about 250

11:37

students from the top 1% of the parental

11:40

income distribution who are in an

11:43

average starting first year class that's

11:46

about 1 1650 students so right there

11:48

about 15% of the class is coming from

11:50

the top 1% of families of those 250 we

11:54

find that about 160 of them are extra in

11:58

the sense that that if everyone attended

12:00

at the same rate when they have the same

12:03

test score there would only be about 90

12:05

students from the top 1% of families and

12:08

so then of That 160 about a hundred are

12:12

coming from the fact that high income

12:14

students are more likely to be admitted

12:16

there are smaller effects coming from

12:19

differences in application rates even

12:21

smaller effects coming from differences

12:23

in matriculation rates but primarily the

12:26

differences are coming through the

12:27

admissions process and even before we

12:30

get into specific policies I think that

12:32

that decomposition is incredibly

12:33

important because the admissions process

12:36

is the one part of this that schools

12:38

entirely control themselves if you want

12:40

more people to apply to your school

12:43

that's hard because applications are the

12:44

students decision you have to go out and

12:46

convince a bunch of students to apply if

12:48

you want to get more students to yield

12:49

to matriculate you have to convince

12:51

those students it's their decision the

12:53

choice about who to admit it's just the

12:55

school's choice this is the one lever

12:58

that the schools entirely control and so

13:00

the fact that most of the disparities

13:02

are explained by this set of policies on

13:05

the one hand maybe that's a good thing

13:08

that they control and maybe can directly

13:11

fix what is the source of the problem on

13:14

the other hand it's a little bit

13:16

discouraging that it's in the choice of

13:19

these own universities that these

13:20

disparities are being created despite

13:22

what are typically loudly voiced

13:25

concerns for upper Mobility so it's

13:27

really the admissions process matters

13:30

now we then go down to the next level

13:32

and this gets to the factors that you

13:34

mentioned why is it that a high-income

13:37

student with a 1400 test score is going

13:40

to be admitted at a higher rate than a

13:43

middle-income student or a low-income

13:44

student with a 1400 test score and even

13:46

just to start with in some sense the dog

13:49

that didn't bark here you might have

13:51

thought that students with a 1400 from

13:55

low-income families they might even be

13:59

more impressive that they got to that

14:00

level despite facing all of these

14:02

barriers but we see that admissions

14:04

rates are in fact much higher for

14:06

high-income students and we Trace that

14:09

back to three factors the first and most

14:12

important factor about 40% of what's

14:15

going on is the preference for legacy

14:18

students those are students who are

14:19

children of alumni of the institution

14:23

now Legacy students affect the

14:25

admissions rate of high-income

14:27

individuals for two reasons one is

14:29

pretty obvious the alumni of these

14:31

institutions themselves are just much

14:32

more likely to be high income it's kind

14:34

of the generation before we're getting

14:36

the same positive effect of attendance

14:39

but the second reason I think was a

14:41

little bit more surprising to me it

14:43

turns out that Legacy students from high

14:46

income families receive a substantially

14:49

larger admissions boost even than

14:52

legacies from lower income families so

14:55

there's kind of a preference for high

14:57

income students even within the Legacy

15:00

pool and you put those two things

15:02

together and that accounts for about 40%

15:04

of the admissions difference the second

15:07

factor is the fact that all of these

15:11

schools designate about somewhere

15:14

between 12 and 15% of their class for

15:17

athletic recruits now there's nothing

15:19

inherent in athletics that means that it

15:23

has to be students from high- income

15:25

families and in fact if you look at the

15:27

distribution of athletic recruits at

15:30

public universities those students

15:32

mirror the income distribution of most

15:34

of the other students at the school in

15:36

the sense that there's not a tilt

15:37

towards high-income families but at

15:39

private institutions the share of

15:42

admitted students that are ethletic

15:44

recruits among High income families is

15:46

significantly higher more like 13 14%

15:49

than it is among admitted students from

15:52

low income families where only five or

15:54

six% of those students are athletes now

15:58

why is this the case I was an athlete in

16:00

college myself and I don't think that

16:03

it's just because kids from higher inome

16:05

families are more athletically talented

16:08

I think it has to do first with the

16:11

resources that are available to these

16:14

kids becoming a college baseball player

16:16

isn't just about having good hand eye

16:18

coordination it's about being able to

16:20

attend clinics being part of a travel

16:22

team there's like a lot of stuff that

16:24

goes along with being able to get to

16:26

that level and then I think the second

16:28

factor is that the set of sports that

16:32

are offered by many of these

16:34

institutions go well beyond the

16:37

canonical football basketball baseball

16:40

which may be a little bit more

16:42

broad-based but they also include sports

16:46

like water polo or sailing or equestrian

16:49

and these are sports where I'm sure that

16:52

there are examples of athletes from all

16:54

across the income distribution but I

16:56

think they tend to skew towards more

16:59

High income families so athletic

17:01

recruits are the second major chunk and

17:04

then the third is what my friend David

17:07

leonhard of the New York Times likes to

17:09

call Private School polish a lot of what

17:12

these schools focus on in the admissions

17:14

process goes beyond just how

17:18

academically prepared people are and

17:20

they really like to see somebody who's

17:22

doing interesting things that could be

17:24

as part of extracurriculars that could

17:26

be the way they spend their summer could

17:28

be the way that teachers write about the

17:32

students or that guidance counselors

17:34

write about the students and all of this

17:37

gets channeled through a students

17:39

evaluation on nonacademic

17:42

factors and what we see there is that

17:45

not only are students from high-income

17:47

families much more likely to get very

17:50

strong non-academic ratings that seems

17:53

to flow through things like

17:55

recommendation letters that are really

17:58

centered at the school level and just

18:01

more generally you find if you compare

18:03

high and low income students who are

18:04

attending the same school you no longer

18:07

see this disparity in non-academic

18:10

ratings and so our sense is that these

18:13

other broader factors that kind of seep

18:15

into the admissions process are

18:16

accounting for the Third Leg of this

18:20

tripod that's giving High income

18:22

students an advantage in the admissions

18:23

process and some parents are probably

18:26

sending their students to more elite

18:27

private schools in the hope that that

18:29

will enhance their prospects and the

18:32

schools that accept them recognize that

18:34

one of the reasons students are going

18:36

there is because they prepare them

18:37

better for selection in a more

18:40

prestigious institution I think that's

18:42

exactly right I think it's not just

18:44

parents and schools the fact that

18:47

colleges place a substantial weight on

18:50

these nonacademic

18:52

factors which then can be kind of

18:55

trained for and developed over the years

18:58

I think this is really a major force

19:02

that shapes the way that parents and

19:05

kids and lots of organizations in

19:08

society direct their resources so let me

19:11

just give you an example here I was

19:13

presenting this paper at UC Berkeley the

19:16

economics department and a friend of

19:18

mine who lives in San Francisco who's a

19:20

professor there sent me a picture of an

19:23

advertisement on the side of the road

19:26

like kind of billboard on the side of

19:27

the road for A Fencing Academy it's

19:29

called the saber school and it says a

19:32

safe fun sport that will help colon what

19:36

are the things that doing saber will

19:38

help well number one it will enhance

19:40

performance at work in school okay that

19:42

sounds plausible number two it will

19:44

enhance speed coordination and

19:46

decisiveness number three it will help

19:48

you get accepted at top us colleges it's

19:51

like right there on the billboard and so

19:54

if you want to fence as a kid that's

19:56

totally fine and some people are going

19:58

to really enjoy it but the fact that

20:01

colleges value this and so now all sorts

20:03

of people are spending their time

20:05

fencing simply because they think it

20:07

will help their college application I

20:09

find that to be a little bit silly I

20:10

don't think we would have found that

20:11

billboard in my

20:14

neighborhood although if you brought a

20:16

saber to work you might get more

20:18

attention that raises a whole set of

20:20

other

20:21

issues in your study you also examine

20:24

admission rates at highly selective

20:26

public colleges do there admissions also

20:29

favor students from high income

20:30

households over lwi income households

20:33

when other student characteristics are

20:35

held constant yeah so the public most

20:37

selective institutions they provide a

20:40

really interesting contrast to the

20:42

private schools and there really two

20:44

differences the first difference is that

20:47

it's still true that students from high

20:49

income families with the same test score

20:51

are more likely to be attending these

20:53

places like UC Berkeley or Michigan than

20:55

students from lower income families but

20:58

it's not the Super concentration in the

21:02

top 1% the top 1% are about 20% more

21:06

likely to attend but so are the top 5%

21:08

and roughly top 10% it's more kind of a

21:10

broad top of the income distribution

21:11

than kind of the Uber Rich that are

21:13

benefiting from this then second when

21:16

you do the decomposition that we do at

21:19

the private schools you find that in

21:22

fact it's not the admissions process the

21:25

chances of admission for students with a

21:26

given test score are almost identical

21:28

across the income distribution if

21:30

anything slightly higher for lower

21:32

income students the big differences come

21:35

in the fraction of students who apply to

21:37

these schools you see almost all of the

21:41

over attendance is explained by higher

21:44

application rates of high-income

21:45

students and so that really points to a

21:47

very different part of Pipeline and I

21:50

think there's a whole other set of

21:51

issues that you kind of policy concerns

21:53

that that brings up just to cite some

21:56

fantastic work in this space by by my

21:59

colleague at Harvard Sue dinari she and

22:02

a number of co-authors have worked with

22:04

the University of Michigan over the past

22:05

10 years on something called the hail

22:08

scholarship and this program is really

22:11

focused on this application Phase where

22:14

they reach out to students who are doing

22:16

very well in Michigan public schools and

22:19

who are not from high income families

22:22

and they not just inform the students

22:25

about the University of Michigan but

22:26

they provide a simplified form of

22:29

financial aid that's a tiny bit more

22:30

generous but just mostly clearer there

22:32

basically guarantee of zero for four

22:34

years and that seems to have really

22:37

large effects big increases in the share

22:39

of students who are applying who receive

22:41

these types of Flyers that then carries

22:44

through to those that are admitted and

22:45

those that end up matriculating and so

22:47

first of all it's really interesting

22:49

that there are some sense different

22:51

problems these different schools but

22:53

also I think it's a nice lesson that

22:56

even among two different schools which

22:58

are objectively at the very top of the

23:00

US higher educational sector there are

23:04

really important differences in terms of

23:06

how these different institutions operate

23:08

and what types of policies are going to

23:09

be most appropriate for increasing

23:13

diversity of students and social

23:14

Mobility at those places is the rate of

23:17

return to education significantly

23:18

different between the IV plus

23:20

institutions and Elite public

23:22

institutions the answer is yes but it's

23:26

different in a very particular way so in

23:29

our data what we find using a bunch of

23:31

different empirical approaches is that

23:35

students that attend these IV plus

23:38

institutions are significantly more

23:41

likely to be at the very top of the

23:42

income distribution to attend an elite

23:44

graduate school to hold a very

23:46

prestigious job they're much more likely

23:48

to do that than students who attend the

23:50

very most selective of the public

23:53

institutions those public institutions

23:55

in turn are significantly better at

23:58

propelling students to these leadership

23:59

positions than lower rated less selected

24:03

public institutions and so it is both

24:05

true that those public institutions are

24:06

very good and also true that these IV

24:09

plus schools are really quite a bit

24:11

better that's focusing on these topend

24:15

leadership positions if you look Instead

24:18

at something like what's the chance that

24:22

you'll be in the top 20% of the income

24:24

distribution so for kids in their early

24:27

30s that's earning more than about

24:29

$60,000 so that's a good solid

24:32

professional job you don't have to be a

24:33

hedge fund manager there attending these

24:36

IV plus schools is not really going to

24:37

make that much of a difference and the

24:39

reason is that at that point in the

24:41

income distribution that's just not what

24:44

these schools are designed for you're

24:46

quite likely to get a job that's going

24:48

to pay more than that from an IV Leal

24:49

school you're also quite likely to get a

24:51

job that pays more from that at one of

24:52

these Elite public institutions there

24:54

are differences in average income but

24:56

it's really d driven by this kind of a

24:59

lottery ticket that you're getting on

25:01

maybe you're going to be really just an

25:03

extreme leader again either very top of

25:05

the income distribution very prestigious

25:07

firm so the answer is yes these schools

25:10

differ but they primarily differ in this

25:12

particular way which is why we've placed

25:15

the emphasis on leadership rather than

25:18

just kind of broad Economic Security

25:19

it's not clear that students from IV

25:22

plus schools are just broadly more

25:24

economically secure in that middle of

25:26

the income distribution than from public

25:28

schools you also examined in this paper

25:31

what would be the effects if the

25:32

admission process at the more Elite

25:34

institutions were similar to that at

25:38

highly selective public institutions

25:40

what do you find there in terms of the

25:42

income diversity of students in the IV

25:44

plus institutions if those preferences

25:46

were eliminated yeah so we're able to

25:50

simulate exactly as you say what would

25:53

these classes look like at least

25:54

probabilistically if the admissions

25:56

office were to place less weight on some

25:58

of these factors and it makes a

26:01

meaningful difference so just to give

26:03

you one

26:05

statistic currently on average they are

26:08

a bit less than 60% of students at these

26:11

schools that come from the bottom 95% of

26:14

the income distribution those are

26:15

families making less than call at

26:17

$250,000 a year if you were to get rid

26:20

of all these three preferences that I've

26:22

talked about if you were to remove

26:24

preferences for legacy students just to

26:26

be clear on what that means we're just

26:28

going to admit them based on all the

26:29

other characteristics often times

26:31

they're great students but we're just

26:32

not going to give them an extra boost

26:33

for being a legacy student if we were to

26:37

remove this seeming bias that arises in

26:40

the process where higher inome students

26:43

are getting stronger non-academic

26:45

ratings and if you were to not

26:48

necessarily remove Athletics but just

26:50

make the athletes look like all the

26:52

other students so there's not this tilt

26:54

towards High income students among

26:56

athletes you would increase the share of

26:58

students from the bottom 95% from a bit

27:01

less than 60 up to about 70% a bit less

27:04

than 70 and so what does that mean in

27:06

practice again there about 1,600 1650

27:09

students in the average entering first

27:12

year class we're talking about another

27:14

150 160 students from more modest

27:17

backgrounds and of course this is not an

27:20

enormous change but it's on the same

27:22

order as people are talking about when

27:25

we think about what's the difference in

27:27

student bodies that might come from

27:30

changes in racial preferences and

27:32

admissions flowing from the Supreme

27:34

Court decision it's on a similar

27:36

magnitude we're going to have 100 maybe

27:37

150 fewer students of color on campus

27:41

and I think it not only affects the

27:44

diversity on campus I think it also

27:47

meaningfully affects the role that these

27:49

schools are playing in upward Mobility

27:52

particularly to these leadership

27:54

positions you make some admittedly

27:57

heroic assumptions and kind of flow

27:59

things through this type of change is

28:01

going to make another two or three US

28:04

senators from the middle class instead

28:06

of from very high income backgrounds and

28:10

let's not overstate this like it's only

28:11

two or three Senators but for a set of

28:14

decisions that literally 12 people can

28:16

decide to make if they want to I think

28:18

that's pretty impressive and that

28:19

doesn't even think about well what if

28:21

the Northwestern and the nyus of the

28:24

world decided to make some of these

28:25

changes as well so my sense is that

28:28

we're not going to remake Society by

28:30

doing this but it's a pretty lwh hanging

28:32

fruit and thing to say is it just from a

28:34

policy perspective you can achieve the

28:37

same differences in the admissions pool

28:40

either by getting rid of the preferences

28:42

that are afforded to high-income

28:44

students or by introducing new

28:47

preferences that benefit students from

28:50

low and middle-income families that are

28:52

particularly academically strong and

28:54

what we show in the paper we kind of

28:56

calibrate it we say like if you were to

28:57

introduce a new preference specifically

29:00

designed to get exactly the same mix of

29:02

students that you would get from

29:03

eliminating these preferences what you

29:05

would need is a preference for low and

29:09

middle-income students that is weaker

29:12

than the preference even that current

29:14

admissions offices put in place for

29:16

legacy students so Legacy students on

29:18

average are about three or four times

29:19

more likely to be admitted you'd need

29:22

really strong academic students from low

29:24

and middle- income backgrounds to be on

29:26

average about twice as likely to be ad

29:27

and that would be a big change but it's

29:29

not like these are changes that go well

29:31

beyond the type of preferences that are

29:33

already in place in the admissions

29:36

process and seemingly pretty

29:39

actionable yeah I mean look I think that

29:42

this is a particular moment of fluidity

29:45

in higher education admissions because

29:48

of the Supreme Court decision people are

29:51

not just reconsidering how to think

29:53

about diversity that's kind of the

29:55

direct effect but once you open up the

29:59

the gearbox I think it then becomes

30:01

natural to rethink a lot of different

30:02

things when it comes to admissions both

30:05

because once there's a process it's

30:07

easier to think about other stuff and

30:09

also because I think that having a

30:10

preference for students from

30:12

overwhelmingly high-income families

30:14

becomes increasingly awkward when you're

30:15

no longer allowed to give preferences

30:18

for students who are clearly

30:20

experiencing very large disparities in

30:22

the run up to college so I think almost

30:25

all colleges are really strongly

30:28

considering a bunch of this stuff some

30:30

of them are doing so in publicly

30:32

announced committees here at Brown

30:34

University I serve on a committee

30:36

including both faculty and trustees that

30:39

are thinking about a bunch of these

30:40

issues other universities are doing it

30:42

more internally only trustees maybe they

30:44

include students all the universities

30:46

are doing this in a different set of

30:48

ways and I wouldn't be surprised if we

30:50

see more change in the way College

30:53

admissions works over the next year or

30:56

two than we've seen in a long time and

30:59

so yeah hard to know what will happen

31:00

but these are an incredibly important

31:02

set of issues to consider and I hope

31:06

we've been able to contribute to that

31:07

debate as an academic all you can ask

31:09

for is that people will listen policiy

31:11

is up to them there's a lot of factors

31:13

that go in it that go beyond the

31:14

research but we've been really both in

31:16

public and had a lot of conversations

31:18

with University leaders about how to

31:20

think about these issues so whatever the

31:22

decision is I'm confident it'll be made

31:23

on the basis of what I hope is a better

31:26

set of analyses an understanding for

31:28

what's going on than we had before

31:29

before I think everyone expected that

31:32

these types of results were occurring

31:33

but I don't think it was really clear

31:35

how large the magnitude was and your

31:37

study certainly contributes to that

31:39

knowledge having data like this and

31:42

these results I think will put more

31:43

pressure on institutions to change than

31:46

just the general suspicion that they

31:48

were privileging a very elite group of

31:51

students one of the things you note in

31:53

the study is that making these changes

31:55

will lead to a more diverse leadership

31:57

ship pool but it may not have as much of

31:59

an effect on intergenerational income

32:01

Mobility could you talk a little bit

32:03

about that that's exactly right and I

32:05

think that stems from some of the themes

32:09

we've been talking about where the role

32:12

that these schools play in

32:14

intergenerational Mobility to leadership

32:17

positions that's potentially very large

32:20

but they're just too small to play a

32:23

role in addressing some of the very

32:26

Broad

32:27

differences in equality of opportunity

32:30

that we see in this country other than

32:32

through kind of the indirect Channel

32:34

which is that I think when you have

32:36

individuals in these leadership

32:37

positions that come from a broader range

32:39

of backgrounds you're more likely to get

32:41

policy that's made in a way that takes

32:43

into account some of these effects and

32:45

so that actually leads to some of the

32:49

research that we're really now focusing

32:51

on which is that when you think about

32:54

intergenerational mobility and higher

32:55

education an initial paper that I wrote

32:57

on this decomposed the problem into what

33:01

we called access that's who's attending

33:03

and the success what happens to the

33:05

students that attend you need both of

33:07

them to be working together to have

33:08

intergeneration Mobility if either of

33:10

them is absent then you have less

33:12

mobility and what we found was that

33:15

different types of Institutions seem to

33:17

have problems in different areas so

33:20

institutions that were highly selective

33:23

not only the IV plus institutions but

33:25

honestly also some of the public

33:26

institutions in the country their lack

33:29

of effect on mobility in large part was

33:31

coming from the relatively undiverse set

33:34

of students on an income Dimension that

33:36

were attending their schools many of

33:38

these schools again both public and

33:40

private the share of students Come From

33:42

The Bottom 20% of the income

33:44

distribution is really just three or 4

33:46

perc really not large at all so we

33:48

really wanted to separate the question

33:51

for these institutions of how do you

33:53

improve Mobility through increasing

33:55

access with the situation for what is a

33:59

very different set of Institutions not

34:01

just the elite public institutions but

34:03

some of the Open Access institutions the

34:04

community colleges where there not that

34:07

access can't be improved but I think

34:10

much more the problem is that in many

34:13

cases students are attending these

34:14

institutions and not being propelled

34:16

upwards in the income distribution in

34:17

the way that we would hope and so that's

34:19

really now what we're focusing on how

34:21

can we first measure in a very broad way

34:24

what these different institutions and

34:27

programs are doing in order to propel

34:28

students up the income ladder really

34:31

give them the the skills the human

34:33

capital the social capital in order to

34:34

get good paying jobs and move upwards

34:37

after that in their career and then what

34:40

are the policy levers that you would

34:42

pull in order to improve that the way I

34:45

like to think about this is suppose that

34:47

you gave the governor of california10

34:50

billion to improve upward mobility and

34:52

education in his state would you want to

34:56

get get more people going to Cal State

34:58

instead of the California community

35:00

colleges is it important that you not

35:02

only go to a cal state is it important

35:04

that you go to a particular Cal State is

35:05

it important that you have a particular

35:07

program are some programs much more

35:09

effective than others should we be

35:11

encouraging more people to go to

35:12

community colleges even if that Coston

35:14

has fewer people going to Cal State do

35:16

we want more people to start a community

35:17

colleges and transfer up to Cal State do

35:20

we want more people to not start at

35:22

community colleges because it's better

35:23

if you start directly C State there's

35:25

all these different questions there's

35:27

been some great research on different

35:28

aspects of it but I think with the data

35:31

that we have we're hoping to provide a

35:33

more unifying framework to think about

35:35

what are the particular places where

35:37

there's more or less success for

35:40

students again defined as like the

35:41

causal effect of attending these places

35:43

and how can we expose more students to

35:46

high success environments either by

35:47

moving them around or by changing what

35:49

the programs are in your intro we

35:51

mentioned that you were a member of the

35:53

US Treasury Council on racial equity and

35:56

the co-director ctor of opportunity

35:57

insights could you talk a little bit

35:59

about what these organizations do so

36:02

opportunity insights is a research and

36:05

policy organization that I run jointly

36:08

with my co-authors Nathan hendren and

36:11

Raj chedy and what we're doing there is

36:15

trying to put together a research agenda

36:17

to understand upward Mobility both from

36:20

an academic and a policy perspective

36:23

research that involves this kind of big

36:25

data has evolved over the last decades

36:29

to almost look more like a science lab

36:32

where it's very team oriented it's not a

36:35

professor and her keyboard or chalkboard

36:38

just kind of plugging away in isolation

36:40

anymore and opportunity insights is a

36:43

way for us to organize all of that team

36:45

in terms of there are other faculty that

36:48

we collaborate with there are graduate

36:49

students we collaborate with there are

36:51

research assistants we collaborate with

36:52

there are visitors at all the different

36:54

levels that we collaborate with and so

36:56

opportunity insights is really the

36:58

organization through which we just do a

37:00

lot of This research and try to

37:02

translate it to help policy makers and

37:04

whatever that means depending on the

37:06

research the treasury advisory Council

37:08

on racial Equity is very very different

37:11

treasury is one of the largest agencies

37:14

in the federal government and it has

37:17

many different policies that directly or

37:20

indirectly affect racial equity in ways

37:23

that are obvious or not obvious and the

37:26

purpose of this advisory council is to

37:30

bring together people from many

37:32

different aspects of society that are

37:35

relevant to treasury's financial

37:38

policymaking so there are a couple of

37:40

academics on the committee like me but

37:43

there are also people who run financial

37:46

institutions there are people who run

37:48

nonprofits that deal with financial

37:49

institutions people who run

37:51

non-financial institutions more

37:53

businesses and the idea is to be a group

37:57

that can both proactively offer

38:00

suggestions to treasury in terms of how

38:03

they can change things either out of

38:05

blue sky or on particular policies that

38:08

are undergoing active policymaking as

38:11

well as a resource for them to turn to

38:13

when they say look like we're trying to

38:15

figure out example is a lot of the focus

38:18

of Treasury over the last two years has

38:20

been the implementation of the IRA bill

38:22

which includes a lot of tax incentives

38:25

for green investment how can they

38:28

Implement all of those tax credits how

38:30

can they write all those regulations in

38:32

a way that really does so to support

38:36

racial equity and to make sure that

38:39

Black and Hispanic and Native

38:41

individuals are not left behind in a way

38:43

that unfortunately has been too often

38:45

the case in our nation's history so

38:48

that's far from a full-time role we meet

38:50

once a quarter in public meetings and

38:51

try to offer our suggestions and even

38:53

again the suggestions span how treasury

38:56

should Implement different regulations

38:58

from even how treasury can make research

39:01

on racial Equity more accessible or make

39:04

data more accessible to support more

39:06

research so that there's more broad

39:08

knowledge when it comes time for policym

39:11

you're doing some really exciting and

39:12

interesting things thank you thanks so

39:15

much for your work and sharing it with

39:17

us today but we always wrap up by asking

39:19

what's next so I talked about some of

39:22

the work in the college space but I'm

39:25

trying to think about

39:27

other parts of upward Mobility as well

39:31

to understand how environments or

39:35

policies contribute to these disparities

39:37

or what policies can help alleviate them

39:39

and the big theme in some of my recent

39:42

work is to try to broaden our measure of

39:45

Mobility to go beyond these purely

39:48

economic measures it's a natural place

39:51

to start both because having a higher

39:53

income is something that is kind of

39:55

meaningfully related to the quality of

39:57

one's life and also because it's pretty

40:00

consistent data to measure income but I

40:02

think even economists will admit to you

40:04

that income is not the end of it and

40:07

we're trying to think about other ways

40:09

not only to measure people's well-being

40:12

thinking about health thinking about

40:14

social capital for instance but also to

40:16

measure folks influence on broader

40:19

Society so there are Physicians like

40:22

entrepreneurs or scientists inventors

40:26

that that if we generate more innovation

40:29

in society that's not something that

40:31

just benefits the individual inventor

40:32

it's something that benefits Society

40:34

much more broadly and so I think that's

40:37

not only very important as kind of an

40:39

alternative economic outcome but it's

40:41

important to thinking about why

40:43

something like social Mobility goes

40:45

beyond merely thinking about well each

40:48

individual should have their fair chance

40:50

of success these are ways in which just

40:53

society as a whole is better more

40:56

Innovative more engaged when there's

40:59

more upper mobility and and that way I

41:01

think it's really a rising tide that can

41:03

lift all boats so that's a little bit of

41:06

what I've been thinking about recently

41:07

well thank you for taking the time to

41:09

join us we really enjoyed this

41:10

conversation and we really as Rebecca

41:13

said appreciate all the work that you've

41:14

been doing thank you so much it's really

41:16

been a pleasure to talk with you about

41:18

all this work over the last hour and I

41:20

appreciate

41:25

that

41:27

if you've enjoyed this podcast Please

41:29

Subscribe and leave a review on iTunes

41:32

or your favorite podcast service to

41:34

continue the conversation join us on our

41:37

t for teing Facebook page you can find

41:39

show notes transcripts and other

41:41

materials on tfor teaching.com

41:45

music by Michael Gary

41:48

Brewer editing assistance by

41:55

ganes

42:06

C