Thought of the day: Every member of the Harvard community should be ashamed by the fact that we openly practice affirmative action for rich white people in college admissions.

(Thread based on Arcidiacono, Kinsler & Ransom's forthcoming JOLE paper)

🧵
public.econ.duke.edu/~psarcidi/lega…
(1) 14% of the admitted Harvard class are legacies. That's 1 in 7! For a typical white applicant (legacies are 70% white), the chance of admission is *FIVE TIMES* higher if you are a legacy. 41% of legacies in the class of 2019 came from top 1% income households.
(Oh, and there is literally a "Lineage" field on the application, apparently. Welcome to the US's hereditary aristocracy.)
(2) 10% of the admitted Harvard class are recruited athletes. (16% of white admits).

The admissions rate of recruited athletes is 86%, compared to less than 5.5% for typical admits (& this underestimates the admission advantage b/c doesn't control for other characteristics).
Why are athletes an issue for equity? Recruited athletes at Harvard skew rich and white. 70% are white. Most of these are well off: 3% of white athletes are economically disadvantaged compared to 15% of white admits. 26% of white athletes come from top 1% income households.
FWIW, skewing rich & white isn't necessarily intrinsic to having recruited athletics! (though that is a whole separate conversation). It appears to be partly because of the varsity sports Harvard offers: often expensive, often elite, including fencing, sailing, and skiing.
(3) The Dean's List. I'm unclear exactly what this is, but seems to include relatives of donors or "potential donors". 14% of white admits are on the Dean's list (compared to 2%, 4%, and 5% for African-American, Hispanic, and Asian-American admits respectively).
Dean's list applicants seem to get a big admissions advantage. Based on raw numbers, about 30-50% of applicants on the Dean's list get admission. From AKR's estimates, a dean's list applicant is 7x more likely to be admitted than an otherwise comparable non-dean's list applicant.
(4) Then there is the children of faculty and staff. They only make up a small share of the class (less than 2% of white & Asian American admits, less than 1% of Hispanic & African American admits). But this pool also skews white and privileged & receives preferential admission.
*The most mind-blowing stat*:

In total, 43% of white admits to Harvard college are athletes, legacies, dean's list, or the children of faculty or staff.

Only about 15% of African-American, Hispanic, or Asian-American admits fit into these categories (See Panel D).
What would happen to the Harvard admit pool if legacy, athlete, dean's list, and faculty/staff children preferences were removed?

Based on all the above, almost certainly Harvard college would become more diverse both by race & income background.
Indeed, AKR use their preferred model to estimate that if these preferences were removed, the admitted student pool would see fewer white admits, more Hispanic and Asian American admits, and the same number of African-American admits.
I hope more of us affiliated with Harvard will be having the important conversations about how we can reduce the school's role in perpetuating elite privilege like this. And the other schools who practice these kinds of policies too, while we're at it.

/End
PS. Needless to say, I hope: I have loved and appreciated the many, many students I've taught at Harvard. Statistically, I'm sure many fell into these categories. My comments here have no bearing on my opinions of each of my wonderful former students. ...
... This is a systemic question, not a personal one, and the more of us can recognize the fact that our privilege played a big role in getting us where we are, the faster we can start using that privilege to make things better for those who come next.
Some additions based on discussion in comments:

1/ To be clear I'm against legacy, donor etc preferences REGARDLESS of the racial makeup because they unfairly privilege the privileged. But the *effect* is also racially disparate and I think that's important to highlight.
2/ These preferences privilege kids from rich backgrounds, who are disproportionately white. The losers are kids from non-rich backgrounds, which includes kids from all races, including non-rich white kids.
3/ Afaict it's unclear from the paper whether the racially disparate effect of these privileges *also* occurs conditional on household wealth & prior connections, or whether it is all mediated through those factors. (Which of course are disparate by race).
4/ Understanding the counterfactual without legacy, deans list, athlete preferences is tricky because it requires assumptions about how behavior responds in other areas. For example, if you assume Harvard sets race quotas for intake, getting rid of legacy prefs ...
... would substantially increase socioeconomic diversity but would not change racial diversity. If you assume Harvard has implicit HH wealth quotas, getting rid of legacy could simply lead to selection on wealth by other means. Etc.

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Anna Stansbury

Anna Stansbury Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @annastansbury

13 Feb
Finland in 1991 gave workers the right to elect representatives for 20% of the seats on the board/management body.

No negative effects for these firms, small positive effects. Seemed to "facilitate information sharing and cooperation rather than shifting power or rents to labor"
This is another must-read paper from @simon_jaeger and @Schoefer_B on shared governance!! You should definitely read the paper, including section 8 where the authors discuss implications.

A few brief thoughts on what I take away on this regarding shared governance reforms:
1/ The right to worker representation introduced was around 20%, and often these representatives don't have formal voting rights (see excerpt from p.1 of the paper below).

It's great to see that this resulted in better information sharing and communication. It also seems ...
Read 9 tweets
11 Feb
I think about this statistic every morning.

24 MILLION US adults "sometimes" or "often" didn't have enough to eat over the last week.

80% of these report that the reason is that they *couldn't afford to buy more food*

(according to @uscensusbureau Household Pulse Survey)

1/
More than 1 in 6 Black non-Hispanic adults and Hispanic adults report being in households where there is not enough to eat.

Of all adults in households going hungry, 42% are White non-Hispanic, 28% are Hispanic/Latino, and 20% are Black.

2/ ImageImage
Fully 1 in 5 of adults without a high school degree report being in households where there is not enough to eat.

But hardship is of course not only a function of education. 1/3 of those without enough to eat have some college or an associate's degree; 10% have a BA.

3/ ImageImage
Read 10 tweets
29 May 20
Thank you to @tylercowen for writing up our paper on @MargRev!

Tyler makes a number of thoughtful critiques. @LHSummers and I respond to a few of these here:

[1/N]

(with more detailed responses to Tyler's Qs, & other Qs, at this link: scholar.harvard.edu/stansbury/decl…)
Q: Is this to do with wage getting closer to MPL?

A: Quite plausible that employers becoming more ruthless and increased use of monitoring technology led to pay being pushed down– closer to MPL in perf comp labor market (but may be below MPL in monopsonistic labor market) [2/N]
Q:What about top earners?

A: We're measuring worker power for the *majority* of workers, excl top earners. Definitely possible that the fall in worker power involved a redistribution of rents up to managers/executives/people who start their own companies (See Section V.C)[3/N]
Read 6 tweets
26 May 20
In our paper, @LHSummers and I argue that the *decline in worker power* is behind many of the major trends that have shaped the American economy in recent decades [1/N]

(WP out w/ @nberpubs and presented at @BrookingsEcon Spring 2020 BPEA. Ungated link scholar.harvard.edu/stansbury/rese…)
We argue that the decline in worker power in the U.S. economy can explain:

(1) the entirety of the decline in the labor share,
(2) much of the increase in corporate valuations, profitability, & measured markups,
(3) a large share of the fall in the NAIRU

[2/N]
Of course, our focus on the decline of worker power is not new: we build on a long history of progressive institutionalist work in econ, sociology, and political science, which identifies the decline of worker power as one of the major structural trends in the U.S. economy [3/N]
Read 22 tweets
13 May 20
What do we know about occupational mobility in the US?
As you may know, there isn’t very good existing data on it.

So @gregorschub, @Bledi_Taska & I construct new occupational mobility data, using an amazing new data set of 16 million U.S. resumes from @Burning_Glass. [1/N]
Resumes are snapshots of workers’ career histories. Assigning occupation codes to jobs, we can calculate transition probabilities from each occ to each other occ, from one year to the next.

We use this data to document 6 facts about occupational mobility. [2/N]
Fact 1: Occupational mobility is high. When workers leave their job, about 25% of them *also* leave their SOC 6-digit occupation. (6-digit codes are pretty narrowly defined…) [3/N]
Read 11 tweets
13 May 20
📢 New WP! Monopsony and Outside Options 📢
(@gregorschub, me, & @Bledi_Taska)

How much do workers' outside job options matter for wages?

This is important to understand the degree of imperfect competition & role of employer concentration in labor markets [1/N]
Paper available at:

SSRN: papers.ssrn.com/sol3/papers.cf…

Direct download: scholar.harvard.edu/files/stansbur…

[2/N]
Takeaway 1: Employer concentration reduces workers' job options. For workers in highly concentrated labor mkts, this matters a lot - esp. for occupations w/ low outward mobility. Within a given occupation, concentration can explain ~21% of wage variation across metro areas [3/N]
Read 17 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!