UC Berkeley Admissions: Insight of an Insider

A guest post by Vincent Cook

I met Vincent Cook over three decades ago when I was studying at UC Berkeley. After my recent livestream, he emailed me a small motherlode of intel about Berkeley admissions, followed by another informative email. Reprinted with his permission.

---

Email #1

Hi Bryan:

Yes, it really was the Vincent Cook from 30 years ago (maybe even a few more years than that) that asked a couple questions on your Substack chat today. It was good hearing from you again.

I wound up working in the University of California Office of the President for 30 years, retiring in 2022. For my last five years at UCOP, I was a data analyst in its Institutional Research and Academic Planning (IRAP) department, the last few years being sort of like a libertarian mole quietly embedded deeply within woke central of UC. Some of my coworkers were tasked with arguing for the value of UC degrees (in the absence of any genuine market test of course). I was aware of the existence of your book arguing that college degrees function as signals and not as adding value to human capital, but they weren't, so I was intrigued listening in on their attempts to make a non-signaling case for UC degrees. My IRAP colleagues did have access to some very interesting datasets that shed more light on your signaling thesis, even if they weren't thinking about it in those terms.

One of these datasets correlated degrees awarded (and failures to get degrees by enrolled undergrads) with subsequent career tracks (making use of LinkedIn profiles) and with incomes earned post-graduation (the correlations between our student data and state income tax data being performed for IRAP by California's Franchise Tax Board). One finding was that in many fields, the substantial income boost from finishing a degree (as opposed to dropping out even after scarfing up content for several years as an undergrad) didn't really depend on staying within the particular career track one might expect from the field of study for one's degree. Careers in fields completely unrelated to one's degree were also way more remunerative than the uncredentialed--in most fields the most important thing was just to get a degree regardless of what course your career took afterwards. This supports your signalling thesis.

Another dataset involved correlations between various indicators used in undergrad admissions with degree outcomes, and stratifying the data by race looking for sources of increased disparities (this study was done in the context of the woke push to get rid of SAT scores for admissions purposes). The data clearly showed that SATs made no additional contribution to racial disparities, nor did UC itself display any "systemic racism." By far the most important source of racial disparities in this dataset was the inability of the K-12 public schools to prepare black and Latino kids for meeting UC's High School subject requirements. When these results were presented to the UC Provost (who was already an expert on educational outcomes) and I asked him about it, his opinion was that racial disparities really show up in a big way around the 3rd grade, so he wasn't surprised by what my colleagues found.

Another thing I learned from those data was that the SAT and the various admissions proxies IRAP analysts came up with for SATs (notably those related to high school grades) were strongly correlated to subsequent degree completion rates. All these correlated variables in turn all appear to involve one's ability to study for tests and handle the stress of test-taking. In the context of your signaling thesis, it would seem that employers value such skills. Of course, the on-going validity of this presumes that high schools in California wouldn't start engaging in grade inflation in an attempt to game the UC's admission process once the SAT requirement went away; my colleagues didn't seem to be aware of that danger.

It occurred to me that it would be worth exploring the impacts that state-generated privileges, subsidies, and regulations have on the employer demand for the college degree signal. The most remunerative degrees tend to be in fields where they are prerequisite to gaining professional licenses (doctors and lawyers being the obvious examples), but even outside those fields, the credentialed class is clearly preferred for jobs in large institutions that are in turn tend to be favored by the state. Maybe test-taking skills are a good predictor of an employee's prospects of being a faithful minion in a rule-bound bureaucracy, or of an employee's ability to enable their employer to conform in rule-bound or litigation-prone contexts. Or maybe it is the government's labor laws, not skills that are objectively necessary to perform a given job, that incentivize employers to look for the signal. The implication of this is that without all the government interventionism in the economy, employer demand for a relatively highly-paid credentialed class might collapse. Maybe it's not an accident that credentialed vs. non-credentialed strongly correlates to differences in partisan affiliations, attitudes about higher education and religion, and even in fertility rates. There is a lot more work to be done in this area.

By the way, my Epicurean website moved to

https://epicurus.net

and the email address to use is vcook@epicurus.net -- you should update your links accordingly (links to my website and email appear on my Pol Pot article in your Museum of Communism, and a link to my email address also appears in your Anarcho-Capitalism FAQ). My recent writings can be found at https://mises.org/profile/vincent-cook and as comments on articles at various Substack sites (including yours).

Best regards,

Vince

---

Email #2

Hi Bryan:

Yes, you can incorporate my comments into an article, and you don't have to anonymize my name (one of the benefits of being retired). Even better, you can look at the original value-of-a-degree work IRAP produced and come up with your own analysis of it (and check to see if my impressions concerning the results are correct)--it is still available on-line. IRAP has a page with links to a series of topic briefs and a report summarizing the work in a PDF format. Also, the UC Information Center has a page at with a Tableau dashboard with a tab for the alumni earnings data and another tab for the alumni industries data (the tabs are just above the graphs). The dashboard has several useful filters you can stratify data by, including the degree field I mentioned. There is also a link just below the footnote explaining how to download whatever filtered or unfiltered data you find interesting, including images of graphs and cross-tabbed datasets. The Information Center is a great resource for many other interesting datasets in a Tableau dashboard format too.

You could also get some feedback directly from IRAP itself regarding UC degree outcomes (possibly including details about whatever more recent work they've done). The head of the department, VP Pamela Brown, is approachable and can direct you to someone on her staff who can answer your questions. Another contact to follow-up with would be the economist who worked with IRAP on the value-of-a-degree project, Zachary Bleemer. A Berkeley grad student at the time, Zach is now an Assistant Professor of Economics at Princeton (you may be familiar with that institution too). He is a labor economist who studies higher education, so he may have heard of your book by now and have some thoughts about it, or at least would be able to respond to the signaling thesis in detail.

Best regards,

Vince