Best of #econtwitter - Week of April 3, 2022 [1/3]
Welcome readers old and new to this week’s edition of Best of Econtwitter. Thanks to those sharing suggestions, over email or on Twitter @just_economics.
This is part one of three, with a special Saturday edition in order to smooth the large number of very good papers this week. Part two is here; part three is here.
Paper summary threads
Striking new paper @DrDaronAcemoglu, Alex He, and Daniel le Maire: Eclipse of Rent-Sharing: The Effects of Managers' Business Education on Wages and the Labor Share in 🇺🇸 + 🇩🇰. Managers with MBA ➡️ wages ⬇️ + labor share ⬇️nber.org/system/files/w…
They draw on 🇺🇸 and 🇩🇰 admin data and show event studies of managers with MBA / business degree entering firms. Wages and labor share ⬇️, no noticeable productivity effects. 🇩🇰 event studies below.
@ryanmcdevitt @BetseyStevenson Maybe I’m misreading it, but it looks like positive point estimate on output of about the same magnitude (albeit considerably less precise). Even if identification is solid, I wouldn’t characterize this no effect on output - just not precise enough to say much
^Substack cutting of the image again here
🚨With @ProfSongMa, we recently released an updated version of our paper “The Education-Innovation Gap” as a NBER wp/CEPR dp. Here’s a short thread if you’re curious about what we do & find! 📜1/10
We decided to try to measure the extent to which courses are “at the frontier,” i.e., they cover recently disclosed knowledge. Our conjecture was that this knowledge is key for labor market success, innovation, growth.4/10
We constructed this measure-the E-Igap-as ratio of text similarities bw each syllabus and papers published a)many yrs ago and b)in last few yrs, using the text of 20Mpapers in top journals across years. Pic below summarizes the approach we use to do this text analysis.5/10
Our main finding is simple: access to the economics major (first stage) substantially increases students' wages in their mid-20s (reduced form).
New working paper!
No-strings, 4-yr tuition GUARANTEE increases apps 28 pp & enroll 9 pp
…but requiring students go thru needs-analysis to confirm eligibility increases apps just 9 pp & enrollments ZERO
Paper: tinyurl.com/bpeu8fen
Appendix: tinyurl.com/5dd7376v
NBER @nberpubs
Interesting discussions
Alright, the results are out! Work hours:
Untenured faculty > PhD students > predocs > tenured faculty
anthonyleezhang.eth @AnthonyLeeZhang
^friendly reminder that self-reported data are fake news, but
Most RCT papers I read these days use the post-double selection (PDS) method by Belloni et al (2014). The approach uses the Lasso method to select control variables (X) that are EITHER predictive of y OR predictive of treatment (T). But I think this is wrong. Thread below.
Part of a thread about how the British Journal of Sports Medicine games impact factor (IF) score. Apparently IF divides ALL cites to a journal by a count of QUALIFYING article types. You can game IF by publishing stuff that gets cites but doesn’t qualify.
James Heathers @jamesheathers
