Best of #econtwitter - Week of January 9, 2022 [3/3]
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 three of three (hopefully a rare three-part issue). Part one is here; part two is here.
Paper summary threads
Physicians who switch to a more intensive group practice immediately increase the amount and cost of care they provide to patients, with no detectable change in readmissions or mortality outcomes, according to new research by Joe Doyle & Becky Staiger. nber.org/papers/w29613
🎉Out this week:
Children across societies enforce conventional norms but in culturally variable ways pnas.org/content/119/1/…
@PNASNews
with @EstherHerrmann2 @DBMHaun and the twitterless Marie Schäfer, Henriette Zeidler, & Michael Tomasello
1/n
🚨 Excited to share a new paper in @AnnalsofIM 🚨
Our goal: LaLonde (1986) but in context of assessing heath plans...
We ask: How well measures of plan performance from observational data match RCT-like evidence?
Results were concerning. A 🧵... (1/5)
acpjournals.org/doi/10.7326/M2…
We studied a unique context in which 2/3 of Medicaid enrollees were randomized to 1 of 5 plans ("RCT-like" evid.) and 1/3 selected a plan ("observational" evid.).
Using risk adjustment methods on observational sample, we generally could not recover the RCT-like estimates. (2/5)
New working paper: Mark Egan, Hanbin Yang, and I use a comprehensive dataset of 401(k) retirement plans from 2009-2019 to examine heterogeneity in investment behavior. For example, 44 percent of assets are allocated to U.S. equity funds, but this varies widely across plans... 1/5
NBER @nberpubs
In the US, more educated immigrants raise local vote share for Democrats, less educated immigrants reduce it.
Halving immigration overall would have hurt Democrats in 2016, since former effect exceeds latter
New research by Mayda, Peri, and Steingress—> doi.org/10.1257/app.20…
New research shows that short-term high-interest digital credit in Malawi had no significant impact on well-being. Notably, a financial literacy intervention intending to increase awareness about risks and costs actually increased (!) demand and repayment, increasing overall risk
How can we map predictions of such notional unified consumption models to the real world? In particular predictions about marginal propensities to consume (MPCs)?
This is the question we tackle in a short paper with David Laibson and Peter Maxted
benjaminmoll.com/MPX/
2/
Just got to discuss a compelling new paper on the downstream effects of increasing abortion access from @ninarbrooks and @ZoharTom at #ASSA2022. An (event study) picture is probably worth 1,000 (or 280) words:
More: patent enforcement and litigation; cyclical labor market reallocation; air quality and elections; “green patents”; optimal tax theory; intersectional inequalities in science; raising p=.05?; IO and MP; macro SVAR; monetary policy and sovereign default risk; measuring preferences for public goods
Interesting discussions
A handful of thoughts on flyouts for job market candidates.
One of three 🧵s – threads 2 and 3 have tips specific to in-person flyouts and remote flyouts
#EconTwitter #EconJobMarket #AcademicTwitter #AcademicChatter
^part 2 on remote flyouts; part 3 on in-person
Great thread about how the credibility revolution widened economics’ scope by introducing tools that didn’t depend on economic theory. Sometimes I narrowly feel that requiring clean causality limits the set of questions but often forget just how many domains economics now covers!
scott cunningham @causalinf
^poll, click through to see results (n=190)