Best of #econtwitter - Week of March 19, 2023: paper summaries
Welcome readers old and new to this week’s edition of Best of Econtwitter. Please submit suggestions — very much including your own work! — over email or on Twitter @just_economics.
Paper summaries
Excited to share a new WP with Ran Abramitzky, @JacobCConway, and Roy Mill about differences in economic outcomes by perceived skin tone among African Americans using census data from 1870–1940
[Short slideshow👇 and short thread below that]
@nberpubs WP nber.org/papers/w31016
^the award for ‘innovation in paper summary tweets’ goes to
SVB got in trouble for investing in MBS. It’s certainly not alone in doing so. Why do banks invest in assets like MBS in the first place? 1/
The "Will GPT automate all the jobs?" paper is out
With participation from @OpenAI, OpenResearch and @Penn
🧵 1/9
^and:
They use human labelling and GPT4 to measure task exposure to Large Language Models and aggregate those over occupations. When doing this, they distinguish between exposure to vanilla LLMs and exposure to *tool augmented** LLMs (v. cool).
They find that 19% of US jobs are highly exposed (>50% tasks exposed) to tool-augmented LLMs. Higher education jobs are more likely to be exposed; jobs that require scientific and critical skills less exposed; computer programmers very exposed.
Excited to share this new paper with the world!
Together with @SamiraMarti and @Florian_Scheuer @econ_uzh, we study whether a wealth tax reduces wealth inequality.
A thread 🧵
1/11
EU Tax Observatory @taxobservatory
I was reading Heckman, Urzua and Vytlacil (2006) because in our cookstoves paper/ongoing followup @BerkouwerS and I have a continuous instrument and a binary treatment.
I needed to draw some graphs to build some intuition for this result, and figured I'd share them. 🧵
Delighted to release a working paper with @_geoffliu and @Tris_Sainsbury on Australia's A$38b pandemic stimulus program in which people were allowed to withdraw up to $20k from their retirement accounts, normally inaccessible until retirement. Thread. 1/
static1.squarespace.com/static/59b0bb0…
Updated draft of "Labor Market Power, Self-Employment and Development" - with @pame_medinaq @mmorl89
Link: drive.google.com/file/d/1l70Jh7…
CEPR DP: cepr.org/publications/d…
A small 🧵 with a recap and summary of new results. 1/n
Just out in @j_econ_growth
Religion has been used to justify rules, such as "abortion is illegal" or "women should not work".
We document that this old fact has occurred for all types of societies and religions, throughout time.
And it hurts democracy.
doi.org/10.1007/s10887…
Nice writeup of a study by @I_Am_NickBloom and Alex Finan: "work" from home flattened the curve of golf demand over the week
nbloom.people.stanford.edu/sites/g/files/…
nytimes.com/2023/03/16/bus…
Finally ready to share a lit review 🧵on recent (econ) papers around effects of interventions targeting parenting in LMICs. . .this is an area w/a lot of exciting new contributions + interdisciplinary work (note focusing on interventions targeting parents - not childcare)
Excited to share this research!
@mrfrank5790 and I usually study skills from the demand side. In this paper, led by exceptional Pitt grad student @hungkchau, we look at university syllabi and the skills they teach.
Morgan Frank @mrfrank5790
🧵1/ New paper: We argue machine learning should be used to generate truly novel hypotheses.
The application: judge decision-making. We find a single factor is heavily related to who gets jailed: the defendant’s mugshot.
What are judges looking at?
Our method unpacks this...
^more
Running a study where you use names to signal race?
Be careful! Names signal more than just race alone.
But (!), fret not; we have pre-tested 600+ names for perceptions of race/class/citizenship.
Paper: nature.com/articles/s4159…
Data: github.com/jaeyk/validate…
🚨We're excited to share a revised version of our retitled WP: "Taking Teacher Evaluation Reforms to Scale"🚨
We dedicate considerable attention to exploring WHY we find null effects, on average, of state teacher eval reforms (plus robustness galore).
nber.org/papers/w30995
🧵
Teaser alert: Silence does != no problems. We find that providing plausible deniability about reporting through exogenous garbling increases reporting of workplace harassment by 46-288%, with a larger increase for more sensitive types of harassment.
NBER @nberpubs
🚨 NEW WORKING PAPER 🚨
We study school closures during the pandemic to show that schools are a critical channel through which children with disabilities learn about SSI. Counties with more virtual schooling in 9/2020 had fewer SSI applications in subsequent months.
1/4
More paper summaries
New Cato working paper on how minimum wages reduced the number of au pairs in Massachusetts.
What are au pairs?
How did the min wage undermine au pairs in MA?
Why is the Biden administration thinking of doing something similar nationwide?
🧵
cato.org/working-paper/…
Since @Andrew___Baker called for a break day, let's go back to our favorite Twitter activity of 2020... discussing DAGs! I'm very happy that our paper "Causal Inference and Data Fusion in Econometrics" is finally forthcoming in the Econometrics Journal. academic.oup.com/ectj/advance-a… 1/
This is my last one for a while, but I hope to see some of you tomorrow!
Are you curious to know how the gender gap in EEOC litigation payouts measures discrimination severity in the workplace and its impact on firm performance?
New work by myself, @YikAu from @umanitoba, and
🚨New project alert🚨
As @AnthonyLeeZhang has pointed out, LLMs like GPT-4 or Bing AI are a serious threat to traditional Q&A platforms like StackExchange. @QuinnMrtn and I are currently estimating to what extent LLMs are affecting Q&A platforms.
Anthony Lee Zhang @AnthonyLeeZhang
THREAD. My latest academic paper, with @jeffreypclemens, tries to answer this question: Do higher minimum wages decrease union membership in minimum-wage-intensive industries?
We find no evidence of a change in union membership among high-skilled workers in these industries.
New working paper: Monetary-fiscal policy interactions when price stability occasionally takes a back seat cepr.org/publications/d… # via @cepr_org 1/4
1/6 I’m excited to share my updated paper “HANK beyond FIRE” (joseeliasgallegos.com/uploads/7/5/1/…), where I relax the Full-Information Rational-Expectations (FIRE) assumption in a Heterogeneous-Agent New Keynesian (HANK) model #econtwitter