Best of #econtwitter - Week of May 9, 2021
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.
Paper summary threads
Integrates amazing data on scientific funding, scientific papers (all types, including econ), and paper citations in government docs, patents and media. Good news: papers that are “hits” in the academy (top 1% of citations) are much more likely to be cited in other domains. 2/4
^“massive data gathering exercise to piece together who funds scientific research and who uses it for what”
Super interesting paper finding that the US extending patent protection to plants in 1985 led to a surge in new crop development and increased (especially large) farm profits and land values economics.mit.edu/files/18687
A key finding in our new version:
Because spillovers are so large, the firms that receive R&D grants only capture about 25-50% of the net patent value their R&D stimulates
Very cool paper analyzed Doctor Who episodes to find a way of increasing creativity: regularly shaking up a team by adding new members with new connections. Switching in team members upped creative quality; going against manager’s desire for stable teams. journals.aom.org/doi/full/10.54…
The growing importance of decision-making is widespread. I stitch together two datasets of job vacancies over time and plot the share of job ads that include word stems like “decision making”, weighting it to make it representative. 4/x
Here is a new JEP piece on the rise of research teams, with messages for economics. The short story is:
#1 Research teams have big benefits. Coauthored papers have increasing impact advantages, and solo-authored work is increasingly rare – in economics and other fields.
#2 But teamwork obscures credit, which is central to the reward system of science. Junior economists now produce little if any solo-authored work before tenure, leaving letter writers and promotion committees to decide tenure on increasingly opaque grounds.
Fantastic new paper on religion, science, and economic growth. Also, measuring religiosity with names is super-creative.
Jeanet Sinding Bentzen @JeanetBentzen
Before the pandemic, WFH accounted for ~5% of working days (2017-2018 American Time Use Survey)
During the pandemic it has risen to ~50% in our data, ~10x the pre-pandemic number
After the pandemic, we project it will be ~20%, ~4x the pre-pandemic number
4/n
A big study on COVID remote work using objective data on 10k IT professionals finds pretty bad outcomes for all. The charts tell the story: productivity fell 20%, but output was stable because people worked 30% more hours. Also employees got less coaching. bfi.uchicago.edu/wp-content/upl…
Aaron documents how, via bid-rigging, a cartel supported extremely high insulin prices in Mexican, and how after changes in policy to limit the cartel’s effectiveness, insulin prices collapse by more than 75%.
How should we discount the consumption of future generations? In this new WP, I show that Pareto implies:
(1+SDR)=(1+r)*mu
where:
SDR = social discount rate
r = interest rate
mu = the relative distributional weights of two people born one year apart
In the fashion category, the presence of ad-initiated searches is very prominent – 15% of all clicks and 53% of all website arrivals are a result of clicks on ads. Consumers click ads early in the search process, and spend less time on the website in ad-initiated searches…4/8
Many national safety nets include cash transfers to poor households. We show that giving transfers to some, *but not all* households can ⬆️ key food prices in remote & poor areas.
AND these price increases can ⬆️ child malnutrition in households that *don’t* receive the cash 2/5
This graph remains my favorite part of the Chetty et al. 2011 Project STAR Paper (didn't make it to the final draft)
academic.oup.com/qje/article-ab…
More: lead pipes null result; present bias parameter meta-analysis; there was no housing bubble
Public goods
Put links to each of the full set of lectures here for my applied methods PhD class here:
github.com/paulgp/applied…
causaldata is complete in R, Stata, and Python! Or at least it now has all data sets from both The Effect by me and Causal Inference: The Mixtape by @causalinf (except for judge_fe, it's too big). Install instructions: github.com/NickCH-K/causa…
Nick HK @nickchk
Interesting discussions
On a more serious note, a short 🧵 about desk rejections
I desk reject around 40-50% of the papers I handle at AEJ-Policy (i.e., reject a paper w/o consulting referees), and I always try to describe the thinking that led to my decision
It's usually 1 of 3 reasons...
Matt Notowidigdo @ProfNoto
There is nothing like writing down a simple model and discovering how incomplete and/or wrong many of your intuitions were, even after a lot of hard thought.
It is one of the very best parts of the job, the experience of which is very hard to communicate to modeling skeptics.
Niche post: Doing a lit review, it's amazing how confused the literature seemed to be about identification in BLP (demand side) given how quickly it became the workhorse. As recently as this year (!) I've had smart people claim that some parameters of BLP aren't identified 1/
Hey #EconTwitter! Thinking about including modern diff-in-diff in a 1st year econ metrics class? Here are some ideas from a class co-taught with @kbburchardi and @arash_nekoei. Key elements:
1) MATRIX VISUALIZATION
2) UNIFIED NOTATION
🧵⬇️
Most of what I covered in one slide:
^last week’s edition has links to about 20 of these memes for different fields