Forthcoming at the Journal of Political Economy: We find that consumer product markups increased more than 25 percent from 2006 to 2019.
One contribution is an approach to estimate IO-style models at scale, yielding flexible consumer preferences and estimates of marginal costs.
We've significantly updated our paper on modeling + measuring systemic discrimination! Check it out:
www.dropbox.com/scl/fi/ab5yx...
(cc @aleximas.bsky.social + @aislinnbohren.bsky.social!)
A short 🧵 on what's new...
For more, and full methodology, check our website www.causal.claims/risks-to-rep...
TWO Virtual Mentoring Events Jobs for Economists: Transitioning Between Sectors
The Society of Government Economists is holding back-to-back mentoring workshops on transitioning between sectors
The first one is on March 4th at 6pm EST
The second is on March 13th at 7pm EST
mailchi.mp/a681e5651456...
Useful exam template!
FANTASTIC
ICPSR at U Mich is coordinating the archiving of at-risk federal data (all of it?)
You can upload data you have & search for data you don’t have
www.datalumos.org/datalumos/
An examination of the value managers create finds that managers apply different skills in different proportions depending on the production technology in which they are embedded, from Alan M. Benson and Kathryn L. Shaw https://www.nber.org/papers/w33431
A line graph depicting the average number of days it took to install a gigawatt of solar energy capacity worldwide from 2001 to 2023. The vertical axis represents the number of days on a logarithmic scale, spanning from 1 to 1000 days. The horizontal axis indicates the years from 2001 to 2023. In 2004, the average time was about 1 year, which is marked on the graph. A downward trend shows significant decreases in installation time over the years. By 2010, it took under 30 days, and by 2015, just over 7 days. The most notable point is in 2023, where it dropped to about 1 day on average for installation. The data source is attributed to IRENA (2024), with calculation by Our World in Data, and the chart is identified as CC BY, indicating it is licensed for sharing and adaptation with appropriate credit.
In 2004, it took the world a year to add a gigawatt of solar power — now it takes a day
Using a new panel-data approach to document strong declines in the returns to unobserved skills in the US over the late-1980s and 1990s. Skill volatility rose over this period, from Lance Lochner, Youngmin Park, and Youngki Shin https://www.nber.org/papers/w33407
These inaccessible files include Census / ACS flat files, TIGER/Line shapefiles, and much of the data documentation / guidebooks on the website.
However: you CAN still access data via API as well as on data.census.gov.
I just found out that @epi.org has updated and modernized its invaluable archive of State of Working America data and tabulations. Check it out! data.epi.org
HT: @benzipperer.org
Please consider signing this letter to Congress and Administration officials raising awareness about the risks facing our federal statistics. Please RT.
docs.google.com/forms/d/e/1F...
I wish I was proficient navigating IPUMS, but I always get frustrated with the download interface/process. So, I just go download the micro data from Census.
Downloaded the CPS Basic micro data for the last 5 years as a just in case. I probably should grab the ASEC as well.
Call your representatives and ask them to defend public data
URGENT: APDU urges you to contact your representatives, regardless of their party affiliation, and let them know that you, as a constituent, believe that the loss of public data is unacceptable.
☎️ is most effective and takes just two minutes of your time.
Contact lookup and sample script in 🧵
In fact, I have coauthored a whole paper arguing that the networked nature of modern production can amplify the harm of a small or moderate shock...
way beyond what standard models predict.
bengolub.net/snff-2/
4/
Here are the first five sets of slides:
01 Introduction: psantanna.com/DiD/01_Intro...
02 Classical 2x2 setup: psantanna.com/DiD/02_two_b...
03 Clustering issues: psantanna.com/DiD/03_Clust...
04 Functional form: psantanna.com/DiD/04_Funct...
05 Covariates: psantanna.com/DiD/05_Covar...
Lecture 02 is here: psantanna.com/DiD/02_two_b...
This covers the 2x2 DiD setup without covariates + some clustering issues.
OK. As the kids were sleeping, I managed to clean up the introduction lecture.
See Lecture 01 here: psantanna.com/DiD/01_Intro...
I will post these on GitHub later, too!
And polish these slides further
Check it out! (ungated here: www.dropbox.com/scl/fi/y1omw...)
Kirill, Xavier, and I tried hard to make this useful to wide groups of readers: new users of shift-share IVs, experienced users, and even those who don't know they that they are users
So I think R and Stata have their pros and cons. I like both, and I use both. In terms of power, cost, and flexibility, there is no doubt that R has an advantage over Stata. While R and Stata can both do econometric analysis, I just find Stata easier and more concise to code and implement.
Can anyone recommend papers that compare RCT and quasi experimental estimates for the same research question?
#econsky
Eating too much cake is the sin of gluttony. However, eating too much pie is okay because the sin of pi is always zero.
🖖
Just a general thought, but the experience on Bluesky feels like how Twitter used to be. I think I’m ready to delete my Twitter profile after 15 years.
🚀 I’m looking for a Postdoctoral Fellow for 2025-2026! Join me at Penn to study e.g. monopsony power in the labor market, technology, jobs & inequality, and income support programs. Apply: sp2.upenn.edu/employment/#.... Please share with your network! 🙏
Welcome new #Economists moving to #EconSky!
I have a thread of Starter Packs in the link if you are seeking your people. The lack of algorithm means *you need to follow people* otherwise you will experience BlueSky as dead.
If I've missed an Econ-relevant Starter Pack, let me know in my DMs.
I think I got them (location: Howard County, MD)
What kind of childhood makes a top scientist? Is it enough to have all the right traits (brilliance, grit, etc) or do you need the right family too?
And why should we care? A 🧵 on our paper on the Nobel Laureates.
A teaser: the income distribution of the laureates' fathers.
1/N
The agency improves its initial estimates as slower, more-accurate info arrives. This happens on long-settled schedules using established, proven procedures.
Sometimes political actors dislike the economic news BLS reports & attack it.
They have never had evidence of wrongdoing, just BS.
