Inequality/Mobility. @NBERpubs. 3rd Year Econ PhD, The Graduate Center. 2nd Lt., @USNationalGuard. Former Congressional Staff.
Brooklyn, NY
Joined July 2009
- Tweets8.9K
- Following988
- Followers6.6K
- Likes22.6K
Pinned Tweet
🚨 New WP "The Inequality of Recessions" 🚨
Investigates many #SahmRule #Recession indicators, expanding on the work of @Claudia_Sahm.
Showing downturns are detectable sooner within vulnerable groups and more comprehensive measures of underemployment. Finding that recessions hit some groups/places/sectors first.
papers.ssrn.com/sol3/papers.…
#EconTwitter #JobsReport
Just wanted to put out there to @HarvardEcon that I will 100% be a scab for any/all online classes.
Liberals won an improbable map, largely by running moderates in deep-red states, with swings up to D+26. Proving outreach works and partisanship can be overcome.
Liberals need to use the mandate to foreground economic policies that improve the material well-being of Americans. #EconTwitter
In a D+5 state, in what looks like will end up being a D+16 election, El-Sayed is somehow running a nail-biter race.
My model projected a D+8... This looks like a D+16.
markgsheppard.github.io/2026…
I love that agents have lowered the barrier to entry for interactive visualization, but we need to include mean reversion and/or some measure of polling bias in these models so we don't repeat 2016.
Built my dashboard for the evening ... Starting from my model predictions (based on polls) as priors, it will update and rerun the model as county level returns start coming in, based on a factor-model for polling error. Also reports updated betting market numbers all under one roof. [Did a simulation of the 2022 evening, seemed to work pretty well.] I'll post some updates over the course of the evening, provided it doesn't crash out.
I do the same thing at my Statistics class @BklynCollege411.
I appreciate this exercise in general.
Econ. policy is downstream of politics, and it's wrong to pretend otherwise. I appreciate Prof. @JasonFurman registering his priors, then giving voters a step-by-step guide that punctuates with the limits of his expertise.
Final 2026 forecast: Democrats favored for a thin sweep. House 229–206, Senate 51–49, with Ohio the tipping point.
Dems are running ahead of state baselines almost everywhere; Michigan is the exception.
markgsheppard.github.io/2026… #EconTwitter
1. I love that AI makes making data visualization more accessible
2. The problem here is that the WB's definition of "poverty" is a very, very, very, very low threshold to clear for any advanced economy.
1. More billionaires are associated with less poverty.
2. Increasing billionaires per capita by ten fold lowers poverty by 43 percentage points.
3. Sweden has more billionaires per 10M people than the U.S. (34 vs 30)
4. Correlation is fairly strong (R-squared of 0.6 for two variable cross-section data)
4. Since the graph is only a correlation, what do you think is the cause?
🔗in🧵
Mark G. Sheppard retweeted
Errors in the research of Emmanuel Saez and Gabriel Zucman marginalrevolution.com/margi…
Haven’t seen a post get ratio’d so unanimously by #EconTwitter like this since this site was bought by a trillionaire and turned into 8chan.
Errors in the research of Emmanuel Saez and Gabriel Zucman marginalrevolution.com/margi…
Mark G. Sheppard retweeted
Thank you Jason Furman @jasonfurman for creating this. As noted by many, this AI evaluation of Saez and Zucman is essentially worthless, replete with the treatment of modelling choices and judgments as mistakes.
Overall, very unimpressed with the current use of AI to critique papers. With reference to critics, if someone uses AI and finds an error in a paper, the author(s) should be contacted to see whether there is an explanation before a grand public announcement is made. With reference to journals, protocols need to be developed now so that the products of these exercises, if valid, quickly appear *in the journal* as errata. The existing system for assessing comments on papers, already working poorly, is completely unequipped to address the widespread reevaluations that are coming.
If you believe that mindless AI evaluations are legitimate then why would you spread the evaluation of Saez & @gabriel_zucman without first subjecting it to its own mindless AI evaluation?
I just did and it turns out the critic's errors all go in one direction.
Today is my birthday!
I encourage my students to create group chats and work together. Today my students let me know that sometimes they take pictures of me during lectures, make stickers of my expressions, and then roast me in the group chat 😂 honestly funniest bday gift ever.
#EconTwitter
Mark G. Sheppard retweeted
If you believe that mindless AI evaluations are legitimate then why would you spread the evaluation of Saez & @gabriel_zucman without first subjecting it to its own mindless AI evaluation?
I just did and it turns out the critic's errors all go in one direction.
I made an explainer video for my WP "The Inequality of Recession" that extends the @Claudia_Sahm rule to different labor market data, hopefully the video makes this work a bit clearer for a broader audience.
Website: inequalityofrecession.com
#EconTwitter
