Max R. P. Grossmann

Economist @UniMelb · Formerly @UniCologne @UT_Dallas · Visited @FCEAJaveriana @Mercatus @Vinson_Centre · Dr. rer. pol. · Working on

[Max R. P. Grossmann @gr0ssmann]

Ihr werdet noch gottlos versklavt werden.

[Max R. P. Grossmann @gr0ssmann]

That’s right. Voters in the past read bills, and jurors were able to follow closing arguments. That’s very well-known.

[Dale Chu @Dale_Chu]

“The students who cannot read a 20-page article today are the voters who will not be able to read a bill, or the jurors who cannot follow a closing argument, tomorrow.”

[Mr. Dave @DaveNumber7]

Your students can't read? Neither can I!

[Max R. P. Grossmann @gr0ssmann]

The Pangram bot breaking character is a not-so-great development. Sutor, ne ultra crepidam!

[Pangram Labs @pangram]

@ShinMarginalScr If someone AI-generates an article, you can't know if they've verified the output, if sources are hallucinated, or if the arguments are their own. And AI models have biases, which, if everyone uses them to write their articles, are liable to coagulate into a monoculture

[Max R. P. Grossmann @gr0ssmann]

@nik_wall I'll stay in here until the light comes

[Max R. P. Grossmann @gr0ssmann]

This may be the most controversial thing he’s posted yet. True coal.

[Nicholas Decker @captgouda24]

And here is my tier list of Econ Nobel Laureates

[Hayashi Heikichi @lianda_edu]

Every publisher should ban the use of statistical software like Stata, R, SPSS, etc.! These tools have greatly facilitated academic misconduct such as p-hacking. Scholars should calculate the coefficients, standard errors, and t-statistics themselves using calculators. This would force them to truly engage in empirical work and raise the cost of p-hacking!

[Max R. P. Grossmann @gr0ssmann]

Is Milei the Argentinian Milei?

[Max R. P. Grossmann @gr0ssmann]

Predictable and indeed predicted by many. This is what happens if you outsource:

[Kane Murdoch @CCguerilla]

Unis, signing on to microsoft, are now discovering how the other shoe falls.

More enshitification for us all.

[Max R. P. Grossmann @gr0ssmann]

So fresh, so clean. https://t.co/dIGjctQ9oh

[Max R. P. Grossmann @gr0ssmann]

These are really good. https://t.co/axBxPH0qyd

[Max R. P. Grossmann @gr0ssmann]

A lesson on equilibrium thinking—and its relationship to property rights, I guess:

[Sylvain Catherine @sc_cath]

Something I never quite understood is the claim that Social Security wealth is “not bequeathable.”

The mechanism is simple. Once Social Security benefits are paid, they are fungible with other resources. For any given level of consumption, those benefits protect retirees’ other assets. Without Social Security, retirees would have to draw down those assets more quickly.

So, yes, Social Security benefits are not literally what gets bequeathed. But they are often the reason other assets can be bequeathed.

The causal empirical estimates are strong. Lee and Tan (2019) find that a $1,000 increase in annual benefits raises bequests by $16,740. That is a very large effect, especially given that retirement often lasts around twenty years.

There are important nuances. If you die early, you lose the remaining benefits, and so do your children. But if you live long, you benefit from the longevity insurance embedded in the program.

In that sense, Social Security wealth is not literally bequeathable, but it can very much increase bequests.

[Max R. P. Grossmann @gr0ssmann]

Erklärt ihr euch bereit, für eine totalitäre stalinistische Diktatur zu sterben?

[Michael Christl @chr_mich]

Eigentlich ist das alles relativ einfach: Die „Armutsgefährdungsquote“ ist vor allem ein Ungleichheitsmaß. Der Begriff suggeriert Armut (wie auch im Tweet unten), gemessen wird aber die relative Position in der Einkommensverteilung.
https://t.co/JHwvBQFuwW

[Pangram Labs @pangram]

What?? 45% of psychology survey responses are LLM-generated?

[Max R. P. Grossmann @gr0ssmann]

Such an annoying Claude Code bug: Ctrl+o to inspect the transcript, then it irrecoverably halts when trying to perform another action in the conversation.

The software is clearly made for people who just "YES AND" everything.

[Max R. P. Grossmann @gr0ssmann]

Is Abelardo de la Espriella the Colombian Milei? https://t.co/mDVQHbySpm

[Max R. P. Grossmann @gr0ssmann]

This is why economics really matters. #RevealedPreferences

[Marc Porter Magee 🎓 @marcportermagee]

What's interesting about this trend is how far back it goes. There has been a steady decline in how important people say college is today since 2010.

[Max R. P. Grossmann @gr0ssmann]

This is not an issue. Your experiment is not big data! The first variety has virtually no advantage for small n, but it makes everything including the analysis more painful and error-prone. Do not do it.

[Max R. P. Grossmann @gr0ssmann]

Just LOL at people who still teach

country = models.IntegerField(choices=[(1, "Australia"), (2, "America"), (3, "Germany")])

instead of

country = models.StringField(choices=["Australia", "America", "Germany"])

in oTree (or uproot).

[Max R. P. Grossmann @gr0ssmann]

In Ungarn schlägt die Unseredemokratie jetzt richtig zu: https://t.co/JxNMOW4HYR