Mafalda Pratas

I think we can all agree: "There is only one way to do credible research, and it happens to be the way that I myself do research."

Meanwhile scholars will hit the panel data FE regression with the incantation “parallel trends,” that literally means “I assume away non-causal explanations for a correlation,” and think we learned something scientifically special.

“I personally can’t think of a way” is deficient. “We collectively can’t think of a way” is not that.

Also note the identifying assumption in, say, IV regression is not doing anything besides this. “We can’t think of a way for the instrument to directly affect the DV” is actually the argument.

This is an interesting paper.

I continue to sense that we as a field have concluded, either you have an identification strategy as defined here, or you don’t have a valid causal claim. And that is really not true.

Take: moderating in swing districts in 2026 highly likely to work; moderating partywide in 2026 has some evidence it will work; moderating partywide for the long term as a strategy to rid off Trumpism has way more perils and risks than proponents acknowledge.

Not new! In The Political Beliefs of Americans (1967) Cantril & Free found that Americans were "ideological conservatives" opposed to abstract "big government", but "operational liberals" for lots of specific spending. It doesn't help that they have no clue re distribution of spending.

Do voters perceive the CDU/CSU choices as representation and responsiveness? Or merely as electoral opportunism? Most see it as electoral calculus (and this might be costly), but a decent proportion (many who agree) see it as legitimate representation.

More answers in the paper to come!

7/7

German public is also divided (though not equally) between those who think Germany has special responsibility in dealing with right-wing extremists (the majority) and those who think there is no historical responsibility on this matter (25-30% voters).

6/n

- Voter's immigration attitudes also matter. Those who think immigration is an important problem and want to restrict it saw CDU choices in positive light.

- Interestingly, accomodation (of CDU/CSU to AfD) seems more costly, in electoral terms, in the West than in the East!

5/n

However, AfD voters did like CDU actions and reported a slight increase in prop. to vote for them. Note, however, the AfD voters are particularly committed to vote for the party and it is extremely hard to actually change their behavior.

4/n

Still, we can still interesting trends and changes!

Overall, general electorate did not like the Union's actions and showed a slight decrease in propensity to vote for them. Interestingly, CDU voters from last summer also showed slight decrease in prop. to vote for them.

3/n

Measuring counterfactual vote changes is notoriously hard! Our results show small changes, as expected from a single event in real-world, and in line with the recent polls that came out this week. However, note that being able to break a firewall with small electoral consequences is big!

2/n

Last week, the CDU/CSU breached the so-called "firewall" and attempted to pass a restrictive immigration bill with the AfD. How did German voters react? Together with Chris Wratil and Jorge Fernandes, we ran a survey with 2000 German voters to find out!

Read thread below for insights ⬇️

1/n

Deutscher Bundestag - Unionsgesetz zur Zustrom­begrenzung mit knapper Mehr­heit abgelehnt Der Bundestag hat das sogenannte Zustrombegrenzungsgesetz der CDU/CSU-Fraktion (20/12804) mit knapper Mehrheit abgelehnt. Gegen die Initiative votierten am Freitag, 31. Januar 2025,...

You wonder how the CDU/CSU playing with the "Brandmauer" to AfD last week has influenced German voters? @mafaldapratas.bsky.social @jmfernandes86.bsky.social & I have you covered. We surveyed sample of 2,000 German voters. 🧵with first results 👇. Spoiler: It did not go down as well as CDU/CSU hoped.

‼️ Hiring‼️

One postdoctoral position in my ERC Consolidator in Madrid.

Application deadline: end of February.

Job ad & details: tinyurl.com/3dwm96kb

Come and join me and my team in sunny Madrid for 5 years of intensive research and lots of publications, traveling, and fun!

Please share widely.

sometimes representative democracy means you all choose one guy to chuck over the fence and charge the military

Very proud of my dear husband @jmfernandes86.bsky.social, who just got a big ERC Consolidator Grant (2M€!) to study those we do not see and do not represent in European democracies. Many innovations, but I'm particularly excited about the move beyond the urban-rural cleavage. 🎉

I'd love to be added!

Hi! Great list! Would like to be added, if that's okay. :)

I liked in the morning! Could you please add me? Many thanks for all your service!

Hey! Could you add me to Polisky? Thanks!!

Elon said in a livestream with Netanyahu that he would starting charging all users (though it was not a set policy, just a shower thought he had). Plus the 2 week delay in invite codes, yes.

Transplanting this one from the other network, because it's so on point and so recurrent...

Perfection

Let's see how this works...