Overall, Democratic & Republican states don't perform differently across objective indicators, even though you can find polarized policies with real effects: niskanencenter.org/do-democrats-a…
We consistently overestimate the party control -> policy change -> socio-economic outcomes path
On these 3 issues:
Housing: Dem states are more restrictive, though mostly due to local policies
Taxes: Overall, Dem states are more progressive, despite the outliers here
Education: Limited spending & equalization effects of party control, not enough to matter for outcomes
You can’t just compare current outcomes in current Red & Blue states. You need to study whether changes in party control change policies & how those policies change subsequent outcomes. There is good evidence but it doesn’t support an easy path from party to policies to outcomes
• • •
Missing some Tweet in this thread? You can try to
force a refresh
This confusion seems to be widespread. Thermostatic politics does not require Biden to change his policy proposals from the campaign. It also does not require close voter attention to policy detail. It just requires voters to see or expect a leftward change in policy from Trump
eg voters asked in the Trump era whether they want more or less immigration or more or less health care spending are now being asked more or less from a new (or expected new) status quo. Fewer should now say more & more should say less, even if no one has changed ideal views
How much that (widespread & largely mechanical) pattern causes changes in votes or differential turnout is a more controversial ? But no one should think “he said it in the campaign” “they haven’t passed all of it yet” or “voters don’t know the details” undermines the pattern
The Commission pursued a voting rights strategy that maximizes districts with Black population around 40%. Compared to the computer-generated random maps, this looks quite different. Here are draft maps for state House with the highest Black populations:
On partisan fairness, the maps are between perfect symmetry & what would be expected from randomly-drawn maps (which would favor Reps). For example, here is seat share for 38 member senate based on the 2018 Senate results compared to computer & public maps ippsr.msu.edu/sites/default/…
Across Western democracies, the education divide slowly reversed from higher education voters favoring parties on the ideological right in the 1960s to favoring parties on the ideological left by 2020, easing but not reversing the income divide academic.oup.com/qje/advance-ar…
In multi-party systems, the education divide coincided with the rise of Green parties on high-education left & anti-immigration parties on low-education right. In the US, factions arose within the major parties, making the 2-party education divide stronger & income divide weaker
Globally, party vote share among the highest educated has become more correlated with party platform positions on sociocultural issues. Party vote share among the highest income voters remains correlated with its party platform positions on economic issues watermark.silverchair.com/qjab036.pdf?to…
The average swing against the president's party in the midterm election is -3.8% in share of the national House popular vote & -6.2% in House seat share. If that happened from 2020 to 2022, Dems would end up with only 47.7% of the 2-party popular vote & a 45 seat deficit
So far, Biden’s underwater approval has not translated into any sign of an anti-Democratic wave on the generic ballot. But research finds ballot numbers follow prez approval & a thermostatic ideological reaction against direction of policy. An R wave would be historically normal
We’re having a 2024-appropriate election discussion when the electoral task at hand for Democrats is avoiding a massive wave against them in 2022. 1994 & 2010 were huge & impactful National & state-level waves. They were products of large public thermostatic swings, not messaging
There are real trends in the rise & fall of disciplines, but they are slow. Trends in research university tenure-track faculty do not necessarily match trends in the much larger higher education teaching market researchcghe.org/perch/resource…
A big source of inertia is that most research university departments are aging, with assistant professors making up a small share of tenure-track faculty (the social sciences are on the young side) researchcghe.org/perch/resource…
Clinton & Obama comparisons are more for 2024 than now. Both suffered massive losses in 1st midterms, linked to congressional agendas. By re-elections, they had both generic incumbency & a radicalized Republican foil (including on economics) to enable visible triangulation
Low-education voters were traditionally inattentive, meaning both lower turnout & more nature-of-the-times voting. We haven’t yet run a low turnout election or a democratic incumbent under education polarization. But basic midterm backlash dynamics may overwhelm other factors
We don’t know yet how Republicans will look in 2024 (including on economics). 1995-6 & 2011-12 Rep internal fighting (including primaries) & public image had a lot to do with Dem successes in 96 & 12. Left/center conflict could matter less or allow triangulation with Rep foil