Many thanks to @mattgrossmann.bsky.social for having me on the @niskanencenter.bsky.social #ScienceofPolitics podcast to talk '26 midterms for the U.S. Senate & House!
TLDR; battleground looks expanded for Senate Democrats compared to year ago thanks to candidate recruitment efforts. Take a listen!
in most cases, female state legislators have lower state legislative effectiveness than their male counterparts, in contrast to congressional findings
onlinelibrary.wiley.com/doi/abs/10.1...
Predicting Congressional Election Results
Presidential approval & ideological direction of public opinion, not economics or party conflict, predict election results. That looks good for Dems
New #ScienceOfPolitics podcast/transcript with Carlos Algara
www.niskanencenter.org/what-predict...
I was a guest on The Great Battlefield podcast for a wide-ranging interview, mostly on connections between political science & practical politics:
greatbattlefield.com/episode/the-...
These build on political science research & data collection on topic discussions in city councils & school boards:
civicsearch.org
We may need academic/practitioner/vibecoder coordination as these proliferate, but an era of more data on local governments is coming fast
On a separate project, there is also searchable databases of city council meetings, votes, and comments: district.myhamlet.com
& interesting analyses using these data:
district.myhamlet.com
The 5th edition of Campaigns & Elections (or 9th if you count the midterm updates) is ready for Fall course adoption & review:
wwnorton.com/books/978132...
School board votes are overwhelmingly unanimous & cover humdrum issues.
From a high school student's vibecoded open database of nationwide school board votes:
school-board-votes.up.railway.app
Nice example of now-possible independent research
I accept quick AI timelines but draw the line at predicting end of credit card & housing fees (tech is not the barrier). Working at a university also makes me skeptical about quick realignment of hiring to take advantage of tech efficiency (especially eliminating admin positions)
I (well codex) tried to update the @adambonica.bsky.social spreadsheet of announced DOGE layoffs (showing that they were at liberal agencies) with the actual employment changes at federal agencies. There is still a relationship, though there were also unannounced cuts at conservative agencies.
My new disambiguation page to alleviate Matt Grossman(n) confusion:
www.mattg.org/disambiguation
Legislators Are Raising Money Instead of Making Policy
By rewarding fundraising, parties miss out on diverse leaders & effective legislators. But states that reform make more policy.
New #ScienceOfPolitics with Michael Kistner on Paying for the Party
www.niskanencenter.org/legislators-...
AI's accuracy in censorship decreases with increasing repression, especially during times of political crisis
onlinelibrary.wiley.com/doi/10.1111/...
Despite cutting lots of federal staff, Trump did not get the spending cuts he requested from Congress:
www.nytimes.com/2026/02/17/u...
Research Obsession Day 9: In honor of president's day, thermostatic public opinion -- a very important topic!
w/ recent research by @mattgrossmann.bsky.social & @cbwlezien.bsky.social on its impact on elections
www.matthewg.org/thermostatic...
Democrats use AI more than Republicans, but it is due to education, industry, and occupation
www.nber.org/papers/w34813
Empirical evidence on concentration, markups, and mergers does not show a widespread decline in US competition
www.journals.uchicago.edu/doi/full/10....
Oklahoma governor is eliminating new tenure at public universities & colleges other than Oklahoma & OK State:
www.insidehighered.com/news/faculty...
These universities will operate with existing tenured faculty but current assistants will be converted to fixed-term; regents still have to accept
So few new legal constraints on admin trying it bigger this year. Constraints are mostly political: avoid unpopular actions in an election year with opinion moving against you. Hiring & grant activity will be early tells; have $ to ramp back up but want to avoid.
Lost in the ICE battles is that the parties agreed on government funding levels & provisions. Dems did not win any real constraints on recessions or DOGE-style actions and implicitly OKd some admin restructuring. But Reps did not win any real $ shifts to match Trump budget.
Re-assessing my Jan. 2025 pod with @mattgrossmann.bsky.social
✔️ Jan 6 pardons
✔️ Subdued opposition but eventual rally
✔️ Wield DOJ vs. enemies
✔️ Chilling effects of DOJ abuse
✔️ High-profile firings
✔️ Mass deportation
✔️ Challenge elections in future
❌ didn't foresee DOGE
❌ didn't foresee ICE/CBP $$$
DOGE & early shock & awe may have helped congress acquiescence in some cases (USAID, CFPB) but led to backlash in others (NIH) or little change (Ed). Where goal was reduced capacity (IRS & CFPB), easy for lower staff to = retrenchment. But unclear what will survive divided gov
Other big spending changes were approved by Congress: CFPB & IRS cuts & ICE increases were in OBBBA. USAID move was implicitly OKd in latest spending deal, though program funding remains. Multi-year NIG grants compromised. So won concessions in Congress deals, but lost others
Need update on spending power moving from Congress to President. Personnel reductions, partly through incentives, are large & presidentialized. But still spent more $ last year & Congress has reversed most planned $ cuts (including NIH & Ed). Agencies lose capacity but not funds
So for USAID, they cut to the operating expenses account 93%, which implicitly accepts the move of USAID to State. That seems to be more acceptance than happened at the Dept of Education, where the staff was cut in half but the funding for operating staff remains.
www.usglc.org/the-budget/c...
The Education Department budget also largely rejects proposed cuts & allocates funds for staffing, IES, & admin but doesn't appear to have additional sticks for stopping administration reorganization efforts:
www.edweek.org/policy-polit...
Any explanations of leeway compared to 2025?
Republican governors show higher levels of populist rhetoric than Democrats, but this difference predates Trump’s presidency and shows no evidence of intensifying following his first election.
journals.sagepub.com/doi/10.1177/...
The budget deal extends international aid, not agreeing with the presidents' cuts but spending less than last year:
healthpolicy-watch.news/9-42-billion...
Can someone explain whether they implicitly approved the move of USAID activities to State, disapproved it, or remained silent?
far-right parties benefit electorally when the current government is on the left because there is an ideological shift to the right among the electorate when left-wing parties govern
www.cambridge.org/core/journal...
Reframing policy arguments with opponents' moral foundations did not change policy opinions across 5 issue areas:
www.tandfonline.com/doi/full/10....
