Big city realignment
A failed messaging pivot, the need for moderation, shattered assumptions, downballot Dem survival, high turnout wouldn't have helped Harris, polling error was still bad, mobile is eating the world
No. 340 | December 6, 2024
🇺🇲 2024
155,234,550 votes counted based on the Cook Political tally. Our 153.7M estimate continues to be very close.
U.S. presidential vote swing in the Trump electoral era, 2012 - 2024.
There was a realignment in New York City, with Democratic votes dropping precipitously in large swaths of the city and Trump votes growing or holding steady.
More maps of urban Democratic disappointment.
While Trump’s urban gains were significant, another New York Times analysis finds that they can’t necessarily be tied to failures of urban governance.
David Shor calculates that immigrants swung 23 points to Trump.
Jacobin ran an analysis with a bunch of charts showing Kamala Harris turning away from economic populism and towards “democracy” rhetoric in the final weeks. The point should not be lost on anyone that this messaging pivot almost perfectly with Harris’s October polling decline.
Democratic candidates need to be perceived as more moderate because conservatives outnumber liberals in the electorate. This is a problem that Kamala Harris did not solve.
NBC’s Alex Seitz-Wald breaks down some of the assumptions shattered during the 2024 presidential election
1. Higher turnout benefits Democrats: Democrats have long taken for granted a simple truism: The more people who vote, the better for Democrats. That may have been true once — though that is also unclear — and it’s a feel-good story for a party that aligns itself with democracy.
But in the Trump era, Democrats have become the party of more reliable voters (college-educated, higher-income and older voters), while Republicans often stand to gain by turning out low-propensity voters (non-college-educated and blue-collar voters) who are mostly apolitical but like Trump.
2. Democrats are the party of the working class: For more than a century, Democrats have viewed themselves as the party of workers and the GOP as the party of the bosses. Strongly aligned with labor unions, Democrats have supported welfare programs and populist economic politics like higher taxes on the wealthy.
In 2024, Democrats lost the working class by the two most common measures — income and education levels. NBC News exit polls show Trump won voters without college degrees 56%-42%, while he narrowly won voters with family incomes of $30,000 to $100,000 annually. Kamala Harris won voters with annual incomes over $200,000.
3. Trump can’t expand his base: This is one reason why some Democrats viewed Trump as beatable, especially after his 2020 defeat. But on his third run for the White House, Trump expanded his base of support into major cities, onto Native American reservations and into heavily Latino communities.
4. Latinos and immigrants will vote against restrictive immigration policies: Democrats have based their Latino outreach and immigration policies around this implicitly accepted assumption.
But Trump had the best-ever performance for a Republican presidential candidate among Latinos, according to NBC News exit polls, outright winning Latino men, while increasing support among Asian Americans and in immigrant communities from Dearborn, Michigan to Lawrence, Massachusetts.
5. The Electoral College is biased against Democrats: Until this month, both Republicans elected president in the 21st Century lost the popular vote, leading many Democrats to conclude the Electoral College is structurally biased against them.
The idea has some merit, as big blue states like California and New York are unrepresented relative to low-population red ones like Wyoming. But Trump won both the Electoral College and popular vote this year as Harris’ performance fell in those Democratic bastions.
Democrats played their incumbency and candidate selection hands better this cycle, with the result that while there are ~230 Trump districts per GOP redistricting guru Adam Kincaid, Republicans will end up with just 220 seats in the new House.
Nevada was a win for Trump, but individual cast-vote records show high levels of ticket-splitting supporting downballot Democrats — especially among Hispanics.
Bracing data from Kyle Kondik: Democrats won as many as 13 Senate seats from deep red states this century are now down to zero, with few prospects for getting any of these back. Partisan sorting means that Democrats could be locked out of a Senate majority for a long time to come.
Eye-popping numbers from Nate Cohn:
Overall in Times/Siena data for Clark County, Ms. Harris led registered Democrats who voted in 2024, 88 percent to 8 percent. But she led by a much narrower 71-23 among the registered Democrats who stayed home.
Similarly, Ms. Harris had an 87-8 lead among Black voters in Georgia who turned out and voted, but she led by 54-35 among Black Georgians who stayed home.
Overall, higher turnout would not have helped Harris that much, since nonvoters in core Democratic groups were much more likely to be defectors.
One problem with “the polls were fine” narrative is that while the magnitude of the error was smaller relative to 2020, the polling misses in almost all the individual races over-estimated Democrats.
📰 Media Habits
Michael Beach’s State of the Screens is a must-read. A few charts from his most recent edition drive home how the mobile is eating everything, with implications for both marketers and the polling industry.
Mobile penetration is close to 100% and smartphones are at 91%.
Just 29% of households have a working landline phone…
Connected TV is growing even faster than mobile, and advertisers are racing to keep up…
🤖 Artificial Intelligence
ChatGP’s o1 model is out for everyone, along with a $200 (!) monthly pro plan geared towards math and science researchers.
Great analysis. The swing vote maps are particularly interesting.