Anthony Fox

Pendulum politics

November 2026

At the polling station: the redder the booth, the longer its queue of voters who swung to the Democrats

Key points

The reddest and poorest states swung furthest

Sources: Wikipedia, Bureau of Economic Analysis, Bureau of Labor Statistics, Energy Information Administration, Zillow, US Census Bureau.

About six in ten voters told the AP Voter Poll on 3 November that they were very concerned about the cost of groceries, and roughly half called the cost of living crucial to their vote for Congress.1 Those voters gave the Democrats a lead of about 13.6 points in the House popular vote, the party's largest House popular-vote win since 1976.2

The ready reading is that America voted against its bills. Yet bills did not rise by the same amount everywhere. Household electricity cost 43% more in Washington DC in the first eight months of 2026 than in the same months of 2024, and 1% more in California, so a vote on local bills should have swung furthest where they rose fastest.

The swing barely followed local conditions. What sorted the states was how Republican they had been in 2024 and, after that, who lives in them. Voters appear to have judged inflation as a national price level rather than by their own state's bills.

Did the states hit hardest swing furthest?

The states hit hardest swung no further. In the 46 states compared, the third where household electricity rose most swung 8.0 points towards the Democrats, against 7.9 where it rose least. Rents, home values, natural gas and petrol left no clearer pattern, and neither did federal civilian pay, cut by about 9% across the country since the end of 2024.

Partisanship sorted the swing far more cleanly. The reddest third of states swung 11.7 points and the bluest third 5.0, since the states that had voted most heavily Republican had the most Republican voters left to lose. Poorer states also swung further, but poorer states are also redder, and the link disappears once partisanship is allowed for.

Who lives in a state mattered once partisanship is held still. Within each third of states, sorted by how they voted in 2024, the half with fewer white residents swung 2 to 3 points further than the half with more, 12.7 against 10.5 points in the reddest third and 6.2 against 3.9 in the bluest. A state's Black and Hispanic shares point the same way, which fits the exit polls' finding that Latino voters swung about twice as far as voters overall.3

Farm income is the one local economic measure that adds to partisanship. Farm income per person grew most between mid-2024 and mid-2026 across the Plains, from North Dakota to Kansas, and least across the Southeast, and the states where it grew least swung further than their partisanship alone would predict. Real income per person rose in every one of the 46 states, and its link with the swing is weaker.

What did prices do under Trump?

Prices kept rising under Trump, led by energy. Between January 2025 and September 2026 the BEA's index of consumer prices rose 5.4%, food bought for home 3.0% and energy goods and services 20.2%. Household electricity cost more in the first eight months of 2026 than in the same months of 2024 in 49 of the 51 states, and 11% more across the country, while natural gas for heating cost 17% more and petrol 11% more. Rents rose more gently, by 4.7% across the country in the two years to September 2026 on Zillow's measure, and fell in Denver and Phoenix while rising about 11% in Chicago and Cheyenne.4

Heating and power bills rose fastest, rents slowest

Change in what households paid, %. Natural gas, electricity and petrol from January to August 2026 against the same months of 2024; rent from September 2024 to September 2026. Colour: how far each state sat above (rust) or below (teal) the US change. Rows sorted by the average rise in the three energy bills. Shown: the six biggest and four smallest. The last three columns are the BLS consumer price index for the state's census division, January to October 2026 against the same months of 2024.
Sources: US Energy Information Administration (Form EIA-861M electricity; residential natural gas prices by state; weekly regular petrol by state for nine states and by region for the rest); Zillow Observed Rent Index for each state's largest metro area and Zillow Home Value Index by state. US Bureau of Labor Statistics, consumer price index for all urban consumers (CPI-U, not seasonally adjusted) by census division, where each state shows its division's figure. Natural gas and the CPI columns compare the same months in both years. A dot marks no data.

Did the shocks fall on red states or blue?

The shocks of 2025 and 2026 fell about evenly on red and blue states. Household electricity rose about 12% over two years in both the reddest and the bluest third of states, and federal civilian pay fell about 8% in each. Yet the reddest third swung 11.7 points and the bluest 5.0, which is hard to square with a vote on local hardship.

Prices rose slightly faster in blue states. Red states swung almost twice as far

Sources as in the table above; votes from Wikipedia (2024) and Popular Vote Analytics (2026). Each figure is the plain average across the states in the group. Consumer prices, groceries and shelter are for each state's census division.

So whose inflation was it?

The voters worried about their grocery bills on 3 November were reacting to a national price level about 27% above mid-2019, and about two-thirds of them told the AP Voter Poll that Trump's policies had made the economy worse.1 That verdict was national, which is why the swing reached every state counted, from Vermont to Alabama.

Local conditions decided only how far each state moved. The reddest states had the most Republican voters to lose, and the states with fewer white residents swung further than their partisanship alone would predict. Local shocks, from power bills to federal job cuts, left little mark on the vote.

The inflation that decided the midterms was the country's. Who lived in each state, more than what its bills did, decided how hard it swung.

Notes

  1. AP Voter Poll, survey of about 80,000 midterm voters in all 50 states. PBS NewsHour, "How Trump and Republicans lost the 2026 midterm elections, according to AP Voter Poll", 4 November 2026, and "AP Voter Poll: How key demographic groups voted in the 2026 midterms", 3 November 2026.
  2. Popular Vote Analytics, 2026 House popular vote tracker, unofficial count from 434 of 435 districts, 10 November 2026; Wikipedia, "2026 United States House of Representatives elections", on the largest win since 1976.
  3. The New York Times voter analysis, as reported by the UCLA Latino Policy and Politics Institute, "Four Ways Latinos Shaped the 2026 Midterms", November 2026.
  4. Zillow Observed Rent Index, all homes, smoothed, by metro area, September 2024 and September 2026.

Method and sources