In a November 5 note, Citadel Securities' Frank Flight1 reads the US midterms as more than a political event. The result, he argues, reshapes three things at once: the AI investment cycle, the Iran conflict and the 2028 political map. The equity and rates implications do not point the same way, and that divergence is the heart of the piece.
Why Does a Vote Matter for Data Centres?
Voters turned local. Flight cites exit polling showing 56% of voters called AI and data centres a crucial or major factor, and voters opposed development in at least 39 of 45 local ballot initiatives. About 45 GW across 350 projects in Texas, Ohio, Pennsylvania and Georgia is exposed between 2026 and 2030, including 24 GW in Texas alone.
The conclusion is deliberately measured. Capital is mobile, so restrictions in one jurisdiction push construction toward others. Flight expects slower building, higher costs and execution risk, but a much smaller hit to aggregate investment. As he writes, the outcome implies "a more challenging investment environment for the AI build out."
Who Wins and Who Loses?
The supply chain splits. Flight expects markets to chase a "peak AI capex" theme, rotating out of memory, neoclouds and energy, and into staples, healthcare, banks, housing, retail and the hyperscalers. Slower deployment favours those doing the spending over those receiving it.
Hardware suppliers face the sharpest risk. Components may be built faster than powered buildings can absorb them, leaving order deferrals and excess inventory, most acutely in memory, optical and other semiconductors. Operators already selling energised compute gain the opposite: delayed competition supports utilisation, pricing power and margins. CBRE's 1.4% vacancy and 80.4% of capacity under construction already preleased reinforce that scarcity. Texas exposure becomes a key stock-selection variable.
Is the Election a Clearing Event?
Not in Flight's view. Removing political uncertainty helps, but a larger-than-expected swing that complicates the AI build-out weakens the clearing argument. Rates are the cleaner trade: slower investment, modest financing deferrals and lower energy prices all favour lower yields, though he cautions against extrapolating a lasting collapse in spending or issuance.
What Happened to the Republican Coalition?
The swing is broad. Flight attributes it to the Iran war, tariffs and fuel prices that hit rural and Midwest communities hardest. Versus 2022, the 14-point Republican lead among men is gone. The lead among white men falls from 28 to 12 points, and among white non-college men from 46 to 30. Non-college voters move from 12 to 4 points, and voters over 65 from 12 to 3. The Democratic margin among women widens from 8 to 18 points, and among college graduates from 10 to 26.
Prediction markets reprice accordingly. DeSantis rises from roughly 4% to 19% for the Republican nomination, while Vance slips from about 36% to 31%. Flight reads this as a premium on distance from the administration's record, and a sign that a personalised coalition may not outlast Trump's presence on the ballot.
Did the Left Gain Too?
Less than the centre did. Across 17 states with Senate races in both cycles, the average Democratic swing is 6.7 points, but 3.1 for Sanders-endorsed candidates against 8.1 for others. Flight cautions that state and candidate differences may distort this, yet he concludes the result is a repudiation of the administration, not an endorsement of a more progressive economic agenda, which strengthens moderates ahead of 2028.
Does the Election Shorten the Iran War?
Possibly. Both Trump and Vance partly blame the war for the losses. Flight revisits his earlier "Peak Gamma" thesis, that Tehran's leverage resembled a long, high-strike oil call expiring around the midterms, and suggests Tehran may now monetise what remains while Washington's incentive to settle strengthens. Options markets agree: the three-month implied probability of Brent above $120 falls from 23% to 9% since October 28, and the one-month probability from 27% to 7%, with upside skew flattening.
Five Key Takeaways for Advisors and Investors
- Expect rotation, not retreat, in AI. Slower buildouts favour operators with energised compute and the hyperscalers over hardware suppliers.
- Check geographic exposure. Texas and other tight-permitting states now drive execution risk and valuation differences.
- Rates carry the clearer signal. Lower yields are favoured, with caution against extrapolating.
- Treat energy risk premia as falling. Option-implied escalation odds have dropped sharply, and a diplomatic settlement could compress them further.
- Watch the centre. A moderating politics may shape fiscal, tax and regulatory policy into 2028.
Footnote:
1 Flight, Frank. "Houston, Do We Have A Problem?" Citadel Securities, 5 Nov. 2026, www.citadelsecurities.com/news-and-insights/global-macro-strategy/houston-do-we-have-a-problem/. Accessed 10 Nov. 2026.
