US Midterm 'Blue Wave' Triggers AI Selloff, But Divided-Government History Favours Stocks

How a Democratic Sweep Turned Into an AI Stock Rout

US equities suffered their sharpest one-day decline since April on Thursday, as the Democratic sweep of both chambers of Congress triggered a violent selloff concentrated in artificial-intelligence and semiconductor stocks. The Philadelphia Semiconductor Index plunged 12.2%, falling below the 10,000 level, while the three main US indices dropped between 3.4% and 5.5%, all closing at their session lows. More than 5,000 individual US stocks traded lower, with 83% finishing in the red — a breadth that signals broad-based selling rather than an isolated sector rotation.

The damage was overwhelmingly concentrated in AI-linked names. Nine of the ten worst-performing S&P 500 constituents were AI-related, led by Sandisk (-22.3%), Lumentum (-18.7%) and Micron (-17.5%). Nvidia, the flagship of the AI trade, fell 9.3% and slipped below a $4.5 trillion market capitalisation. Apple was the lone standout, briefly touching a record high and remaining the only company valued above $5 trillion.

The proximate cause, according to most analysts, is political. A Democratic-controlled Congress is expected to scrutinise or slow data-centre construction and potentially tighten AI regulation — both of which would weigh on the earnings outlook for the companies that have driven the market's recent gains. That prospect prompted a wave of position unwinding in the most crowded AI trades. However, the article cautions that extrapolating the election result into a two-year bear case for equities may be an overreaction, given that President Trump retains substantial executive authority and policy tools even with a hostile Congress.

Divided Government, AI Fatigue and What History Says About the S&P 500

Why the AI Trade Was Vulnerable Before the Vote

The selloff did not emerge from a vacuum. Before the election, investor scepticism about AI was already building, particularly as technology giants reported massive capital expenditure commitments without clear evidence that earnings could catch up to elevated valuations. The Democratic victory acted as a catalyst that accelerated an existing rotation out of the sector rather than creating a new one. In this reading, the election gave bearish traders a convenient trigger, but the underlying fragility — stretched multiples, uncertain return on AI investment, and heavy concentration in a handful of names — was already present.

What Divided Government Has Meant for Stocks

Since the Second World War, there have been seven prior instances of a Republican president facing a Democratic-controlled Congress, occurring in 1954, 1958, 1970, 1974, 1986, 1990 and 2006 — roughly 40% of the relevant election cycles. In each case, the S&P 500 was initially choppy before grinding higher. By the 150th trading day after the election, gains ranged from 10% to 25%, averaging about 19%. The worst early drawdown came in 1974, when the oil crisis, recession and political turmoil combined to push the index down roughly 10% within a month of election day; it recovered those losses in about two months and turned positive by the 57th trading day, in late January 1975. The article notes that today's political and economic environment bears some resemblance to that period, which could provide conditions for another significant dip — but argues that the medium-to-long-term outlook does not warrant pessimism.

Trump's Executive Tools and the Limits of Gridlock

Even with Congress lost, the president retains considerable unilateral levers — executive orders, regulatory appointments, trade policy and agency direction — that can shape outcomes for AI, energy and data infrastructure. The assumption that a divided government automatically translates into sustained equity weakness may therefore be overstated. For context, markets have historically adapted to gridlock because it constrains the most disruptive legislative changes while allowing executive-driven policy to continue.

Where This Leaves the AI Complex

The episode exposes how sensitive AI-linked valuations remain to any perceived change in the policy or spending environment. Companies with heavy data-centre exposure — chipmakers, networking suppliers and cloud infrastructure providers — face a period of heightened scrutiny over whether their capital spending plans can proceed at the pace investors have priced in. Conversely, Apple's resilience suggests that investors are willing to reward businesses whose earnings are less dependent on the AI capex cycle. The medium-term question is whether the regulatory threat from Congress is real or rhetorical; if legislative action proves slow or symbolic, the selloff may look like an overreaction.

Positioning for a Politically Gridlocked but Historically Rising Market

The single most important insight: the AI selloff was driven by a political catalyst acting on pre-existing valuation anxiety, not by a fundamental change in AI demand — and history suggests divided government has been followed by S&P 500 gains averaging about 19% over 150 trading days.

For senior management and investors with exposure to the AI complex, the immediate priority is distinguishing between regulatory rhetoric and legislative reality. A Democratic Congress can hold hearings and propose bills, but passing meaningful AI regulation through a divided legislature is slow and uncertain. The more tangible near-term risk is a slowdown in data-centre permitting or federal spending that directly affects infrastructure buildout timelines.

• 0–6 months: Track whether Democratic committee chairs introduce specific AI or data-centre legislation, and monitor Nvidia's next earnings report for any change to its data-centre revenue guidance — that will be the clearest signal of whether end-demand is actually weakening or whether this was purely a positioning unwind.
• 6–24 months: Watch federal data-centre permitting and any Congressional appropriations that touch AI infrastructure or semiconductor subsidies. Companies with concentrated AI capex exposure should stress-test their plans against a scenario where regulatory friction adds 6–12 months to project timelines.
• 2–5 years: The structural AI investment cycle remains intact unless Congress passes binding restrictions. Position for the possibility that the 2026 midterms shift the balance of power again, which historically has been the point at which divided-government market rallies mature.

Metrics worth watching: Nvidia's next quarterly data-centre revenue and guidance; the Philadelphia Semiconductor Index's ability to reclaim the 10,000 level; the S&P 500's performance at the 57th and 150th trading days post-election, against the historical 10–25% gain range; and any formal AI regulation bill emerging from Congressional committee.