How the Stock Market Reacted to the Midterms, and What History Says Comes Next

The stock market rallied hard on the midterm result. The day after Democrats won both chambers of Congress, the Dow Jones Industrial Average rose 1,569 points, or 3.03%, to 53,356, its biggest post-midterm single-day gain in 44 years according to CFRA Research. The S&P 500 gained 1.58% to about 7,689 and the Nasdaq Composite 1.44%. Underneath, it was a rotation, not a broad melt-up: consumer and big tech stocks surged, oil fell about 8.7%, energy stocks dropped about 4.8%, and the companies that build data centers lost 17% to 23% after a Democrat won the Texas governorship on a promise to pause new ones.

History leans the same way. In each of the 19 midterms since the middle of the last century, the S&P 500 was higher 12 months later, by 15.5% on average in our count. But 11 of those 19 years also included a fall of 10% or more along the way, and 19 data points are a thin base for a forecast.

Watch it play outDemocrats gained seven Senate seats, and the Dow added 1,569 points the next day

Each dot is a Senate seat. Before the vote, Democrats held 47 of the 100.

They flipped seven, reaching 54, with Mississippi and Alaska still too close to call. The House flipped too.

The day after the vote, the Dow jumped 1,569 points, about 3%, its best post-midterm day in 44 years.

Oil did the opposite: Brent fell about 8.7% on hopes that the Iran war ends sooner.

Replay

What happened on the day after the vote

Counting is not finished. NBC News has Democrats at 54 Senate seats to 44, a net gain of seven, with Mississippi and Alaska uncalled. In the House, CNBC's election desk projects Democrats will win at least 250 of 435 seats, and CBS News estimates they could end with 251 to 268 once dozens of close races are counted. Either way, President Trump faces a Congress run by the other party for his last two years.

Here is how the main markets closed on the first full trading day after the result, against the prior close:

Index closes are from Yahoo Finance; the energy sector and yield moves are from Reuters figures compiled by INDmoney.

Three details put the day in context. It was the Dow's biggest gain of the year, and its best day since the tariff-pause rally last year. The S&P 500's 1.58% was its second-largest post-midterm daily gain in 44 years, behind only the midterm eight years ago. And none of the indexes set a record: at midday, CNBC noted the Dow still sat about 2.7% below its intraday high, the S&P 500 about 1.7% and the Nasdaq 6.6%.

For a sense of scale, $10,000 in a Dow tracking fund gained about $303 that day. The VIX, a gauge of how much volatility traders expect over the next month, fell almost 20%.

The result also went against much of the pre-vote advice. JPMorgan had estimated that a blue wave would knock about 0.5% off the S&P 500 on the day, and Bank of America's Michael Hartnett had warned that Democrats winning the Senate and Texas could bring a "big slump" of more than 10% by year end. Both outcomes happened, and the index rose anyway.

Winners and losers: a rotation, not a rising tide

The headline gain hides a sharp split. Money moved out of anything tied to building data centers and into the giant companies that pay for them, plus consumer names that benefit from cheaper fuel.

Big tech rose 3% to 7% while data center builders fell up to 23%

Microsoft+6.7%

Amazon+6.1%

Apple+5.3%

Alphabet+3.2%

S&P 500+1.6%

Nvidia−3.2%

Energy sector−4.8%

GE Vernova−9.3%

CoreWeave−9.9%

Vertiv−12.9%

Comfort Systems USA−15.1%

Modine−21.4%

Sterling Infrastructure−21.5%

Hut 8−23.2%

The winners. Microsoft rose 6.7%, Amazon 6.1%, Apple 5.3% and Alphabet 3.2%, according to StockTitan's close-to-close figures. Apple closed at a record that took its market value above $5 trillion. Consumer discretionary was the best sector. CBS reported that solar companies, dollar stores and hospital operators led the rally. Spot gold rose about 3.4% and bitcoin about 3% to just above $80,000.

The losers. Energy fell with oil, and refiners were hit hardest. Chipmakers slipped: Nvidia closed down about 3.2%, and memory makers Micron and Sandisk fell about 7% and 8%. Stocks tied to Elon Musk fell on his Texas exposure, with Tesla down about 3.3% and SpaceX about 3.8%. The biggest falls were in data center construction.

Why data center stocks fell 17% to 23%

The trigger was Texas. Democrat Gina Hinojosa beat Republican Governor Greg Abbott after campaigning for an immediate statewide moratorium on new data centers. A moratorium is a temporary ban: no new approvals until the rules are rewritten. Before the vote, Hinojosa had asked Abbott to sign an emergency order putting a moratorium on data centers until the legislature meets.

Texas matters more than any other state here. Wells Fargo estimates it holds about 27% of planned US data center capacity. Goldman Sachs expects new Democratic governors in Texas, Georgia and Michigan "to impose conditional moratoria pending new legislation restricting data centers," and puts the share of planned capacity in states likely to add restrictions at 45%. Local voters pushed the same way: 38 of 39 called ballot measures on data centers limited development.

Watch it play outStates that may restrict data centers hold about 45% of planned US capacity

Picture all the data center capacity planned in the US as 100 dots.

About 27 of them are in Texas, where the new governor campaigned for an immediate moratorium.

Add Georgia and Michigan, and Goldman Sachs puts about 45% in states likely to impose limits.

The six hardest hit builders closed down 17% to 23% on that risk, before a single rule has changed.

Replay

The stocks that fell hardest were the ones whose order books depend on new construction. StockTitan's closing figures show Hut 8 down 23.2%, Sterling Infrastructure 21.5%, Modine 21.4%, Cipher Digital 20.0%, INNIO 19.5% and Solaris Energy Infrastructure 16.9%. Hut 8 is the clearest case of concentration: its Beacon Point site in Texas accounts for 1,000 of the 1,330 megawatts it is building. Larger suppliers fell less: Comfort Systems USA 15.1%, Vertiv 12.9% and GE Vernova 9.3%. On $10,000, a Modine holder lost about $2,140 in one session while an Amazon holder gained about $610.

The logic behind the split: if fewer data centers get built, the companies that pour the concrete, cool the halls and wire the power sell less. The cloud giants paying for them spend less, which leaves them more free cash flow, the cash left after capital spending. That is why Fidelity's Jurrien Timmer said a slower build-out "might therefore be good for the entrenched Mag 7".

Yet the businesses themselves were growing fast. Vertiv's third-quarter sales rose 61% to $4.3 billion, and Comfort Systems' backlog, the value of signed work not yet done, reached $17.8 billion against $9.4 billion a year earlier. Nothing has actually been banned. Hinojosa has not taken office, it is unclear whether any pause would cover projects already under construction, and before the vote Goldman judged that "a national moratorium looks unlikely in any election outcome".

So the market priced a lot of damage in a day, on a policy no one has written yet. Whether that was right comes down to three questions: what a Texas governor can legally do without the legislature, which companies have most of their backlog in the states at risk, and how much of the drop was a crowded trade unwinding rather than lost profit. Our member issue, The vote that knocked almost a quarter off data center stocks, works through each one, ranks the exposed names and says whether the sell-off overshot.

Why the overall market liked the result

Gridlock removes policy surprises. "Markets like gridlock and that's what we got," said Jurrien Timmer, Fidelity's director of global macro. Gridlock means a president and a Congress from different parties, so big new laws are hard to pass. Deutsche Bank's George Saravelos said the outcome "should act as a break on extreme policy outcomes and bring down risk premia", the extra return investors demand for uncertainty. The limit runs both ways. As Saxo's Charu Chanana put it, Democrats "would not have a free hand to rewrite policy either, because Trump retains veto power."

Oil and the Iran war. Brent settled at $94.62, its lowest since the end of the summer. Nationwide's Mark Hackett put the drop down to "hopes for a deescalation in the Iran conflict, even as the diplomatic signals remain ambiguous". Vice President JD Vance said "the reason we lost is the war", and Goldman Sachs' Jan Hatzius expects Democrats to "use fiscal deadlines to seek to limit the war with Iran". Cheaper oil cuts costs for airlines, retailers and households, and eases the inflation that has the Federal Reserve raising rates.

Lower yields and possible tariff relief. The 10-year Treasury yield slipped from about 4.95% to 4.90%. Lower yields make future profits worth more in present terms, which helps stock prices. Hatzius also expects Democrats to push to "roll back selected tariffs, and reverse spending cuts enacted last year", which would be good for company profits if any of it passes.

The uncertainty is over. Vital Knowledge's Adam Crisafulli summed it up: "Investors are happy to have the midterms out of the way."

What history says comes next

The record after midterms is unusually consistent. Various firms measure it slightly differently, but they agree on the direction:

• Edward Jones data reported by CBS: the S&P 500 gained an average of 15.5% in the six months after each of the last 14 midterms.
• The Stock Trader's Almanac: after the past 19 midterms the index was up six months later every time, by 14.18% on average.
• J.P. Morgan Asset Management: positive in the 12 months after every midterm in its data, by about 15% on average.
• Fidelity, using a longer history: higher in the following year about 95% of the time, averaging about 14%.

We checked this ourselves using S&P 500 daily closes from Yahoo Finance, measuring from the close before each of the last 19 midterms. The returns are price only; dividends would add a little more.

Our six-month average over the last 14 midterms comes to 13.7%, a little below Edward Jones' 15.5%. The gap likely comes from dividends and start dates. Either way, all 14 were positive.

What the averages hide is the ride. In each 12-month window we also measured the worst fall from a high to a later low:

All 19 years ended higher, but 11 included a drop of 10% or more

12-month returnWorst drop along the way

7616.2%−8.1%

7233.2%−10.6%

6811.1%−9.2%

6424.8%−4.1%

6012.9%−6.6%

5613.7%−11.4%

5222.0%−14.1%

484.9%−10.2%

4421.7%−7.1%

401.3%−33.2%

3624.0%−6.1%

3227.8%−4.4%

2821.9%−12.1%

2415.8%−14.7%

206.9%−9.4%

164.5%−19.4%

124.2%−12.4%

812.4%−16.4%

415.1%−10.3%

In money terms, $10,000 in an S&P 500 index fund at the close before the vote would have been worth between $10,130 and $13,320 a year later, and about $11,550 on average, before dividends. The worst case is the reminder that matters. Forty years ago the index was up almost 18% after six months, then the worst one-day crash in the index's history wiped out nearly all of it, leaving a 1.3% gain at the one-year mark after a 33% fall from the peak.

Not everyone expected a rally this time. Before the vote, BTIG's chief market technician Jonathan Krinsky noted that stocks usually rise after midterms because the index "typically is weak INTO the elections", and that "we have the opposite setup this year." The S&P 500 had closed at a record about four weeks before the vote. His conclusion: "while consensus is that we see a post midterm rally, we think the opposite is actually more likely." The first day proved him wrong, but his point was about the months ahead, and it still stands.

What 19 elections can and cannot tell you

They can tell you that a midterm has never, in this data, been followed by a losing year, and that the uncertainty relief is real and repeatable. They cannot tell you that the pattern will hold. Nineteen observations is a tiny sample: one more bad year would cut the hit rate to 95%, and the market rises in most 12-month periods anyway, so part of the "midterm effect" is just the normal upward drift of stocks. Each of those years also had its own oil price, interest rates and valuations. This one starts with a war, rate hikes and stocks near record highs, so treat the history as a tilt in the odds, not a forecast.

What to watch now

Oil and the Iran talks. Oil drove a large part of the day after the vote, and it can reverse just as fast. Rapidan Energy Group's Bob McNally said traders expect the administration to abandon the war, then added: "We're not buying it. We think escalation is more likely." If Brent climbs back above $100, the rotation from energy into consumer stocks would likely unwind.

The Fed. At its last meeting, the Fed held its benchmark rate at 3.75% to 4% after a quarter-point hike at the meeting before. Two officials voted for another hike, its preferred inflation gauge was projected above 3% even excluding food and energy, and futures point to a strong chance of a hike at the next meeting. Cheaper oil is the main thing that could change that.

AI spending and data center policy. Watch what Hinojosa does once in office, and whether Georgia and Michigan follow, and what the cloud giants say about their capital spending plans. The market is betting a slower build-out is good for Microsoft, Amazon and Alphabet and bad for their suppliers. If the giants keep spending at the same pace and simply build in other states, the suppliers' sell-off could reverse. Our hyperscaler capex tracker follows the spending plans company by company.

Earnings. In the end, profits move stocks more than politics. Fidelity's advice after the vote was to keep watching earnings, business spending and economic conditions, and Capital Economics said the result does "not change too much for the near-term policy outlook".

What would extend the rally, and what would undo it

Extends it

• Iran talks progress and Brent stays below $100
• Inflation cools enough for the Fed to stop hiking
• Data center limits stay local and slow, not statewide bans
• Earnings keep growing into the new Congress

vs

Undoes it

• The war escalates and oil returns above $100
• The Fed hikes again and yields rise
• Several states impose moratoria and the cloud giants cut orders
• A drawdown like the one in 11 of the last 19 midterm years

Our read

The day after the vote was a rational repricing, not a celebration. Investors decided that a divided Washington means fewer policy shocks, that the Iran war is more likely to end, and that a slower data center build-out shifts profit from suppliers to the cloud giants. The broad index move is well supported by history: 19 of 19 midterms were followed by a higher market a year later. But the typical year also brought a double-digit drop along the way, and this one begins with oil, inflation and rate hikes all unresolved. For a long-term investor, the midterm is a reason to stay invested, not a reason to buy more.

The more interesting question is the data center sell-off. A 17% to 23% fall in a day for companies with record backlogs, on a policy that has not been written, is the kind of move that either marks the start of a real slowdown or overshoots badly. We think the risk is real for a handful of companies that depend on Texas, and overstated for suppliers with orders spread across many states. That is where we would look first.

Go deeper

• The vote that knocked almost a quarter off data center stocks: what a Texas moratorium can really do, who is exposed, and whether the sell-off overshot.
• Hyperscaler capex tracker: how much Microsoft, Amazon, Alphabet and Meta plan to spend on AI infrastructure.
• Does copying Congress's stock trades work?: what happened when we tested 4,967 stock buys disclosed by members of Congress.

Join for the full data center issue and every member analysis.