Stocks After Midterms Vs Presidential Elections: Which Year Is Better?
Start here: will the midterms move your stocks?
The midterms are over, and an old piece of market folklore is making the rounds: the year after a midterm is the best year of a president’s four. If that’s true, 2027 should be a good one. It’s a fair question and an easy one to check, so we checked it, every election since 2006, two different ways. The short answer: in our 20 years, it didn’t hold. But there’s a better question underneath: what will actually drive 2027, and does your plan work if next year turns out flat instead of great? We’ll come back to that.
“The calendar doesn’t pay you. Earnings do.” – Anonymous
What Does The Folklore Say?
The idea goes like this. A president takes the hard medicine early, in years one and two, and pulls out every lever to get the economy humming before the next presidential vote. So year three, the year after the midterm, should be the strongest. The folklore leans on many decades of history. We can’t check those early decades here: our study starts at the end of 2002. What we can check is the last two decades, the ones that look most like today, with the Fed, index funds and Big Tech running the show.
Measure One: The 12 Months After Election Day
Start from the close on each election day and look one year out. That’s the honest version of “buy after the vote,” since nobody knew the result the day before.
| Midterm | S&P, next 12 mo | Presidential | S&P, next 12 mo |
|---|---|---|---|
| 2006 | +6.6% | 2008 | +4.1% |
| 2010 | +2.1% | 2012 | +24.0% |
| 2014 | +4.5% | 2016 | +21.3% |
| 2018 | +12.0% | 2020 | +38.3% |
| 2022 | +13.6% | 2024 | +16.4% |
| Average | +7.7% | Average | +20.8% |
S&P 500 (^SPX), election-day close to 12 months (252 trading days) later, from Trefis platform price data. Averages are our arithmetic.
All ten years were positive. But the years after presidential elections came out ahead, by about 13 points on average. Pair each midterm with the presidential election two years later, and the presidential year won four times out of five; only 2008, in the middle of the financial crisis, lost to 2006.
One catch: the 12 months after a November midterm run mostly through year three, but not neatly. So we measured it the way the folklore is usually told, too.
Measure Two: The Calendar Year
Here’s year three of each cycle, from the last close of one year to the last close of the next. Starting from the close of Dec 31, 2002 gives us six year-threes, from 2003 to 2023.
| Year 3 | S&P 500 | Nasdaq | Worst S&P drop in year |
|---|---|---|---|
| 2003 | +26.4% | +50.0% | −14.1% |
| 2007 | +3.5% | +9.8% | −10.1% |
| 2011 | 0.0% | −1.8% | −19.4% |
| 2015 | −0.7% | +5.7% | −12.4% |
| 2019 | +28.9% | +35.2% | −6.8% |
| 2023 | +24.2% | +43.4% | −10.3% |
| Average | +13.7% | +23.7% | −12.2% |
Index levels, no dividends; worst drop is from the highest close of the year so far. Trefis platform price data; averages are our arithmetic.
Three big years and three flat ones. Not bad. But here’s how year three stacks up against the other three years of the cycle.
| Year of the cycle | S&P, average | Years up | Best / worst |
|---|---|---|---|
| Year 1, after a presidential vote (2005-2025) | +19.8% | 6 of 6 | +29.6% / +3.0% |
| Year 2, midterm (2006-2022) | +2.4% | 3 of 5 | +13.6% / −19.4% |
| Year 3, after a midterm (2003-2023) | +13.7% | 4 of 6 (2011 flat) | +28.9% / −0.7% |
| Year 4, presidential (2004-2024) | +5.5% | 5 of 6 | +23.3% / −38.5% |
S&P 500 calendar-year price returns, Trefis platform price data; averages are our arithmetic. The 2026 midterm year is not included: +10.3% through Nov 9.
So does the folklore hold? Honestly, half of it. Year three was a solid second, and with dividends it never lost money in our record: the SPY fund returned between +1.2% and +31.2% in all six year-threes. But year one, the year after a presidential vote, was better on average, never had a down year, and beat the year three that followed it in three of five terms. The Nasdaq tells the same story: about +26% on average in year one, about +24% in year three.
The part of the folklore that did show up is the weak middle. Midterm years averaged the smallest gain, and they carried deep dips: the S&P fell 15% from a high, on average, somewhere inside each midterm year, including 25% in 2022. Then the 12 months after the vote were positive all five times. That’s the pattern worth remembering: the midterm year tends to wobble, and the market tends to look past it.
Two honest caveats. Six years per bucket is a tiny sample: one crash like 2008 swings an average by six points. And the spread inside each bucket swamps the gap between them. Year three ran from −0.7% to +28.9%, a 30-point range; the gap between the year-one and year-three averages is about 6 points.
Why Did Each Year Turn Out The Way It Did?
The data gets interesting when you ask why. Go down the list of year-threes and the White House barely shows up. The Fed and earnings do.
- 2003, +26%: the market climbed out of the dot-com bust as the Fed cut rates to 1% and a tax cut passed, even with the Iraq war starting in March.
- 2007, +4%: housing cracked and credit tightened; the Fed started cutting in September and the S&P peaked in October, just before the 2008 crash.
- 2011, flat: Washington’s debt-ceiling standoff led to the first-ever downgrade of US credit in August; the S&P fell about 19% from its high.
- 2015, flat: oil collapsed, China devalued its currency in August, and in December the Fed raised rates for the first time in nearly a decade.
- 2019, +29%: the Fed stopped raising rates and cut three times, after the late-2018 selloff.
- 2023, +24%: ChatGPT kicked off the AI spending race and the Fed made its last hike in July; Big Tech stocks led the way.
See the thread? The three big year-threes came when the Fed was cutting or finished hiking. The three flat ones came with a credit crack, a debt fight in Washington, or the first hike in years. The year-ones tell the same story from the other side: 2009, 2013 and 2021 had near-zero rates and Fed bond buying behind them, and 2017 ended with a tax cut signed in December. The weak one, 2005, came while the Fed raised rates at every meeting. The presidential cycle mostly sorted who happened to be in office when the Fed moved.
What Does That Mean For 2027?
2027 is year three, and it starts with divided government: Democrats won both chambers of Congress on Nov 3. Some will call that good news for stocks, some bad. Our record says the bigger drivers sit elsewhere, and they’re already on the table.
AI spending, first. It helped power 2023, and it’s still a big driver of the market. Last week showed the other side: on Thursday, Nov 5, the AI trade unwound and the S&P 500 fell 4.4% in a day, as data centers turned into a local fight over power and electric bills. Texas’s governor-elect pledged a moratorium on new ones, and the Wall Street Journal reported last week that the backlash could play into the hands of Big Tech giants like Amazon and Microsoft. Will the US actually stop building data centers and hand the AI lead to China? We don’t see it. The four biggest AI spenders put about $1.15 into capital spending last quarter for every $1 of cash their businesses brought in, and even Nike has cut jobs while it keeps paying for AI, as we laid out in Why Big Tech Spends $1.15 Per $1 Of Cash. If that spending holds, it’s the 2027 tailwind to watch, not the calendar.
Inflation and the Fed, second. That was the swing factor in every year above. Our Oct 20 piece, Inflation Is Back. Is A Stock Market Crash Next?, boiled the story down to war, immigration and taxes, with AI as a fourth fuel: all those data centers soak up power, chips, equipment and building materials. If inflation keeps the Fed from cutting, 2027 can look more like 2015 than 2019, whatever the folklore says.
What you pay, third. Going into 2023, the S&P 500’s earnings yield paid about 1.3 points more than a 10-year Treasury. Today it pays about 0.5 points less: 4.34% for the S&P versus 4.88% for the Treasury. Stocks start 2027 priced for good news, which leaves less room if the Fed or AI spending disappoints.
| Midterm year | S&P, calendar year | Worst drop in year | Next 12 mo from vote |
|---|---|---|---|
| 2006 | +13.6% | −7.7% | +6.6% |
| 2010 | +12.8% | −16.0% | +2.1% |
| 2014 | +11.4% | −7.4% | +4.5% |
| 2018 | −6.2% | −19.8% | +12.0% |
| 2022 | −19.4% | −25.4% | +13.6% |
| 2026 | +10.3% (to Nov 9) | −9.1% | ? |
Where 2026 sits: an ordinary midterm year so far, with a shallower dip than most. Trefis platform price data.
What’s The Bigger Number?
Step back from the four-year cycle. Over the 23 full calendar years in our record, 2003 through 2025, the S&P 500 rose in 18, was flat in one and fell in four, and the SPY fund turned $1 into about $11.80 with dividends. Every year of the cycle was positive on average. That’s the number that will still matter in 2035: not which year of the four was best, but whether you stayed in for all of them.
Do You Have A Plan For 2027, Whatever Kind Of Year It Is?
Here’s the question we promised to come back to. You don’t need to know whether 2027 is a 2019 or a 2015. You need a plan that works in both. One concrete thing to do this week: write down three lines.
1. Your mix. What share of your money sits in stocks, and how much of that is the same handful of AI names through your index funds.
2. Your bad-year line. The worst S&P drop inside a year-three in our record was about 19%, in 2011. Decide now what you’d do at a 15% or 20% fall: add in set amounts from cash you won’t need for years, or simply hold.
3. Your rebalance rule. If a strong year pushes stocks well above your target mix, trim back to it. No forecast required.
What this replaces: loading up because “year three is always good,” or sitting out because of who controls Congress. The risk: if inflation forces the Fed’s hand or AI spending cracks, 2027 can be flat or worse, as 2011 and 2015 were. A plan made in advance keeps that from becoming a scramble.
What Are The Pros Watching?
The Fed’s path into 2027 and each month’s inflation report. The biggest AI spenders’ next spending plans, due with their next quarterly reports, after the year ends. Whether any state actually passes a data-center ban. And earnings: the S&P’s yield is already below the 10-year Treasury’s, so earnings growth has to do the work. Each one is a reason to check your mix. None is a reason to trade on the calendar.
The folklore didn’t hold in our 20 years. From election day, the S&P 500 averaged +20.8% in the 12 months after presidential votes and +7.7% after midterms. On the calendar, year three averaged +13.7%, second to year one’s +19.8%, and was flat or down twice in six. What decided each year was the Fed, earnings and AI, not the cycle. For 2027: watch inflation, the Fed and AI spending, and set a mix, a bad-year line and a rebalance rule.
What Should You Ask Next?
- What mix of stocks fits a year that could be great or flat? S&P 500 Or Dividend Stocks: What Mix Fits Your Plan For 2027?
- Do you have a plan for a bad week, not just a bad year? Will Stocks Rebound Today, Or Do You Have A Plan?
- Energy had its worst election week in our 20-year record. Do the losers bounce back? Buy Or Sell Energy Stocks After Their Worst Election Week?
Follow this story with us: we’ll come back a year from now with 2027’s number in the table, whichever way it goes. Data: SPY · QQQ.
Is Your Plan Built For Any Kind Of Year?
One week rarely decides a year. If you have $1 million or more invested, see whether your plan fits what comes next in a free 15-minute call with a Trefis partner firm.
For questions about this article or underlying research, email help@trefis.com.