Do Stocks Go Up on Fed Days? Every Fed Decision Since 1994, Tested
Buying stocks before the Fed was one of the best-known edges in the market. We tested every scheduled Fed decision since 1994, and split the recent ones at the 2 p.m. statement.
By Pavel Penev, MScFounder, TradeWize · 10+ years trading the marketsEight times a year, the US Federal Reserve announces what it's doing with interest rates. Traders call it Fed day.
There's an old idea that stocks go up on those days. It isn't folklore. Two economists at the New York Fed showed that the S&P 500 had made a big share of its gains in the hours before Fed announcements. Their study ran in the Journal of Finance in 2015.
So does it still work? We tested every scheduled Fed decision from February 1994 to October 2026: 262 of them.
The short answer
It used to. From 1994 to 2011 the S&P 500 averaged +0.34% on Fed days, against +0.02% on other days, and rose 60% of the time. Since 2016 it has risen on only 44% of Fed days, less often than on an ordinary day. What's left is the size of the move: after the 2 p.m. statement, the S&P 500 moves 2.2 times as far as on a normal afternoon.
What this article is, and isn't
This is a test with public data, not personal financial advice, and there are no affiliate links on this page. Every source is linked at the end.
What a Fed day is
The Federal Open Market Committee, or FOMC, sets the main US interest rate. It meets eight times a year on a schedule it publishes in advance. Each meeting ends with a decision: raise the rate, cut it, or hold it.
The decision comes out at 2:00 p.m. New York time on the meeting's last day. Since 2019 the Fed chair has also held a press conference at 2:30 p.m. after every meeting. That's the day we call a Fed day.
How we tested it
We took every scheduled meeting from the Fed's own calendars, and every rate change from its records, so each Fed day is marked as a hike, a cut or a hold. We left out emergency moves, like the two in March 2020, because nobody knew they were coming.
Then we measured the S&P 500 from the close before to the close on each Fed day, from 3 January 1994 to 13 November 2026. That's 8,273 trading days, and 262 of them were Fed days.
To check for luck, we drew the same number of ordinary days at random from the same years, 20,000 times. If random days often match the Fed-day average, there's no real effect.
1994 to 2011: the best days of the year
For 18 years, the old idea was right. The S&P 500 averaged +0.34% on the 144 Fed days from 1994 to 2011, against +0.02% on every other day. It rose on 60% of them.
Average move from the close before to the close on the day. 144 Fed days from 1994 to 2011, 86 since 2016.
That wasn't luck. Random days matched the Fed-day average only 0.14% of the time.
It also added up. Fed days were 3% of trading days, but from 1994 to 2011 they carried 48% of the S&P 500's gain. The New York Fed economists, David Lucca and Emanuel Moench, measured it a different way: the 24 hours before each announcement, against T-bills. They put the share at more than 80%.
It wasn't just the 2008 crash, either. The biggest Fed day in our test was 16 December 2008, when the Fed cut rates to near zero and the S&P 500 rose +5.14%. But from 1994 to 2007, before the crisis, Fed days still averaged +0.22% and rose 59% of the time.
Since 2016, the edge is gone
The 86 Fed days since the start of 2016 averaged +0.03%. Ordinary days averaged +0.055%. The S&P 500 rose on 44% of Fed days and 55% of other days.
So a Fed day has been slightly worse than a random day. Random days matched the Fed-day average 59% of the time, which is what you'd expect if there's no effect at all.
The S&P 500's moves on each year's scheduled Fed days, added together. Green when the year's Fed days made money.
Year by year, Fed days added up to a loss in four of the 11 years since 2016. In 2026 so far, the S&P 500 has risen on one of seven Fed days.
Researchers saw the same thing. A 2021 paper by Kurov, Wolfe and Gilbert is called, simply, "The Disappearing Pre-FOMC Announcement Drift".
We can't prove why it faded. Two things changed. The finding was published and widely read, so traders could buy ahead of it. And the Fed started saying much more in advance about what it planned to do, so less was left to learn on the day.
Learn it by doing
Reading about it is one thing — it clicks when you do it. Learn it hands-on with free, interactive lessons on TradeWize.
Try the free lesson →Where the move went: after 2 p.m.
A close-to-close number hides what happens inside the day. So for the Fed days since 2016 we used one-minute prices and split each day at the 2:00 p.m. statement.
The S&P 500 on 82 Fed days from January 2016 to October 2026, against the same hours on ordinary days.
Before the statement, the S&P 500 still rose a little on average: +0.18% from the prior close to 2 p.m., against +0.06% on ordinary days. But that average rests on a handful of days. The typical Fed morning moved +0.04%. Take out the five best and the average falls to +0.07%. Random days matched it 11% of the time.
After the statement is where things happen. From 2 p.m. to the close, the S&P 500 moved 0.64% on an average Fed day, up or down. On an ordinary afternoon it moved 0.29%. That's 2.2 times as far, and no set of random days came close.
On average those afternoons went down: −0.13%. A few big falls do most of that, so don't read it as a reliable short. Read it as a warning. The statement is a coin flip with a bigger coin.
| Fed day | Decision | Close to 2 p.m. | 2 p.m. to close | Whole day |
|---|---|---|---|---|
| 18 December 2024 | Cut | +0.21% | −3.21% | −2.95% |
| 19 December 2018 | Raised | +1.57% | −3.10% | −1.54% |
| 4 May 2022 | Raised | +0.41% | +2.60% | +2.99% |
| 21 September 2022 | Raised | +0.73% | −2.49% | −1.71% |
| 2 November 2022 | Raised | −0.33% | −2.20% | −2.50% |
| 15 December 2021 | Held | −0.18% | +1.72% | +1.64% |
S&P 500. The first two columns are one-minute prices; the whole day is the official close to close.
A cut doesn't mean stocks go up
You'd expect a rate cut to help stocks. It doesn't work that cleanly. The worst Fed day in our test was a cut: on 18 December 2024 the S&P 500 fell −2.95%. The Fed cut, but its forecasts showed fewer cuts ahead than before, and the market sold off after 2 p.m.
It goes the other way too. On 19 December 2018 the Fed raised rates. The S&P 500 was up +1.57% going into the statement and finished the day −1.54%.
Since 2016, the 9 cut days averaged −0.29% and the 20 hike days rose only 30% of the time. The decision itself is usually expected. What moves the market is what the Fed says about next time.
What it means for the December meeting
The next Fed day is 9 December 2026, with the statement at 2:00 p.m. New York time. This year's Fed days have mostly been quiet holds: six of the seven.
Based on the last ten years, don't count on a rally. Expect a bigger afternoon than usual, in either direction. If you trade around it, size your position for a move about 2 times the normal one, and set your stop before 2 p.m., not after.
If you invest for the long run, there's nothing to do. Over the whole test, $10,000 in the S&P 500 only on Fed days grew to $18,116. Left in every day, it grew to $162,111.
What this test can't tell you
- It's the S&P 500 price index, without dividends. Dividends add a tiny amount to every day, Fed day or not.
- The before-and-after split only covers 2016 on. Before that the Fed's pages don't give the release time, and it moved between 12:30, 2:00 and 2:15 p.m.
- The one-minute prices are a feed of the index (HistData), not the official tape. Its 4 p.m. price sits a median 0.05% from the official close. It's missing March to July 2023, so four Fed days drop out of that part.
- There are no trading costs here. A trader moving in and out around eight days a year pays spreads and commissions each time.
- In the 1990s the Fed didn't always put out a statement after a meeting. We count every scheduled meeting day either way.
- 262 days is a small sample for something this noisy. One or two big days can move an average a lot, which is why we show medians and random tests too.
So do stocks go up on Fed days?
They did, for a long time. From 1994 to 2011 the S&P 500 averaged +0.34% on Fed days and made 48% of its gains on 3% of its days. That's one of the strongest calendar effects ever found in stocks.
Since 2016 it's gone. Fed days have risen 44% of the time, less often than ordinary days. What hasn't gone away is the risk: after 2 p.m., the market moves about 2 times as far as usual, and the decision alone won't tell you which way.
Do stocks go up on Fed days?
Not any more. From 1994 to 2011 the S&P 500 rose on 60% of Fed days and averaged +0.34%. Since 2016 it rose on 44% of Fed days and averaged +0.03%, worse than an ordinary day.
What is the pre-FOMC drift?
It's the tendency for US stocks to rise in the hours before a Fed announcement, found by New York Fed economists David Lucca and Emanuel Moench. It was large from 1994 to 2011. Since 2016 the typical morning before the statement has moved +0.04%.
What time does the Fed announce rates?
At 2:00 p.m. New York time on the last day of the meeting. The Fed chair's press conference starts at 2:30 p.m.
Do stocks go up when the Fed cuts rates?
Not reliably. Since 2016 the 9 cut days averaged −0.29%. The worst Fed day in our test, 18 December 2024, was a cut. Markets react to what the Fed signals next, not just the decision.
How much does the stock market move on Fed days?
Since 2016, the S&P 500 moved an average of 0.64% between the 2 p.m. statement and the close, up or down. On an ordinary afternoon it moved 0.29%.
Learn how markets really move
Our free lessons cover interest rates, the S&P 500 and why calendar tricks rarely last, with exercises in every lesson.