The Nasdaq 100 hit an all-time high of 31,224 on October 6. Four weeks later it was 10% lower. It bounced to 8.9% below the high on November 4, then fell to 14.7% below on November 5. On Monday, November 9, it closed 10.6% below the high.
This is the third 10% dip this year. The first two, in March and July, were back above their highs within two months.
Is this one routine, or the start of something worse? I went back through every 10% dip in the Nasdaq 100 since 1986. There are 45 before this one.
In this article
How often a 10% dip was recovered within a year
How often a dip became a bear market
How Fed hikes and cuts changed the rebound
How the S&P 500 compares
Within a year, it usually recovered. In 41 of the 45 past dips, the Nasdaq 100 was back above the high it fell from within 12 months. The median dip took about three months.
A year after its first close 10% below its 52-week high, the median gain was 20%. The two earlier dips this year are too recent for a full year of returns, so the return figures cover 43 dips.
Rebounds were smaller when the Fed had been raising rates. In those dips, the median gain was 14%.
Fed in the prior six months | Dips | Higher a year later | Lower | Median gain |
|---|---|---|---|---|
Cutting | 12 | 9 | 3 | 23% |
On hold | 18 | 17 | 1 | 28% |
Hiking | 13 | 10 | 3 | 14% |
All dips | 43 | 36 | 7 | 20% |
Where is the Nasdaq 100 now?
The sell-off has been fast. The index fell through its 50-day average in mid-October. Since November 5, it has closed below its 200-day average as well.
The Fed raised rates in September. Fed Funds Futures put the odds of another hike in December at near certainty.
Source: YX Insights
What did the Nasdaq 100 do after a 10% dip?
I counted a dip as the first close 10% below the 52-week high. Each sell-off counts once, until the index makes a new 52-week high. I measured returns from the next day’s close, without dividends, against any stretch of the same length. I use the 52-week high because the all-time high would hide 2008, when the index was still below its 2000 high. Measuring from the all-time high or the three-year high gave similar results.
Measured from | Dips | Higher a year later | Median gain | Median gain, Fed hiking |
|---|---|---|---|---|
52-week high | 43 | 36 | 20% | 14% |
156-week high, from 1989 | 38 | 32 | 20% | 13% |
All-time high | 32 | 26 | 23% | 13% |
Source: YX Insights
Six months after a dip, the index was up an average of 16%, versus 8% over any six-month period. That gap held up when I resampled the data, meaning I reran the test thousands of times on random draws of the same dips. Nearly every rerun still showed a gap above zero.
Twelve months after a dip, the mean was 22%, against 17% for any year. That edge is less reliable. The results ranged from a 66% loss after the 2000 dip to a 110% gain after the 1999 dip. With results that far apart, the reruns put the gap anywhere from about -5% to +14%. So the true edge at 12 months could be zero.
How often did a 10% dip become a bear market?
About one in five dips turned into a bear market of 20% or more. Five went past 30%: the 1987 crash, 1990, the dot-com bust, 2008, and 2022.
Source: YX Insights
The dot-com collapse went all the way to 83%. The index did not get back above its 2000 high until 2015.
That is the trouble with dips. Most are over in about three months. A few take years. But they all start the same way.
Does it matter that the Fed is raising rates?
I sorted each dip by what the Fed did to rates in the six months before it. Most dips were higher a year later, regardless of what the Fed did. However, the rebound size did change.
What was happening when the Fed hiked or cut?
The hiking dips split in two. I call it a boom when the index rose more than 37% in the year to its high. That is the median rise before all 46 dips, so half came after a bigger run-up and half after a smaller one. Any line between 37% and 53% splits the hiking dips the same way.
Six hiking dips came after a boom, in 1987, 1999 and 2000. Three of the six were lower a year later. The other seven came without a boom, in 1994, 2005, 2006, 2018 and 2023. All seven were higher a year later.
Fed hiking | Year to the high | Higher a year later | Median gain |
|---|---|---|---|
After a boom | +54% to +119% | 3 of 6 | -6% |
No boom | +13% to +36% | 7 of 7 | +16% |
Today | +26% |
Most cuts came before any sign of trouble in jobs. In 9 of the 12 dips while the Fed was cutting, unemployment had risen by less than half a point over the prior year.
Cuts into a recession gave mixed results. In February 2020, the Fed was already cutting when the pandemic hit. It kept cutting into the recession. The index was 57% higher a year later. In November 2007, the index closed 10% below its 52-week high with the Fed already cutting. A recession began the next month. A year later the index was 44% lower.
Today's dip came after a 26% rise in the year to the October high. That is below the 37% boom line. Unemployment was 4.1% in October, below its level a year earlier.
Under my rule, a dip counts as "hiking" when the Fed Funds Rate rose by more than a quarter point in the prior six months. One hike in six months sits right on that line. This dip counts as "on hold", by a hair.
So two groups are close matches. The on-hold group is where my rule puts this dip. In that group, 17 of 18 dips were higher a year later, with a median gain of 28%. If the Fed keeps hiking, the hikes without a boom fit better, as in 1994 and 2018. All seven of those dips were higher a year later, with a median gain of 16%.
Is a double dip before a new high a worse sign?
Not so far. A double dip here means the index climbed back to within 8% of its high, then fell 10% below it again within a year, before making a new high. That happened 20 times. The index was higher a year after the second dip in 18 of them. Most of those years overlap with the first dips, so this adds little new evidence.
The November bounce stalled 8.9% below the high. That is short of the 8% line, so it does not count as a double dip under my rule.
Is the Nasdaq 100 different from the S&P 500?
Yes. I ran the same test on the S&P 500 back to 1951. Its 10% dips were rarer but nastier.
After a 10% dip | Nasdaq 100 | S&P 500 |
|---|---|---|
Dips | 46 since 1986 | 34 since 1951 |
Became a bear market | 20% | 32% |
Median time to recover | About 3 months | About 7 months |
Mean return over the next year | 22% | 9% |
Mean return over any year | 17% | 9% |
For the Nasdaq 100, a 10% dip has been routine. It came more than once a year on average. For the S&P 500, it was more often the start of something bigger. A year after a dip, the S&P 500 made about what it made in any other year.
Most 10% dips in the Nasdaq 100 were recovered within a year. The weakest years came when the Fed was hiking after a boom, as in 2000. This time, the run-up was no boom.
At YX Insights, we use a range of models to build the Systematic Portfolio, which decides what names to hold each week. See how the Systematic Portfolio works.
DISCLAIMER: This newsletter is intended for educational purposes only. Any information or analysis in this note does not constitute an offer to sell or a solicitation of an offer to buy any securities. Nothing in this note is intended to be investment advice, nor should it be relied upon to make investment decisions. Any opinions, analyses, or probabilities expressed in this note are those of the author as of the note's date of publication and are subject to change without notice.
