If this year’s US midterm election cycle is anything like previous ones, the stock market will hit a major bottom right around Election Day in November.
That should give some hope to struggling investors, whose shares have suffered double-digit losses so far this year. A significant rally could be just a few weeks away.
I’m referring to the stock market’s historical pattern of weakness before the middle and strength after the middle. This pattern is plotted in the chart below, which is based on the July-December average performance of the Dow Jones Industrial Average
DJIA,
in the last 17 years of by-elections (since 1954).
Although the average date on this chart is in October, the actual bottoms in the historical record may come sooner or later. A lot depends on when the stock market starts anticipating the outcome of the interim periods and therefore gives it back. It’s a good guess that the bottom this year will be later, given the uncertainty surrounding the outcome of the elections – especially in the US Senate.
It’s always possible that the pre-midterm lows will happen before Election Day. In fact, it wouldn’t be inconsistent with the historical record if this year’s low occurred the day after Labor Day. As of September 9, the S&P 500
SPX,
was more than 4% higher than that low.
It’s worth noting how remarkably each pattern emerges when multi-year stock market swings are averaged. Although each year charts a unique path, the highs and lows usually offset each other, leaving the average to be a gradually upward-sloping line. A pattern has to be quite pronounced in the historical data for a deviation as pronounced as the one in the accompanying chart to appear.
This pre- and post-intermediate-term pattern is so pronounced that it is the source of the famous seasonal pattern known as the “Halloween indicator,” according to which the stock market is strongest between October 31 and May 1 and weakest during the remaining six months of the year. Yet take away the six months before and after the midterm elections, and the Halloween indicator disappears.
The main data is shown in the table below. The cell marked with one asterisk
| refers to the current six-month period, while the cell marked with a double asterisk (**) corresponds to the six-month period starting at the end of October 2022. | Year of the 1954 presidential cycle |
Dow average gain from Halloween to May 1st |
|
Average Dow gain from May 1 to Halloween |
1 |
6.4% |
|
1.5% |
2 |
4.7% |
|
-0.2%* |
3 |
15.1%** |
|
1.1% |
4 |
4.3% |
0.5%
So if you’re tempted to bet on the Halloween indicator, your time is fast approaching. If you miss it, you won’t have another chance until the 2026 midterms. Credit for discovering that the Halloween indicator tracks the months before and after midterms goes to Terry Marsh, professor emeritus of finance at UC Berkeley and CEO of Quantal International, and Kam Fong Chan, senior lecturer in finance at the University of Queensland in Australia.Their study of this model appeared in July 2021 in the Journal of Financial Economics
.
The likely source of the pattern, the researchers say, is the uncertainty that exists before the midterm elections and the resolution of that uncertainty after the elections. They note that it does not seem to matter which party dominates Congress before the midterm elections and which becomes the majority party afterward. They believe the pattern exists because the stock market craves certainty, even when the source of that certainty may not align with each investor’s political preferences. Mark Hulbert is a regular contributor to MarketWatch. Its Hulbert Ratings tracks investment newsletters that pay a flat fee to be audited. He can be reached at