Momentous Momentum

Medium-term Momentum just experienced an extraordinary weekly move

It is no surprise that equities were under pressure last week. Large and small cap indices declined over 3%. Tariffs, extended valuations, and slowing growth were the main culprits of the selloff.

From a factor perspective, last week produced some very rare returns, particularly among large cap momentum stocks.

We measure momentum in two ways - short-term (STM) and medium-term (MTM). Medium-term momentum measures returns from six months prior to the measurement date to four weeks before the measurement date. We exclude the four weeks prior to measurement date because that is our measure of STM, which we believe is a separate phenomenon.

For both measures, we rank stocks from the largest returns to the smallest returns (or greatest underperformance) over the defined time period. Then, the stocks are grouped into deciles. The top ten percent are Decile 1, the second ten percent are Decile 2, and so on. To create the factor return, or spread, the average return of stocks in Decile 10 is subtracted from the average return of stocks in Decile 1. This is the factor's return.

We track factor data on a regular basis. MTM is one of the factors we pay close attention to as it 1) is well-known and well-documented; 2) has a relationship with our institutional equity strategies; and 3) has a behavioral component to it.

Simply put, momentum "works" as a factor because stocks that are rising tend to keep rising. So the theory goes. Intuitively, it makes sense. Stocks that rise tend to have positive attributes - positive earnings, strong sales growth, improving margins, gaining market share, disruptive product, or any number of good things happening. As their share prices rise, additional investors take notice and also want to own those stocks, which puts more upward pressure on the price. The cycle continues.

Eventually, the price rises so much that whatever metrics or catalysts that originally made the stock a buy are less appealing in light of the now high share price. Even though the price has run up, there are still investors that do not want to miss out and keep buying no matter what the price. Momentum becomes more of a behavioral phenomenon and less of a fundamental one. This is the point where new reasons are used to justify continued buying - "they will keep growing!", "the founder is a visionary!", "what they are doing will change the world!" or the very dangerous, "this time is different!". Whatever the motto, stocks do go up and also down. And the higher they go, the greater they can fall. No company grows forever. Ultimately, something upsets the company's "good thing" or there is some external factor that disrupts the company's ability to continually exceed market expectations.

This leads to a momentum crash, which is what we witnessed last week. The return difference between Decile 1 and Decile 10 of medium-term momentum last week was greater than three standard deviations below the weekly average since 2000. There have been only 17 weeks (out of 1,314) where MTM was more negative than last week. That's less than 2% of all the weeks in our data.

Source: S&P Global, Jackson Creek

Last week was the first week of at least a two standard deviation move (positive or negative) in over two years. The last negative three standard deviation event occurred in the week ended June 24, 2022.

The chart below shows the cumulative weekly returns to MTM with weeks where there was a +/- 2 or 3 standard deviation return.

The above return stream is created by a portfolio that is rebalanced weekly and buys the top 10% of stocks ranked on MTM and simultaneously sells short the bottom 10% of stocks ranked on MTM. The portfolio assumes the transactions are made at the close of business on Friday and the positions are held for the following week then rebalanced again on the next Friday.

We can see that extreme moves are not that uncommon. The most volatile MTM returns occurred during broad market volatility and recessionary periods. MTM can experience long periods of positive returns, but there are also very quick reversals that can erase those gains quicker than the time to earn those gains. The latest quick reversal puts the cumulative return from the beginning of 2000 in negative territory.

It is too early to tell if this is the beginning of another prolonged period where we see multiple large weekly spreads, as was the case in the early 2000s, 2008-9, 2020, and 2022/early 2023.