Trend following is old.
The Turtles started their experiment in the early 1980’s, but they were far from the first to take advantage of serial correlation in markets. Legendary trader Bill Dunn had been at it for at least a decade at that point.
The basic idea of trend following is often attributed to David Ricardo (1772-1823), who “amassed an immense fortune” via “his own three golden rules:”
- Never refuse an option
- Cut short your losses
- Let your profits run
Ricardo never described his system, but these are the general principles of trend following.
Researchers looked into the question of how does trend following hold up over the long run? Looking at over 200 hundred years of data - going back to the end of Ricardo’s life - they found that it does very well.
They argue that trends exist across all asset classes and all time periods making this an exploitable edge.
Why?
That comes down to human psychology.
There are two distinct but related biases at play: conservatism and anchoring biases.
Conservatism Bias: Why Change is Often More Profound than We Think
The conservatism bias causes people to underestimate changes in the long-term. We think we react quickly to fundamental news such as rate hikes, unemployment reports, wars, natural disasters, and so forth, but in reality, it takes time for the consequences to be felt. This leads to trends.
It reminds me of Amara’s Law: “we overestimate the impact of technology in the short-term and underestimate the effect in the long run.”
Take the internet for example. In the 1990’s, there were vague ideas about how the technology would revolutionize the economy, which led to the tech bubble boom and bust. Despite the fervor of the time, many tech companies went on to provide far greater value than was ever expected, even at the peak of the bubble.
Look at Apple, from its tech bubble peak to market close on December 30, 2022, it is up 11,739%. Amazon is up 1,475%. Microsoft 544%. Even Oracle - one of the most over-hyped real businesses of the era - is up 112% vs its tech bubble peak in 2000 (and this is at the end of a steady, one-year downward march for tech stocks in 2022).
All this to say that even though valuations were absurd at the absolute peak of the tech bubble, they were still well-priced relative to the long-term value they created.
Anchoring Bias
If I’m negotiating with you over the price of something, say a used car, I might open up and offer you 50% of what your asking price is (I’ve done this successfully at car dealerships in the past). This is called an extreme anchor and leans into our cognitive biases (as an aside, the negotiating techniques in this book work).
Research shows that people value things higher when they recently saw a high number or value them lower when they recently saw a low number - regardless of the relevancy of the number. It could be a high or low temperature, a high or low speed limit, price, or whatever - it works its way into our brain and we stick with it.
In finance, this can lead to the emergence of trends. Market participants tend to value a highly-priced stock more than a low-priced stock.
Have you ever read analyst reports and price targets on stocks? They tend to follow the price. While they may list a stock as being overpriced on quarter, if the price rises, then they will repeat their overpriced analysis but with a higher price target!
Here’s an example from a major research firm which set a price target of $98/share on an oil and gas producer, and then later revised that upward to $112/share as the price continued to increase.
Reading the updated report, they maintain that all of the fundamentals they called for previously are still in effect and that the high price reflects “market optimism” and overvalued oil prices. The 14% bump in the firm’s value is equivalent to an extra $8 billion dollars, which is chalked up to being “too bearish” on an investment the company had announced two years ago. That same find saw the stock price rise 20% at the time.
I don’t see any fundamental drivers in the price upgrade. Instead, I see fundamental analysts anchoring to new, higher prices.
200-years and Going Strong
The best trading strategy of 2022 goes to trend following.
Many funds saw returns in excess of 50% for the year, with the "laggards" still turning in double-digits in a year the major indices were deeply negative.
Thankfully, we don't need to just jump on what's hot today. With trend following's strong and lengthy history we can trade this strategy confidentially in 2023, and 2024, and beyond (or at least until our hunter-gatherer brains get updated for modern financial markets).