I was going through my daily routine of looking through stock lists and scans, and one of the lists I like to watch is the 52 week low list. Buffett and Walter Schloss have both said that they look at this list each and every day, so I thought it would be good to copy that part of their routine. However, I will mention a few warnings to heed if anyone wants to use this list as a watchlist. There have been numerous momentum studies that have shown that the 52 week high list is actually a better place to look for stocks that are likely to go up in the near future. A couple studies that discuss this anomaly are:
- Buying Winners and Selling Losers, Jegadeesh, Titman (1993)
- 52 Week High and Momentum Investing, George, Hwang (2004)
These studies (there are a number of others that come to similar conclusions) show that stocks that have outperformed the averages over the last 3-12 months continue to outperform in the next 3 months. In other words, momentum tends to persist when measured on an intermediate time frame. I’ve studied momentum and found that it does in fact exist, but it’s very difficult to practice. You must have a well defined method for trading and you must define yourself as a speculator that defines an edge by exploiting the human tendency to “herd”. That is why momentum works.
Momentum works until it doesn’t (mean reversion always wins in the end)
However, a more interesting aspect of momentum is the fact that it reverses at a certain point. Some of these same studies that discuss the success of using the 52 week high (or 12 month momentum), also discuss that if you go out 2 years or longer, momentum tends to reverse…. this means that the best stocks over say the last 2 years often underperform the averages over the next year or two. 3-5 years is even more reverting.
So there is a battle between momentum and mean reversion. Both work, and that’s why you see many traders that have similar long term results that use seemingly opposite approaches. You might think that this doesn’t make sense. It does in fact make sense: it’s just that they are using different time frames.
Value investing is by its nature a mean reverting investment strategy. The objective of the value investor is to find undervalued assets. This can only occur if the market price is depressed below its fair value, and often times this means the stock in question has underperformed, or is at new lows. Of course, value investors can sometimes judge a stock to be undervalued even if it has risen significantly recently, but often times this is not the case.
Another point I’d like to make: I am not personally interested in using momentum nor mean reversion as a strategy in an of itself. I simply think it’s helpful to understand the anomalies so that you can devise a filter or a way of tracking stocks that may be undervalued.
Use the 52-week low list to look for longer term lows
So as I mentioned at the beginning of the post, I shamelessly copied what Buffett and Schloss did by looking at the 52-week low list each day. I do this with the understanding that these stocks often tend to go lower. But I put one spin on this watchlist. I first look for stocks I recognize, and then I look for stocks that might be at multiyear (2 years or longer) lows. These are the stocks that academics have shown to be likely outperformers in the coming years, but more importantly these are the stocks that are likely to be undervalued. Multiyear lows (and multiyear highs) are places where stocks tend to be mispriced based on human irrationality. So take a look at the 52 week low list, and look for longer term lows. It takes less than 10 minutes to do this scan, and occasionally you might find something that you didn’t notice in your other screens.
I started this post with the idea of discussing more details on a specific stock I noticed today. Since the post is getting long, I’ll simply mention that one stock on today’s list jumped out at me: Jos. A Bank (JOSB).
JOSB hit a 2 year low today, and it might be a stock that I’d get more interested in if it continues to get a bit cheaper. One thing I like about JOSB is the fact it’s been compounding its book value at a rate of 24% per year for the past 10 years, gaining net worth each year without interruption since 2003. It’s now selling at 2 times book value, which is less than most of its peers. It also has a very strong balance sheet, and no debt.
Here is a brief 10 year view of the compounding assets and book value:
It’s not what I would consider to be extremely cheap at 13 times earnings and 2 times book, but it looks to me (from the numbers) to be a good company with some sort of competitive advantage (10 years of 24% growth in book value).
I analyze stocks by the numbers, and try to buy a basket of stocks that appear to be above average companies at below average prices. JOSB could be one of these if it gets a bit cheaper in the near future.
And the idea came from a simple scan of the 52 week low list.