Euclidean Technologies letter to investors for the fourth quarter ended December 31, 2016; titled, – “How Does One Prepare For An Unknowable Future?”
We are pleased to write our ninth year-end letter to Euclidean’s limited partners. During the 12-month period ending December 31, 2016, Euclidean returned 19.5% after fees to investors. During this same period, the S&P 500’s Total Return (that is, the index return including dividends reinvested) was +12.0%.
During the past several years, the stock market’s rise was driven in part by investor enthusiasm for fast-growing companies with no, or marginal, profits. In this context, many value-oriented strategies — which seek to own profitable businesses when they are offered at low prices — did not participate in the market’s returns. Euclidean’s results were a case in point.
However, as we noted in last year’s letter, when these types of environments emerged in the past, they consistently ended with big rewards for value-oriented investors who maintained their discipline through the cycle. In 2016, we received some validation of this view. We are hopeful that the first stages of a supportive value environment are underway.
Yet, if that sounds like a prediction for the future, then you would have good reason to be skeptical. If you ever needed a year to demonstrate just how worthless it is to devote time to predicting the future, you received one in 2016. The experts failed to predict Brexit, the path of interest rates, the US election, and the stock market response to that election. They also counted out my (Mike’s) hometown Cubs when they were losing to the Indians in the World Series.
On all of these fronts, it has been interesting to watch so many experts tout their views with articulate confidence and yet be proven wrong.
It seems that while it is easy to make predictions, it is hard to predict the future with accuracy. This gets us to where we want to go with this letter. There is no question that so much beyond companies’ fundamentals impact share prices in the short term. This is a big reason why many investors devote attention to expert predictions regarding how political, industry, and macroeconomic dynamics will impact their portfolios. But to what end? Even prior to forecasters’ dismal record in 2016, expert forecasts have never been very reliable.
There are clearly great uncertainties that exist in our world today and how they play out will impact the share prices of what you own. However, perhaps the right takeaway is not to invest energy trying to know the unknowable. An alternative — the one we wholeheartedly embrace at Euclidean — is simply to prepare for an unknowable future by attempting to purchase sound companies at very low prices such that negative future scenarios are largely baked into their valuations.
Consider why this alternative can be fruitful. When companies are very inexpensive or expensive in relation to their fundamentals, they are priced in such a way for a reason. That reason is investors as a whole have established a consensus view on how the world is going to look going forward. That view embeds assumptions that prop up the prices of expensive companies and hold down the prices of inexpensive ones. But investors aren’t very good at accurately predicting the future. Thus, eventually, something surprising happens that challenges the consensus view and proves some of its embedded assumptions to be wrong. This erodes some of the lift — and alleviates some of the pressure — previously causing certain companies or industries to be priced, respectively, at premiums or discounts to their fundamentals.
So, although predicting the future is hard, you can prepare. And when you prepare by owning a collection of companies with good fundamentals and low prices, surprises can be very good things.
How Euclidean Is Different
We suspect that quantitatively oriented value investors, as a class, more or less subscribe to what we wrote above. So, it may be helpful to describe what is different about Euclidean.
To begin, if you want to do better than average, your portfolio has to look different than the market. The greater the difference, the greater potential you have for exceptional returns. Of course, this can work in either direction, with the potential for both exceptional over- and under-performance.
Therefore, good measures of an investment process are the magnitude and frequency that its recommendations outperform the median stock over some period of time. As others have extensively documented, sensible places to look for stocks with this kind of potential are companies that are inexpensive in relation to their fundamentals. Across long periods of time, it seems that owning a portfolio of companies that are inexpensive in relation to their earnings, sales, or book values would have yielded better results than owning the market as a whole.
The challenge, though, is that there are quite a few companies that prove to be rightfully cheap; these companies are sometimes referred to as “value traps.” Therefore, to achieve a reasonable probability of achieving good long-term results across a portfolio constructed using simple proxies for value, you need to maintain a very large number of positions. The catch, of course, is that as you own more positions, you also own a greater percentage of the market, and you progressively limit your potential performance.
So then, a reasonable objective is to craft an investment process that demonstrates better outcomes with individual investments than does a simple filter for screening for cheap stocks. If achieved, such a result would give you a sound basis for further concentrating your holdings and pursuing higher returns. The desired results would look something like this:
We believe that this objective can be achieved by taking a nuanced look at companies’ operating histories. Traditional value investors attempt to do this, to varying degrees, by conducting rigorous assessments of individual companies to understand the true character of their businesses. At Euclidean, we do this instead using the tools of machine learning to evaluate individual companies in the context of how similarly situated opportunities played out in the past.
In this context, you can visualize the difference between how Euclidean forms its portfolio and the way many mainstream quant funds form their Smart Beta, or Factor, funds. As they attempt to isolate one or a few statistical factors, you often see them employ a shotgun approach, where they invest in many names, often going long and short hundreds of positions. Euclidean, on the other hand, balances a variety of major concepts to examine a company’s operating history and evaluate it as a potential investment. As we do so, there are many inexpensive companies that our process excludes because of other qualities.
To use an analogy, consider this difference in the context of how you might go about building a winning basketball team. Index investors (Team A) are essentially making a statement that it is really hard to form a better-than-average team, so they do not evaluate individual players and simply let everyone play. Factor investors (Team B), on the other hand, home in on statistical anomalies