Why Value Statistics Aren't Enough to Find Good Investments

Value investing at it's core is a simple concept. You try to identify companies which are selling for less than their intrinsic value, and then you buy them and hold them until the stock appreciates back to it's intrinsic value.

Simple right?

The difficulty lies in the details.

One problem of course is how do we identify a company's intrinsic value?

Identifying statistics which give us clues as to a company's intrinsic value are not hard to find. We can easily find Price/Earnings ratios, Price/Sales ratios, Dividend yields, Free Cashflow Yields, etc.

These might give us clues as to value, but it would be a big mistake to stop there.

Michael Mauboussin and Dan Callahan recently published an article entitled ROIC and the Investment Process (link below in comments). ROIC stands for Return on Invested Capital, and ROIC represents another common statistic which value investors use to identify what a company's intrinsic value might be.

It is a mistake when value investors are overly reliant on these statistics and what they currently say and fail to appreciate that the current ROIC is not important as what the future ROIC will be.

That's one of the key takeaways from the article written by Mauboussin and Callahan. In order to outperform, it is not sufficient that a company has a high ROIC, but rather that ROIC must either unexpectedly improve over time, where the improvement is not reflected in the current market price, OR, the high ROIC must persist for much longer than is currently expected by the market as reflected by the market price.

In other words, in order for you as an investor to outperform, you need to have a differentiated view on a company on why it is going to get more profitable (improve its ROIC) in the future or why its competitive advantages (as evidenced by a sustained high ROIC) will persist for longer than the market is expecting.

If you are solely reliant on statistical cheapness, then you run the risk of falling into a "Value Trap". Sometimes companies trade cheaply because the business is just not that good. Unless the companies profitability improves, then it will continue to trade cheaply.

This gets complicated as identifying and assessing a company's competitive advantages is not simple like calculating a company's ROIC. It involves assessing a lot of qualitative information about the company and the management and then making judgments based on experience and logical reasoning. It is far from an exact science.

Famed investor Bill Miller actually figured this out before most value investors. While running the Legg Mason Value Trust in the early 1990s, Miller did research on what factors actually drove stock market returns. From that research he concluded that you couldn't rely on statistical cheapness to outperform. Instead you needed to pair statistical cheapness with a view on when a company's return on capital would begin to improve. (Source below in comments).

This leads to a very important further insight for prospective value investors. The insight is that a company may be currently cheap, even though it displays little of the statistical cheapness which is typically associated with value stocks. Amazon has rarely if ever looked cheap by historical value statistics, but it is undeniable that there were many points in time where Amazon would have delivered outperformance. Amazon was a value stock that was hidden.

One could argue that talking about Amazon now is hindsight bias and that identifying Amazon prospectively would have been very difficult to do. I AGREE. It is very difficult to do, but not impossible.

Bill Miller identified Amazon early and is now the largest individual shareholder not named Bezos. Nick Sleep of the Nomad Fund also identified Amazon early and likely continues to hold it to this day. Identifying Amazon's competitive advantages and extrapolating from those competitive advantages that Amazon would continue to grow and raise it's potential ROIC for a long time to come was a difficult, but not impossible task. Could Sleep and Miller have been wrong about Amazon? Of course! They made other investments on these same principles which did not work out nearly as well. But the outsized gains generated by a winner like Amazon allowed Miller and Sleep to outperform markets almost on that one decision alone. It is a feature of this style of investing, and not a bug.

Mauboussin/Callahan and Miller have demonstrated that focusing on what actually drives outperformance is a key factor to actually outperforming. Utilizing Bill Miller's method of combining statistical cheapness, and a view on when a company's profitability will improve can be a very successful strategy. It can also help us to avoid value traps. In addition, it may be that in some companies like Amazon, statistical cheapness might be hard to identify, but nevertheless the strength of their competitive advantages makes the company cheap even at statistically expensive prices. This is very difficult to do, but it's important to realize that it's not impossible and in fact it might be one of the best ways to outperform as an investor.