Never Say Never

By Jack Forehand, CFA (@practicalquant)

Never tends to be an overused word in investing.

Don’t get me wrong, I am big believer that the past is often the best guide to the future. I think base rates are the best tool we have when trying to predict what will happen going forward. But relying on the fact that something has never happened in the past to predict that it will never happen in the future can sometimes get you in trouble.

Successful investing involves trying to do everything you can to put the long-term odds in your favor. We invest in equities because they have produced long-term returns that have exceeded other asset classes and inflation. Those of us who invest using factors do so because the long-term evidence suggests they produce excess returns over the market. Relying on the past to help figure out what will happen in the future is certainly a recipe that can lead to investing success. But long-term evidence can rarely provide absolute certainty. That balance between following what works in the long-term, but also understanding that sometimes even the best odds don’t lead to a winning hand can be a very difficult one.

Let’s look at an example.

If you were an investor in the mid-90s and were looking at the market’s valuation from a long-term perspective using the CAPE ratio, what you would have seen would have been something like this.

https://www.multpl.com/shiller-pe

As valuations were rising, it might seem from this chart that a CAPE of 30 was something that would not be exceeded. After all, it had never happened before.

But here is what happened next.

Not only did the PE blow through its all time high, the market went on a multi-year run that ended with a CAPE ratio that was 50% higher than it ever had been before. And since then, we have again exceeded that 30 threshold that had never been exceeded before in the current bull market. So something that never happened in 120 years of market history happened twice in a span of 15 years.

For those of us who are value investors, another example of this same idea might hit a little too close to home since it is happening now. The chart below looks at the 10 year rolling returns of the HML factor using data from Ken French’s data library. The HML factor measures the performance of high Book/Market (low Price/Book) stocks relative to low Book/Market stocks. It is a way to look at the performance of cheap stocks relative to expensive ones. When the line below is above 0, value stocks are outperforming expensive stocks over the trailing ten-year period. When the line is below 0, expensive stocks are winning over 10 years.

The first obvious conclusion one can draw from the chart is that cheap stocks beat expensive stocks far more often than the other way around. The second is that there have been some periods where expensive stocks have won historically, but they have been short in nature.

Looking at this chart prior to the past five years, one might conclude that when the ten-year return of the HML factor turns negative, it only stays there for a short period of time. As a value investor in 2015, if you had only considered the history you would have had available, you would have assumed the odds of value bouncing back in the near term were close to 100%. But it didn’t happen. And then more years went by and value continued to struggle. As a result, we now find ourselves in the longest negative period for 10-year value returns that exists in the dataset. Once again, something that has never happened before has happened. And anyone who bet that it was impossible for the past to be exceeded has lost.

Keeping History in Context

And these are only a couple of examples of this type of thing. There are many others. For example, 2017 was by many measures the least volatile year for the US market ever. Investors who bet that volatility couldn’t last as the year went on because it had never happened before were proven wrong over and over again. And if you think the CAPE ratio the US market reached in the dotcom bubble is high, it is nothing compared to the 90+ CAPE the Japanese market reached in the late 80s. It is not impossible that the US market could reach a CAPE like that someday.

None of this is meant to suggest that events that have never happened before are common. None of it is meant to say that predicting the future using what has happened in the past isn’t a good idea. In fact, I think the opposite it true. I think studying the past is the best way to understand what might happen in the future. But it is also important to understand that nothing is foolproof. It is important to understand that the odds of anything happening in investing are rarely 100%. Some really smart investors have seen their careers destroyed by betting that the improbable was actually impossible. So next time you hear that something in investing is never going to happen because it never has in the past, keep in mind that investing is one place where never can be a dangerous word.