Most of the time when we're asked questions, we don't have answers, because our answers are in the portfolio, or they may soon be in the portfolio, and they are the result of lots of focused attention and work that compounds over time to make us true experts. This can only be done in certain areas. As a result, outside of those areas, we may have hypotheses, but we cannot claim that we are the expert.

And at times, though, there are obvious things that have become non-obvious because of the complexity of financial markets, and there is one topic that has emerged which surprises me, and that is the concept of **mean reversion**.

Now we have, say, for simplicity, a stock market in the U.S. that rises 14% a year. I have interacted with several allocators who would consider that well above the historical average returns, which can be calculated in different ways, but they typically end up in some area of high single digits.

For simplicity, I'll just say that the equity returns over time might be a long-term average of 7%. Okay, once you develop a concept that something will revert to the mean, it's important to understand that **it's not a place**. The mean doesn't exist. It is discussed like it's a destination, like a road trip back home. However, it is no place. It is a calculation.

## What Mean Reversion Really Means

And so we need to be mindful of what we know and what we don't. If one expects some mean reversion, which is not an unreasonable beginning point, and one, for example, were to pencil in equity returns over the future of 7%, and equity returns have been 14%, that doesn't mean that we will go to a 7% in that allocation bucket.

It means that we must spend some time **well below 7%** so that mean reversion occurs. In simple terms, 14% plus what divided by 2 equals 7? Well, of course, that would be 14 plus 0.

And so, again, while I'm simplifying, it's pretty simple to see that if you believe in mean reversion, and if you think that the markets generally grow earnings or their value by 7% a year, and if we've been doing 14%, we'll have to spend some time, many years in fact, at 0% in order to so-called revert to the mean.

## Statistical Dissonance

In a way, this is a statistical dissonance. It is as if there's a concept that spreads, and this is natural because markets typically overshoot to the upside. And if we believe in mean reversion, it suggests that this momentum will not continue. Now one can make that claim about mean reversion five years ago, and it had not yet occurred. However, over longer periods, it may occur.

I have no horse in the race of mean reversion. However, it's not an unreasonable starting point, but if done, it should be done accurately. And yet, because of so high equity return period, it's difficult for the mind to accept what mean reversion implies. We use mean reversion all the time.

## The Mean Is Not a Place

When we look at an equity allocation from the other point of view of an allocator, we doubt that any equity allocation ever really achieved the mean. It would suggest an allocation that was held for such a long period that it converged on the mean. And yet, of course, we all know that that would require leaving something untouched for a long period of time.

The mean, again, is not a place and it's not really a price that anybody ever really buys. So, I would say to **beware** because the mean may not hold. Reversions will take place and they might settle down, but they might settle down in a much lower place. And that place would be below the mean.