There is perhaps no piece of information that is more misused in equity investing than the valuation of the market. Whether someone has an agenda to show that the market is expensive, or that it is cheap, it is usually very easy to find data or a chart to prove their point. Even for those who don’t have an agenda, it can be very easy to fall into the trap of misusing or misinterpreting valuation information, and reaching an incorrect conclusion as a result. This is especially true at a time like now when there are extreme opinions on both sides of the debate as to whether the market is overvalued.
Despite the limitations of market valuation data, it can be useful if put into the proper context. Here are a few things to keep in mind when looking at market valuation data.
Time Frame is Very Important
There are two very important questions to ask when looking at historical market valuation data and how it relates to the present:
[1] What past time frame does it cover?
[2] What future time frame am I trying to predict?
The valuation of the market at any point in time isn’t that useful unless it is compared to a period in history. The period chosen is exceptionally important and can have a major impact on the conclusion that is drawn. If the period is too short, you end up missing the historical context required to properly analyze the data. For example, a chart measuring the market PE over the decade ending in 1999 would have significantly overstated the average valuation. But on the other hand, the period can be too long as well since things like improvements in technology and productivity over time can justify higher average valuations than the full historical dataset. For instance, if you look at a market valuation chart going back 100 years and compare the current level to the average, you would conclude that the market has been overvalued for almost all of the past 30 years, but most experts agree that higher valuations are justified today relative to the average over the full history.
In addition to the historical time frame covered, what you are trying to predict using market valuation data is also very important. Despite the fact that many pundits try to make predictions using valuation data, the fact of the matter is that there is very little you can reliably predict.
Valuations have little or no value when trying to predict where the market will go over the short-term. It is very tempting to see a very expensive or a very cheap market and think you can catch the top or the bottom, but the reality is that valuations tell us essentially nothing about where the market will go in the next year, or even the next three.
This chart from Vanguard shows the proportion of the variance of future stock returns explained by a variety of metrics. As you can see from the chart, the short-term PE has almost zero predictive power for one year returns and the 10-year PE (CAPE) isn’t much better. When you bring the time frame out to 10 years, the predictive power improves a lot, but it still is far from strong.
The moral of the story is that if you want to use valuations to try to predict the future, you better have a very long-time frame and even then you should understand that any predictions you make will be far from exact.
Source: Vanguard, “Forecasting stock returns: What signals matter, and what do they say now?“
The Metric Matters
If you think the market is expensive, you can almost always find a metric that will back up that hypothesis. If you think it is cheap, the same thing is true. There are many different ways to measure the valuation of the market, and each one has both strengths and weaknesses.
For example, I could value the current market using a standard TTM PE ratio, but we all know that earnings this year will look nothing like earnings next year because of the COVID-19 crisis. To correct for that, I could use a forward PE, but then I have the problem that analyst estimates are notoriously unreliable, and that is likely especially true during a period of uncertainty like this. Or given that profit margins are at all- time highs (or at least they were before the recent crisis) and tax rates and interest rates are at historic lows, I could make the case that a better way to value the market on a normalized basis is using the Price/Sales. This would lead me to conclude that the market is far more expensive than it was using an earnings-based approach. I could go on and on, but the key point to keep in mind is that different metrics can paint very different pictures.
Calculation Methods Can Vary Significantly
One thing that is often overlooked when analyzing the market valuation is the calculation itself. Not only can different methods lead to very different conclusions, but you would probably be surprised how many market valuation calculations that are presented publicly utilize flawed methods.
To illustrate this, let’s start by considering the PE ratio of the S&P 500. Many investors believe that the proper way to calculate its PE multiple is simply to average all the PE ratios of its underlying holdings. But this is incorrect on two different levels. The first is that you can’t just average PE ratios to calculate the PE of a portfolio of stocks. I won’t go into all the details here, but this calculation requires something called a harmonic PE, which averages the earnings yields instead of the PE ratios. If you want more detail on how to do that, there is an excellent post on the blog Econompic that explains why this is important.
The second reason you can’t simply average the PE ratios of the individual companies in S&P 500 to obtain its PE ratio is that each stock has a different weight in the portfolio. The valuation of the larger stocks has much more of an impact on the portfolio’s valuation than the smaller ones do. This is especially true during a time like this when the larger stocks in the index are the most expensive and have the greatest weights.
Some aggregate market valuation calculations don’t use a mean at all. Some use a median, which is just calculated by ranking all stocks and selecting the value that falls in the middle. For an index like the S&P 500, this doesn’t make much sense because of the different weightings among the positions, but if you want to look at the valuation of a typical stock and not a market-cap weighted index, the median can provide some useful info. For instance, the fact that the median PE Ratio of all stocks right now is well below the mean of the S&P 500 is evidence that the average stock is much cheaper than the largest stocks.
Times Can Change
Over time, major shifts can occur in markets. These shifts can make valuation indicators that were useful historically far less so in current times. The Price/Book ratio is probably the best current example of that. If you look back 50 years, intangible assets were a much smaller portion of the total assets of companies than they are today. Since book value only accounts for assets that are on firms’ balance sheets, it has not risen nearly as much as actual assets have when intangibles are included. As a result, the market will look more expensive now using the Price/Book ratio than it has historically because the denominator is smaller than it should be. As a result, this metric is of limited value in looking at the current market valuation relative to history. This is just one example, but in general, it is important to consider whether structural changes have occurred in the market that limit its value when evaluating any market valuation indicator.
The Limits of Market Valuation
So what does all of this mean from a practical standpoint?
I think there are two key takeaways. The first is that there are many ways to look at market valuation, and it is helpful to assume that whoever is presenting the data has some sort of agenda or point they are trying to make and has presented the data in such a way that it supports them. As a result, digging into the details behind what you are seeing is very important. The second is that unless you are trying to predict a range of likely returns over a long-period of time, the market’s valuation just isn’t that valuable as a predictor of the future. It offers almost no useful information over a time frame that is less than five years. The valuation of the market can be very interesting to look at and to analyze, but as you do that, it is also important to keep its limitations in mind.
Photo: Copyright: 123rf.com / donskarpo
Jack Forehand is Co-Founder and President at Validea Capital. He is also a partner at Validea.com and co-authored “The Guru Investor: How to Beat the Market Using History’s Best Investment Strategies”. Jack holds the Chartered Financial Analyst designation from the CFA Institute. Follow him on Twitter at @practicalquant.
