Understanding P/Es (Part 2)
In my last article, I talked about some of the subtle, but significant differences in how investors calculate a Price to Earnings ratio and how those differences can lead to very different conclusions. While none of the three approaches (forward P/E, trailing P/E, and normalized P/E) will ever be perfect return forecasting tools, I did cite the normalized approach as the most valuable to me personally. In this article we’ll dive a little deeper into why and incorporate some real world examples.
Before we spend time on the normalized approach let’s briefly review the two other methods and discuss why each falls a bit short, in my opinion, relative to the normalized.
Forward P/E Ratio
One of the primary conceptual advantages of the forward P/E ratio actually ends up being a consistent negative in reality. A forward ratio uses analyst’s estimates for the “E” portion of the ratio. In concept, this should allow analysts to account for expected market dynamics, new products or events, anticipated cost savings, or countless other items they think could happen and be impactful. In reality, this flexibility given to the analyst has historically resulted in overly optimistic figures during good times to overly pessimistic ones during bad times. Analysts after all are human and subject to some of the same behavioral biases as the rest of us, in this case recency bias.
Recency bias is the inclination to use our most recent experiences or results as the foundation or baseline for what will happen in the future. Positive trends beget more positive trends and vice versa. For investors looking for a dose of reality and understanding when markets may have risen too high or fallen too far, this version of the ratio is more likely to justify or rationalize market tops as still attractive and market bottoms as still too expensive. Heading directly into the start of the tech bubble, I can vividly recall analysts tripping over themselves to raise estimates each time there was another positive data point to justify it. Estimates and price targets were raised for no other reason than the old ones had already been met.
Trailing P/E Ratio
Unlike the forward “E”, the trailing “E” is less subjective and relatively free of behavioral biases. Using the last four quarters of data also makes its data current enough to adjust fairly quickly for most relevant changes to the business, industry or market environment. This short window of actual observations, however, can fluctuate significantly from one quarter to the next and result in temporary under/over statement of a company’s real earnings power and sustainable growth rate.
Both the tech bubble and the financial crisis provide great illustrations (see following graph) of this dynamic at work. In both instances, as the market reached its highest points the trailing P/E ratio had actually been falling. Then as markets fell, the P/E ratio was increasing (indicating it was getting more expensive, not less) and in the financial crisis reached its zenith at nearly the same time as the market bottomed. The reason for that is that while the market price was indeed falling, the earnings figures were falling at an even faster rate. The trailing “E” was not just falling due to a slowing economy, but it was being further impacted by significant “extraordinary” accounting write-offs and adjustments. The end result is that while market lows were offering up the best opportunities U.S. equity investors have seen this century, the trailing P/E ratio was sending not just a useless signal, but a sell signal that was far too late.
While many professionals are knowledgeable enough to understand this aspect of trailing P/Es it doesn’t take much for a nervous investor or advisor to rationalize selling when markets are in full blown panic mode. Few like to admit it, but I’ve talked to quite a few investors who were sellers at market lows and their advisor was either unwilling or ill prepared to save them from themselves.
Now on to the normalized methodology. When compared to the other two calculations we’ve discussed, it has at least pointed investors in the right direction when the opportunities were the greatest. Any of the three are fine measures when things are normal (whatever that might mean); the test should be how they behave when markets and investors reach extremes. To be fair, the normalized approach has its own set of real and perceived weaknesses. One of the more consistent pushbacks, and rightly so, is that accounting rules have changed over time and when incorporating older and older historical data (remember CAPE Shiller uses 10 years of historicals) this difference is going to impact not just the current reading, but the historical comparisons, as well. The other weakness is one related to the ratios timing efficiency (ability to market time). The first argument I fully accept, but the second is one I struggle with. Not because it isn’t true, but because 1) I’ve yet to find a metric that was a good market timing tool and 2) this specific metric by its very design was never expected to or championed as any type of market timing tool. It has always been intended to be used as an indicator of longer term return potential.
This last concern on timing is a difficult one for many to accept. While most every investor, from novice to professional, acknowledges the futility of market timing embracing an approach that portends no short term timing ability is simply uncomfortable. Advisors are increasingly evaluated on a shorter and shorter timeframe and every day we are bombarded with an endless stream of geopolitical and financial information. The natural reaction is to feel like we or our advisors must do something with it. It almost feels irresponsible not to. While I’d never go as far as to say that daily news and information is irrelevant, the majority of it is, however, simply noise. While tempting, ignoring headlines and short term fluctuations typically proves not only profitable, but would keep most investors and advisors from committing the investing sin that is almost impossible to recover from – selling out at the bottom or getting too aggressive at the top.
Let’s go back to how this has played out historically and look at the yearly normalized “E” for the S&P 500 when graphed in comparison to the yearly S&P 500 index price level. The power behind this ratio is that smoothing the earnings figures emphasizes the notion that the true earnings power and growth of a market is far less volatile than either the last quarter or year of earnings or the price (see S&P 500 Normalized “E” vs. Trailing “E” chart below). There will obviously be some ebb and flow with the earnings figures over time, but in general it helps keep investors focused on the notion that as the price of an investment falls it becomes more attractive (or vice versa, of course). Sounds too simple, I know, but so does “buy low, sell high”. Not many are able to follow that mantra either as fear or greed end up overruling reason and we search for numerical data or metrics to rationalize what we want to do and not what we know we should do.
The normalized earnings figures (dark grey line) are obviously much smoother and predictable than the year by year figures (blue line). The resulting normalized P/E (shown in the next graph) therefore much more closely resembles the movement of the market price. These are yearly earnings and index observations, so there is even far more volatility in the actual daily S&P index level than is shown here, but the argument would not change and this is far easier to follow!
I will pause with one additional caveat to this idea that an approach emphasizing normalized over trailing or forward ratios is superior. As an allocator of capital across large, diversified asset classes, styles, and strategies our focus is on searching for opportunities at a very high or big picture level. Analyzing individual stocks or investments is indeed a different exercise requiring a much more balanced and nuanced approach. Individual stocks can and do go to zero, so simply being cheap is not enough to warrant an investment. Entire markets, however, are much more balanced and diversified and while they may seesaw in value we can be significantly more confident that they are not “going out of business”.
I hope this, and our previous article on the Price to Earnings ratio have shed some light on one of the more basic ways investment industry professionals attempt to value equity markets. One of our industry’s running jokes is that Wall Street has never seen a graph or metric it couldn’t find a use for (see the yearly discussion about the Super Bowl winner and equity returns). Understanding not just what a metric means, but how it was constructed can only help investors make more informed investment decisions. Our clear preference is to focus on ratios or metrics that emphasize a longer term, valuation based view. That runs counter to most metrics that are quoted on TV or in articles and in our next article we’ll jump into the arguments of why valuation matters.
At Smith & Howard Wealth Management, we believe that our clients are best served over the long term through a combination of our understanding and close monitoring of valuations and overall market conditions, our approach to diversified investments across asset classes and financial planning that is consistently monitored and updated as needed. It is our goal to help you achieve your financial goals. Please call me at 404-874-6244 or email me with any questions.
Unless stated otherwise, any estimates or projections (including performance and risk) given in this presentation are intended to be forward-looking statements. Such estimates are subject to actual known and unknown risks, uncertainties, and other factors that could cause actual results to differ materially from those projected. The securities described within this presentation do not represent all of the securities purchased, sold or recommended for client accounts. The reader should not assume that an investment in such securities was or will be profitable. Past performance does not indicate future results.