STOCKS ARE CHEAP, BUT THIS METRIC DOESN’T WORK?
I’ll be frank – I have a special place in my heart for the PE ratio and it is the same place where all the things I hate are stored. This simple to understand metric has, in my opinion, resulted in more misguided Wall Street thinking than just about any metric in existence. A quick glance at the breakdown of the PE ratio shows serious flaws at work here. It is basically a moving price target (which is never correct unless you still believe in EMH) divided by the earnings estimates that are created by analysts who have literally no idea where future earnings will be. In other words, you might as well pick random numbers out of a hat and divide them and then go buy or sell stocks. Naturally, proponents of the PE ratio will say that you shouldn’t use forward PE’s, but to those people I have to respond: do you always drive through your rear view mirror? The numerator (or market price in the PE equation) could care less about past earnings so it’s less than helpful in telling us where future prices might go.
What disgusts me even more about this metric is its incessant use in selling buy and hold strategies. You can’t read a book on value investing or buy and hold without running into the PE ratio. “The market is cheap – stocks for the long-run!” You’ve probably seen this slogan on every mutual fund pamphlet you’ve ever read. Your stock broker no doubt thinks the market is “cheap” right now. The PE ratio has become the sales pitch of an entire generation of sales people who are just herding small investors into fee based products. “Did you know Warren Buffet is a value investor?” “Just buy cheap stocks and hang on. Your status on the list of the world’s richest is in the making!” Or so goes the old sales pitch.
So, I wasn’t surprised to open Yahoo Finance this morning to see the following headline arguing that stocks are cheap according to the PE ratio. But just two articles down is an article from the WSJ arguing that the PE ratio doesn’t work in this environment. You can’t make this stuff up. According to the article:
“Not only is the P/E ratio dropping, it also is in danger of losing some of its prominence as a market gauge.”
Losing its prominence? The only reason this metric ever gained prominence was in large part due to the Wall Street sales machine that has spent the last 25 years selling the idea of buy and hold to a generation of investors who thought they were on their way to becoming the next Warren Buffett. Unfortunately, investors like to latch onto what’s simple. Few things are easier to understand in the complex world of investing than the PE ratio.
I generally adhere to the slogan: “you know less than you think you know”. And I consider myself a pretty savvy investor. I understand the complexities of monetary theory, I’ve traded just about every instrument under the sun, I have seen the inner workings of Wall Street’s slimiest firms, I’ve been lucky enough to outperform the markets for years on end and yet I still am consistently humbled by the markets. I certainly know less than I think I know. In the case of your average Main Street investor you know less than less than you think you know. This idea that anyone can become the next great investor is like me walking into a Harvard Law class and proclaiming that I can talk, therefore, I am the second coming of Johnny Cochran. It’s just not that simple. Yet, the sales pitch continues and the small investor continues to eat up the dream…
The PE ratio is certainly simple. In fact, it’s too simple. Buying stocks or convincing yourself that the market is “cheap” based on the results of two incorrect moving targets is not only a recipe for disaster, but it’s sheer lunacy in my opinion. The contradicting headlines above show just what a confused and bewildering place Wall Street is to begin with. I can only imagine what your average Main Street investor is beginning to think of this place we call “Wall Street”.
Trying to dumb down a system as complex as the market only misleads small investors into believing that they can swim with the sharks without being eaten. The PE ratio has only helped herd more and more small investors into equities with the belief that they too can become the next Warren Buffett. But hey, where will Wall Street be if the little guy ever figures out that Wall Street might not be the right place for their money? Wall Street might actually shrink. This unproductive facet of the US economy might actually sink back to the meager corner that it once maintained. And while that might be a bad thing for the big banks it would likely be great for the long-term productivity of the US economy and the wealth of Main Street citizens – all of whom are invested in the United States though not necessarily in the form of Wall Street products.
TPC, unfortunately, I’m not sure how to paste graphs into the comment section, so i can’t show the data. However, the CAPE has a 75% regression with 15 yr future market returns during the post war period
(Expected Return = -.007*CAPE + .1826).
I am not claiming that one can trade on this information, but, in my opinion, it goes a long way towards helping to understand where we are at in the larger market cycle. And for long term investing, it explains a good chunk of the variation in returns.
I am guilty as charged — just a small time investor that has a whole lot to learn. But CAPE is a metric that gives me some orientation in very volatile seas.
Here’s what you’re looking for I think:
I’ll agree that this is definitely useful in gauging the macro picture of where we are in the cycle. Shiller has certainly made progress on the average PE ratio usage I am generalizing about above, but I still have no idea how to use this in my investment plan. I tend to adhere to the belief that PE ratios don’t matter a heck of a lot so consider me biased.
There’s a conflict in valuation metrics. They’re all meant to show long-term future trends in the macro outlook, but the implementation of a strategy involving something like this requires superior market timing skills.
Valuations don’t matter (new paradigm) until they do. gold, leveraged buyouts, junk bonds, emerging markets, dot.com, real estate, treasuries are the examples of valuations not mattering since I started investing. PE’s do matter if you have an idea of where you are in a economic or industry cycle.
There are certainly some excellent metrics for valuing stocks or companies. I don’t believe the PE ratio is one of them….
P/FCFE?
I like to consider the P/E ratio as a macro sentiment indicator. Most of the % gains in equities during a secular bull market are the result of an expanding average P/E ratio (most easily viewed through Shiller’s CAPE), not a simple increase in earnings.
From memory- so may be off — I believe that from 1982-2000 earnings increased by roughly 250%. But, the P/E ratio went from about 6 to 40 — nearly 700%. if it weren’t for the P/E expansion – the result in a change in investor sentiment toward equities, it would not have been much of a bull market.
Thus, during the long cycles where the average P/E ratio is expanding; Buy and Hold can be an effective form of money management. During the long periods of time where the average P/E ratio is contracting, Buy and Hold is more likely to underperform.
my 2 cents.
they’re definitely no importance to me in the 10th year of an average 16-18 year secular bear market.
Terrific article! In 1970 I took a grad course in technical analysis of the stock market while completing an MBA. The prof, a young Phd, assigned papers to us early in the class so we could investigate different popular theories of stock prices. I picked P/E ratios because I had covered them in other finance classes and never had any reason to question their value. It was a set-up, an ambush because the prof knew they were not reliable and wanted to see if any of us would question what we had already learned. Fortunately, I came to the conclusion they were of very limited value especially on individual equities. Over the ensuing 40 years I am constantly amazed at the misinformation given by Wall Street especially when P/E ratios are used. The point the prof was making is that P/E ratios are popular because they are simple to understand and since the market is not simple, one must rely on many indicators.
Assuming that you consider P/E ratios in analyzing individual companies, you would never look at it as the end-all ratio, in a vacuum. The P/E is (questionably) only useful when taken in the context of growth (PEG ratio). What’s silly is that you never hear analysts state that a lower nominal GDP growth forecast merits a lower P/E multiple?? It’s always about being “average” or cheap versus the historical average” — based on next years hyped up forecasts. As the chart in the first comment clearly shows, P/E rarely rest at the “average”. It’s constantly cycling between over and undervalued.
Given the sovereign debt situation and other structural issues with our economy, I believe secular growth expectations are low. Secondly, the demand for equities (as measured by the ERP) is less given the demographic shifts (income) and the fact that stocks are despised by most “savers” (40-60 yo) after being burned twice in a decade while playing against a stacked deck.
Ultimately, I expect that even if we stabilize earnings in the $80 range, the current multiple of ~ 13 is likely to come down and could even reach historic “puking stocks” lows of 6 or 7. How confident are you all that we’ve reached a multi-generational low at 666 on the SPX?
I have been trying to condition people around me off P/Es for years now and has never worked. Seriously, the P/E is the most flawed metric one could use and will only lead to bad investment decisions.
A good place to start for any small investor would be to start considering things such as return on equity and earnings retention instead…as these are the things that ultimately determine value you get from any investment.
echoing the comments of Mathew Claasen above, I’ve run regressions on Schiller’s data to indicate that the variation in returns is about 80% due to multiple expansion and 20% due to earnings growth. I really try to not be dogmatic (and I have NEVER been accused of favoring simplicity, quite the opposite) but the statistical relationships are what they are — beyond argument or opinion. Not to be cliche, but i think it is really all about time horizon. (though i feel that enterprise value/Ebitda is a better metric and have been trying to build historical datasets on this with limited success)
Of course valuations matter. The price at which you buy or sell something always matters. Trying to figure out whether something is too cheap or too rich is not, however, an easy thing.
Taking the big picture view, the body of evidence shows that buying when valuations are high, no matter which measure you use, tends to lead to substandard returns. That’s why the own-equities-no-matter-what message of ten years ago was so damaging to investors. Conversely, muted expectations, which we seem to be working toward, offer the odds of better long-term returns.
No matter the specific valuation measure, be it a simplistic one like P/E or an intricate one like DCF, there are embedded assumptions, and if you don’t know what they are or their limitations, you are likely to misuse the tool. More importantly, any valuation metric should be looked at not as an answer, but as a jumping off point for analytical questions.
I agree that the P/E is too simple as a starting point on its own, and it is perverted even further in the popular PEG ratio, as I pointed out in this piece: http://researchpuzzle.com/blog/2010/02/10/unpegged/.
The problems with P/E ratios are more in their misapplication than anything else. A P/E that is based on the trailing twelve months is of some value. Indeed to invest without at least looking in the rear-view mirror is reckless. The mistake most investors make is in the assumption that the next 12 months will look much the same as the previous twelve months. Technical analysts especially like to look at recent historical trends and attribute them to what will happen in the future. The obvious problem with trends is that we have absolutely no way of knowing how far into the future a trend will continue.
P/E ratios are valuable in comparison to themselves alone. A company like Procter & Gamble that consistently generates large revenues year-in and year-out, with no indication that things are deteriorating, can be said to be cheap at one price and expensive at another; but only compared to where it has historically traded.
Does this mean you should buy PG when the P/E is 7, I’d say maybe, but you certainly need to perform comprehensive due diligence before buying a stock that just took such a nose dive.
At best, P/E ratios–especially those based on the trailing twelve months–give you an indication of investor sentiment. If you’re to buy a company trading at 50 times it’s earnings(ttm), you better be a prophet! Likewise, if your going to buy a “cheap” (read low P/E) stock you better be sure it’s not a bear in bull’s clothing.
It’s idiotic to use valuation on an index of stocks to time that index. Grow the heck up and use a moving-average crossover! Duh.
PE is very valuable in one context, however, just as many of the other valuation and fundamental metrics are: take the universe of stocks and hold all those with X, Y, and Z. Or rank them by X, Y, and Z, then hold the top N ranking stocks. Etc. Best to use valuation metrics relative to an industry or sector instead of relative to the whole index, since different industries tend to trade at different valuations.
Where it blows up is when nimrods take something that works on a statistical basis and try to apply it to the five stocks they want to hold after investigating the heck out of them, instead of dispassionately holding the 40 to 100 or so stocks out of the S&P 500 that rank best fundamentally or valuation-wise.