BUNK 26
LOW P/ES MEAN LOW RISK
People tend to think low price-to-earnings (PIE) ratios mean low risk for single stocks and the market as a whole. It’s an almost universal, near-religious belief that “everyone knows.” But it’s just wrong. Using P/Es to forecast risk and return over any reasonable time period is about as useful as using a Ouija board.
I debunked the “high P/Es are risky, low P/Es aren’t” myth pretty thoroughly in my 2006 book, The Only Three Questions That Count. Taking it apart multiple ways took up about half a big chapter. I won’t restate that here, but instead take you through yet another way to know P/Es—high or low—aren’t predictive on their own. (I also co-authored a scholarly piece on why P/Es don’t forecast risk or return with my buddy Meir Statman that you can find on the Internet if you’re academically oriented—“Cognitive Biases in Market Forecasts,” Journal of Portfolio Management, Fall 2000.)
My criticism here doesn’t mean P/Es aren’t useful. They can be—but not as forward-looking predictors of market or stock returns. Like all commonly used valuations, their predictive power is long gone since anyone can get them lightning fast on the Internet and markets pretty efficiently discount all widely known information. Use them to compare stock peers, sure. Flip the P/E over into an E/P (the earnings yield) and compare that as a form of an interest rate to going bond yields to know if a firm or sector or the entire market has more incentive to issue new shares of stock or engage in cash-based stock buybacks—that can be useful to know too. Stock buybacks reduce the future supply of stocks and, all else being equal, are bullish. Stock offerings are the reverse. There are many ways to incorporate P/Es into other things that are useful. But don’t think a P/E alone says where a stock or the market is headed.

Fear of Heights

A major problem with P/Es is our natural fear of heights. Modern humans hit the world about 250,000 years ago—and we’ve been dealing with basic problems of survival since. Back then, if you were up high and fell, that was pretty much the end of it. Dead, or crippled—which then was about the same thing. So we developed a very healthy fear of heights. And P/E is expressed as a heights framework—high/low.
Modern capital markets came around only in the last few hundred years. Investing didn’t become broadly available to average Joe types until the last few decades with the innovation of 401(k)s, IRAs, and discount trading outfits like Charles Schwab. Now, anyone with income can easily open a brokerage account and start buying stocks, mutual funds, bonds, etc. Even just 50 years ago, that wasn’t so. Back then, mostly, only those with accumulated wealth did so, and there were precious few of them as a percentage of society.
Simply: Modern investing has evolved much faster than our brains’ ability to deal with it. From our past, we believe height is dangerous. Falling a long way is crippling if not deadly—we believe that cold. We naturally think about high P/Es the same way. Anything high is scary—low is safer, so we think low P/Es are less risky, high P/Es more risky. Not so.
Just for fun, and to go off on a tangent, let me show you another simple way to know our ancestral fear of heights can be misleading. Is it safer to have more space between you and lower elevations or less? Depends! Take flying in planes, another thing our brains weren’t ancestrally set up to do. If you’re in a small plane at 150 feet in elevation and the engine starts acting up, I bet you would rather be at 2,500 feet—more time to get the engine going or coast to a safe place to land. Falling from 150 feet doesn’t make you any less dead than falling from 2,500 feet—splattered either way. But sometimes in some frameworks higher is safer. Higher can be and often is misleading when it comes to safety.

Two Moving Parts to a P/E

A basic problem with P/Es is folks fail to remember there are two simultaneous variables—the P and the E. Then they fail to remember that the P/E is telling you about where you are, not where you’re going—and all the market ever cares about is where you’re going over the next 6 to 24 months.
P/Es were very high at the end of 2008—the S&P 500 P/E was 60.7.1 Whoa! Way above average. But if you thought high P/Es were risky, you would have avoided stocks and missed the big 27 percent surge in US stocks and 30 percent in world stocks in 2009.2 The P/E was high then because earnings were temporarily and hugely depressed because of the recession—very low relative to the P, giving it a very “high” P/E. Fact is, year-end 2008 P/Es were also way above where year-end 2007 P/Es were. But 2007 marked the peak of the market—before the recession and earnings started to decline. So if you thought lower P/Es meant lower risk and higher P/Es meant higher risk, you would have presumed the market was lower risk at the peak of the market than at its subsequent low, which has to be wrong and backwards.
Another example: The S&P 500 P/E was 31.9 on December 31, 2002.3 Sky high—but also a great time to own stocks. US and world stocks returned 29 percent and 33 percent in 2003, respectively.4 Again, P/Es were high then because earnings had crashed during the immediately past recession. It’s very common to see high P/Es during and after recessions—usually a fantastic time to own stocks.
At both those points in time, the P/E was reflecting the here and now and the trailing 12 months past—not where earnings were going. But stocks look forward and typically move ahead of the earnings—they fall before earnings do and go up before earnings rise. The P/E doesn’t help you see forward earnings and future P/Es, unless you dig down to see why earnings are so low relative to prices and make forecasts about the future—and you can’t do that if you don’t get that those two variables don’t necessarily move in sync.
But the feature of the overall market P/E being high at market bottoms is almost universal—just when stocks are very low risk and about to rise hugely. That doesn’t mean high P/Es only coincide with bear market bottoms. Sometimes stock prices overshoot even strong earnings—and go higher still. Why? Rapid earnings growth can lead to an even faster rise in stock prices as investors price in boom times ahead. This leads to higher P/Es. Eventually, too much investor optimism can cause stock prices to get out of whack with economic reality, but don’t expect P/Es alone to indicate that the tipping point has been reached. That P/Es are high doesn’t mean stock prices must fall (any more than low P/Es mean stocks must rise). P/Es were pretty elevated all during the mid- to- late- 1990s, and stocks kept roaring ever higher. The S&P 500 P/E was 32.6 at 1998’s close—then the S&P rose 21.0 percent in 1999.5 P/Es were way above average at the close of 1995, 1996, and 19976—then stocks did 23.0 percent, 33.4 percent, and 28.6 percent respectively in 1996, 1997, and 1998.7
At the close of 1999—just a few months shy of the 2000 peak—prices were indeed high relative to earnings, finally “too high.” But the P/E at the close of 1999 was 30.58lower than previous yearend P/E. Lower! Yet stocks were riskier in 2000 than 1999. And that 30.5 was lower than at year-end 2002. I’ll say that differently—P/Es were higher at the tail end of the major 2000-2003 bear market than they were at the start—but stocks were actually much, much less risky at the end.
What about low P/Es? Do they signal low risk? Another way to ask that: Were P/Es low or high in 1929? Folklore says they were high, but as I documented in my 1987 book, The Wall Street Waltz ( John Wiley & Sons, pages 68-69), they were actually pretty low, about 13—certainly not the sky-high levels of mythology. And that didn’t keep stocks from crashing into a massive bear market. In fact, my sense of history is the low P/E gave folks a false sense of security, lulling them into complacency just before earnings went to hell. The P/E was low because earnings were too high.
A low P/E can mean just what mythology implies—that stock prices are “too low” relative to earnings and might rise. Year-end P/Es for 1981, 1982, 1983, and 1984 were low—and stocks did fine each year following, as most folks would expect.9 Sometimes a low P/E is parallel to a good deal. But a low P/E might mean earnings have run up too far and are unsustainable at those levels, as was the case with 1928 and 1929.
So it’s perfectly normal for stock prices to crash before earnings do—which gives you a low P/E, like year-end 1980 when a low P/E (9.2—very low) preceded falling stock prices in 1981.10 Low yearend P/Es preceded stocks falling in 1930, 1931, 1939, 1940, 1957, 197611—I could go on and on. Those low P/Es didn’t cause falling prices. They just reflected a snapshot of stock prices falling before earnings fell.
Put another way: As my grandfather’s one-time girlfriend and noted literary phenomenon Gertrude Stein once famously said about her hometown of Oakland (across the bay from my grandfather’s hometown of San Francisco), “There’s no there there.”
There’s simply no reliable predictive pattern to P/E alone—any way you slice it in any time frame anyone cares about. It’s an incomplete snapshot in time. P/E says nothing about forward-looking risk or return over the next one, three, or five years. You must dig deeper to understand why a P/E is high, low, or middling, and what that means (if anything), and where future earnings are going—which isn’t easy. Stocks can rise and fall on low and high P/Es equally. And P/Es can be high and stay high for a long time—all while stocks rise. Same with low P/Es.
Many won’t want to believe this because they want to believe what they want to believe—a basic part of what behavioralists define as confirmation bias. They’ll want to go with their guts (a cognitive error—Bunk 7). They’ll want to look for examples that prove they’re right and ignore or explain away those contradicting the bias. But for every point anyone can find in history showing that P/Es work the way their guts believe, there are an almost equal number of examples showing the exact reverse and in the same magnitude. (If you’re curious, I show this mathematically in my 2006 book, The Only Three Questions That Count.)
Let me stretch this a bit further. As bewitching as the idea is of a single valuation or indicator that will tell you where a stock or the market is going, it doesn’t exist. P/E, price-to-book (P/B) ratio, my long-beloved price-to-sales ratio (PSR), dividend yield—it doesn’t matter. Your single silver bullet has never been shown to work—at least not yet. And if such a thing surfaced, everyone would quickly discover it, and it too would soon become powerless. Unfortunately, investing is never that simple. If it were, everyone would beat the market. And you know that will never happen.
..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
44.210.151.5