When something you expect to happen, doesn’t happen, you get surprises. In markets, the result is an adjustment in prices. Some people might call it volatility.
Of course, volatility carries a negative connotation (confusing it with risk hasn’t helped its image either). I’ve heard nobody is comfortable around asset prices that jump around a lot. It’s apparently so bad, that people purposely build strategies to minimize it in their portfolio. Their goal: avoid surprises.
A smart contrarian might see it differently: volatility — surprises — are a feature of markets. Expectations are often wrong, unexpected things happen, prices adjust sometimes dramatically, which creates opportunities for those less fearful of volatility. In other words, embracing surprise leads to opportunity.
Suppressing volatility is no different than attempting to remove surprise from markets. It’s not only impossible but it — probably, eventually — leads to more of the thing you’re trying to avoid. How does it go — “Stability leads to instability.”
If the goal is stability, what’s the reaction when the thing that was built to be stable becomes surprisingly unstable? Fear comes to mind first, which brings a slew of fear-based decisions because something happened that was never expected (unfortunately, investor behavior tends to be fluid when it ought to be stable while portfolios are built to be stable or rigid but better off somewhat fluid).
Peter Bernstein takes the long way to explain all of this better than I can in one of his Economics & Portfolio Strategy letters from 2004. I share it because it fits well with last weeks post on hidden variety in averages and the consistent fear toward price volatility since the Financial Crisis and the willingness to smooth everything to avoid it.
In Eugene O’Neill’s “Moon for the Misbegotten,” Jim Tyrone wails, “There’s no present or future, only the past, happening over and over again. You can’t get away from it. Ever.” If this is the inherent frailty of the human psyche, how are we, who earn our livings by forecasting an ever-changing future, going to overcome it?
We are doing a poor job of overcoming it, that’s for sure. We can see how poor a job it is by looking at the behavior of forward markets like the stock market, which supposedly reflect expectations more than current events or ancient history. In those markets, prices change when events are different from what the market has expected them to be. These prices are notoriously volatile. In short, reality is continuously taking us by surprise…
Our expectations of the future are not unbiased and do not reflect all available information. Too many of us are Jim Tyrones, and Jim’s expectations of the future fail the test of rationality.
Why is this? The primary explanation, I would suggest, is our overwhelmingly stubborn reluctance to accept change, even when it is right under our noses. We naturally fear the unknown, and the future is always unknown. Hamlet touched a deeply sympathetic chord in all of us when he said, “We would rather bear those ills we have than fly to others that we know not of.”
The awful consequence of this feature of human behavior is that it is inescapably self-defeating. The habit of making decisions whose origins are so heavily weighted with retrospection is part of the process of change itself and part of the forces that shape the future. Rational expectations models so often go astray, in fact, because they are unable to accommodate this dynamic aspect of the process. More important, it is the main reason why surprise is so frequent in its occurrence and so shattering in its impact.
Although human beings are always being taken by surprise, we have learned not to be surprised by surprise. Consequently, we devote a great deal of effort to protect ourselves from surprise – which leads me to the second major portion of this paper: the critical role played by smoothing.
What do I mean by this? Smoothing removes the bumps and wrinkles from the data. In a more fundamental sense, smoothing assumes that variations from a specified path are either transitory or systematic, rather than the first step toward a different and possibly less familiar path. Much smoothing is therefore either naïve, because it assumes change away, or disingenuous, because it is designed to hide changes that might be unpleasant to observe…
We enthusiastically massage the raw data to make them appear to be well behaved. We adjust them for seasonal variation and use year-over-year rather than monthly or quarterly figures when the latter seem too erratic to suit our purposes. Moving averages, trendline calculations, and even the hallowed dividend discount model itself, are among our everyday tools for smoothing short-term variations into manageable shapes.
In more complex fashion, portfolio managers spend much of their time in trying to smooth portfolio performance. We diversify, we hedge, we shift asset mix. Volatility is a no-no, and, all other things being equal, low beta is more virtuous than high beta.
We even try to smooth that most unruly variable of them all — stock prices. Specialists and market-makers earn their keep by containing price moves to the greatest extent that their guts and their capital will permit. We require this in the hope that stock prices will better reflect long-term values rather than the crazies induced by the beauty contest phenomenon in which managers select stocks only on the basis of what other managers are selecting. Nevertheless, there is evidence that the stock market is more efficient in processing information about what other investors are doing than it is in processing fundamental information about the underlying assets, which is why stock prices so often turn out with hindsight to have been crazy rather than rational…
This brings me full circle to the beginning of the argument. The more persistently we smooth the data, the easier it is to sink into the habit of making autoregressive forecasts, or staring into the rear-view mirror. And the more we do that, the more vulnerable we are to surprise.
The truth that we all hate to face is that the data are NOT smooth. It is a cliché to say that change is inevitable, but we still hope that change will arrive in uniform packages, tied in readily recognizable pink ribbons. Unfortunately, it does not: smoothed data are an illusion. The new car you buy loses a third of its value the instant you leave the showroom, even though you may proceed to depreciate it on your books with straight-line methods.
But worse still. The effort to keep the data smooth results in even greater turbulence when the ultimate reality works its way through them…
The meaning of this phenomenon for investors should be clear. While some have argued that fluctuations in stock prices are far too volatile to make any sense in terms of changes in the underlying fundamentals, the case I am making here is quite the opposite.
This is the nub of my entire argument here. Investors may wish that the world were less turbulent, but they sense that smoothing masks reality. So long as this difference between appearance and reality prevails, they will be prey to every kind of rumor, tip, and even more sophisticated forms of information. This in itself makes stock prices volatile and nurtures the beauty contest phenomenon, because the anchor to the fundamentals lies on a soggy bottom. Add to that what happens when surprise does overtake us and reality finally bursts forth — and the volatility of stock prices should indeed be the norm rather than the exception.
The most fascinating part of it all is that this process itself becomes endogenous within the larger forces: the unfortunate consequences of surprise lead to greater smoothing efforts to avoid surprise, followed in turn by even greater turbulence as those efforts inevitably fail. The consequence of smoothing is coarseness in the extreme…
The analysis suggests that the riskiest stocks are the ones with the smoothest fundamentals, not the ones with the most variable data! The companies with the lousy fundamentals may be riskier than the companies with clearly defined trends, but…stocks and companies are not necessarily the same thing.
The market recognizes the vulnerability of companies to surprise when their fundamentals are choppy. We see this recognition in correspondingly low P/Es, high yields and high implied discount rates, short durations, and — most helpful in many ways to the imaginative investor — analyst neglect. The smoothies, the companies to which the market awards high P/Es, low yields, and low implied discount rates, long durations, and intense analytical scrutiny are the stocks that are the riskiest because of what happens to them when the data begin to show bumps.
Look at the falls from grace of Eastman Kodak, Xerox, Polaroid, and even IBM that had unruffled growth trends for so long but not forever! The same thing happened to the steel companies after 1959 – at a moment when steel companies were actually considered growth companies – and to the public utilities after the mid-1960s. High P/Es are extraordinarily slow in recovering after the data reveal that the company is, after all, ultimately going to be subject to variability just like everyone else.
Humanity has constructed a set of convenient devices for avoiding surprise by staring into the rearview mirror to forecast a volatile future and by smoothing data that are inherently bumpy. But there is no wisdom in mere convenience. These stratagems have always turned out to be self-defeating and will continue to be self-defeating. Worse, they doom us to discover over and over that they have neither suppressed nor eliminated the inevitable surprises we hope we have evaded. They serve only to make the shocks bigger.
