NYT: When the Financial Past Is Not Prologue
When the Financial Past Is Not Prologue
Imagine that I flip a coin five times, and with every flip, the coin comes up heads. Now let’s say I offered you a bet on the next toss. Which side would you pick?
It has to be tails, right? I’ve flipped the coin five times and seen heads. The odds are definitely in favor of a bet on tails.
Except they aren’t, and we’re fooled into thinking otherwise because of something called the gambler’s fallacy. It’s a cognitive bias we show by taking a long-term average and assuming it applies to every observation. As Daniel Kahneman describes it, “Chance is commonly viewed as a self-correcting process in which a deviation in one direction induces a deviation in the opposite direction to restore the equilibrium.”
For our purposes, it means that the odds of that next flip turning up tails are still 50-50. Flip a coin long enough and the results will bear out those odds. The gambler’s fallacy, however, leads us to think that what happened most recently bears some weight on what will come next.
In reality, the sample size we’re referring to is usually too small, and as Kahneman notes, we’re inclined to over-interpret the results. Long-term averages don’t fit short-term windows, but that doesn’t seem to stop us from trying to make them fit. A classic example is the debate over when we’re due for a market correction (a drawdown of at least 10 percent) or a bear market (a drawdown of 20 percent or more).
John Prestbo, a retired editor and executive director of Dow Jones Indexes, did the math and figured out that “the market hits a pothole of one size or another about every 20 months on average.” In our pursuit of certainty, we’ll latch on to that number (roughly 600 days) and become convinced that when we start closing in on that date, another drawdown must be around the corner. But that’s the problem with averages.
By focusing on the average experience, we’re ignoring other factors, like how many corrections happened in the 1970s (almost one quarter of the bear markets). The reality of averages is they are a blend of varying numbers and multiple factors that may or may not be relevant at this particular time. However, despite knowing that the past is not prologue, we still seem to expect averages to produce clockwork-like results that we can bet on with certainty.
So as the date of the average correction approaches, we wave the number around, and the discussion becomes, “We need to do something. We’re due a correction any day now.” And maybe we are, but one piece of information isn’t enough to tell us one way or the other.
One of the main reasons I’m a big advocate of the financial planning process is because it removes making a decision based on a single piece of information, like a market correction. Instead, we’re making decisions based on our values, our goals and our resources, the things we can control.
The question of whether we’re due a mafarket correction next week or next month doesn’t matter because we have a plan that accounts for the possibility that it will happen someday. The best part? When we plan, we get to ignore the nonsense that swells up around 600 days because we know better. We also know that the next coin toss isn’t guaranteed to come up tails. That seems like pretty valuable information to me. Now, what will you do with it?