January was not a pleasant month for stock markets. Stocks endured their worst- ever start to a year, and the number of investors expecting a catastrophic market crash hit its highest level in three years.
Instead of crashing, however, stocks soared higher – a double-digit bounce that erased those early-year losses.
It's easy to be wrong-footed by these swift market turns. In fact, it seems that ordinary investors are hardwired to overreact to market threats, judging by a new Yale study co-authored by Nobel economist Robert Shiller.
Investors, Shiller found, hugely overestimate the odds of market crashes. Fears invariably escalate after, not before, market declines.
Crash survey
Shiller has been surveying individual and institutional investors regarding their expectations for equities since 1989. One question is the basis for the widely followed Yale Crash Confidence Index.
The question asks respondents to estimate the probability of a “catastrophic” market crash over the following six months, one resembling those seen in October 1929 or October 1987, when indices suffered one-day plunges of 13 and 23 per cent, respectively.
One-day crashes of this magnitude are extremely rare, occurring in just 1 per cent of all six-month periods. But investors habitually deem such an event to be all too possible. On average, they estimate the odds of a one-day crash to be about 19 per cent, according to the Yale study, Crash Beliefs from Investor Surveys.
These figures are skewed by some periods of especially apocalyptic sentiment; the median estimate (10 per cent) is lower, although both estimates show investors’ crash probabilities to be wildly inflated.
Remarkably, institutional investors are almost as bad when it comes to estimating crash probabilities, indicating that their grasp of market history is not what it should be.
Crash estimates are routinely too high, but sentiment gets especially ugly in the aftermath of serious market declines. The most bearish readings of the last 26 years occurred in early 2009 and in November 2002, both near the bottom of two major bear markets.
The recent market volatility catalysed another spike in bearishness. In January, the percentage of ordinary investors expecting a market crash hit its highest level since early 2013.
Shiller suggests that investors are prone to acting on a mental shortcut known as the "availability heuristic", to use the term coined by Nobel laureate Daniel Kahneman.
In essence, when people evaluate the odds of something happening, they tend to rely on easily recalled events. Recent events invariably come to mind in most cases, which means investors are liable to rely on recent investment outcomes when assessing future returns.
Media negativity
The damage caused by this mental bias is likely to be greater in the case of ordinary investors. Institutions are likely to access different investment information on a daily basis, while individuals are more likely to be dependent on the media for their financial outlook.
Unfortunately for them, the Yale study makes clear that the media suffers from an inbuilt negativity bias, emphasising investment negatives and thus making these negatives more mentally front and centre to retail investors. Negative stock market days are more likely to make front-page headlines while positive market days are typically relegated to the inside pages.
Similarly, a big down day for stocks is more likely to influence media coverage for a number of days, whereas coverage of a big one-day gain tends to be confined to a single day.
Other studies confirm this negativity bias, which is by no means confined to the tabloid press. One 2014 paper, The Kinks of Financial Journalism, examined stock market coverage in the Wall Street Journal and the New York Times. It found that when markets dropped from a gain of 1 per cent to a decline of 1 per cent, the number of negative words used the following day jumped by 50 per cent.
When markets fell by an amount ranging between 1 and 3 per cent, the number of negative words used doubled. In contrast, the number of positive words used barely budged when stocks registered strong gains.
This pattern is long-standing; the asymmetry in journalistic coverage, the study said “has barely changed from the 1920s to the 1990s, and virtually all authors exhibit the same pattern, emphasizing negative returns, ignoring large positive market moves”.
The evidence is “conclusive”: “Negative market returns taint the ink of typewriters, while positive returns barely do.”
More influenced
Another study, this one of China, confirms that media coverage, as noted earlier, is more likely to influence retail investors.
In China, trading is temporarily suspended in stocks that rise or fall by more than a given amount in a single day. The policy is designed to allow investors a cooling-off period.
The irony is that stocks hitting their upper price limit are more likely to be receive media coverage. As a result, ordinary investors – especially first-time buyers of stocks – are net buyers of such “attention-grabbing” stocks the following day.
Experienced traders, on the other hand, are the ones doing the selling, having accumulated the shares prior to the stocks hitting their price limit.
A similar phenomenon is reported by behavioural finance experts Brad Barber and Terrence Odean in their paper, All That Glitters: The Effect of Attention and News on the Buying Behaviour of Individual and Institutional Investors. Ordinary investors, faced with searching the thousands of stocks they can potentially buy, opt for "attention-grabbing stocks". No such "attention-driven buying" is evident among institutional investors.
It’s not that institutional investors are invariably savvy folk not given to emotional investing.
As noted earlier, the Shiller study found little difference between the crash probability estimates of institutional and ordinary investors, with both groups likely to become especially skittish in the aftermath of serious market declines. But professional investors are traditionally less likely to be swayed by media coverage, and this finding is replicated in the Shiller study.
The Yale study found the survey responses of ordinary investors were significantly associated with negative media coverage during market downturns; this was not true of professional investors.
Importantly, positive media coverage during upswings does not tend to significantly impact ordinary investors’ crash probability assessments, according to the Yale study.
There’s an obvious reason for this asymmetric reaction. A ton of research shows that both humans and animals give more weight to negative events and experiences; negative experiences are more likely to arouse, more likely to be remembered and more likely to influence our evaluations.
Little wonder, then, that media negativity carries much more weight than media positivity.
Next downturn
There’s little doubt that the next market downturn will catalyse the usual media hysteria, for two reasons.
Firstly, if it bleeds, it leads – plummeting stock markets and panicked investors can make for an emotional and gripping story. Secondly, it’s likely that even the most sober of journalists will give more weight to the potential downside, given that it appears to be human nature to overweigh the negative in times of trouble.
All this has important implications for ordinary investors.
Even in good times, investors seem to greatly overestimate the odds of a market meltdown. In bad times, fear levels escalate further, with excessively negative media coverage only amplifying anxieties. A market correction suddenly becomes 1929 redux, in the mind of some investors.
Of course, the next downturn cold well be a “catastrophic” one. Who knows? Though such a scenario is possible, it is not probable. As the Yale study shows, fearing a market downturn can prove more costly than the downturn itself.
