Stock market volatility has soared. The CBOE’s Vix index – a measure of the expected volatility of the S&P 500 – has almost doubled since early November, from just over 15 to almost 30 per cent. One interpretation of this is that the probability of a 10 per cent monthly fall in the index has risen from around 1 per cent to 12 per cent. Reflecting this increased risk, the S&P 500 index has fallen almost 10 per cent so far this year.
Conventional explanations for these moves, however, don’t make sense. The idea that they have happened because investors fear rises in US interest rates or Russia invading Ukraine runs into an obvious problem: these were clear threats weeks ago when the S&P 500 was sailing blithely upwards, so why should the market start worrying about them now?
The explanation comes in two parts.
Part one lies in the fact that we should not regard a share price as the present value of future cashflows: as the Nobel laureate Robert Shiller pointed out in 1980, prices are much more volatile than this would have it. Instead, think of a share price as, in economic jargon, a state-contingent security. The simplest such security is a bet. A bet on Laughing Boy in the 3.30 at Haydock pays out well if Laughing Boy wins and nothing if it doesn’t, and its price is the probability-weighted average of those two possible states of the world. In the same way, a share price is the probability-weighted average of its payoffs in all possible future states of the world, ranging from disaster through to a boom.
To see how this generates volatility let’s simplify and assume there are just two possible future states – one in which we see a benign environment in which the FTSE 100 is at 8000, and one in which trouble strikes and the index is at 6000. If there’s a 75 percent chance of the benign outcome and a 25 per cent chance of the bad, then the index will be at 7500: (0.75 x 8000 plus 0.25 x 6000). If though, the probability of the nasty outcome rises to 30 per cent then the index will fall 100 points to 7400. The market can thus fall a lot even if the benign scenario remains likely.
Prices can therefore be volatile not necessarily because traders become pessimistic and then optimistic, but simply because of varying – but perhaps still low – probabilities in the chances of disaster.
But how do these probabilities change? Here’s the second part of the story. The object of trading is not to anticipate an asset’s intrinsic value: in volatile markets this value – even if it can be discovered – will only occasionally coincide with its price. Instead, it is to anticipate what other traders think its price will be. Maynard Keynes said trading is a “battle of wits to anticipate the basis of conventional valuation a few months hence.” He was wrong only in saying “months” and not “minutes”.
Given that prices depend upon perceived probabilities, this means that traders are trying to anticipate the probabilities that others will attach to various scenarios. In this respect, trading is like comedy: timing is everything. If you worry about something when others don’t, you’ll lose: being right eventually is no use if the margin calls you face by being short wipe you out before then. As Keynes said, trading is like a game of snap, “a pastime in which he is victor who says Snap neither too soon nor too late.”
Which explains why volatility has increased now and not earlier. Before Christmas nobody seemed much worried by Russian troops massing on the Ukraine border or by the prospect of rising interest rates, so worrying oneself about them was risky: being short in a bull market is nasty. Today, however, traders are worrying that others will worry (or are worrying that others will worry that others will worry and so on). The facts haven’t much changed, but perceptions of others’ likely moods have. Quite small changes in the “economic fundamentals” can be a trigger for traders to change their opinion about what others will believe.
Which is not to say that traders are stupid. The best ones are certainly not. Mark DeSantis and colleagues at Chapman University in California have shown that their talents lie not so much in understanding economics but in their theory of mind skills: their ability to work out and anticipate what others are thinking. They know to buy just before others get complacent and sell just before they get nervous.
What they have done in the last few weeks is to anticipate that others attach an increased probability to nasty scenarios, such as war or shares being hit by rising interest rates, and a lower probability to some benign scenarios, such as shares being supported by cheap money.
It’s sometimes said that this ability to anticipate others’ moods is an ability to read the market. This isn’t quite true. The market is not a person. If it were, we would think them astonishingly skittish as they veer from euphoria to panic and back again in the space of a few hours.
Instead, the market is emergent. Prices are the unintended outcomes of individual actions. And so is volatility. It does not rise because anybody wants it to, but because people who want to sell cannot find a buyer, with the result that prices must fall. Volatility is a sign of agreement: if everybody agrees that there are downside risks they’ll all sell, causing big price falls. Conversely, stability is what we get when traders disagree, when buyers find sellers and vice versa without prices having to change much.
What we know about volatility is that it rises and falls as a result of triggers which, if they are perceptible at all, can be seen only in hindsight. Short-term volatility, and the share price moves that accompany it, is unpredictable. Retail investors especially should not try to trade it, or act upon it.
