On the face of it, equity markets have never been safer. The cost of hedging against declines in major equity indices during the past three months has been the lowest on record, even lower than in 2006, when everyone knew that good times would last forever and that equity markets would do nothing but go up.

The standard measure of equity market risk is the implied volatility of index options, that is, the amplitude of the trading range implied by options prices. If equity markets never moved, options would be worthless; if they doubled or halved every day, options would be worth almost as much as the index itself.

The VIX index of implied volatility on S&P 500 index options is the oldest continuous measure we have; it began trading in the 1980s. As the chart below indicates, the VIX index traded around 15% during 2013 through 2015, but fell to just 10% during most of 2017. The trouble is that that the average volatility of the S&P’s major sub-sectors (as measured by options on sectoral ETF’s) rose, from about 25% during 2013-2015 to about 30% in 2016 and 2017.

VIX

Why this occurred is a matter of arithmetic. A lot of major stocks tanked, for example the ill-fated General Electric and retail stocks in general, while the so-called FAANG combination (Facebook, Apple, Amazon, Netflix and Google) soared. If I own $100 each of two stocks, and one goes to zero while the other goes to $200, my portfolio value is unchanged, and its volatility is zero, even though the individual stocks are extremely volatile.

Tech volatility has fallen to just around 10%, or the same level as the index. That’s because big tech companies like Google, Facebook and Microsoft are unregulated monopolies that extract revenue with the regularity of a toll both. The volatility of retail stocks, meanwhile, has risen from 15% to over 20%, as the Amazons of the world hollow out the Macy’s and Target’s.

retainvstech

In other words, the individual stocks within major indices are less correlated and more diversified. The Chicago Board Options Exchange calculates the implied (that is, expected) correlation among the 50 largest stocks in the S&P 500. As the chart shows, the CBOE’s measure of co-movement among stocks has collapsed during the past four years, from about 80% correlation to 40% correlation.

Implied

In a way, lower correlation, or greater diversification, is a good thing. The trouble is that an outsized share of the gains comes from a very small number of stocks that are priced to perfection. 60% of the gains in the Dow Jones Industrial Average this year, for example, came from just five stocks: Boeing, Apple, Caterpillar, 3M and United Healthcare. Returns to the S&P depend disproportionately on the tech sector, up 31% year to date.

The market’s unusual degree of diversification of returns, and its low volatility, depends on the performance of a tiny universe of stock. Apart from this universe, the equity markets are churning away. A slip in Apple’s Chinese sales, or a new market entrant to challenge Hulu, or a federal move towards regulation of Google could pull the pedestal out from a market that seems above criticism.