By Steve Sosnick

The Cboe Volatility Index is one of the most important metrics for interpreting Mr. Market's mood. Increased correlation among index components is a key reason why he has recently become quite jittery.

The VIX isn't designed as a "fear gauge," even if it seems to play one on TV. Instead, its calculation uses S&P 500, or SPX, options to measure the market's best estimate of volatility over the coming 30 days. An underappreciated factor in that volatility assumption is the correlation among the components of SPX. High correlations spur higher index volatilities, and vice versa.

Let's use a highly simplified example to illustrate that point. Suppose we have an equally weighted two-stock index. If the stocks correlate perfectly and both move 10% higher, the index will be up 10%. But if the stocks correlate inversely, with one moving up 10% and the other down 10%, the index would be unchanged.

In both cases, the daily volatility of the index components is 10% -- remember, volatility measures up and down moves equally -- but the index volatility differs wildly based upon the stocks' correlation.

Of course, when we're discussing a 500-stock index with wildly varied weightings and volatilities, the calculation becomes much more complex. Fortunately, the Cboe has a suite of correlation indexes that attempt to quantify the difference between SPX's and the average single stock's implied volatility. Because correlations differ over time -- for example, some items that might be correlating poorly in the short term may have a closer longer-term relationship -- the exchange offers their correlation measures over different periods. I have found the Cboe 1-Month Implied Correlation Index, or COR1M, to offer a useful comparison with the VIX; both use very similar time frames.

A rule of thumb is that correlations increase during market downdrafts and decrease during steady rallies. Considering that the largest technology stocks (that is, the Magnificent Seven) dominated the market's performance for most of the year, even as the vast majority of SPX components lagged behind, it's hardly a surprise that COR1M was mired near longtime lows for much of 2024. In fact, an all-time low was reached on July 12. It is also no coincidence that the VIX was flirting with post-Covid lows at that time, with the absolute low reading of 10.62 coming a week later.

Yet something interesting was occurring at that time -- a "tell," perhaps. As COR1M plunged during June and early July, the VIX held somewhat steady, albeit at low levels. Because the VIX can function as a proxy for institutional investors' demand for short-term protection, I noted at the time that it indicated that there appeared to be an underlying element of caution that kept it from following COR1M even lower. Indeed, SPX peaked on July 16, though the initial declines were modest.

Things changed dramatically in August. The "carry trade," in which hedge funds borrowed low-yielding yen to buy better-performing assets like tech stocks, imploded spectacularly, causing a broad, if brief, selloff. The VIX spiked dramatically, and so did correlations. While the VIX was touching 65, COR1M hit a high of 76.91. Yet even as stocks have largely recovered, COR1M has returned only to the mid-20s. The rotation among leading sectors has suppressed correlations, which helps explain why the VIX remains in the high teens or low 20s instead of reverting to the midteen levels that prevailed earlier this year.

To be sure, investors also recognize the potential for volatility in the coming weeks -- a period that includes a Federal Reserve meeting and the run-up to an election that is a toss-up. A firming VIX can be analogous to the rising price of umbrellas when rain clouds form on the horizon. Consider correlation to be a barometer.

Steve Sosnick is the chief strategist at Interactive Brokers.

Email: editors@barrons.com

To subscribe to Barron's, visit http://www.barrons.com/subscribe

(END) Dow Jones Newswires

September 13, 2024 21:30 ET (01:30 GMT)

Copyright (c) 2024 Dow Jones & Company, Inc.