You can't refer to the cheapness or the richness of the Chicago Board Options Exchange Market Volatility Index (VIX) outside the context of historical volatility. The VIX is, first and foremost, a measure of the anticipated volatility of the near-term options on the S&P 500 Index, and the most important clue over the years to the current level of the VIX has been the short-term (10-day or 20-day) historical volatility of the S&P (the VIX calculation doesn't exactly equate to a historical volatility but it is close enough).
So the use of the VIX level as a "fear gauge" must always be assessed net of the major influence of historical volatility on the VIX. While it is tempting after Friday's VIX implosion to less than half its peak levels in January to suggest that investor fear has dissipated to dangerously low levels, this assessment is seriously complicated by the fact that the VIX is not 'low' in the context of the recent volatility of the S&P.
It's a point I've always tried to hammer home. Implied volatility attempts to divide the volatility of the underlying over the stated time frame (usually the next 30 calendar days). There are always expectations of impending news events, and so on, but the best "prediction" tool is simply the volatility of the underlying over the past 2-4 weeks.
What else would you base your markets on besides what you "feel" right now? And as we know, realized volatilities utterly caved early in the cycle, and those "cheap" options actually overbid. That's not uncommon.
Equity volatility is seasonal. According to a recent study by Larry McMillan: 'There is also a seasonality to VIX patterns ...You can see several patterns (over the past 20 years). Early in the year, there is typically a small peak in VIX in January, followed by a slightly higher one in March. "
"Then it goes into a decline during the spring and into mid-summer. It probably comes as a surprise to no one that the low in volatility occurs around July 1 of each year. What might come as a surprise though is that volatility typically rises quite a bit during July and August. Then it really gets going in the fall -- In September and October, when the stock market typically has major declines. It peaks in October."
"After that, volatility becomes surprisingly docile for the rest of year, until by Christmas, it is almost back at the July lows. Not every year follows the pattern exactly, but most are a reasonable approximation. 2008 followed quite closely, and this year the pattern is typical as well.'
"So is it time to assume that the VIX will rally according to these past seasonal patterns? Perhaps, but I'd also suggest the possibility, based in part on the ongoing compression of historical volatility, that the VIX might continue to decline to surprisingly low levels -- perhaps as low as 20 -- before it bottoms."
Well, first off, I feel great that the absolute master in this business agree with some data I came to similar conclusions regarding the early July cycle trough.
Which is also why it drives me nuts when I see "analysis" on the cheap VIX that fails to put it in its proper context, which took the form of assertive directional calls in both directions. On the one hand, analysts said "smart" option money was right to lower bids, as they "knew" the bull would continue this 3rd Q. On the other hand, they said 26 VIX was a red flag of complacency, and bearish.
In truth, neither interpretation was correct. It was a prediction that the always-slow start of 3rd Q was upon us, plain and simple!
No comments:
Post a Comment