S&P Volatility

Last week, I mentioned in passing that the recent bout of extremely low volatility was being used an argument for lower prices from bears (churn signals a top) and also as an argument for higher prices from bulls (consolidation suggests a move in the direction of the trend).

I mentioned that historically, the evidence doesn't really support either case.  Several subscribers emailed to say, basically, "prove it".  So here we go.

Over the past week, the historical volatility on the S&P 500 has dropped to its lowest level since March, while at the same time the S&P itself chops around near a multi-month high.

This is an easy test, so let's go back to the inception of the S&P 500 futures and see how often this has occurred, and how the S&P performed over the next 1 and 3 months.

We can see that over the next month, the S&P tended to performed below any random time.  The conditions didn't always precede a peak - only half of the time the S&P sported a negative return - but the maximum loss averaged almost twice what it did any random time.

Only once did the S&P show a return greater than +5%, but five times it gave a return worse than -5%.

Three months later, performance was quite a bit better.  Nearly 70% of the time, the S&P's returns were positive...but that's just barely above random, and the median return was actually a bit less than random.  Once again, the S&P's maximum loss during the trades was significantly worse than random, while the maximum gain was less than random.

Short-term confirmation seemed to make a longer-term difference.  If, after coiling into such a tight band, the S&P broke to the upside and showed a positive return a week later, then three months after that it added to its gains 81% of the time and sported a median return of +5.3%.  The max gain averaged +7.6% while the max loss averaged -3.1%.

But if the S&P broke to the downside instead, and was negative a week later, then three months later it was positive only 43% of the time and had a median return of -2.7%.  The max loss averaged -6.1%, which was larger than the average max gain of +4.7%.

So it seems as though the bulls and bears can both make a decent case for this chop/consolidation...but depending on where we end up a week from now, one side or the other will have a much stronger case for the intermediate-term.

Options Speculation Index

Last week, options traders decided to get quite a bit more aggressive about the chances for a rally.

Across all US options exchanges, traders bought to open 11.6 million call options, but only 7.3 million put options.  They also sold to open 6.6 million puts (bets that the market won't drop precipitously) and 7.8 million calls (bets that the market won't rally hard).

When we combine the bullish bets and compare those to the bearish bets, we get an Options Speculation Index of 1.21.  Basically, that means traders were 1.21 times more likely to engage in a bullish strategy than a bearish one.

The chart below plots this Index over the past decade, with the red arrows highlighting the other times the Index rose over 1.2.

It's pretty clear from the chart that the market struggled after the other times we've seen this.  While it did manage to chop higher for a couple of months in 2006, all of those gains were given back in a sharp correction.

Obviously, we only have a few instances, so we have to be suspect about relying too much on the numbers.  But over the next six months, the S&P showed a negative return each time.  The maximum gain over the next six months was a median +2.3% (nothing more than +5%), while the maximum loss was a median -12.3%.

Again, it's just a few instances, but this is not a good sign.