A New Low in the VXO

Bottom Line: The S&P’s behavior the week when many traders return to work can be a fairly good predictor of future performance.

In an intraday note today, I gave some stats that suggested that any major move that occurred the day after Labor Day tended to be a good predictor of the performance in the S&P for the rest of the week – in the opposite direction. That means that when the S&P was up 1% or more the day immediately following the holiday, it tended to do poorly for the balance of the week. Conversely, when it was down 1% or more following the holiday, then it had a habit of doing well the rest of the week.

The common wisdom is that once traders return from vacation, the real trading for the fall season begins, as they put their money to work based on their anticipations for the rest of the year. To test that theory, I looked at the performance in the S&P 500 from 1950 through 2003 to see if its return the week after Labor Day had any apparent influence on its return for the rest of the month and the rest of the year. The table below outlines the results.

	If the week after Labor Day is UP…	If the week after Labor Day is DOWN…
Then the odds the rest of the month will be positive are…	50%	27%
Then the odds the rest of the year will be positive are…	82%	73%

We can see from the table that if the week after the holiday was positive, then there was a 50/50 chance that the S&P would be higher at the end of the month than it was after the first week after the holiday, so that doesn’t tell us much. On the other hand, if the week following the holiday was negative, then there was only a 27% chance that the balance of the month would be positive.

There’s a pretty good spread there, and I think it has some merit. We’d have to go back to 1989 to find a year when the week after Labor Day was negative yet the S&P managed to find its way higher over the next few weeks. But we’d only have to go back one year, to 2003, to find a time when the week after the holiday was positive, but the next few weeks were negative. In fact, out of the past 10 times the week after was positive, the next few weeks were negative the majority of the time (6 out of the 10).

It’s much more of a stretch to suggest that this week is so important that it decides the performance for the rest of the year. While there is a difference in rest-of-year performance depending on whether this week is positive or not, it’s not particularly wide. So I think the most use this week may have in telling us about future performance is if it closes negatively, then there is a greater probability that we will see more weakness the remainder of the month. If this week is positive, however, then it doesn’t tell us much at all (about the rest of the month).

Bottom Line: Implied (future) volatility on the S&P 100 is at its lowest point since February 1996. In addition to the simple fact it is so low, it is also diverging from its underlying index at the same time we are entering its most volatile time of year.

On August 24^th, I pointed out a divergence between the NDX and the implied volatility measure of its options, the VXN. I noted that I thought it was a short-term negative, as several previous divergences (positive and negative) had played out well. Since then, the NDX has chopped higher and lower, closing today a little above where it was then. Now we have a similar situation with the S&P 100 and its volatility measure, the VXO. The VXO is the new name for what we used to call the VIX.

As of today’s close, the VXO hit a new low for the year, closing slightly below where it did on June 23^rd (which happened to coincide with the exact day the S&P topped on its way to the August lows). So, from a theoretical standpoint, we can say that traders are more optimistic now than they were a couple of months ago when the S&P 100 was about 15 points higher than it is now. What’s even more unusual to me is that this is occurring in September, a month when volatility normally picks up substantially from the summer doldrums. The fact that traders are not pricing in additional volatility is even more unusual to me than the fact that the S&P 100 is well below its high for the year. In fact, this is only the second time in its history that the VXO has hit a new yearly low in September when the S&P 100 wasn’t also making a new yearly high. The other time was in 1988, but it’s difficult to compare with that time due to the highly volatile period after the 1987 crash.

If we look at those times when the VXO hit a new yearly low at the same time the S&P 100 was at least 2% below its own yearly high (showing us something of a divergence between expectations and reality), the market was lower after 30 days 7 times out of 10, with an average return of -2.5%. After six months, it was lower 5 times, but the average return declined to -5.2%. All of that average loss, however, can be explained by only two distinct occurrences, in August 2000 and March 2002.

Regardless of the fact that we are seeing negative divergences and entering the time of year when volatility tends to increase, just the simple fact that traders are pricing in the expectation of less future volatility than at any other time over the past year is troubling enough - looking at all times the VXO hit a new yearly low, the S&P 100 showed an average return of under 1% up to 90 days later. While it has happened before and probably will again, it’s highly unusual to see substantially higher prices after traders have seemingly forgotten just how volatile the market can be.

Last time I mentioned that I didn’t see much of an edge in either direction, and that continues to be the case. Most of our short-term measures are still relatively neutral, and that usually tells me to stay with whatever trend is in place at the time until we either see more evidence of excessive optimism or price begins to break down. I would much prefer to see some sort of meaningful pullback before looking for long entries, and the discussions above suggest that should occur soon, but betting on short-term declines over the past couple of weeks has been fruitless for the most part.

Longer-term, the market continues to reaffirm the thought that we have seen an important low. This type of grinding higher, not allowing anyone in with convenient short-term pullbacks, is classic. While I would welcome the opportunity to establish more longer-term positions should we get a healthy whack lower (with a coincident decline in bullishness), we have not yet had that opportunity.

This disclosure is not intended as trading advice in any form. It is meant as a note to subscribers that the author may have a position directly affected by the market outlook reflected in the commentary. Although the author takes great pains to remain objective in any commentaries, it is only fair that readers should know that the author may have taken positions in accordance with his market outlook. Positions can and do change at any time, without notice to the reader.