Saturday, April 19, 2003
So much has been made of the VIX recently that many of you are probably
sick of hearing about it. However, I believe that studying market
volatility is one of the basic foundations of determining probable future
movement, so I think it's an important concept on which to focus.
I've received several e-mails over the past couple of days suggesting that the VIX should not be a concern of ours right now because it was below 20 around this time last year. With the VIX currently hovering around 25, there's plenty of room for it to fall. My contention is that such a low VIX was reflecting the volatility seen at the time. On April 17, 2002, the VIX was at 20.18. However, the 22-day historical volatility on that date was 17.58, meaning the VIX was trading with a 2.6 point PREMIUM to actual volatility. As of this past Thursday, the VIX closed at 24.59, while historical volatility reached 24.27. This is a premium of only 0.3, suggesting traders are now more certain of future movement than they were last year, even though the absolute level of the VIX is significantly higher now than it was then.
Why am I using 22-day historical volatility? According to Robert Whaley of Duke University (who created the VIX and VXN indexes in association with the CBOE), the implied volatilities of the options which make up the VIX are weighted and constantly adjusted in such a manner as to create a synthetic 30-calendar-day (22-trading-day) at-the-money option. So the implied volatility on this synthetic option (i.e. the VIX) is an aggregation of all market participants' assessments of the expected annualized volatility of the OEX over the next 22 trading days.
Let's see just how accurate these options traders are. The discussion below is oversimplified, but the general concept holds true. The chart below looks at 22-day historical volatility versus the implied volatility from 22 days ago. What this means is that the implied volatility (VIX) has been shifted back by 22 days. So, today's historical volatility of 24.27 is being compared to the VIX close of 35.78 from 22 trading days ago (March 18th). We can see that they track pretty closely:
The following scatter plot shows the correlation between historical volatility and implied volatility from 22 days prior. With a correlation of +.57, it's moderately strong (on a scale of -1 to +1, a correlation of +1 would indicate a perfect correlation). This means that the VIX is a fairly good predictor of actual future volatility.
So, the VIX reading of 24.59 from this past Thursday means that options traders are expecting volatility over the coming 22 days to be about the same as it has over the previous 22 days. This concept is the crux of the VIX Fear Premium, which is posted to the site. As you can see from that chart, options traders are now about as confident of future volatility as they were at this time last year, regardless of whether the VIX is at 25 or 20 or 15 or whatever. This should not be used for specific "buy" and "sell" decisions, rather it should be used to determine if the current environment is positive or negative for equities, based on the potential risk versus probable reward. In that sense, currently the risk of declining prices outweighs the potential reward of rising ones.
This is confirmed by the VIX Transform, an indicator I presented for the first time last weekend. Again, this measures the distance of the 10-day VIX from the 200-day, transformed into a reading that is statistically valid. The "transform" part is extremely important, as it gives true meaning to the word "extreme". The Transform is now very close to maximum overbought, and this has indicated a negative environment for equities since 1986 on a very consistent basis.
The total score for our short-term indicators reached an extremely low 27% at Thursday's close. This is one of the lowest readings in the past three years, meaning that there is a very large confluence of our short-term indicators in extreme overbought territory. In the past three years, the score has dipped below 30% only 7 times. The S&P 500 return 1, 3 and 5 days later has averaged (0.8%), (1.1%) and (1.5%) respectively, with only 1 of those occurrences resulting in an up market. That 1 occurrence was 8/14/00, after which the market made a steady climb of about 2.5% over the ensuing week before finally collapsing into September of that year.
I want to quickly mention that the STEM model hit an all-time low at the close on Thursday. I touched on the low absolute level of the model a couple of times this past week, which showed the market's performance afterward, so I'm not going to harp on it again. But needless to say, a new record low in the model while price has done nothing should not be taken as a positive sign.
Something interesting happened in the Rydex mutual fund complex on Thursday. Even though the NDX was up more than 2%, total assets in the leveraged Velocity fund, which is a long-side fund, dropped. Conversely, assets in the leveraged, short-oriented Venture fund rose. This type of activity happens less than 10% of the time when the NDX rises on the day, and it occurs less than 3% of the time when the NDX is up more than 2%. So when we see these wrong-way Rydex timers betting against a rising market, it should be a bullish sign, right? Not really. I've pointed this out before, but took another look this weekend. When the NDX rises on the day, and these timers bet against the rise, it is actually a better non-contrarian signal than contrarian. Meaning, the market dropped afterwards more often than it rose. The bias isn't that great, but I've had several e-mails suggesting it was a very positive omen, when in fact precedent does not bear that out.
One positive from an intermediate-term perspective is the positioning in the large S&P 500 futures contract. For the first time in the history of this data (going back about 16 years), the Futures Balance Matrix, which compares the aggression of commercial traders versus small speculators on the long side, has been at 100% - the maximum bullish reading - for 5 straight weeks. Also, for the first time in over three years, commercial traders are more net long than the small specs. Just a couple of months ago, I thought it was possible we would NEVER see that again, so now that it's here, the potential message should be noted.
I should temper that with a note about the e-mini once again. The total number of contracts being reported just by the commercials and small specs has approached 1 million, twice that seen just a couple of months ago, and five times what it was a year ago. As I've been saying recently, this contract has done a decent job at tipping off short-term turning points over the past six months or so, and while its long-term history is suspect, its recent record should be noted. In that vein, the commercials added heavily to their record short position during the latest reporting period, while the small specs increased their record net long position just as aggressively. This must be considered a negative for the short-term.
Our short-term measures are suggesting in the strongest tone possible that we should expect weakness over the next 1-5 days. This does NOT mean that one should go out and randomly short the market - it DOES mean that when a valid setup presents itself, one should be more aggressive on the short side than the long side. The only possible exception to this should be if we trade above the 905 level on the S&P. If that happens and the market holds, I believe there may be enough trend-followers looking at that level to make an upside continuation a self-fulfilling prophesy. If so, I would then concentrate on the next level of resistance (930/940) to establish short positions, only at that time I would likely look longer-term.
Our intermediate-term measures have not yet seen the confluence of extremes that the short-term ones have. We are still seeing positive indications from the large S&P 500 futures contract, as noted above, as well as the general level of NYSE specialist shorting and short interest in NYSE and Nasdaq issues. All of those are lending some support to the market here. On the negative side of the ledger, longer-term market breadth is overbought, Rydex mutual fund timers have been aggressive in pursuing this rally, put/call ratios are beginning to tick lower, the sentiment surveys are showing an overall complacent attitude (note the low relative level of the AIM model), and the low volatility is *extremely* troubling (see above). Oh, and lest we forget, we're STILL in a downtrend. When everything is taken together, it suggests that any move up over the 905 level is more likely to set up a high-odds, multi-week shorting opportunity than it is the launching pad for an extended ramp higher.
- Jason Goepfert
Disclosure: long OEX puts
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.
© 2003 Sundial Capital Research, Inc. All Rights Reserved.