Ratio Worries

Bottom Line: The SPX/VIX ratio is something that has been getting a lot of attention, though much care should be taken before reading too much into it.

This week, there has been much made of a ratio taken between the S&P 500 and the CBOE Volatility Index, or VIX. Taking a ratio of the two supposedly shows how complacent traders are given the current level of prices.

I’ve discussed before the slippery slope we enter when taking ratios of things. One of the most used is the TRIN, which is actually a ratio of a ratio. The problem with something like the SPX/VIX indicator is that the S&P 500 can theoretically go up forever, while the VIX for all practical purposes will hit a ceiling (and a floor). When the S&P eventually makes it to 1500, and the VIX is still at 10 or 12 or whatever, the SPX/VIX ratio will hit a new high. It doesn’t mean that traders will be even more complacent then than they are now, it is just a factor of the numerator being non-cyclical and the denominator being cyclical.

One way to correct for this is to make both parts of the ratio cyclical. A very simple way is to compare the current reading to a moving average, like the 50-day. Even while the S&P has gained nearly 7000% over the past 54 years, it has still cycled back and forth above and below its 50-day average during that time. Even this type of correction is no panacea, since the S&P has spent about 65% of its time above its 50-day average, while the VIX has spent about 44% of its time above, so we still have a positive bias to the top of the ratio that affects the results. Generally, though, this helps to correct for the non-cyclical nature of the trend in the market. If we look at where the S&P 500 is in relation to its 50-day moving average and at the same time compare it to where the VIX is in relation to its own 50-day average, then it should give us a better read on the sentiment behind the numbers than the straight SPX/VIX ratio.

The chart below shows both ratios. The middle pane is the standard SPX/VIX ratio, and the lower pane shows how much different the S&P 500 is from its 50-day average than the VIX is from its average.

From the middle pane (the blue line), we can see what has been getting so many people so excited. The SPX/VIX ratio peaked at around 90 in August 2000, the very top of the bull market. It just about reached that level on October 1st, and once again on November 11th. The assumption, then, is that there is as much complacency now, given the price of the S&P, as there was at the very top of the previous bull market.

But the bottom pane (the red line) shows what happens when we do a very simple correction for the trend in the market. Currently the S&P is about 4% above its 50-day average while the VIX is about 8% below its average. This difference of 12% is well below the 22% difference seen in August 2000, and is nowhere close to the highs of 40% or 50% differences seen at the peaks in the ratio.

From the implied volatility imbedded in options prices for broad-market indexes, it is apparent that traders are not forecasting any great uncertainty. I would be more comfortable on the long side if it was the other way around, but as we saw from 1993-1995, implied volatility can remain very low while the market continues higher. During that time, the VIX spent 442 out of 757 days (58% of the time) below its current 13.0 level, while the S&P went on to rise more than 40%. The SPX/VIX ratio is a misleading gauge, and I would not count on it going forward. Instead, it makes more sense to concentrate a combination of the two that takes the upward trend of the S&P into some consideration.

Bottom Line: Since the Nasdaq 100 bottomed in October 2002, it has been very unusual to see breadth and price diverge to as large a degree as it did today. It has preceded short-term weakness most of the time.

It is relatively rare to see the breadth on popular market indices like the Nasdaq 100 diverge very much from the breadth of underlying technology shares. In other words, on days the NDX is up more than a marginal amount, most tech shares also close higher. On very rare occasions, like today, we see a divergence where the index will close higher but a solid majority of other tech shares close lower on the day.

Since most tech shares bottomed in October 2002, there have only been a handful of days where we have seen the Nasdaq 100 close higher by 0.5% or more while the breadth of the broader Nasdaq market closed at -200 or lower (meaning at least 200 more stocks closed lower than closed higher).

Date	NDX Close	Breadth	NDX 10 Days Later
01/07/2003	1.0%	-206	-6.1%
10/10/2003	0.6%	-220	-2.4%
12/04/2003	0.9%	-215	-0.1%
03/24/2004	0.9%	-266	7.2%
04/23/2004	0.8%	-235	-6.1%
07/07/2004	0.5%	-202	-4.5%
07/22/2004	1.6%	-472	-3.9%
09/14/2004	0.5%	-204	-3.1%
Average Return			-1.8%
Number Positive			1

Your breadth figures my differ from mine, as various quote vendors will close with different figures. Still, from the table we can see that when traders showed a clear preference for the prominent tech stocks in the Nasdaq 100 to the relative exclusion of others, it has tended to precede a down market within two weeks. The 10-day return after such occurrences – all since the low in October 2002 – was negative 1.8% with only one being positive. That one outlier is notable since it came very near the March low, but otherwise we see a pretty consistent pattern.

Bottom Line: Short interest in Nasdaq stocks has been extremely high over the past few months, which is yet another sign that things are very different from several years ago.

A subscriber appropriately reminded me this morning of something that should have been included with the discussion of cash at customer accounts at NASD firms that I mentioned in the last comment, and that is the amount of short interest in Nasdaq stocks.

The Nasdaq, like the NYSE, on the monthly basis makes available the number of shares that have been sold short of stocks listed on their exchange. To sell short, a traders borrows stocks from someone else and sells it, hoping to buy it back later at a lower price (and a profit), so it is a bet on falling stock prices. The more shares of stock that are short, the more negative traders are. There are all sorts of games hedge funds play with short sales, so the above statement was a gross generalization, but from a contrary point of view this is bullish, both because of what is says about traders’ attitudes and also due to the simple fact that many of those shares will need to be bought back at some point in the future, potentially spurring a rally.

The chart below shows the data that we post to the site, which is the short interest ratio on the Nasdaq. This ratio takes into account both short interest and volume, and as we can see, it registered a very low reading of 1.47 in March 2000. Now, however, the ratio is more than double that, at 3.21. Since 1995 the ratio has been higher than that only four months – February 2003, March 2003, July 2004 and September 2004.

The most recent short interest figures will be released in the coming week, so we will get to see then how traders have reacted to the rally during October and early November. If we see short interest plunge to the lower end of its range, then that would not be a good sign for the bulls. On the other hand, if it stays relatively steady, then that is yet one more long-term positive sign to add to the others mentioned earlier this week.

Nothing at all has changed since the last comment. Many of our measures are showing extreme bullishness, yet the S&P and NDX have gone on to hit new highs. Until we see a “euphoric” spike higher (not yet) or a breakdown in price (obviously not yet), I think the risk remains uncomfortably high for short positions. And, due to the extreme bullishness of our measures, the risk also feels uncomfortably high for new long positions. As yet another example, our de-trended equity put/call ratio, which accounts for what type of market environment we are in, has moved down to one of its lowest readings in 7 years. The only other time it was this low, early January 1999, the market suffered a short-term decline before resuming its advance. It’s hard to not realize that so many others are also looking for some short-term weakness, but this type of data is what continues to leave us on the sidelines awaiting a better 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.