Oct 12, 2003

In contrast to the beliefs of the smallest of options traders (see below), some of the “smart money” options traders are betting heavily the other way. My mention of OEX options traders being “smart money” always elicits responses, most of them in disbelief, but it is information like that below which tells me that these traders know what they are doing, at least certainly more than small equity options traders.

The chart below is that of something I’ve talked about before – what I called the OEX Determination Index. This is created by looking at how much of the daily OEX option volume is going into new positions. For example, suppose there were 10,000 call contracts and 10,000 put contracts traded on a given day. That would give us a traditional OEX put/call ratio of 1.0. Now suppose that on that same day, open interest in OEX call options increased by 2,000 contracts but put open interest increased by 5,000 contracts. So, only 20% of the call volume went into creating new positions, while 50% of the put volume did.

The OEX Determination Index tracks that very thing, and subtracts the call positions from the put positions – so, for example, on the day mentioned above, the Determination Index would be 30% (i.e. 50% - 20% = 30%). The higher the number, the more OEX traders are opening new put positions in relation to call positions. The chart below is a 60-day moving average of this data, and I have put red arrows near those times the Index has been as high as it is now over the past six years.

This Index gave a questionable signal in 1998, as the S&P 500 churned for several months, then rallied sharply before plunging in the Fall. In early 1999, it gave a bad signal. Even though the market didn’t continue its rally for several months after the Index reached an extreme, it also didn’t decline. After a couple of months, the uptrend resumed in force. Since that time, each of the other peaks in the Determination Index have coincided quite well with weakness in the market over the ensuing several months. Conversely, when the Index reached very low levels, telling us that OEX traders were opening new call positions to a much greater degree than put positions, the market performed well afterwards. This is further evidence that OEX traders do, indeed, know what they are doing to a certain degree. I don’t believe the traditional OEX put/call ratio reflects this as well as it does in combination with open interest.

The argument could be made that just because more put positions are being opened than call positions, it doesn’t mean the traders are bearish. After all, they could be selling these puts to open, which is generally a bullish strategy. That could be the case, but I doubt it, at least for the majority of the volume. OEX options have an American-style exercise, which means that they can be exercised against you at any time prior to expiration. This could open a trader up to unexpected risk, since they may have an option put to them when they weren’t expecting it. This is not the case with certain other options which have a European exercise, such as SPX options, and it is the main reason I believe these OEX traders mostly buy their options to open instead of sell them.

Small options traders are at it again. For the week just past, the R.O.B.O. put/call ratio^TM dipped to 0.45, which tells us that these smallest of options traders (where the average trade size is somewhere between $200 and $2000), bought to open more than two calls for every put. More notable than the volume, perhaps, is the fact that they paid an average of 34% more for their call purchases than they did their put purchases, a “demand premium” higher than any seen since September 2000 – the height of the bubble. The chart below plots this demand premium for calls since 2000:

We can see that this demand is greater than any seen over the past couple of years, and obviously significantly above what was seen this Spring when the demand premium was NEGATIVE 28%. It is clear from the chart that this type of scrambling for calls can become much more extreme – after all, in March 2000, the premium had reached almost 100% (meaning these small traders were paying twice as much for bullish call options than they were willing to pay for the protection of put options).

We can also see their bullish posturing by the distribution of their strategies. Call buying was the most popular recipient of their attention, as it accounted for 35% of the volume for trades of 10 contracts or less. Call selling was next at 32%, followed by put selling at 17%. Protective put buying brought up the rear with just under 16% of the volume. So, overall, the bullish strategies (call buying and put selling) accounted for 52% of the volume. I’ve discussed several times in the past how rare it has been for bullish strategies to get most of the volume over the past couple of years - in fact, this is only the eighth week since October 2001 where bullish strategies have overwhelmed bearish ones in terms of volume.

Barring a 5% drop on Monday, the S&P 500 will mark its 100^th consecutive day of being at least 5% above its 200-day average. This is the 21^st time since 1950 that the S&P has enjoyed this long of a streak. The logical question is…”so what happened next?”. The table below shows the performance of the S&P 500 the given number of days after the 100^th consecutive day of it being at least 5% above its 200-day moving average.

	30 Days Later	60 Days Later	90 Days Later	120 Days Later
Avg Ret	1.6%	3.2%	4.8%	5.4%
# Instances	20	20	20	20
# Higher	14	14	15	17
Max Gain	8.1%	14.9%	17.2%	19.9%
Max Loss	-6.8%	-6.9%	-3.8%	-16.2%

Overall, the performance was quite positive. We can see that six months later, the S&P was higher 17 out of the 20 times, sporting an average gain of over 5%. While it also showed a very nice maximum gain of nearly 20% (in 1954), it also suffered a very large correction of over 16% (in 1987).

Interestingly, the S&P went on to a string of at least 130 consecutive days every time but three. Over the course of the 20 occurrences, the average length of the streak came out to be 178 days, or nearly 8 ½ months. The longest streak was an incredible 436 days in 1954/1955 and the shortest was 104 days in 1951. If we take out that outlier of 436 days, the standard deviation was 50 days, meaning 68% of the occurrences should have lasted between 127 days and 228 days. These streaks were fairly evenly spread among the decades, but of course became more common during the great bull run of the last 20 years. The 1950’s had 3, the 1960’s had 3, the 1970’s had 3, the 1980’s had 5 and the 1990’s had 6. Obviously, this is the first one so far in the 2000’s.

An interesting phenomenon is apparent when we look at the time of year these streaks occurred. By the time the streak first reached 100 days, the month was May or June in 11 out of the 20 instances. No other month had more than 2 occurrences. So if most of the streaks hit 100 days in May or June, then that means that 100 days prior would be sometime in January or February. So we can conclude that the majority of the time, the S&P first stretched more than 5% above its 200-day average in the very beginning of the year.

MONTH	# OCCURRENCES
Jan	1
Feb	2
Mar	0
Apr	2
May	6
Jun	5
Jul	1
Aug	2
Sep	1
Oct	1
Nov	0
Dec	0

We can see that the streak hit 100 days in October only one other time, which was 1958. Then, the S&P first stretched 5% above its 200-day average in early June. It hit 100 consecutive days in October of that year and kept on going for another 156 days, finally ending its streak one year later in June 1959.

As far as applying this to our current situation, my take-away from this is that this type of trend persistency tends to, well, persist. It would be unusual to see the streak end before 130 days, and it would be even more unusual to see the market lower than it is now six months hence. In the near-term, like over the next 30 days, the market was higher 70% of the time. The average gain was 3.4% while the average loss was 2.7%. This is in conflict with the study I showed last week, which presented the performance of the market after October began with five straight up days. In that case, the results were decidedly negative, as the market was lower by 4% in nearly every case after 30 days.

Statistics like this can be found to support pretty much whatever conclusion the author wants to make, so how do we actually use this information to make objective investment decisions? I think both studies have merit – this one shows that such positive market conditions as we have now tend to keep going, and the other one highlights a consistent seasonal tendency when October begins so positively (perhaps due in some part to an initial flurry of fund flows). Frankly, I’m torn between the two and am unsure why I would give one more weight than the other, other than the fact that the October study only had seven instances (four in you include dates since 1950), while the one presented above has 20, so statistically this one may be somewhat more reliable. I suppose one could say that the October study is more short-term and that we could decline 4% or so within the next 30 days, after which time the study above could kick in and carry us higher longer-term. That seems awfully “scripted” to me however, and the market typically doesn’t work that way.

One thing that may give the October study a bit more juice is the fact that we will soon be entering the historically weakest time of the month over the next couple of weeks. The chart below shows the performance of each individual day in October for the S&P 500 from 1950 through 2002. We can clearly see that the strongest upward tendency is toward the beginning and the end of the month (both in terms of average return and percentage of time positive), while all the weakest days are concentrated in the middle of the month. In fact, during the 12 days from the 16^th through the 27^th, only 3 days have an overall positive bias.

Seasonality is difficult to use in isolation, and I wouldn’t recommend it in the least. But when combined with other factors, it is nice to know if you are fighting a headwind or have the wind in your sails. For the next couple of weeks, it appears as though there will be more wind pushing us down than up.

One sentiment indicator that has been steadfastly bullish for the market since March is the level of NYSE specialist shorting as a percentage of total shorting activity. For the latest week available, that percentage dropped to a very low 28%, which is the lowest since in 7 years except for immediately after the markets opened after 9/11. This data is aged, as it covers the week ended 9/26/03, during which the S&P 500 declined about 4%.

I’ve discussed the issues I have with this indicator several times, such as the high level of convertible debt that has been issued (causing the hedge funds which buy the converts to short the underlying shares as a hedge, which in turn would “artificially” depress the Specialist Short Ratio), and the tendency for this data to form long-term multi-year trends. There isn’t much we can do about the hedge fund activity distorting the numbers, but we can account for those long-term trends by “de-trending” the data. When we do that, then this de-trended ratio is only in neutral territory, and not near the other extremes seen over the past few years (this de-trended chart is posted to the site each week). So, while this data does appear to be quite bullish, we must consider the fact that the S&P has already rallied 4% from the time these numbers were created, and that it is not especially extreme in the context of recent history.

For the first time in three weeks, commercial traders have picked up their shorting in the large S&P 500 futures contract. Overall, they increased their net short position by nearly 11,000 contracts. At the same time, small speculators reduced their shorting enough to increase their net long position by nearly 3,000 contracts. The changes are neither large nor especially notable. As I’ve been saying for the past few weeks, I’m not reading a whole lot into this information at this point, especially with the dominant action in the e-mini contract (where commercials reduced their net short position by 42,000 contracts and small specs increased their net longs by 5,000 contracts).

For the week ended October 8^th, AMG Data reports that investors put $3.59 Billion into equity mutual funds, which is one of the larger inflows seen over the past year. Contrary to logic, which would suggest that this liquidity adds fuel to stock prices, the market has usually declined after such large inflows. Since July of last year, the week following an inflow of $2 Billion or more has seen the S&P 500 decline 62% of the time, for an average loss of 0.7%. Conversely, when equity funds suffered an OUTFLOW of $2 Billion or more, the S&P rose the following week 58% of the time, and enjoyed an average gain of 1.3%. Four out of the five largest weekly gains in the S&P in the past year followed some of the largest equity fund outflows. This is nothing dramatic, but I wanted to point out that such large mutual fund inflows such as we saw this past week do not necessarily translate to an immediately higher market, and in fact may be a better short-term contrary indicator if anything.

The broader market is now solidly overbought according to several consistent breadth measures. In addition, we are seeing a fairly large dose of speculation from normally wrong-way traders (note the “demand premium” for calls discussed above, and the extremely high level of the Rydex Beta Chase Index for two examples), and some possibly negative seasonality at work. This is not a good combination, so I continue to prefer concentrating on the short side for the time being. Should we quickly recover and make (and hold) new highs, however, I would not want to remain in short positions.

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.