One Indicator Is Not Enough

By: Henry To | Mon, Oct 3, 2005
Print Email

Dear Subscribers and Readers,

Before we begin this commentary, I want to make a very important announcement with regards to our transition to a subscription model. We had previously stated that the "go live" date was October 1st. Unfortunately (for us), we lost contact with our New Orleans-based attorneys (who were setting up MarketThoughts LLC) for nearly two weeks due to Hurricane Katrina. And then, there was Hurricane Rita... well, you get the idea. As a result, our "go live" date will be postponed to October 15th. Starting on that date, all our recent commentaries will be accessible to subscribers only, although our discussion forum will still be accessible to the public at least until the end of this year. All our archived commentaries and everything else, however, will still be available to the public. Going forward, there will most probably be a time delay on our commentaries for at least three weeks, although we haven't made a final decision yet at this stage. Moreover, there will be a free first-time 30-day trial subscription available to everyone - including to our current subscribers. Finally, for the first six months only, we are offering a discount to all first-time subscribers - a bargain subscription price of US$39 for the first six months. Going forward, the subscription price will be US$99 a year, with appropriate adjustments for a month-to-month or a six-month subscription. Since we are basically a two-person operation, we are only accepting Paypal for payments - no exceptions - as I am sure our subscribers would like me to spend my time on the markets instead of sorting through billing problems. More details will be available when the time comes.

We switched from a neutral position to a 25% short position in our DJIA Timing System on the morning of July 14th at DJIA 10,616. As of Friday at the close, the Dow Industrials stood at 10,568.70 - giving us at a slight gain of 47.30 points. As a trader and speculator, I am still very reluctant to go long here - unless one is already in a winning position in the commodity markets, semiconductors, or international equities, just to name a few. The problem is two-fold, and surprisingly, high energy prices are not one of them. What are those concerns, you may ask? Well, here they are:

  1. The lack of an oversold condition dating back to late October 2004. Sure, there was an oversold condition back in mid April of this year, but according to some to our traditional indicators such as the ARMS Index, the short interest ratio, the equity put/call ratio, etc, this was only mildly oversold compared to the May or the August 2004 bottoms. Moreover, the market has so far failed to enjoy any sustainable rally since January 2004 - suggesting that even the May or August 2004 oversold conditions were not oversold enough.

  2. The persistence of a Federal Reserve board that is intent on continuing their series of rate hikes dating back to June 2004. From a trough of 1%, the Fed Funds rate is now at a target of 3.75%. This is now projected to rise to 4.25% by the end of this year. At some point, the market should sell off given this persistent hawkishness of the Federal Reserve.

Given the fact that we are still in a trendless market - and given my inherent nature as a cautious person, I only like to buy stocks when the market is generally in an oversold condition. History has shown that at various points in history, the market will always sell off to a sufficiently oversold condition - a condition that will allow the market to enjoy a sustainable rally going forward. The most recent periods when the market has gotten to such a condition was March 2003 and August 2004. Since then, the market in general has not been kind to either trend followers or buy-and-holders (unless one was holding commodity stocks or shorting the airlines). History has also shown that while the market has most often risen when the Fed is raising rates, the market will inevitably suffer a significant sell-off before the Fed will stop. The prudent man (or woman), IMHO, should not go long here.

Let's now get on with the theme of our commentary. Many pundits in the popular media and in the financial newsletter world are now either calling for a significant rally or a stock market crash going forward. There is nothing wrong with this, per se, as long as their conclusions are backed up by sufficient analyses (and disclaimers). This author does not claim to do a good job of predicting the market either - but what I do have a problem with is this: The analyst or the newsletter writer who - by just discussing one indicator - is calling for a huge rally or a crash going forward. Such "analysis" is dangerous at best; and unethical at worst.

Let me illustrate why. As I have stated many times before, trading stocks or the stock market successfully requires the ability to properly gauge possible future events and assigning appropriate probabilities to each event occurring. One does not need to understand or know all the variables - just the ability to recognize the important ones and the not-so-important ones. Whatever the case may be, the ability to predict future market movements is severely impaired if one only depends on one indicator. It doesn't matter if this indicator has worked flawlessly for the last one hundred years. Readers who are not familiar with probability theory may be surprised - but one hundred years does not provide a sufficient sample to gauge the future usefulness of one technical indicator. However, if a trader or a speculator can identify another indicator that is at least as promising, then one has effectively doubled one's ability to predict future stock market movements. With three important indicators, one triples his or her predictive ability, etc. This is what we seek to do - looking at as many indicators as we could - along with the courage to filter out the not-so-important indicators and to develop new ones.

As an aside, one of the potentially interesting investments that have recently come up on my radar screen is palladium. The primary reason why palladium came up on my radar screen is the jump in jewelry demand of palladium. During 2003, jewelry demand consumed 250,000 ounces of palladium. During 2004, this figure soared to 920,000 ounces, due in no small part to an increasing demand by Chinese jewelers. Moreover, there is a negligible above-ground stock of palladium, as well as the fact that demand from the global autocatalyst sector increased by 10% from 2003 to 2004. What is not obvious, however, is that while purchases of palladium in the North American autocatalyst sector increased by 20%, consumption actually decreased by 10%, as the trend of "thrifting" continues to grow in the light vehicle markets (during the period from 2000 to 2004, the average amount of palladium used in autocatalysts have been halved). At the same time, there was still a global surplus of approximately a million ounces in 2004 - and this surplus is still projected to exist for the foreseeable future. For readers who are currently bullish on palladium, I would definitely like to hear your views.

Okay, I admit. I am still currently in the "catching up" process - both with the markets and in my full-time job. As part of my "catching up" process, I have been updating a great number of the many indicators that I track, and given the theme of this commentary - this is very timely indeed. To put it briefly - and as I mentioned in the beginning of this commentary - I am still short-term bearish, but for the first time in awhile, I also believe things are starting to look better for the long-term investor. The signal to commit on the long side in a substantial way would come when the market sells off in a significant way - and when the time finally comes, you can bet that this author will be there for our subscribers.

I will begin our "lack of an oversold condition" argument by first taking a look at the NYSE short interest ratio, along with the total outstanding short interest on both the NYSE and the NASDAQ. Following is a chart showing the NYSE short interest ratio vs. the Dow Industrials from January 1994 to September 2005:

NYSE Short Interest Ratio vs. Dow Industrials (January 1994 to Present) - Note that the SI ratio during October of last year spiked to a level not seen since the October 2002 level - giving us a decent rally during the November to December 2004 timeframe.  Please also note the most recent spike to 6.2 in June and now 6.1 in September - is this enough to give us a bull rally?  My guess: Probably not although it is a positive long-term development for the bulls.

While the current NYSE short interest ratio of 6.1 is high on a relative basis - we have not seen a spike up in this ratio since June 2005 - a spike which will indicate to us that we are now at the elusive oversold condition. We got such a spike during October 2002, as well as during October 2004. Moreover, those two spikes took the NYSE short interest ratio to a reading of 6.7 - a reading which equates to, all things equal - another 10% increase (or 850 million shares) in the total outstanding short interest on the NYSE. This will take at least a few more months to play out.

Now, we take a look at the absolute short interest on both the NYSE and the NASDAQ. First, short interest on the NYSE. Following is a monthly chart showing the NYSE short interest vs. the Dow Industrials from November 2000 to September 2005:

NYSE Short Interest vs. Dow Jones Industrials (November 15, 2000 to September 15, 2005) - For the month ending September 15, 2005, total short interest on the NYSE decreased 23 million shares, after rising 228 million shares for the month ending August 15, 2005.  Total short interest on the NYSE is now at 8.56 billion shares - still close to an all-time high of 8.59 billion shares. However, this near all-time high does not signal an imminent rally, as the four-month increase in short interest is actually negative 0.08% - substantially lower than the record of 10.79% set on May 15, 2005.

The same story applies when we take a look at the total amount of short interest outstanding on the NYSE - that while the short interest is relatively high on a historical basis, we have not had a spike in recent months, and thus - the lack of an oversold condition. For the month ending September 15, 2005, the total short interest outstanding on the NYSE actually decreased by 23 million shares, compared to an increase of 228 million shares for the month ending August 15, 2005. While this is close to an all-time high, it is interesting to note that the four-month increase in short interest is actually NEGATIVE 0.08% - substantially lower than the record high of POSITIVE 10.79% set on May 15, 2005. Bottom line: This author will not consider going long until the total short interest outstanding on the NYSE reaches a significantly more oversold level. At this point, I want to see an additional increase of 850 million shares, or approximately 9.4 billion shares in total short interest.

Second, we now take a look at the short interest outstanding on the NASDAQ. Following is a monthly chart showing the NASDAQ short interest vs. the value of the NASDAQ Composite from September 1999 to September 2005:

Nasdaq Short Interest vs. Value of NASDAQ (September 15, 1999 to September 15, 2005) - Short interest on the NASDAQ decreased 9.5 million shares in the latest month but is still very close to its all-time high of 5.84 billion shares (a record made three months ago).  Short interest on the NASDAQ now totals 5.81 billion shares.  This author, however, is still waiting for another spike in NASDAQ short interest (a spike which will generate a sufficiently oversold condition) before he is willing to commit on the long side.

As mentioned on the above chart, the total short interest outstanding on the NASDAQ actually declined by 9.5 million shares in the latest month. The four-month rate of increase in short interest has now declined to 1.68% - the lowest reading since January 2005 (when the four-month increase was actually negative 3.94%). Again, while the total short interest outstanding on the NASDAQ (at 5.81 billion shares) is still close to its all-time high (at 5.84 billion shares), this author would prefer to see a similar spike in the NASDAQ short interest before committing on the long side in a substantial way.

The lack of an oversold condition can also be witnessed when one takes a look at margin debt data. A sufficiently oversold condition is virtually always accompanied by a substantial liquidation of margin debt. Following is a monthly chart showing the Wilshire 5000 vs. the change (three, six, and twelve-month) in margin debt from January 1998 to August 2005.

Wilshire 5000 vs. Change in Margin Debt (January 1998 to August 2005) - As shown on this chart, the rise in margin debt got slightly out of hand in late 2003 but it has certainly been dying down since then (with the exception of November and December of last  year).  However, this author would still like to see both the three-month and the six-month change hug the flat line (it is currently at $11.9 billion and $8.2 billion, respectively) before I would call for a sustainble uptrend here.  The solution?  A significant correction along the lines of the July to August 2004 correction.

As can be seen on the above chart, the change in margin debt over the last three and six months (an increase of $11.9 billion and $8.2 billion, respectively) is still relatively high - relative to the late 2000 to early 2003 period, and even relative to the July to August 2004 period. Again, this author would like to see a substantial liquidation in margin debt before committing on the long side (preferring to see both the 3-month and the 6-month increase hug the zero line), and with the Wilshire 5000 having actually risen since the end of August, my guess is that we will need to wait at least for a couple of months before we will be able to witness such an oversold condition.

Finally, the lack of an oversold condition is also evident from the following chart showing the 10-day and 21-day moving averages of the NYSE ARMS Index vs. the Dow Industrials from January 2003 to the present:

10-Day & 21-Day ARMS Index vs. Daily Closes of DJIA (January 2003 to Present) - The 10-day MA of the ARMS Index is now at 1.03 - pretty much stuck in neutral at this point.  The most oversold reading in recent months was the 1.28 reading we obtained on August 26th - definitely not low enough to act as a base for a sustainable rally going forward.  This author would like to see this reading surpass the 1.5 level at some point before he will be willing to committ in a huge way on the long side.

The most recent "oversold condition" occurred on August 26th - when the ten-day moving average of the NYSE ARMS Index reached the 1.28 level - a somewhat oversold level but definitely not oversold given the many similar readings we have witnessed over the last two years. For instance, this reading touched a level just north of 1.3 in mid-January and in late April 2005, and yet the market has failed to enjoy a sustainable rally coming out of those two oversold readings. Before this author would commit on the long side, I would like to see a ten-day reading of 1.5 or higher. The current reading of 1.03 is still very far from such a level. Of course, any oversold reading in the NYSE ARMS Index will have to be confirmed by similar oversold readings from all the other technical indicators that I am currently tracking, including the McClellan Oscillator and Summation Index, the number of new highs vs. new lows, the Rydex Cash Flow Ratio, and the equity put/call ratio (which is currently at a lowly reading of 0.59).

So Henry, what are you saying? Are you saying that we are now ripe for a stock market crash, given that you looking for a substantially lower market before being willing to commit on the long side?

My answer: From my perspective, a substantial sell-off - where subsequently I could purchase relatively cheap shares - would be attractive. Moreover, since I am inherently cautious and since history is suggesting that we will witness an oversold condition sooner or later (see concerns #1 and #2 in the beginning of this commentary), I am not willing to commit on the long side in a substantial way before we have witnessed such a sell-off. That being said, crashes are inherently rare - and there is a good chance that the market could get to the oversold condition that we were looking for with "only" a 5% to 8% decline. Such a relatively small correction is not unprecedented - and anyone calling for a crash here is, to say the least, irresponsible (especially if one is only basing his or her justification with a single indicator). If, on the other hand, the market does not sell off, then I am willing to wait.

Okay, Henry, let me ask you another question. Waiting for such an oversold condition in the stock market is relatively easy to understand, but what makes you confident that the market will enjoy a sustainable uptrend going forward? What is there to keep the market from being range-bound, similar to what we have witnessed over the last two years?

My answer: Some of the other longer-term indicators I am tracking are falling into place. For example, if the Dow Industrials closes below 10,000 sometime in the next few months, there is no question in my mind that the Federal Reserve will seriously think about pausing their rate hikes (or even cutting rates). Such a move is friendly to the stock market in the long-run - even though it usually takes awhile for the rate cuts to affect the market in a positive manner. Serious readers may recall what I wrote in our last commentary - in our discussion of the Conference Board's Consumer Confidence Index. As I had mentioned, the latest decline in Consumer Confidence was the biggest decline since October 1990. I also stated: "As I have consistently mentioned for the last 12 months, any rally in the markets will not be sustainable unless it is preceded by a sub-90 reading in the Consumer Confidence Index, and for the first time since March 2004, we have gotten such a reading. More importantly, it is interesting to note that the 3-month, 6-month, and 12-month returns of the Dow Industrials subsequent to the October 1990 plunge were approximately 12%, 18%, and 25%, respectively."

Anything can happen in the stock market, and therefore this author will not be surprised if we managed to get a subsequent 12-month performance - starting within a few months - similar to the performance during the October 1990 to September 1991 period. I have also mentioned that domestic equities have been pretty much out of favor with retail investors recently - so much so that at some point in the near future there is a possibility that retail investors and speculators would shift back to domestic equities en masse. The following quarterly chart shows the historical amount of equities and mutual funds as a percentage of total financial assets as held by U.S. households from 1Q 1992 to 2Q 2005 (source: the latest Flow of Funds rate released by the Federal Reserve late last month). Please note that domestic equities are still relatively out of favor with U.S. households today:

Equities and Mutual Funds as a Percentage of Total Household Financial Assets (1Q 1992 to 2Q 2005) - 1) The 'blow off phase' in late 1999 and early 2000 saw the equities and mutual funds as a percentage of total household financial assets ratio topping out at 37.37% - with a concurrent increase in this percentage of over 3% in just one quarter! 2) Equities and mutual funds as a percentage of total household financial assets bottomed at 24.70% in the third quarter of 2002 in the recent 2000 to 2002 cyclical bear market.  The current bounce of 2.10% so far in this percentage is mediocre (at least on an absolute basis) - suggesting that the current rally in the equity markets is not over yet - especially given the fact that we have not had a final 'blowoff phase' so far - a phase which has historically meant an increase of at least 2% in this percentage in just one quarter.  Also note that this percentage as of the end of 2Q 2005 at its lowest level since the third quarter of 2003. 3) Bottom line: Not only is the 'blowoff phase' yet to come, but this development is long-term bullish.

As shown on the above chart, this percentage most recently bottomed in the third quarter of 2002 at 24.7%. While we have witnessed a 2.1% bounce since the third quarter of 2002, the current reading of 26.8% is still relatively low - and post-bubble period aside, is actually the lowest reading since the fourth quarter of 1995! The trend in this ratio tells me two things. One is that the cyclical bull market is not over yet. The other is that this cyclical bull market still has much more room to run before ending. The longer-term picture remains friendly to the bulls - although in the short-run, we are still relatively bearish.

Let's now turn to the most recent action in the U.S. stock market. Many of the sectors within the market continue to diverge from one another, and while the action of the Dow Industrials and the Dow Transports don't tell the whole story, it is always important to keep track of the two most popular Dow indices:

Daily Closes of the Dow Jones Industrials vs. the Dow Jones Transports (July 1, 2003 to September 30, 2005) - Over the last two weeks, the Dow Industrials declined 73 points while the Dow Transports rose 106 points - signaling another small divergence in the two popular Dow indices.  The million-dollar question is: Is the latest bounce in the Dow Transports the beginning of a sustainable uptrend or is it merely a bounce?  This author will argue the latter, given that the Dow Transports (and the market in general) had not experienced a sufficiently oversold condition since late mid April (or late June in the case of the Dow Transports).  For now, we will stay with our 25% short position in our DJIA Timing System.

Over the course of the last couple of weeks, the Dow Industrials and the Dow Transports continued to diverge from one another - with the former declining by 73 points while the latter rose 106 points. The million-dollar question is: Is the latest bounce in the Dow Transports the beginning of a sustainable uptrend or is it merely a bounce? Given the lack of an oversold condition since late June, my guess is that we will at least need one more substantial decline - most probably one that will make a new 52-week low in the Dow Transports before this author will be willing to commit on the long side in general. For now, we will stay 25% short in our DJIA Timing System.

Let's now quickly take a look at our most popular sentiment charts, as this commentary is getting very long already - starting with the Bulls-Bears% Differential readings in the American Association of Individual Investors Survey vs. the Dow Industrials:

DJIA vs. Bulls-Bears% Differential in the AAII Survey (January 2003 to Present) - The Bulls-Bears% Differential in the AAII survey declined 22% to negative 7% in the last couple of weeks - rendering the weekly reading into an oversold status.  However, the ten-week moving average - currently at 9.2% - is still not as oversold as we would like.  Please also don't forget that the most substantial sell-offs in the stock market has occured while the AAII survey is oversold.  For now, this author will wait - and stay with our 25% short position in our DJIA Timing System for the foreseeable future.

Over the last two weeks, the weekly bulls-bears% differential reading in the AAII survey moved back to an oversold condition - decreasing from 22% to negative 7% for a 29% down move. While the weekly reading is oversold, please keep in mind that the ten-week moving average (currently at 9.2%) is not as oversold - suggesting that there should be some more downside for the stock market in the weeks ahead. For now, we will remain 25% short in our DJIA Timing System - we will keep our readers up-to-date throughout the week should we start to change our minds regarding this 25% short position.

The Bulls-Bears% Differential in the Investors Intelligence Survey, meanwhile, again held relatively steady in the last couple of weeks - as the bulls-bears% differential remained steady at 26.6%. Meanwhile, the four-week moving average also held relatively steady - decreasing from 26.6% to 26.5% in the last couple of weeks:

DJIA vs. Bulls-Bears% Differential in the Investors Intelligence Survey (January 2003 to Present) - The Bulls-Bears% Differential in the Investors Intelligence Survey again remained steady in the last couple of weeks at 26.6%.  The four-week moving average of this reading remains somewhat oversold at 26.5% - but definitely not oversold enough for a sustainable rally going forward.  The 'optimal' oversold reading that this author is looking for is a four-week moving average below the 20% level.  For now, the message remains the same: We will remain 25% short in our DJIA Timing System.

While the four-week moving average at 26.5% is relatively oversold, it is still not as oversold as we have seen during the corrections of the past 18 months - suggesting that more downside is in order. This lack of an oversold condition argument in both the AAII and the Investors Intelligence Survey also applies to the Market Vane's Bullish Consensus:

DJIA vs. Market Vane's Bullish Consensus (January 2002 to Present) - The Market Vane's Bullish Consensus decreased by two percentage points in the last couple of weeks - from 65% to 63% in the latest week.   Meanwhile, the ten-week moving average of this reading declined slightly 65.9% to 64.8% - which is still not close to oversold territory.  This author would like to see a weekly reading at the 50% to 55% level (which we have not seen since November 2003) and a four-week moving average of below 60% before he is willing to commit on the long side.

During the last couple of weeks, the Market Vane's Bullish Consensus declined from a reading of 65% to 63% - a relatively oversold reading given the history of this survey over the last two year but definitely still not oversold enough for a sustainable rally going forward. For months, I have mentioned that the "optimal oversold reading" would be a weekly reading of approximately 50% - and I still stand by this view.

Conclusion: While some of our longer-term indicators are now getting friendly for the bulls, this author is still very worried with the short and intermediate term action of the stock market, given the lack of an oversold condition over the last two years and a persistently hawkish Federal Reserve. Moreover, this author is still not entirely convinced that energy prices have topped out, especially given that most of the Gulf Coast production of oil and natural gas is still shut-in at this point. As I have communicated over the last month or so, this author is getting increasingly worried with respect to natural gas - as a colder-than-expected winter could trigger a further rally in natural gas prices (not to mention real shortages in the NE parts of the country). For now, we will remain 25% short in our DJIA Timing System, as we are still looking for a substantial decline in the next several months. Once we have entered the so-called oversold condition that we have been looking for, however, this author will not hesitate to go long in a substantial way.

Signing off,


 

Henry To

Author: Henry To

Henry K. To, CFA
MarketThoughts.com

Henry To, CFA, is co-founder and partner of the economic advisory firm, MarketThoughts LLC, an advisor to the hedge fund Independence Partners, LP. Marketthoughts.com is a service provided by MarkertThoughts LLC, and provides a twice-a-week commentary designed to educate subscribers about the stock market and the economy beyond the headlines. This commentary usually involves focusing on the fundamentals and technicals of the current stock market, but may also include individual sector and stock analyses - as well as more general investing topics such as the Dow Theory, investing psychology, and financial history.

In January 2000, Henry To, CFA of MarketThoughts LLC alerted his friends and associates about the huge risks created by the historic speculative environment in both the domestic and the international stock markets. Through a series of correspondence and e-mails during January to early April 2000, he discussed his reasons and the implications of this historic mania, and suggested that the best solution was to sell all the technology stocks in ones portfolio. He also alerted his friends and associates about the possible ending of the bear market in gold later in 2000, and suggested that it was the best time to accumulate gold mining stocks with both the Philadelphia Gold and Silver Mining Index and the American Exchange Gold Bugs Index at a value of 40 (today, the value of those indices are at approximately 110 and 240, respectively).Readers who are interested in a 30-day trial of our commentaries can find out more information from our MarketThoughts subscription page.

Copyright © 2005-2009 MarketThoughts.com

All Images, XHTML Renderings, and Source Code Copyright © Safehaven.com