What is the ARMS Index Currently Saying?
(June 18, 2006)
Dear Subscribers and Readers,
We entered a 50% long position in our DJIA Timing System on Thursday morning, June 8th at a DJIA print of 10,810. As mentioned in our last weekend's commentary, the author was getting ready to shift to a fully-invested 100% long position in our DJIA Timing System. We had been waiting for an oversold condition in the market to do so, and the market did not disappoint. In a real-time email that we sent to our subscribers, I noted to our subscribers: We have just shifted from a 50% long position to a 100% long position in our DJIA Timing System at DJIA 10,800. The NYSE intraday ARMS index just touched a hugely oversold reading of 2.46 while the VIX spiked up another 15%. Based on Friday's close of 11,014.15, our 100% long position in our DJIA Timing System is 209.15 in the green. Again, readers who are interested in our historical signals can see more (and learn about our rationale behind those signals) at our MarketThoughts DJIA Timing System page.
As of Sunday evening, June 18, 2006, this author has no intention of shifting our 100% long position in our DJIA Timing System unless the decline over the last six weeks resume or accelerates. In terms of timeframe, we are definitely close to an intermediate bottom, but between now and that ever-elusive bottom, anything can happen (readers should note that stock market crashes come when the market is very oversold). For the first time in the history of this commentary, we will be placing a stop on our 100% position at our average entry point 10,805 in order to avoid the possibility of a crash. As this author will illustrate in the following commentary, however, we do not believe that the market will crash from current levels. In fact, chances are better than 50-50 that the market has put in an intermediate bottom last week at a DJIA print of 10,706.14 at the close last Tuesday.
As many of our long-time subscribers should remember, this author has been calling for a significant top in the world's markets since early this year. Since early May, many of the world's markets have literally collapsed with the U.S. markets performing relatively the best (since it had declined the last). In a liquidity-challenged environment, U.S. denominated assets almost always perform the best and precisely because of this reason, this author believes that the leadership will shift from international equities and commodities to U.S. assets (especially large cap stocks) going forward. That is, while it may continue to be fruitful to short copper or the Indian Closed End Fund (IFN), for example, this author would definitely not be shorting the S&P 500 or the Dow Industrials right now. Folks who continue to watch for a crash in either of these two indices may have to wait longer than usual readers please stay tuned.
Readers who have been with us for awhile should know that this author uses the NYSE ARMS Index extensively as an overbought/oversold indicator in order to try to pinpoint market tops and market bottoms. The former is usually a thankless endeavor as overbought/oversold and sentiment indicators are usually not that great in trying to catch market tops (to try to catch market tops, spotting key divergences is usually the key). Market bottoms, however, are usually easier to predict, as bottoms tend to be formed over a shorter period of time since fear is a much greater emotion than greed. In developing the ARMS (or the TRIN) index, Richard Arms has discovered that the ARMS index works best as an oversold indicator and the ARMS Index did not disappoint as it hit a very severe oversold level prior to the big rally in the Dow Industrials last Wednesday and Thursday. As an aside, the 1996 edition of his book The ARMS Index is a must-read.
In particular, this author likes to use both the 10-day and the 21-day simple moving average of the NYSE ARMS Index as a gauge on how oversold or overbought the market may be at any point in time. Following is the daily chart showing the 10-day and the 21-day moving average of the ARMS Index vs. the Dow Industrials from January 2003 to the present:
At their most recent peaks (June 12 and June 13, respectively), the 10-day and the 21-day moving average of the NYSE ARMS Index touched a level of 1.52 and 1.36, respectively representing the most oversold readings since August 2004. To put this in perspective, the 10-day moving average of this reading with the exception of the 2001 to 2002 bear market rarely touches the 1.5 level, as can be seen from the following chart showing the 10-day moving average of the NYSE ARMS index from January 1949 to the present:
The red line on the above chart represents the 1.5 level of the 10-day moving average of the NYSE ARMS Index. As can been on the above chart, the 10-day moving average of the NYSE ARMS Index does not usually touch the 1.5 level. In fact, with the exception of the early 1950s and the bursting of the technology bubble during 2001 to 2002, there has been less than a dozen occasions during which this reading has touched or surpassed the 1.5 level. The first of the two most oversold conditions occurred during President Eisenhower's heart attack on September 24, 1955, while the second should be one which stock market students should be most familiar with that of during Black Monday, October 19, 1987.
So Henry, what are you saying? Are you claiming this oversold condition in the NYSE ARMS Index is giving way to a significant bottom?
It is still too early to say, but I believe that we are now very close to a bottom, if we haven't already had one at the close on Tuesday of last week. That being said, there is no way to know how far the ensuing rally could go but chances are that this will lasts for at least six to eight weeks. The chances of the market having already touched at least a ST bottom is further reinforced by the strong up day last Thursday, as the NYSE ARMS Index closed at a hugely powerful (on the buy side) reading of 0.19 (which was also confirmed by the fact that it was a Lowry's 90% upside day). Going back to the mid 1990s, there were only four prior instances when the NYSE ARMS index closed at a level lower than 0.25. As shown by the below chart, two of those readings were blow-off days while the remaining two readings gave us a significant bottom:
Please note that three of these readings came in the midst of the great 2000 to 2002 bear market in technology stocks and unless the bird flu hits Japan or Western Europe tomorrow, it is difficult to envision that the major market indices will continue to decline from current levels. In this author's opinion, October 28, 1997 may actually be a better comparison (and even better given that the October 28, 1997 bottom occurred while most of Asia was bottoming at the same time similar to the situation today).
Perhaps a better comparison may be to look at such days (when the NYSE ARMS Index closed at a level lower than 0.25) prior to the mid 1990s. Following is the chart showing the Dow Industrials vs. days when the NYSE ARMS Index closed at a level lower than 0.25 from January 1984 to December 1992 (please note that there were no such readings during the 1993 to 1996 period):
Note that there were three such readings from January 1984 to December 1992. Two such readings came on May 31, 1988 and again on September 2, 1988 the year when Wall Street was still barely recovering from the October 1987 crash. In retrospect, these two readings foretold the buying power that was lying dormant in the stock market. There was one more quick correction following the September 2, 1988 reading but from late 1988 onwards, the Dow Industrials was to appreciate nearly 50% before topping out on July 16, 1990. The final reading came on May 11, 1990 which in retrospect was a precursor of a final blow-off rally which would take the Dow Industrials up another 7.1% before topping out.
In other words, the record of the ARMS Index given the hugely oversold condition last Tuesday followed by a powerful upside day last Thursday suggests that we have at least made a significant short-term bottom in many of the major market indices. This study is especially interesting when one takes into account the fact that last Thursday also happened to be a Lowry's 90% upside day which was preceded by two 90% downside and one near-90% downside day. The last time that such a combination occurred was March 17, 2003. October 28, 1997 also actually came close. While the powerful NYSE ARMS Index reading was not accompanied by a 90% upside day on October 28, 1997, it is interesting to note that it was preceded by one 90% downside day. While no indicator is infallible, the historical study of the NYSE ARMS Index suggests that the market touched a significant low last Tuesday a low which should give way to a rally lasting at least six to eight weeks long.
But Henry, the Dow Industrials is still trading at over 11,000. How could you call that oversold?
That is certainly true, but please note that many of the small cap and mid cap indices have already sold off significantly from their early May highs. Moreover, the NASDAQ Composite topped out as early as April. Same goes for the Philadelphia Semiconductor Index and the Dow Utilities the latter of which actually topped out in October of last year. Finally as I have been pounding on the table for the last few months many of the U.S. large cap brand name stocks have continued to sell off significantly in recent weeks and months, such as MSFT, INTC, DELL, EBAY, YHOO, WMT, HD, GCI, UNH, AMAT, APOL, DJ, IBM, and TYC.
The oversold condition is also evident when one takes a look at the relative strength of the retail HOLDRS (RTH) vs. the S&P 500:
As mentioned in the above weekly chart, the relative strength of the RTH vs. the S&P 500 declined nine weeks ago to a level not seen since April 2003 suggesting a severely oversold condition. But more importantly for us bottom-seekers, the relative strength of the RTH has typically bottomed two to eight weeks prior to a general bottom in the stock market since this cyclical bull market began in October 2002. Since it has been nine weeks since the most recent low, the timeframe for the stock market to make a lower low in the coming days has probably run out and chances are that we have seen a significant low in the stock market last Tuesday at the close.
Let's now take a look at the most recent action of the Dow Industrials vs. the Dow Transports, as shown by the following chart from July 1, 2003 to the present:
Last week, I mentioned that:
given the severe oversold condition in many of our technical indicators and given the strength in the Dow Utilities over the last few weeks, probability suggests that a short-term market bottom will develop sometime this week followed by a tradeable rally that should play out over the next couple of months. Readers should continue to monitor the NYSE A/D line, the NYSE McClellan Oscillator and Summation Index, along with NYSE and NASDAQ volume for any signs of a stronger-than-expected recovery, but given the non-confirmation of the NYSE A/D line and the McClellan Oscillator on the downside last week, it now looks like that breadth should continue to hold up in the coming weeks. Time will tell, but this author will now look for any weakness in the Dow Industrials to shift to a 100% long position in our DJIA Timing System sometime this week (and possibly even tomorrow).
Our stance doesn't change this week as last week's bounce is most likely a precursor for a more sustainable rally in the next six to eight weeks. For now, we will continue to maintain a 100% long position in our DJIA Timing System. However, since no indicator is infallible, we will also put in a stop at our average entry point of 10,805.
I will now end this commentary with a quick discussion of our popular sentiment indicators those being the bulls-bears percentages of the American Association of Individual Investors (AAII), the Investors Intelligence, and the Market Vane's Bullish Consensus Surveys. The latest four-week moving average of these sentiment indicators nearly touched the 0% level last week declining from 7.7% to 2.7% in the latest week a low not seen since April 2003! (Editor's note: I apologize for an error in this reading in our previous commentary, as last week's reading was supposed to be 7.7%, not 4.0%). This reading has now convincingly surpassed the oversold levels that are consistent with short-term bottoms in the markets over the last few years. Following is a weekly chart showing the four-week moving average of the Market Vane, AAII, and the Investors Intelligence Survey Bulls-Bears% Differentials from January 1997 to the present week:
Conclusion: Based on the readings of the NYSE ARMS Index and based on the current sentiment backdrop, chances are good that last week's action resulted in a significant low in the stock market a low which should provide a basis for a rally lasting at least six to eight weeks. As for the Fed, this author maintains that the Fed will hike one more time on June 29th, and that will be it for this cycle. The combination of higher borrowing costs and additional supplies coming into the markets should put a lid on crude oil prices which is all the more probable given that the chances of a significant disruption in crude oil and natural gas production during this hurricane season is highly unlikely.
Henry K. To, CFA