In a US stock market-related Commentary on August 16th, we said:

As of the close Wednesday, the US stock market was positioned right on top of major support while amid ideal technical / behavioral conditions to resume its 2002 uptrend.  However, last night's breakdown in the Nikkei 225 Japanese stock market suggests that long term US interest rates are likely to continue declining in the weeks ahead.  If they do, so should the US stock market, which has been tightly correlated to the yield of the US 10-Year Note since the end of last year.

Since then, the benchmark S&P 500 and DJIA rallied sharply into late August, but has essentially been drifting sideways since then, indicating indecision.  At the same time, the tight eight-month correlation between US stocks and interest rates has completely disappeared.  In today's Commentary, we take an updated look at many of the same charts and market relationships that we discussed on August 16th, and provide an updated analysis on what these indicators are currently saying about US stock market direction through the end of 2007.

When a financial market is at an important inflection point, a strong technical argument can usually be made either way while the market is trying to make up its "mind".  The following paragraphs first take a look at the bullish and bearish arguments separately, and then attempt to logically fit them all together into one comprehensive outlook.

The Bullish Argument

2002 Broad Market Uptrend Is Still Intact:  Our first chart (below), which displays the S&P 500 (SPX) since 2006, deals strictly with trend, and measures it in a few different ways.  The SPX moved appreciably below its (orange) 200-day moving average,  a widely-watched major trend proxy, in mid-August.  At the time, this suggested that the 2002 cyclical uptrend in the US stock market may be over.  However, the green March 2003 uptrend line was not broken on the mid-August decline. 

Since then, the the SPX has been essentially been trading on both sides of its 200-day moving average, indicating that there is still indecision as to whether the 2002 uptrend will continue from here.  However, the index's successful test of its 2003 uptrend line indicates that this trend is still intact.

The blue highlights point out that the decline from the July 16th peak at 1,556 retraced between 50.0% and 61.8% percent of the SPX's larger June 2006 to July 2007 advance. According to retracement theory, this is where the June 2006 advance should resume if it is still valid, so the mid August rebound is seen as more evidence that the larger uptrend in the SPX is still valid.  The red highlights, however, point out that the rise from the August 16th low has thus far only retraced around 61.8% of the larger July 16th decline.  Thus, according to the same retracement theory, the 1,484 area ( which is key near term overhead resistance) would have to be appreciably broken to confirm that the larger 2002 uptrend is indeed resuming. 

The combined message is that the 2002 US broad market uptrend is still intact, but a rise above the contract's recent early September highs would be necessary to confirm that it has resumed.

Investor Sentiment Suggests The August Rally Is About Half Over:  The next chart (below) displays the S&P 500 in the upper panel, and a daily, survey-based measure of retail investor sentiment in the lower panel.  The gray highlight points out that these retail investors are currently coming off of a multi-year "least bullish" extreme (of less than 22% bullish), reached in mid August, which had previously coincided with the end of literally every corrective decline in the SPX since the 2002 uptrend began. 

The chart also shows that the indicator in the lower panel is still only about halfway to moving back to previous "most bullish" extremes (of 83% bullish or greater), which had coincided with most every near term top during this period.  Therefore, these data suggest that the August rally is only about halfway completed.

We also note that the the directional implications of this chart are very similar to those of several other retail investor oriented investor sentiment gauges that we monitor.  Together, they make a very strong case that the August rally in the S&P 500 should continue through at least the end of Q3.

The US Broad Market Is Still Oversold:  Our next chart (below) displays the percentage of NYSE stocks trading above their 200-day moving average during the past decade in the upper panel, and the S&P 500 in the lower panel.  The green highlights point out that "washed out" periods in the market, when just 33% or less of these stocks were trading above their 200-day moving averages, coincided with most every near term bottom in the SPX during this entire 10-year period. 

The rightmost green highlight points out that, with the indicator in the upper panel at 37% through September 11th and coming off of a low extreme of 27% on August 16th, the US broad market is amid favorable conditions for another near term rally now.  We also note that similar oversold extremes coincided with at least a several week rally during both uptrends and downtrends alike as, even in February 2000 while the US stock market market was peaking, the SPX still rallied for several weeks before finally setting a top in late March.

Section Summary: The first chart indicates that the 2002 broad market uptrend in the US stock market is still intact.  The second and third charts indicate that the market is at multi-year extremes in both bearish retail investor sentiment and in the small number of NYSE stocks currently trading above their 200-day moving average, both which have historically  coincided with at least a several week rally in the SPX.  These charts not only suggest that the mid-August rally in the US stock market is likely to extend through at least the end of the quarter, but also suggests that, in the event that the market turns lower again from its current level, its downside is probably limited until these too-bearish, oversold extremes are worked off.

The Bearish Argument

Recent Decline In Japanese Stock Market Should (Eventually) Continue:  The next chart (below) displays the Nikkei 255 Index (black bars) and the S&P 500 (blue line) since 2000.  These two series have been very tightly correlated (92%) since October 1998, which was an important bottom in the Asian stock market.  The uppermost red highlights on the chart point out that the Nikkei 255 has recently (in mid August) confirmed a bearish chart pattern, a double top, that targets a decline to at least 1,475. 

Thus, assuming that the nine-year tight correlation between US and Japanese stocks remains intact, and also that the minimum downside target of this pattern will eventually be met, this chart suggests that more weakness is also eventually in store for the S&P 500.  (This bearish pattern in the Nikkei was one of two primary reasons why we thought US stocks could continue lower a month ago.)  

Technology, Small Cap Poised For Relative Underperformance:  Our next chart displays the relative performance of the NASDAQ 100 (NDX, blue line) and Russell 2000 (RUT, green line) versus the S&P 500 in the upper panel, and the SPX itself in the lower panel.  The 2002 broad market uptrend in the US stock market was led by technology stocks and, to a lesser degree, by small cap stocks. Accordingly, it stands to reason that there should be a broad market decline when these two areas of the market are underperforming the S&P 500.

The red highlights point out that this has indeed been the case as the coincident periods of relative underperformance by both the NDX and the RUT between January and August 2004, December 2004 and April 2005, and April and July 2006 coincided with three of the most significant countertrend corrections in the SPX since its the 2002 uptrend began.

Our next chart (below) displays the relative performance of the NASDAQ 100 versus the S&P 500 in the upper panel (blue line), and quarterly momentum (according to the RSI, red line) on that relative performance line in the lower panel.  The gray vertical bars on the chart point out that the blue relative strength line in the upper panel is at a multi-year overbought extreme that had coincided with or led every multi-month period of relative underperformance by the NDX (technology stocks) in the past four years. 

In addition, the red highlights in the upper panel point out that this overbought extreme is being reached just as the relative strength line is bumping up against a four-year extreme in relative outperformance, which we view as overhead resistance.  Combined, these factors suggest that another similar period of relative underperformance by technology stocks is on the horizon, probably within the next few months.

Our next chart (below) displays the relative performance of the Russell 2000 versus the S&P 500 in the upper panel (green line), and the S&P 500 in the lower panel (black bars).  The chart points out a few important things.  First, the uppermost red highlights show that the 1994 extreme in relative outperformance by RUT was tested and held in April 2006, and that small cap stocks have generally underperformed since. 

Next, the blue and lowermost red highlights show that the Russell's December 2000 trend of relative outperformance versus the S&P 500 was broken in August 2006.  More recently, after retesting this trend line as overhead resistance, the green relative performance line has just set a new "low" (a new extreme in relative underperformance).  This entire process is a textbook example of what typically takes place during a trend change, which in this case means a change to a new trend of relative underperformance from one of outperformance.

Section Summary:  The first chart implies that the recent decline in the Nikkei 225 Index is not over and, by way of the tight correlation between the Japanese and US stock markets, suggests that further weakness should also eventually be in store for the S&P 500.  The following three charts show that the S&P 500 has historically declined when both the technology and small cap stocks were underperforming at the same time.  Both technology and small cap are amid favorable technical conditions to underperform over the next several months.

What Has Changed From A Month Ago

Back in mid August, our expectations for a continuation of  the recent decline in US interest rates was projected to put downward pressure on US stock prices because, up until that point, the yield of the US 10-Year Note had been tightly (76% positively) correlated to the S&P 500 since December (green highlights on our final chart, below).  We think the cause of this correlation was, once Treasury yields bottomed in December, the stock market had been depending on the bond market as an indication of economic strength or weakness (and, as such, a reason to either buy or sell stocks). 

However, since August 17th, which is the day that the Fed cut the discount rate by 50 bps to 4.75%, the correlation between these two series has completely disappeared (blue highlights).  This strongly suggests that "opening the discount window" has (at least thus far) succeeded in restoring some confidence in the Fed's ability to keep the economy afloat, despite the housing / sub prime crisis.

So, although stocks and bond yields could very well come back into correlation with one another at some point in the future, at least for the time being it looks like we can no longer attempt to anticipate the direction of the US stock market based on the direction of benchmark US interest rates.

Making It All Fit Together

The charts above make a strong case for both a rally and a decline in the US stock market as we head into the end of 2007.  However, we think that this discrepancy in technical signals is probably more of a time frame issue, rather than one set of indicators being "right" and another being "wrong".

The first series of charts not only indicates that the US broad market is resting on major support, but more importantly also suggests that there is probably a near term floor under US stock prices right now.  This is because recent history tells us that market declines during periods when; 1) retail investors were this bearish on stock prices, and 2) when so few NYSE stocks were trading above their 200-day moving averages, were very infrequent, and were short lived when they did occur.  This leads us to believe that the August rally is likely to extend through the end of Q3, and perhaps into early Q4.

The second series of charts displays an unmet downside target in the Nikkei 225, plus what appears to be impending relative underperformance in both technology and small cap stocks.  These conditions have both been historically bearish for the US stock market.  However, these types of indicators have historically been a much less immediate indication of market direction than those that appear on our first set of charts.  As such, their bearish implications could very well "kick in" later on, perhaps sometime in Q4, after the extreme conditions indicated in our first set of charts is worked off by rising prices.