Time to Stay on the Sidelines

Key indicators are very close to issuing major long-term sell signals

By Mark Arbeter

The stock market accelerated to the downside early last week, moving very close to key intermediate-term support before stabilizing towards the end of the week. From a shorter-term perspective, the market is extremely oversold on many measures, including price, volume and internal measurements, and we believe a rally could occur at any time. However, we think there is little evidence that a substantial low has been put in for the intermediate term, so we would remain safely on the sidelines.

The major indexes have suffered some major technical damage and have moved quickly away from their recent cyclical highs in August. At its intraday low on Thursday, Oct. 13, of 1168.20, the S&P 500 index had dropped 6.2% from its closing high of 1245.04 on Aug. 3. The Nasdaq composite index, at its low on Thursday of 2025.58, had fallen 192.57 points or 8.7% from its closing high of 2218.15 on Aug. 2. The Philadelphia Semiconductor Index (SOX.X ) was 10.9% off its August highs, while the Oil Service Index (OSX.X ) was down almost 15% during the week and the Housing Index (HGX.X ) was off over 100 points or 17.3% from its late July closing high.

The carnage has not missed the small cap arena, as the Russell 2000 was down as much as 10.7% from its early August closing high. The weakness in the market has come primarily from leadership areas, not a good sign in our view.

Our technical work indicates that a bear market began earlier this year, which has been a very slowly evolving process, and that many indexes are now in their initial downward phase. Historically, market tops have taken time to play out. All stocks, industries and indexes typically do not peak at the same time. As you will note, many of the above mentioned indexes peaked during the summer months. However, the Dow Jones Industrials and the Dow Jones Transportation indexes topped out way back in March, while the Philadelphia Bank Index peaked all the way back in December, 2004. These nonconfirmations, where some indexes peak earlier than others, are part of the process of a bull market top and represent one of the many characteristics of a late-stage bull market.

There has been quite a bit of internal deterioration during the last year and it is quite common that many technical indicators crest well before the major indexes start to decline, sometimes giving lead times of 12 to 18 months. For instance, the annual rate-of-change or ROC on the S&P 500 during the cyclical bull market posted its maximum level of 36% in February, 2004. Since that time, the S&P 500 has put in higher price highs while the annual ROC has failed to come close to the February, 2004, peak. This long-term indicator is one of the many that has traced out a clear negative divergence with respect to price strength.

The percent of NYSE stocks above their respective 30-week moving averages, which is another measure of price momentum, climaxed in January, 2004. This indicator has also put in a series of lower highs since that point, despite new price highs by the index. This is another example of a longer-term negative divergence. The percentage of NYSE stocks at 52-week highs hit its crest of 18.2% way back in December, 2003, indicating fewer and fewer stocks are benefiting from the indexes new highs over the past two years.

While we will admit that the transition from a bull market to a bear market has been difficult to call on an historic basis, other long-term indicators are very close to issuing major long-term sell signals. Because these signals are very long term in nature, they occur very infrequently, but are extremely important from our perspective. The monthly moving average convergence/divergence (MACD) based on the price action of the S&P 500 is very close to crossing its signal line and therefore, near a major sell signal. The monthly MACD has been on a buy signal since May, 2003. The last sell signal from this momentum indicator occurred in February, 2000, near the top of the bull market.

The monthly stochastic oscillator is also very close to going bearish after getting extremely overbought earlier this year. Like the MACD, the monthly stochastic oscillator has been on a buy signal since early in 2003. The last sell signal from this indicator was late in 2000. To confirm the signals given by these two long-term momentum indicators, we combine them with the super long-term exponential moving average of 20 months. Like the above indicators, very few signals are generated from this moving average. Currently, the 20-month exponential moving average is sitting at 1163, not far from the price of the S&P 500. A break below this key long-term average would be the first sell signal from this average since November, 2000. The S&P 500 has been trading above the 20-month since July, 2003.

Bond yields moved higher for the third straight week, with the 10-year Treasury moving to 4.53% on an intraday basis Friday, its highest yield since early April. The 10-year treasury broke above the August highs this week, and since early June, has traced out a series of higher highs and higher lows. The next area of support for the 10-year is at 4.55%. This is from a trendline drawn off the yield highs from May, 2004, and March, 2005. There is also chart support, from the highs in March, in the 4.5% to 4.7% range.

Short-term interest rates also continued higher, indicating to us that the Federal Reserve is likely to continue to raise rates into at least early next year. The 1-year Treasury bill has moved above 4%, and has risen steadily since March 2004. The yield on the 1-year is at its highest rate since April, 2001. In our view, there is added pressure on the stock market with both short and long-term rates finally rising together.

Arbeter, a chartered market technician, is chief technical strategist for Standard & Poor's

Before it's here, it's on the Bloomberg Terminal.