LONGER TERM INDICATORS SHOW THAT THE BULL MARKET MAY BE NEARING COMPLETION -- SEASONAL TRENDS HAVE TURNED NEGATIVE -- WHY THE FOUR-YEAR CYCLE HAS ME WORRIED
SHORT-TERM SLIPPAGE... Wednesday's price action pushed the S&P 500 below its 50-day moving average on a closing basis for the first time since mid-May and confirmed recent signs of weakening in its short-term trend. The SPY also closed back below its March peak at 122. The fact that yesterday's selloff occurred on heavy volume is another negative factor. Those earlier signs of weakness were seen in the Rate of Change (ROC) indicator trading under the zero line and negative MACD lines. The market is attempting to bounce back today from a short-term oversold condition. But given the fact that the cyclical bull market that started nearly three years ago is very old by historical standards, with the market having entered a dangerous seasonal period, and with crude oil reaching new records, this is an opportune time to take a longer-range view of the market.
Chart 1
MAJOR TREND STILL UP, BUT LOSING MOMENTUM... The most important line on the S&P 500 weekly bars in Chart 2 is the red line. It's the 80-week moving average. Normally, I'd use a 40-week average which is used by most chartists as the main gauge of the market's major trend. However, the 40-week has given some false signals over the years. I found that by doubling it to 80 I get much more reliable trend signals. The S&P rose above the 80-week line in the spring of 2003 and remains above it. The spring selloffs in 2004 and 2005 stayed above that long-term support line. As long as the S&P stays over that line, the major trend will remain up. That's an important point. The major trend is still up. All of the indicators that I'm going to focus on in this report are anticipatory in nature. In other words, they measure the direction of market momentum. Historically, these indicators give early warnings that a major trend is gaining or losing momentum. Virtually all of those weekly indicators are giving the same market message -- namely, that the cyclical bull market is showing signs of tiring.
Chart 2
RSI IS WEAKENING ... The top line in Chart 2 is the 9-week RSI line. The two most important numbers on the RSI are 30 and 70. Readings under 30 measure an oversold market; over 70 means an overbought market. The two readings under 30 in 2002 gave an early warning that the market was bottoming. The RSI reached two overbought readings at the start of 2004 and end of 2004. But there's more to the RSI than just the numbers. There's the direction of the peaks and troughs in the RSI that lead to positive or negative divergences. Starting with the lowest low in the summer of 2002, each successive trough in the RSI in late 2002 and the spring of 2003 was higher than the one before. That set up a positive divergence with the price action and gave an early warning that the market was bottoming. Since the start of 2004, we've seen just the opposite. The RSI reached its highest level in early 2004. The two other RSI peaks since then (late 2004 and the summer of 2005) have been successively lower. That pattern of lower peaks has set up a negative divergence and suggests a market that's losing upside momentum. The Commodity Channel Index (CCI) at the bottom of the chart shows a slightly different picture. But the message is the same. Readings under -100 are oversold while readings above +100 are overbought. The spring 2003 low in the CCI (green arrow) was higher than the one formed in the summer of 2002. That signaled higher prices. The last move over 100 (this summer) is lower than the one in late 2004. That's potentially bearish. Notice also that the CCI line has slipped back under 100. That last two times it did that the market entered intermediate corrections (see circles).
WEEKLY MACD LINES ARE SHOWING NEGATIVE DIVERGENCE ... Chart 3 applies one of my favorite market indicators -- Moving Average Convergence Divergence (MACD)-- to the weekly S&P 500. The two MACD lines are a combination of exponentially smoothed moving averages. Buy and sell signals are given when the two lines cross. But, here again, there's more to them than that. The MACD lines also turn well ahead of the market and give us positive and negative divergences. In other words, the first buy or sell signal is often just a warning that the market is turning. It's usually the second or third signal that's the real thing. An initial buy signal was given in October 2002 (see green circles). A second buy was given in the spring of 2003. Notice that the second MACD buy was much higher than the first. That's positive divergence. Since the start of 2004, the MACD has given two sell signals -- one just as the 2004 correction was starting and the other near the start of 2005 as the market was weakening (red circles). At present, the MACD lines are weakening again. No actual sell signal has yet taken place, but it's getting close (red arrow). Notice also that the last two MACD peaks have been successively lower (see falling trendline). That's been happening as prices have hit new recovery highs during 2005. That's negative divergence. That's also why I would take any sell signal on the weekly MACD at this point in time as a very negative development.
Chart 3
WEEKLY MACD IS NEAR SELL SIGNAL ... Chart 4 gives a closer look at the weekly MACD negative divergence since the start of 2004, and shows just how close it is to an important sell signal. The last two moves to new highs by the S&P were accompanied by lower MACD readings (see red arrows). I've put questions marks on the last peak because no actual sell signal has yet been given. But it wouldn't take much to turn the weekly MACD lines negative. The green bars below the chart show the MACD histogram. Those bars plot the difference between the two MACD lines and show us whether the two lines are converging or diverging. Right now they're converging. That's an early warning of an possible impending sell signal which occurs when the histogram bars fall below zero. It hasn't happened yet. But it's getting very close. And the fact that September is historically the worst month of the year isn't very encouraging. [It also looks to me like the August 2004 to the August 2005 rally has taken place in five waves, but I'll leave that discussion for another time].
Chart 4
THE FOUR-YEAR CYCLE CARRIES A WARNING... While the monthly seasonal pattern between now and October is negative, there's a longer-range cycle pattern that has me even more concerned. I've written before about the four-year presidential cycle and the tendency of the market to form major bottoms every four years. The monthly bars in Chart 5 show how well that four-year cycle has worked. The green arrows show important bottoms in 2002, 1998, 1994, 1990, 1987, and 1982. The only year that didn't follow the four-year pattern was 1987. That bottom came a year late. The next major bottom is due in the autumn of 2006 (most bottoms have occurred during October). Unfortunately, the early part of the fourth year is usually down. The market may present a great buying opportunity in October 2006. It's the period between now and then that I'm worried about. One final thought. The October 2002 four-year bottom was lower than the 1998 bottom. So far, the 2005 top is lower than the 2000 top. If the market fails to reach its 2000 high, that will be the first time that the long-term secular trend of the market has shown "lower lows" and "lower highs" since the 1970s. Those analysts who suggest that the current situation is similar to the mid-cycle pauses that occurred in the mid-1990s or the mid-1980s are missing one crucial point. The market was in a secular bull market during those two decades. That's no longer the case.
Chart 5
SUMMING THINGS UP ... We have a nearly three-year cyclical bull market that's very old. Longer-term momentum indicators are weakening and are dangerously close to important sell signals. The market may have completed a five-wave advance. Monthly seasonal patterns have turned negative. The four-year cycle is entering its most dangerous phase between now and the second half of next year. Oil prices are hitting record highs. Retailers are weakening. And the housing industry may be peaking. Sounds to me like a good time to turn more defensive. No major sell signals have been given yet. But, all things considered, I'd be inclined to use any short-term August bounces to prepare for what could be a tough autumn.
