SEASONAL PATTERNS SHOW THAT SEPTEMBER IS A DANGEROUS MONTH -- PRESIDENTIAL CYCLE, HOWEVER, FAVORS STRONG 2008 FINISH -- CPC AND VIX LOOK TOO LOW TO SUPPORT A STRONG STOCK RALLY -- EMA LINES ARE STILL NEGATIVE
SEPTEMBER IS USUALLY WEAKEST MONTH ... Earlier in the week I wrote that September and October were two of the most dangerous months of the year. A number of readers asked me to elaborate on that. Most seasonal patterns are taken from the Stock Trader's Almanac which contains a wealth of information on seasonal and presidential cycles. What complicates normal seasonal patterns this year is that 2008 is also a presidential election year. Let's deal with the seasonal cycle first. According to the Almanac, September has been the weakest month of the year since 1950 with an average S&P 500 loss of -0.6. October is also a tricky month. It often starts off weak and ends strong. October has included some of the market's worst declines. The good news, however, is that October has often marked major bottoms. Charts 1 through 3 show three examples of that autumn pattern. That makes September and October dangerous months. We're now heading into that period. The good news is that November to January is usually the strongest period of the year. That's especially true during presidential election years.

Chart 1

Chart 2

Chart 3
PRESIDENTIAL CYCLE ... According to the Almanac, presidential election years are usually up years. Most of those gains come in the second half of the year. The last seven months have seen gains in 13 of the last 14 election years (the only exception being in 2000 when the election results were delayed). That would seem to bode well for this year ending strong. A study of presidential year patterns seems to show two distinct patterns. Several bottoms took place by mid-year (1996, 1988, 1984). Others took place during October (2004 and 1992). That doesn't help us determine whether this year's bottom is behind us or ahead of us. It does mean, however, that odds favor the market ending the year on a stronger note once it gets beyond the dangerous September/October period.

Chart 4

Chart 5

Chart 6

Chart 7

Chart 8
VIX AND CPC LOOK TOO LOW TO SUSTAIN A STOCK RALLY ... One of our readers asked me to revisit the VIX Index and the CBOE Put/call ratio. The messsage of both looks the same to me. Chart 9 shows the CBOE Volatility (VIX) Index having fallen back below 20 since mid-July. That decline has supported a modest stock market rally. The best market rallies usually take place when the market is falling from higher levels. At the moment, the VIX is close to its May low and is showing signs of stabilizing. The VIX appears too low to support a strong market rally. The same is true for the CBOE Put/Call Ratio in Chart 10. The CPC is currently in the lower end of its 2008 range. Previous market bounces (as in March) usually take place from much higher levels. The current low level of the CPC doesn't seem consistent with a strong market.

Chart 9

Chart 10
13 AND 34-PERIOD EMAS ARE STILL NEGATIVE ... One of the main values of moving average lines is that they are totally objective and remove emotion from determining market trends. That's why I'm ending this week's final message by looking at the 13- and 34-period exponentially smoothed moving averages on both daily and weekly charts to determine if anything changed with this week's volatile market action. Nothing did. Chart 11 plots the two daily EMA lines for the S&P 500 (not shown). The two daily lines in Chart 11 turned negative in early June. To signal a short-term uptrend, the faster blue line needs to cross over the slower red line. As of Friday's close, no upside crossing had taken place. The more important lines are the weeklies which are shown in Chart 12. They determine the market's longer-term trend. The weekly EMAs turned negative at the end of last year (for the first time since 2003) and are still negative. Although the distance between them has narrowed a bit since March (see lower line), they're not even close to turning bullish. Based on that moving average crossover system (which has a strong track record in calling market turns), no trend changes took place this week. Volume should return to normal after the Labor Day weekend which should give us a better read on the market's mood. A big question heading into September is whether the positive presidential election year cycle is able to overcome that month's normal negative seasonal pattern.

Chart 11

Chart 12