The good news is:
• Last weeks rally was strong enough to turn some of the intermediateand short term indicators upward.
The NASDAQ new high indicator ( OTC NH) is a 10% trend (19 day EMA) of NASDAQ new highs. Its direction, not level is a good short term indicator. The chart below shows the NASDAQ composite (OTC) in red and OTC NH in green, dashed vertical lines are drawn on the first trading day of each month. OTC NH turned sharply upward as of Tuesday's close, however, it went flat as of Friday's close following two consecutive down days for the OTC. The current value of the indicator is 76 so more than 76 NASDAQ new highs will keep the indicator moving upward.
Summation indices (SI) are running totals of oscillator values.
The chart below shows the OTC in red, OTC AD SI calculated from NASDAQ advancing issues - declining issues, OTC HL SI calculated from NASDAQ new highs - new lows and OTC UD SI calculated from NASDAQ upside - downside volume. When all of the SI's are headed in the same direction it is imprudent to bet against them. Last week OTC HL SI turned upward while the others went flat.
The previous two charts hint that last weeks rally could be the beginning of something bigger.
The next chart casts a little doubt on that hint. The chart covers this year showing the Russell 2000 (R2K) in red and in indicator constructed by subtracting the percentage of the component issues of the R2K making new lows from those making new highs. A 20% trend (9 day EMA) is applied to the result. For this chart, new highs and new lows were calculated over the previous 6 weeks rather than 52 weeks as reported by the exchanges.
In the previous two index lows (late January and late April) the indicator made a higher low at the price low.
If the pattern holds, we should see a lower index low along with a higher low in the indicator.
September is widely understood to be the worst month of the year and by some measures it has been. Measured by the OTC, September is the only month that has had an average (mean) negative return (-0.3%), however, since 1963 the OTC has been up during 60% of Septembers.
As measured by the percentage of the months that have been up September beats July - 51%, February - May - June - 56% each, March - 58% and October 59% putting it firmly mid pack. The problem is that when September is bad it has been really bad.
Breaking OTC performance down further, to Septembers in the 1 st year of the Presidential cycle, the results are even better. During the 1 st year of the Presidential cycle Septembers have been up 70% of the time, but the average return has been a loss of 0.9%. Septembers 70% winning percentage has only been surpassed by April and July at 73% each and, measured by average return, September is ahead of March at -1.9% and June at -1.5%.
When measured by the S&P 500 (SPX) from 1928 it is easy to see where September got its reputation as the worst month of the year. For all years, September is the worst of 3 that have averaged negative returns at -1.2%. The other two months with negative returns, February and May have had average returns of -0.2%. September is the only month that has been up less than half of the time at 43%, the next worst are February and November, both up 54% of the time.
Breaking SPX performance down further to the 1 st year of the Presidential cycle September has been much worse, up only 26% of the time with an average loss of 2.5%. Three other months have up less than 50% of the time, February (40% avg. rtn. -1.7%), June (42% avg. rtn. +0.8%) and April (45% avg. rtn. +2.5%)
The table below shows September returns during the 1 st year of the Presidential cycle for the SPX from 1929 and OTC from 1965.
| Septembers in Presidential Year 1 | |||
| SPX | OTC | ||
| 1929-1 | -5.2% | ||
| 1933-1 | -12.5% | ||
| 1937-1 | -11.9% | ||
| 1941-1 | -1.4% | ||
| 1945-1 | 4.2% | ||
| Avg | -5.34% | (5 period) | |
| 1949-1 | 1.8% | ||
| 1953-1 | -0.3% | ||
| 1957-1 | -6.6% | ||
| 1961-1 | -2.1% | ||
| 1965-1 | 3.2% | 4.2% | |
| Avg | -0.82% | (5 period) | |
| 1969-1 | -2.5% | 3.2% | |
| 1973-1 | 3.8% | 5.4% | |
| 1977-1 | -0.3% | 0.4% | |
| 1981-1 | -5.6% | -7.8% | |
| 1985-1 | -3.1% | -5.5% | |
| Avg | -1.55% | -0.86% | (5 period) |
| 1989-1 | -1.3% | 0.3% | |
| 1993-1 | -0.9% | 2.2% | |
| 1997-1 | 2.1% | 4.2% | |
| 2001-1 | -8.1% | -15.4% | |
| Avg | -2.05% | -2.15% | (4 period) |
| Pres Yr 1 summary | |||
| Averages | -2.5% | -0.9% | |
| Win% | 26% | 70% | |
| All years summary | |||
| Averages | -1.2% | -0.3% | |
| Win% | 43% | 60% | |
| SPX | OTC | ||
The yellow line in the chart below shows the average of all Septembers for the OTC from 1963, the white line shows the average September during the 1 st year of the presidential cycle during the same period.
The chart is constructed by averaging the returns of the 1 st 11 days of the month and the last 10 days of the month.
The OTC loss in 2001 was larger than all other September losses during the 1 st year of the presidential cycle combined.
The following chart is similar to the one above except year 2001 has been removed.
Voila, September becomes a positive month during the 1 st year of the presidential cycle.
The next chart is similar to the two above except it has been constructed using SPX data from 1928 - 2004. The average of all years is in maroon and the 1 st year of the Presidential cycle is in white.
This year has been a little weak as presidential 1 st years go so I think it is likely this September will be down.
During down markets, sharp rallies, like we had last week, often occur during seasonally strong periods. The seasonally strong period is over so last week's rally is likely to be at or near its end.
I expect the major indices to be lower on Friday September 9 than they were on Friday September 2.
This report is free to anyone who wants it, so please tell your friends. If it is not for you, reply with REMOVE in the subject line.
