As the S&P 500 steadily creeps up toward the next major resistance line of 875 (representing the high from February and minor high from January), it also gets more and more overbought. As we have seen throughout this bear market, technical indicators have had no problem reaching extreme conditions and staying there for prolonged periods.
I like to follow the percentage of stocks over their 40-day moving average (T2108 in Worden's Telechart). It is now sitting right at 89.3%, close enough to 90% to compel me to check the historical statistics. Since 1986, there have only been 7 times that T2108 has crossed above 90%. All seven of these times were during 1991 and 2003. These years were major turn-around years for the stock market as the economy transformed from recession to recovery.
Since I am still leaning to the bearish side right now, I am not one to claim that a third trip above 90% will be a charm. As always, time will tell. Approaching resistance levels in an overbought condition right before earnings seems like an explosive combination - Wells Fargo's (WFC) earnings surprise notwithstanding. We also have the "sell in May" period rapidly approaching, but I am not as concerned about that marker. When I looked at the last 46 years of data, May turned out to be less lethal than the old adage suggests. Several months after May provide ample opportunity to exit the market at the May highs. (2008 was one of those rare years where the May high was never seen again for the remainder of the year!)
Assuming the market soon takes the tiny baby step needed to push T2108 over 90%, this signal will contradict the bearish one showing that the S&P 500 topped out the last two times T2108 stumbled over 80%. I have no way to resolve this conflict except to remain cautious: generally avoid establishing new long positions and sell existing longs into strength.
Also note that we are now on day number seven with T2108 above the overbought threshold of 70%. The average time spent above this level is 9 days and the median is between 4 and 5 days. 75% of the time, the T2108 has dipped back below 70% after 9 days. T2108 has remained over 70% for more than 50 days four times: 1987, 1991, 1997, and 2003. Clearly, we are not likely in an "average" situation here. But if the market were to snap back to historical patterns, it would suggest a large and rapid sell-off within a week. Given the current bullish enthusiasm in the market, I am guessing that traders and investors would happily greet such an event with fresh bids.
I do not expect a retest of the lows until the market awakens to the myth of this year's tales of "second half recovery." Given the bullish enthusiasm in place, this awakening will not happen until the data are staring everyone right in the face...right around September or so. For the few of us who remain stubbornly bearish, we experienced one of those "no kidding" moments when the minutes of the latest Federal Reserve meeting indicated a downward revision in GDP forecasts:
The staff's projections for real GDP in the second half of 2009 and in 2010 were revised down, with real GDP expected to flatten out gradually over the second half of this year and then to expand slowly next year as the stresses in financial markets ease, the effects of fiscal stimulus take hold, inventory adjustments are worked through, and the correction in housing activity comes to an end.
It seems to me that in public statements the Fed often forecasts what it HOPES will happen and not what is likely to happen. I suspect these latest revisions remain over-optimistic. For now, the stock market does not care either way. (See Bill Fleckenstein and Jon Markman for good descriptions of the unrepentant bear case).
Be careful out there!
Full disclosure: long SDS. For other disclaimers click here.