It was a week for the history books, and we were glad it finally ended. We have been working at S&P since August, 1987, and the drama this week ranks right up there with the crash of '87, even though the market did not crash.
It sure fell like it was about to. Panic was evident in many of the technical indicators we monitor, as it seemed like the entire financial system was on the brink of collapse. But as has happened in the past, the government rode in on a white horse and rescued the financial markets from the brink of disaster. Never underestimate what our elected officials will do to avert a catastrophe.
Early in the week, the market had a very bad hangover and proceeded to cleanse itself, violently, on Wednesday. That seemed to do the trick, as the market rebounded on an intraday basis Thursday and, along with historic government actions, finished with nice gains on Friday.
We think evidence of a climax bottom last week, and an initial low for the bear market, were overwhelming. When we look for signs of panic, we look at price velocity to the downside, overall trading volume levels, the number of new 52-week lows, put/call ratios, and the volatility indexes. During panics, or capitulations, prices move straight down. This is just the opposite of a blowoff top, when prices basically move straight up.
Interestingly, when the slope gets very steep, either up or down, many times it signals a longer-term turning point for the markets, as the moves are simply unsustainable. During the 3-day period ending on Wednesday, September 17, the S&P 500 had plunged 7.6%. This was the worst three-day performance for the "500" since August 5, 2002, which was right near the bear market bottom.
This type of price velocity to the downside has many times marked a key trough for the stock market. On a 3-day basis, the "500" fell 9.5% on July 23, 2002, right at the initial bottom for the last bear market. The 9.6% decline ending on April 14, 2000 represented an intermediate-term low within the confines of a bear market, but the rally was still very good. The index fell 11.7% on a 3-day basis into August 31, 1998, and that turned out to be the initial bottom of that correction. We saw a 9.5% 3-day plunge in October 1997, another nice time to be aggressive towards the market. We have to go back 10 years from there, when we witnessed a 26% wipeout during the crash in '87.
Just as significant, the 7-day performance by the S&P 500 (ending 9/17/08) was abysmal, falling 8.8%. When the index has fallen by this amount or more in the past, this has generally been either a good long-term time to move back into the market, or marked a good time to invest in a nice counter trend move to the upside. The last time we saw this type of price damage on a 7-day basis was in September 2002, right before the bear market bottom. We saw this kind of wipeout in July 2002, three times in 2001, twice in 1998, once in 1990, a couple times in 1987, and twice in 1974. Interestingly, many of these occurred during the 4-year cycle lows, and many happened in the fall months when the market so many times has bottomed.
NYSE 52-week lows surged last week, another indication of panic, as institutions and investors were throwing equities out to raise cash. In the process, funds moved to the safety of Treasuries, particularly short-dated ones. Two-year treasury yields fell from 2.21% on September 12 to 1.62% on September 17, or 59 basis points. It was the largest three-day drop in yields since September 2001, and the second biggest drop going all the way back to 1992.
Three-month treasury yields were reported to have fallen below 0%, which we think is a pretty odd concept to grasp.
As funds were piling into Treasuries, money was coming out of stocks in droves. NYSE 52-week new lows as a percentage of issues traded hit 39.1% on Tuesday, 37.4% on Wednesday, and 33.6% on Thursday. According to Jason Goepfert at sentimentrader.com, we have not seen three straight days of 30%-plus new lows since September 1974. The 39.1% posted on Tuesday was exceeded slightly on June 15, 2008 at 39.5%. That day represented the highest percentage of lows since the crash in 1987.
Volatility indexes spiked to extreme levels last week, another sign of capitulation. The VIX, or CBOE volatility index, based on option prices of the S&P 500, soared to 42.16% on an intraday basis Thursday, the highest and most fearful reading since October 8, 2002, when the VIX hit 43.4% intraday. Readings at least as high as Thursday's over recent history have marked excellent times to jump back into the stock market. For instance, we saw a 45.2% reading on August 2, 2002, 48.5% on July 24, 2002, 49.3% on September 21, 2001, 49.5% on October 8, 1998, 48.1% on September 11, 1998, and 48.6% on October 28, 1997. All these dates corresponded with a good time to get back into equities.
Due to the huge price thrust, current chart construction, and sentiment data, we do not think the closing low this week will be tested. It appears that the "500" is working on an inverse head-and-shoulder, which would translate into further strength to perhaps the 1300 area, followed by a 3/8 to 5/8 retracement of the rally. There is a confluence of resistance in the 1265 to 1305 region that we believe will act as a strong barrier to this initial rally. To confirm the reversal pattern, we need to see a strong breakout above 1305. In addition, we would like to see the 17-week exponential cross back above the 43-week average, but we think that could take a few months.
At least for the next couple of weekends, I do not think we will have to monitor financial websites looking for the next problem to happen.