My negative opinion on the market - which I have had since the beginning of the year - was strongly reinforced by the dramatic sell-off on February 27, in which the S&P 500 fell 50 points, or 3.5%, and 498 out of the 500 stocks in the index were down on heavy volume. With all the problems in housing, with the collapsing submergingprime market, with the overvaluation of commercial real estate, with all the risk being squeezed out of virtually every asset class on the planet, with the manic stock market behavior in China, with central banks tightening credit around the world, with record high corporate profit margins, with the ridiculous deals being done by private equity, and with margin debt hitting record highs, I figured I would step aside at the beginning of the year and wait for the market to tell me that the party was officially over. And I figured that since the market had been rising on declining volume after marching relentlessly up since June, the heavy volume sell-off that occurred at the end of February and into March was, indeed, that very tell.
But perhaps February 27 wasn't a tell. Perhaps it was a routine, albeit compressed sell-off that has been a common occurrence in this cyclical bull market. And instead of dragging out over several weeks, maybe the market was being courteous enough to not only scare a whole lot of people, but also shorten the time in which we would have to experience the indignity a down market. How polite of the market!
So I decided to deconstruct this cyclical bull. The absolute low occurred on October 10, 2002 but was tested before finally taking off. The final low occurred on March 10, 2003, and from that point is where we will begin.
I wanted to understand the characteristics of each significant move up and down. (The date is followed by the return and the average daily volume in billions of shares. All figures are for the S&P 500.)
Mar 10/03 - Mar 5/04, return 47.7%, ADV 1.25 billion
Aug 14/04 - Mar 7/05, return 14.0%, ADV 1.20 billion
Apr 21/05 - Aug 3/05, return 9.5%, ADV 1.21 billion
Oct 14/05 - May 8/06, return 13.6%, ADV 1.43 billion
June 15/06 - Feb 22/07, return 19.9%, ADV 1.26 billion.
Average return 20.9%, ADV 1.27 billion.
Thus, the most recent run-up ending last month looks similar to the previous four. In fact, it looks downright average.
Mar 6/04 - Aug 13/04, return -7.5%, ADV 1.09 billion
Mar 8/05 - Apr 20/05, return -7.1%, ADV 1.36 billion
Aug 4/05 - Oct 13/05, return -6.2%, ADV 1.23 billion
May 9/06 - June 14/06, return -8.1%, ADV 1.65 billion
Feb 23/07 - Mar 14/07, return -6.7%, ADV 1.45 billion
Average return -7.1%, ADV 1.36 billion
Thus, it appears the most recent sell-off - if it did end on March 14 - was about average on both return and volume, though duration was the shortest of the five downward phases.
Maybe I was overreacting.
Since everything appeared so average, I wondered if the behavior of the market was similar at the breaks of each phase, i.e. around each inflection point at the break, did volume tail off before the sell-off, and was the sell-off heavy? If that was the case, then the most recent sell-off was going according to script.
We can see this graphically. Each graph details the market and volume a few months before the break and in the month following. I put a space between the high and the market break for clarity.
As we can see, volume began to decline at the beginning of February, which is what you would expect before a downdraft. Volume did pick up after the break in March, and was generally heavier than in February but lower than January.
During the second break, volume may have been lower in February than in January, but only slightly. You really couldn't foresee a break coming based on volume. After the break, though more volatile, daily volume was about average compared to the previous two months.
Volume was perhaps a little stronger in June than in July, but not appreciably, especially considering that July is summer. August was even lighter, but Wall Street goes on vacation in August.
Volume is heavier after the fourth break in May, but volume actually picked up in the second half of April and the first half of May compared to the prior six weeks, and is about the same as the volume after the top in May. It doesn't look like investors were leaving the market beforehand.
Finally, the action this year certainly looks like classic topping behavior. Volume is clearly declining as the market is rising at the beginning of the year. Then, when the market breaks, it does so on big volume. And as shown a few days ago, volume has been much higher on down days than on up days.
I think it is fairly safe to conclude that based on price action and volume, only the last of the five phases looks like classic topping action.
Wednesday, the market was flat and volume on the Big Board was 1.55 billion, the same amount as Tuesday. However, on Tuesday about 800 million shares traded in the last 1 3/4 hours of the trading day. Though the market was up strong, all the gains occurred by 3pm, even though volume continued to rise methodically into the close, which may mean that selling pressure equaled buying pressure, a conclusion that may have been reinforced by Wednesday's action.
Wednesday was neither bullish nor bearish. A truly bullish day would have been a rising market on rising volume. In fact, volume on both days was below average since February 27. It wasn't outright bearish either, though, because the market held. A very bearish day would have taken all Tuesday's gains back and more on heavier volume. That didn't happen.
However, nothing has changed my mind. I believe we are currently at the beginning of a topping process. We may go higher, I don't know, and I will always change my mind when the facts change. But as for now, I remain cautious.