Tuesday, November 26, 2013

Every Zig Has a Zag

With SPX approaching 1800, we suggested a reaction of 4-10% was probable, allowing for an overshoot to 1820-30. This was the historical precedent at prior round numbers and it was moreover supported by other market data (post here).

Further supporting evidence is in the latest Weekly Market Summary (post here).

This post looks at the length and extent of the current rally relative to others over the past 20 years, a period which includes the late 1990's tech bubble, i.e., the strongest rally ever seen in the market.

From the November 2012 low a year ago, SPX has risen 34%. A long, strong rally, but one which is not unprecedented.

The chart below uses 1-year log-scale boxes with a 34% rise in price. In the past 20 years, there have been six other comparable rallies, four of which were in the late 1990s.



We have not included rallies off of a significant low (1994, 2003, 2009) as the start of a bull market is expected to be long and powerful.

After each rally, SPX has corrected 13% to 23%, typically over the next one to three quarters.

There is one exception, and that is the rally in 1999. It rose 50%. It is different in that the start of the rally in 1998 came after the largest correction shown: 23%. From the 1998 peak to the 1999 peak, it rose 19%; in comparison, the current rally from the October 2012 peak has risen 26%. It was nonetheless followed by a 13% correction.

We see the same pattern in the small cap index, RUT. Four prior rallies were similar in either/both duration or gain in the past 20 years to the current rally. RUT is a higher beta index. The reactions after each rally have been much larger than those in SPX. In each case, all of the gains were given up in the following year.



Goldman has looked at the current rally and formed a similar conclusion: a reaction of 10% or more (to SPX 1600-1650) is likely after a 25% rise (read here). 

In summary, there are several conclusions to note:

First, the current rally is not an anomaly. Similar rallies have occurred in the past. The notion that the current rally is an unprecedented Fed/QE driven event is unsupported by the facts.

Second, the current rally is not necessarily the end of the bull market. In fact, the odds are it is not at all. But it is likely near a cyclical end (as opposed to a secular end). This is the normal pulse and rhythm of the market. Every zig has a zag.

Third, this is not a call to go to 100% cash, hide in a bunker, expect a crash or exclaim "bubble!" SPX hasn't even closed below its 13-ema since October. The trend is up. But after a long and powerful rally, investors are exuberant (more here and here). Note where SPX is relative to other similar rallies and be alert to a change when (if) it comes.

Finally, the rally in the 1990s was also, at the time, regarded as Fed induced and prolonged. You can go back 90 years and find similar sentiment towards the Fed as today during any bull market. This is not new. And the current rally is not unprecedented, even for the paltry earnings growth over the past two years. But, should this rally continue uncorrected to 1900 and higher without a remarkable increase in earnings, Fed critics will probably have their strongest evidence in history.