Testing the Bottom–technical analysis



In the post-WWII period, inventory-cycle bear markets have tended to last about a year.  More serious bear markets, like 1773-74, 1981-82 or the one we are in now, have tended to last about two years (although this is generalizing from a small number of instances).  In virtually every case, however, the bear market ending has followed a set formula, containing two elements.  

*Enough time must have passed that the worst of the downturn has already run its course and profit growth will resume in six months or so.  Market participants only recognize this in hindsight, however.  At the time,  investors act as if they have lost conviction that profits will ever recover.

*The market establishes a low, bounces up from the low point by perhaps 20%, but then returns after six weeks to three months to “test” the low at least once, before beginning a relatively steady climb upward.  It looks to me as if this “testing” process is going on now.

Fundamental analysts like me usually regard technical analysis as being something like fortune-telling, a dubious enterprise at best.   So, on the one hand, it shows how bad sentiment has become that our thoughts turn to the charts at all.  On the other, a market bottom is by and large a psychological phenomenon.  So it probably makes some sense to try to figure out what the mindset of the market is.  And, as I just mentioned above, for whatever reason, market bottoms seem to follow a clear pattern.  Here goes:



The index to watch in the US is the S&P 500.

Use daily charts.  The low to look for can be an intraday low, not necessarily a closing low, so use a chart form that shows daily high, low and close.

The rate of market decline usually accelerates as the low is being reached.  Many times, stocks that have been the star performers of the down market–that is, relatively resistant to decline–underperform during this period.  (Market veterans have traditionally talked about three phases of a bear market:  hope (or denial), boredom (steady drip, drip, drip of decline), and despair.  During this final phase, investors give up hope and lose control over the fears that have been plaguing them during the down market and discard even the crown jewels of their portfolios.

Volume may not matter.  Technicians ofter talk about a selling climax, which occurs on high volume.  This may be corroborating evidence, but I don’t think it’s necessary.

During the “test,”  the market may fall below the low previously established.  This is scary, because it suggests that the previous low won’t hold.  In the US, if the market rebounds, I’ve always thought of this as a good sign (in Asia, where investors “read” charts differently, breaking below the old low, even for a few seconds, is almost always a bad thing.)  


 We can mark the top of the market as being either in early July 2007 or in October.  There’s a difference of about nine points on the S&P.  July is when the sub-prime crisis first broke.  The intraday high on July 16, 2007 was 1555.90.  The intraday high on October 10th was 1565.42.

On November 21, 2008, the S&P hit an intraday low of 741.02, a drop of 52.7% from the October high and 52.4% from the July high.  The market rebounded to 934.73 on January 2, 2009, a gain of 27.4% from the November low.  It has since declined to 742.37 on February 23, before rising to 775 and closing at 773.14 yesterday.


On purely mechanical criteria of a bottom–the extraordinary depth of the market decline, the three months between lows, the 20%+ rebound off the initial low–what’s going on now seems to fit the bill.  It would have been a bit nicer if the S&P had dropped a few more points intraday last Monday, but you may not be able to have everything.

The BIG issue, as I see it, is time.  Has enough time passed since the highs for bear market psychology to have played itself out and for the economy to have begun to heal itself? If we take the October high as our starting point, the answer is probably no.  

But nothing is ever that easy.  Perhaps just coincidentally, Fed Chairman Bernanke said yesterday that he thinks the economy will decline for another six months but that the recession will be over before yearend.    If we were to say the economy would begin to rebound in October, and acknowledge that the market begins to discount this six months in advance, we should be looking for a low right about now.

 More important, I think, is the fact that this is the first twenty-first century recession and is being played according to somewhat different rules than last-century ones (more about this in subsequent posts).  In short, I think the pain of recession may be front-loaded, but the market doesn’t realize this and is overreacting negatively.  In addition, part of our economic problem is due, not to the price of credit, but the availability of credit.  We have little experience of the latter phenomenon, other than for a brief period in the 1981-82 recession.  At least then, the availability problem showed itself to be hugely damaging but much faster to fix than “normal” cyclical pressures.


It’s possible that the bear market is making its ultimate lows.  I wouldn’t bet the farm that this is the case, but I would watch carefully for three things:

*anecdotal evidence that things at least aren’t getting any worse–I think we’re hearing this already

*a pattern of higher highs and higher lows in the market

*the gradual dissipation of bear psychology, i.e. strong negative reaction to bad news, no reaction to good news, and its replacement with its opposite, a more bullish mindset.

Leave a Reply

%d bloggers like this: