1998/99 versus 2006/07

Barron’s Michael Kahn looks at 2 periods of Dow Charts, and finds them surprisingly similar: 



Kahn notes that the sizes of the 1998 decline and the 1998-1999 rally were twice as big as their current counterparts, but the structure of the action is very similar.

I continue to look at 1973 as my best historical analogy: After the
1966 peak, we had a major selloff, a rally towards new highs, and then
a 30% correction. 

Kahn reaches this conclusion: If this model holds for 2007, then the market is in for a very choppy few months followed by the return of the bear . . .

History Paints a Somber Picture
Micahel Kahn
Barron’s MARCH 7, 2007   

Print Friendly, PDF & Email

What's been said:

Discussions found on the web:
  1. lurker commented on Mar 16

    Important caveat: M. Kahn’s job is to fill column inches between ads, not to generate alpha.
    In short, if he could trade and make money, he would, as it pays much better than journalism.


  2. alexd commented on Mar 16

    OF course the lat thing we want to hear is this time it’s different.

    But why are we so willing to accept this time it’s the same?

    Remeber history does not always repeat itself, but it is incllined to stutter.

    These arguments without alternative scenerios and plans are there just to juice our adrenaline. Any one want to make some calls on what to look for and how they might make money?

  3. Walker commented on Mar 16

    It is charts like this that convince me that technical trading is below astrology in its rigour.

    What makes the double bottom identified in the bottom chart any more significant than the one below the words “Steady Rally”? Was any curve fitting done to make one more significant than the other? Or was it just eye-balling?

  4. Lauriston commented on Mar 16

    It’s Friday Barry, you are forgiven, especially after the fantastic articles this week wrt subprimes mortgage issue!

  5. Steve C commented on Mar 16

    I remember in the Fall, 2002, charts of ‘2000-’02 were superimposed on ’29-’31. As I remember the similarities were amazing.
    Well, from that point onward the charts diverged significantly. I look at similarities in the market over different time periods as comparing ocean waves. No two will remain alike for long.

  6. wally commented on Mar 16

    His prediction sound right; he may be quite correct. However, it will not be due to the chart analysis.

  7. KP commented on Mar 16

    Seems to me the big given in this comparison is that people never learn. Not so unreasonable to me.

  8. Fred commented on Mar 16


    Of course it makes no difference that the economic conditions, valuations, and interest rates were all dramatically different, and that the Fed was RAISING interest rates then compared to now.

  9. Michael C. commented on Mar 16

    Comparison charts offer almost no value, IMO. That’s because after 75+ years of charts, you can almost always find a period that is similar to the one today. And yet it is only one instance, so it’s value is highly suspect if not meaningless.

    I’m all for historical data, such as finding 80 other instances of such and such an intraday reversal with a 65% positive outcome with average gain of 0.6% for example. That is statistically meaningful.

    But comeon. After 75+ years of charts, you will always find a similar period. A year ago it looked like so and so year. 3 years from now it will look like so and so year…

  10. S commented on Mar 16

    I haven’t read the article. But note that the ’98 double bottom coincided with the Russian debt default and LTCM crisis, which led to investors panicing and re-pricing risky assets.

    Today, investors are panicked about another debt crisis, sub-prime mortgage defaults. Risky assets are getting repriced.

    Who knows, maybe today is closer to Oct. ’98 than Oct. ’99.

  11. Gary commented on Mar 16

    What is never “different” this time is human nature. We still haven’t swung the full range to black pessimism. Stocks have never gotten extremely undervalued. We will get there eventually. All the fed’s money printing and massaged data may lengthen the process but human nature will run its course. There is going to be one heck of a buying opportunity sometime in the future. We’re not even close at this time though.

  12. greg0658 commented on Mar 16

    90’s & 00’s = apples and oranges
    90’s = upstart China, India, Mexico
    00’s = war economy

  13. John F. commented on Mar 16

    That’s the lamest piece of chartology I’ve seen in a long time. If this offends Michael Kahn, he can kiss my rounded bottom.

  14. Alex Grey commented on Mar 16

    My background is in economics however I have been reading my way into technical analysis for a few years. I think there is something to chart patterns because they do reveal behaviour of investors and past behaviour of the market is a factor that influences the current behaviour of the market. So you can view the stock market as a continuous feedback mechanism. That being said it seems very risky to try to find parallels in the past where both circumstances and behaviour were similar. I prefer to focus on the recent past in gathering insight from market patterns. One has to be very cautious about longer-term patterns unless one can identify economic trends that have produced a pattern of economic reactions. In this respect the long boom of the market since the early 1980s can be viewed as a reaction to the Reagan-Thatcher revolution whose prime accomplishments in my mind lie in their impact on financial markets. These are: 1) deregulation of financial markets; 2) lower nominal interest rates because of disinflation; 3) permitting financial innovation. The end result was a sharp increase in the credit to GDP ratio. By my estimation this may have more than doubled from 3.7 times GDP in 1978 to around 8.3 times GDP in 2002 (these figures include both private and government debt though the former is more relevant to the economic cycle). This boosted aggregate demand and hence GDP growth but at the cost of greater risk of a deeper recession because of the large amount of debt. Against this backdrop it is understandable (in retrospect) why stock markets boomed during this period. I would argue that the housing debacle the U.S. is entering represents an end to this long credit cycle. In its late stages credit growth becomes closely tied to asset growth – the case of housing today and falls in asset prices will lead to a contraction of credit. In this context it seems unwise to take comparisons from the period neither during the credit boom (e.g. in 1998-1999) nor before it i.e. the 1960s and 1970s. Unfortunately the most comparable period is the 1920s which also experienced a similar credit boom though the credit to GDP ratio peaked at about 5.5 times. But then again finance was not as advanced in the 1920s as it is today.

  15. Michael C. commented on Mar 16

    It hurts just looking at the above comment.

    Try paragraphs, man.

  16. samuel commented on Mar 17


    I agree. Both periods had massive asset price inflation fueled by debt growth. This should be followed by asset price deflation and deleveraging. We know what the result was after the debt bubble of the 1920s. The big question is what will be the result of the debt bubble of 1990s through 2007?

Posted Under