one and a half years in advance! The pound devaluation simply took place during the termination phase of a (nominally) 18-month cycle that has been going up and down with the regularity of clockwork for nearly a century.

HOW NATIONAL CRISIS SHOULD AFFECT YOUR DEQSIONS

How about when President Eisenhower was stricken by a heart attack while in office? Could you spot this occurrence in the history of the market if all markings were removed from Figure IX-2? The market dropped, yes. But if you expand this period of time, you will find that the drop perfectly coincided with the end of the first of three 18- to 20-week cycles in the second of three 18-month cycles of the 52-month cycle that began in late 1963!

The U.S. and Cuba had a big falUng out in late 1960. What did the market do about it? It was time to start an 18-month (in this case 20-month) cycle, so it did. In fact, the market disdained even to show the courtesy of a mild reaction to this momentous turn of events!

Can anyone ever forget the 15 minutes of panic selling that hit the market on the news of President Kennedys assassination? There is certainly no doubt whatsoever as to the relationship of that frenzied few moments of price action to the tragic news that inspired it. The powers that be even closed down the exchange. Yet, in retrospect, it cannot be denied that the sUght drop in the market at this time signaled the expected end of the fust 18-month (17-month in this case) cycle of the 52-month cycle which started in 1962. Here was a case where a popular president was cut down violently and unexpectedly in office and the presidency of the U.S. was assumed by a new administrative head, yet the market barreled uninterruptedly upward until the pullback m 1965 which terminated the second 18-month cycle of the 52!

Try to relate the significance of the Suez Canal crisis, the Sputnik launching, the attack on U.S. destroyers in Tonkin Gulf, thepohtical demise of Premier Khrushchev, etc., to market activity. The unavoidable conclusion that must be reached is major world and national historical events have negligible impact on stock prices!

HOW THE GNP AFFECTS THE MARKET

Yet, we notice one thing more from this figure. Prices moved from about 160 in 1949 to over 1000 m 1966, but the sum total of all the cyclicality of the price-motion model accounts only for price motion within the envelope surrounding the 52-month highs and lows (as demonstrated in Figure 11-10). Our observed cyclicality does not account for the price motion represented by a center tine drawn between the bounds of this envelope. If random events, cyclicality, and major world affairs do not account for this panoramic price change-wAa/ does?

Take a look at Figure IX-3. Overlaid upon our now familiar plot of the DJ 30 Industrials from 1949 through 1968 is a plot of the Gross National Product of the U.S. in billions of dollars. Now it is an interesting coincidence that the scales of the DJ 30 and the GNP in billions of dollars match almost perfectly. The scale on the left therefore represents both the value of the average and the value of the GNP expressed

Which Came First: The Dow "Chfcken" Or The GNP Eggi

in billions of dollars. The correlation is striking. There is even an indication in the GNP of the ups and downs of the 52-month cycle! We must be cautious to the extent of remembering that the observed correlation does not necessarily imply a cause and effect relationship. In fact, it may well be saying that both the DJ 30 and the GNP simply react to the same causes. Or the relationship may even be purely coincidental. However, here is a "fundamental" factor which undeniably does correlate to market action over the time period shown, and at the same time seems to account for the part of price motion not accounted for by cyclicality. It certainly stretches the unagination less to impute a relationship here than in the case of historkal events!

NOW COMPARE CYCUCALITY VS. HISTORY

Consider Figure IX-4. This may well be one of the most informative charts in this book. Its certainly a well belabored one, requiring more than three million separate computations to produce.

The weekly closing prices of the DJ 30 Industrials are used this time-from 1935 through 1951. Again, the major world and national events that were told make market history are accurately overiaid. Now, without peeking at the dateline or captions, can you pick out the period of time in which World War II was in progress from market action alone?

Taking events one at a time again:

1. No events of major significance could be found for 1935, 36, and " ?. Yet the market roared mightily during this period and subsided with a thump in 1937 and early 1938. What caused all this price motion-far more extensive than any that took place during all of WW II?

2. WW II started in 1939 and no immediate effects are seen at all. For nearly a year the market drifted sidewise to dighlly down.

3. The Pearl Harbor hammer blow hh in 1941 and the market plunged some. But it had been going down for months before-at the same rate!

4. The German and Japanese surrenders show no impact at all. The market had been rising sharply for three years, and neither accelerated nor decelerated its pace because of these monumental events.

5. The fall of France in 1940 is closely associated in time with a fierce and very rapid plunge in the market. Could this be an impact of a fundamental event? It is noted that the market plunge took place before the event, but perhaps the market was simply "discounting" the possibility in advance (a favorite phrase of the fundamentalist). Well, lets take a closer look.

THE IMPACT OF THE FALL OF FRANCE

Figure IX-4 also displays a series of six wavy, dotted lines. These are cycUcalities that have been identified for this time period and were removed using the same analytical techniques that were employed with Warner Co. in Figure II-l 1.

Now lets see if any of these look familiar. The periodicity marked "6" shows 92 samples over 884 weeks for an average of 9.6 weeks per cycle. When this form of analysis is carried through to more recent years, it is found that this oscillation slowly