Rally After the San Francisco Disaster

Continued from...1906 San Francisco Earthquake and Stock Market Panic.

Seriousness of the Disaster

At this distance of time we may easily forget how serious the San Francisco disaster was. The loss direct has been estimated at $600,000,000. The Aetna Fire Insurance Company admitted that the conflagration had cost it the savings of forty years. If that was the effect on the strongest fire insurance company in the United States, and one of the strongest in the world, how severe must have been the consequences elsewhere. It was all very well for the shallow, half-taught optimist to say that broken windows made work for glaziers and manufacturers of glass. But it cost you something to put in a new pane, and the money you spent on it would have been spent on something else, while, as Bastiat said, you would still have your window. If that sort of reasoning were good, the quick road to prosperity would be to burn down all the cities in the United States.

We see that the railroad stocks suffered more than the industrials, and we should remember that they were in a higher class, both relatively and positively. But in a sudden and demoralizing break people sell the things for which there is some market in order to protect those for which there is no market. As The Wall Street Journal put it at that time: "The first decline in a panic is scare, and the second and slower decline is the demonstration of the general shock to confidence;" going on to say, in speaking of the market on July 2d, that the line of prices was well below the line of values and that the indications were bullish.

Rally From a Break in a Bull Market

This inference proved correct, and it has been the custom, which is followed in these discussions, to consider the bull market which began in September, 1903, as actually terminating and turning to the bear side, not in January, 1906, but in December of the same year. At the time the bullish inference quoted was published the market was making a line which proved, as the analyst correctly surmised, to be one of accumulation. The forecast was soon verified, and on August 21st The Wall Street Journal again discussed the market from the point of view of the averages. There was a greatly active market at that time, and it remarked how absurd it was to suppose that within two hours of trading on a Saturday one single interest could possibly manipulate one million six hundred thousand shares. This is a useful confirmation, coming out of the past of fifteen years ago, of what we have already seen for ourselves in demonstrating the relative unimportance of manipulation. In that discussion The Wall Street Journal went on to say: "We can only suppose that the long decline between January 22d and July 2d represented a somewhat extended bear swing in a bull market."

Average Deductions Uniformly Correct

Remember that this correct inference was drawn at the time, and not after the event. I could easily go back and show how trustworthy these deductions have been over the twenty-odd years since Dow formulated his theory. It would be absurd to say that it was possible to call the exact turn in the major swings, much less anticipate the unexpected. But these studies in the price movement did what was much more useful from the point of view of those using the barometer from day to day: they were continually right when they said of a major movement that it was still in progress, even when a deceptive secondary movement had made superficial observers bearish in a bull market or bullish in a bear market.

There is a story, probably apocryphal, of James R. Keene saying that he would be well content to be right 51 per cent of the time. I don't believe he ever said it. He must have found a much larger percentage necessary. The balance in his favor would not have paid operating costs, to say nothing of keeping a racing stable. But the deductions from the evidence of the price movement have been right, as the printed record proves, much the most of the time. After searching both the record and my conscience I can find no instance of a radical misinterpretation of the meaning of the barometer. The studies based upon its use were uniformly able to anticipate what the public was thinking about business before the public knew its own thoughts. The errors, where any occurred, were mainly due to the almost impossibility of forecasting the secondary movement of the market. This is really much more difficult than the interpreting of the major swing, just as it is easier for the Weather Bureau to forecast weather for a large area than it is to say whether it will rain in New York to-morrow morning.

Previous...1906 San Francisco Earthquake and Stock Market Panic. Next...1907 Bear Market after the San Francisco Earthquake

From The Stock Market Barometer by William P. Hamilton, published in 1922