There is a large seasonal fluctuation in building operations varying, for twenty leading cities, from about thirty million dollars in January to sixty million dollars in June. This variation must be eliminated before the cyclical movement can be seen clearly. When this correction has been made, building statis tics will be found to keep very close company with the fluctuations of standard railway stocks. They have an important peculiarity, however. They usually make their top for the cycle and begin a slight decline before shares reach maximum figures. The recovery of building after a panic is very prompt. Building is sensitive to banking conditions, since the building of large structures usually requires either the sale of real estate bonds to investors or the in crease of capital by a corporation.
The statistics of the number and estimated cost of buildings as appearing from building permits are compiled by the F. W. Dodge Company and the Bradstreet Company. The list of cities is over one hundred, but it is not identical in the two reports.
Statistics of the number and amount of actual build ing contracts let are compiled and issued as a special publication by the F. W. Dodge Company of New York City.
6. Employment, wages and gen eral, the statistics of employment are not exact enough to use for barometric purposes. Even the experts are at great variance as to the number of unemployed per sons in periods of distress. However, the New York Labor Department for some years has published at monthly intervals the statistics of unemployment re ported by the trade unions of the state.
The statistics of strikes, altho published from time to time by the United States Bureau of Labor Statis tics, are not made available promptly enough for bar ometric purposes. The general theory with refer ence to strikes is that they are most frequent at the climax of good times when the rising cost of living pursues the wage-earner, when the sight of luxury everywhere arouses in him the desire to participate in it more largely, and when large corporate earn ings and a generous state of public opinion make the chances for winning appear good. At such times the demand for labor promises the speedy installation of any strikers who may be discharged in positions with other employers. Strikes are also frequent at the end of a period of depression. Here improvement in business is clear, and conditions seem to make possi He wage advances which will compensate for reduc tions made during the preceding crisis or depression.
The statistics of the number of persons out of work are better than those of the number of strikes. The most significant object of a strike to the investor is a demand for increased wages.
A sample of the material reported by the New York Labor Department is as follows: The incoming of alien immigrants has long been reported monthly by the United States Government. The statistics of net immigration—arrivals minus de only been compiled since July, 1907. These figures are of no value for prediction. They record a slow response of the movement of popula tion to industrial changes. They are highly seasonal, involving a large spring flood and a lesser fall flood. They represent only in part the changes of the Amer ican labor market, for they are complicated by the change of social, political and economic conditions in the European supplying countries.
7. Commodity prices are al most endless in quantity. The trade press of each line of industry carries daily, weekly and monthly prices of the raw materials and finished products in which the trade is interested. Reference is made to these statistics in the chapter on Industrial Securities. The statistics of crops and of the steel trade have al ready been referred to in this chapter. Copper is an earlier barometer than steel, but it is complicated by foreign influences, since one-half of the metal is ex ported. In general, metal prices follow the history of the equipment industries. The statistics of the lumber trade are not gathered in a very satisfactory way. Individual commodities show erratic fluctua tions in price such as the rise of sugar in the latter part of 1911 and of 1914, of cotton in the early part of 1904 and the middle of 1910 and of copper in the winter of 1906-1907. Turpentine, made a peak in early 1911 and rubber in 1906 and 1910. If we study the prices of individual commodities it is difficult to find a clear and direct connection between them and security prices. In the year 1907, whiCh was char acterized by a steady and radical decline of stocks, we find no prominent commodity prices giving clear warning. Certain commodities such as pig iron, bil let steel, turpentine, copper and rubber fell as stocks declined. Others such as sugar, cotton, yellow pine and petroleum closed the year higher than they entered it, while wool made no downward movement until March, and coffee continued its moderate down ward trend of 1906.