Knowledge that Transforms

IN IIg an analysis of New York bank clearings and railroad and industrial stock prices for the period I903-I4 showed that these three speculative series displayed similar cyclical fluctuations, preceding corresponding fluctuations of outside bank clearings and money rates by several months.2 With the object of testing the synchronism of the cyclical fluctuations of the speculative series for a longer period of years, the study has been extended back to i866; and comparisons of the series for the entire period I866-I9I4 are set forth in this paper. For the period 1903-I4 the data underlying these comparisons were taken from the previous study. For the period I866-I902 new data were assembled. Monthly items for bank clearings for I866I902 were obtained from the Commercial and Financial Chronicle.3 In order to obtain a continuous series of monthly railroad stock prices beginning with I866, it was necessary to join three indexes. For the period I866-80 there was available the index of the prices of Io railroad common stocks previously published in this REVIEW.4 For the years I88I-96 no index was available and a new monthly index of the prices of Io railroad common stocks was constructed. This index is described in detail in the Appendix to this paper, pp. I93-98. For the remainder of the period under examination, we used the Dow-Jones index of railroad stocks, available from I897.5 These three indexes, which are on different bases, were adjusted to form a continuous series on the level of the Dow-Jones index by the use of factors based on overlapping data. Indexes of industrial stock prices were available in monthly form beginning with I872. For the years I872-96 the series compiled by the New York Federal Reserve Bank was available, and from I897, the Dow-Jones index.6 As the series for I872-96 was obtained by working backward from the Dow-Jones list of I2 stocks in I897, adjustment of the series for the two periods was unnecessary. No indexes of industrial stock prices were available earlier than I872 because of the lack of representative organized markets. Similarity in the cyclical fluctuations of the three speculative series would probably have been revealed by plotting them on the same logarithmic chart. But comparison would have been difficult at times owing to the diverging or crossing of the curves the result of varying trends. In New York bank clearings the general movement was upward throughout the period (Chart 2). A moderate, steady rate of growth persisted from i866 to I896, followed by a sharper upward trend during a transitional period which lasted until I903. Thereafter until the end of the period, the rate of increase was less marked. The series for railroad stock prices was subject to both changes in the character of the long-time movement and pronounced shifts of level. During the years i866 to about I882 the trend was uncertain; the succeeding movement, not far from horizontal, persisted until I897, and was followed by an abrupt transition in I897I903 to a markedly higher level, which was maintained through I9I4 (Chart 3). Industrial stock prices showed one sharp

THE object of this paper is to assemble and weigh the statistical evidence bearing on one phase of the pre-war history of pig-iron prices, namely, the amplitude or extent of pig-iron price movements. In effecting this object some very familiar and perhaps shop-worn data are employed but, it is believed, with new inferences. In previous studies,' these same data, the cycle figures for pig-iron prices, have been used to interpret business conditions and to illustrate the contrast in amplitude between production and consumption goods or between raw, semi-finished, and finished goods; but a conclusion that pig iron, because it is a production good and a raw material, is in the class of things subject to wide price movements, or that the general character of these movements is of unusual significance in the interpretation of business conditions, does not exhaust the meaning of the data. There are forces to be found particularly in tariff changes and in the rate of growth of successive stages of the iron and steel industry which exercise an important modifying influence on the course of pig-iron prices throughout the business cycle. It is toward a statistical analysis of these forces that this paper is directed. For the purposes of investigation, the selected period, I898-I9I4, is advantageous because of the stability in the art of making iron. Throughout the seventeen years, which cover five price cycles of varying intensity and duration, including one of unusual severity, the blast furnace has seen little modification in technique. Such changes as there have been may be quickly summarized: The industry has grown; blast furnaces have increased in size and in capacity; mechanical devices for handling the ore, coke, and limestone have displaced the older hand methods; and refinements in process have obtained a marked economy in fuels. Yet these changes have been no greater than might be expected in the history of any matured industry, and have been so gradually worked out that they have attracted no attention to their effect on the cyclical movements of the iron trade. The economic organization of the industry during the period demands more careful scrutiny than its technique. The most striking change known to haye taken place concerns the merchant furnaces which produce iron for sale in the open market (see Chart 3). With the growth and integration of large steel companies merchant furnaces have become a less and less. important factor in the production of iron. That part of the industry which makes iron principally for sale, the part represented by merchant furnaces, has expanded little during the period while the number of furnaces which produce chiefly for the maker's own use has increased at a very rapid rate. Since a price is quoted only on the iron which actually enters the market, this change in the organization of the industry is important. However, its consideration is postponed for the moment while a survey of the history of the period is undertaken.