An Index of General Business Conditions for Germany, 1898-1914

Abstract

IN earlier issues of this REVIEW, statistical indexes of business activity for the United States and Great Britain have been developed. It is the purpose of the present study to construct a similar index for the German Empire in the decade and a half preceding the outbreak of the Great War. A statistical picture of German business is of interest primarily because it provides additional examples of individual business cycles. Unfortunately the number of cycles in the United States, or elsewhere, concerning which adequate statistical information is available is so limited as seriously to restrict the conclusions which may be drawn from their study. Monthly or at least quarterly data representing volume of business, prices, and interest rates are essential to the construction of an index, and these are obtainable only for short periods. The years I903-I4, cOvered by the indexes for the United States and Great Britain, include in the first instance only three business cycles and two in the latter. From the viewpoint of probability, five cases evidently do not constitute an entirely satisfactory basis for generalization. The addition of statistics for the German Empire during the relatively stable pre-war years should be of value, therefore, simply through increasing the number of business cycles available for examination. A second reason for undertaking the study of German business is the prospect that some light may be thrown on the interrelation of business cycles in different countries. During the past several years possible variations in foreign demand for American merchandise have frequently been spoken of as of great, sometimes as of preponderate importance in shaping the future course of domestic business. It may not be amiss to examine how this factor has operated in the past, whether it has been possible to have prosperity in the United States with depression abroad, or the reverse. Any light on the correlation of the cyclical swings in business activity at home and abroad is consequently of direct current interest in the problem of interpreting and forecasting current domestic business conditions. In order to facilitate international comparisons the present study parallels as far as the nature of the data permits the work of the Committee on Economic Research in its studies of the United States 2 and Great Britain.3 In these studies, with which readers of this REVIEW are familiar, various measures of business activity were analyzed by a statistical method developed by Professor Warren M. Persons to isolate the swings of business prosperity and depression. Fluctuations in monthly or quarterly series of data representing business activity, he has observed, are composites of numerous different types of movement. He divides them into four categories: (i) seasonal variation, or movements within the year recurring more or less regularly as a result of the round of the seasons; (2) long-time trend, or the gradual tendency to increase or decrease over long periods of time say, roughly, over periods of ten years or more in length; (3) movements of the business cycle; and (4) miscellaneous movements not falling into any of the above three classes. The statistical technique devised by Professor Persons eliminates the first two of these four types of movement, leaving as a residuum the movements of the business cycle and the miscellaneous variations. In the studies by the Committee on Economic Research of the business indexes of the United States and Great Britain, it was found (after they had been subjected to the above described method of analysis) that series of monthly or quarterly data reflecting business activity fell into three different groups according to the order in time of their movements in the business cycle. Series related to speculative ac-