undoubtedly be accompanied by the development of reserve stocks and storage for stabilizing crop yields from year to year, thus freeing industrial operations from the risk and uncertainty of fluctuating crops. A society in which production is carried on continuously and fairly evenly will be a society with no periodical unemployment crises and hence with less of those socially undesirable concomitants of unemployment. The large shifts of occupations from industry to industry, which occur in periods of expansion and contraction, also will disappear, along with fluctuating wage rates and earnings. The sudden rise of large fortunes and the impoverishment of others will decrease, since the rapid fluctuations of prices will not occur, to make such social changes possible. Looked at in perspective, one might expect to see a return to the condition of social stability which obtained from time to time in England before the nineteenth century (between the various upheavals of lesser magnitude than the industrial revolution). But such stability will be something rather new to western society, since it is probable that, under a so-called democratic government, with an income and inheritance tax, with a decreasing number of hours of work (capable of considerable reduction in a stable industry), and with a large reduction in the number of persons engaged in "pecuniary employments" (Veblen), our social life will be greatly altered. It is difficult to conceive of any part of our social life today which will not be changed by integration of industry with all that such a development implies. Thus, as we see this onward sweep of the industrial revolution fulfilling its inevitable function of reorganizing man's industrial activities and hence altering, pro tanto, his social life, the significance of what we call integration appears very large indeed. Naturally we regard the development of integration as something rather daring and experimental, to be viewed questioningly and made to prove its right to usurp our accustomed life and pecuniary methods; but to our grandchildren or great-grandchildren, our doubts and hesitation and reluctance will probably seem highly amusing, much as we regard the early railroad operations where each shipper had his own car and where the ownership of the roads was limited to small local sections, and the movement of goods and passengers was largely at the mercy of bargaining between rival and connecting owners. The notion of transportation as an industrial activity, involving through movement, regardless of proprietary rights, has scarcely become established even now, as witness the current discussion of the merits of railroad consolidation. But integration of industry and consolidation of railroads and public utilities and the social changes they will necessitate will come; the command of the machine and the tool have ever been supreme in man's affairs and probably ever will be. Social theories and philosophies, like all rationalizations, will arise to justify the new order of affairs after it has become a little more established and, no doubt, an economic theory will be developed in due season to minister to this new scheme, as Adam Smith and Ricardo, with their successors, have variously upheld the virtues and inevitability of the old. NEW YORK CITY LAWRENCE K. FRANK THE EFFECT OF PRICE FLUCTUATIONS ON AGRICULTURE It is the purpose of this paper to consider the effect of general price fluctuations (1) on the purchasing power of farm products during the last 130 years, (2) on the economic status of the farmer as a property-owner; and finally to survey briefly the outlook in view of the present world situation. A word as to the meaning of the phrase "purchasing power of farm products." Let us consider first a similar phrase in a different field. We speak, for example, of the purchasing power of wages, by which we mean the ratio of money wages to the retail prices of the commodities and services purchased by the average wage-earner and his family. The retail prices of these goods and services are weighted according to their importance in the workingman's budget, and thus a cost-of-living index is constructed. A purchasing-power wage index thus gives us the changes in the wage-earner's power to purchase consumers' goods. A similar index for farmers would require (1) an index of changes in the net incomes of farmers, (2) an index of the retail prices of consumers' goods purchased by farmers and their families. Neither of these indices is available except in fragmentary form and for a limited period. The phrase "purchasing power of farm products" obviously refers to a somewhat different matter. It might be used to mean the ratio of farm-product prices (paid at the farm) to the retail prices of consumers' goods weighted according to the purchases made by the average' farmer and his family. Such a "purchasingpower" index would by no means be identical with a series showing the ratio of the incomes of farmers to the retail prices of consumers' goods purchased by farmers. Obviously the fluctuations 'It may be noted that the "average" farmer, owing to the diversity of regions and products, is an even more fictitious person than the "average" wage-earner. What is really needed is a special index for each farming region. in net income differ widely from fluctuations in the exchange value of farm products, owing, first, to changes in the farmer's profit margin, and, second, to changes in his volume of production. Such an index would therefore not reflect accurately changes in the real incomes of farmers. A more reliable index would be a series showing the ratio of farm-product prices (paid at the farm) to the retail prices of all products purchased by farmers (producers' and consumers' goods) weighted according to the importance of each in the farm budget. In such a "purchasing-power" index, cost prices would be taken account of as well as consumption prices, but no account would be taken of changes in output per farmer. Price margins might remain the same, indicating no change in the farmer's economic status, yet his condition might steadily be improving, owing to an increase in production. In actual fact the volume of output per farmer has shown a marked upward trend, particularly during the last sixty years. But it is clear that a purchasing-power index of the type just described would tell an important story with respect to the changing economic status of the farmer. But the data for the construction of such a purchasing-power index are not available. As a substitute, we have constructed a series from the wholesale prices of farm products and the wholesale prices of general commodities. This "purchasing-power" index represents, therefore, the ratio of farm-product prices at wholesale to general-commodity wholesale prices. It is in this sense that the phrase "purchasing power of farm products" has been widely used in recent years. At best, however, it is only a rough indication of the quantity of products that the farmer is able to buy in exchange for his own products. The index in question is, however, more acceptable if we are thinking of the broad question of the exchange value of farm products in relation to general commodities, and not merely of the special interest of the farmer with respect to what he can buy for his products. The general-commodity and farm-product price indices used in this article are each composed of several distinct series joined together to cover a period of one hundred and thirty years. The number of commodities used in the several series varies, and the methods of construction, as explained below, are not identical in all cases. Moreover, splicing the series together necessarily involves the conversion of several of the series to a new base year far removed from the earlier series. All this implies the possibility of a considerable statistical error. The writer does not pretend that the result is any more than a rough approximation, the general outlines of which are believed to be sufficiently accurate to be significant. Chart I and Table I show the purchasing power of farm products (as defined) for one hundred and thirty years. Two main periods and four shorter periods may be distinguished. From 1790 to 1820, the trend of the purchasing power of farm products remained substantially stationary. From 1820 to 1920, except for the severe setback in the sixties, the general trend has been continuously upward. This long period of a hundred years may be broken up into three subperiods which may be distinguished by changes in the rate of increase in the purchasing power of farm products. From the decade of the twenties to the decade of the fifties, the rate of increase was very great, the uniform rate of increase being 1.6 per cent per annum. Then came a sharp downward break in the trend. Starting from a much lower level, the trend rose slowly from the decade of the sixties to the decade of the nineties at the rate of increase of .4 per cent per annum; while from the first decade of the twentieth century to the decade 1910-19, the rate of increase was 1.1 per cent per annum. In brief, then, the trend of the purchasing power of farm products did not change materially from the time of the Constitution to the decade of the War of 1812; rose rapidly from this period to time of the Civil War; dropped abruptly from 1859 to 1862; moved slightly upward from the Civil War to the Spanish-American War; and then rose with considerable rapidity from the late nineties to the end of the world-war. These periods correspond fairly well with the main periods into which the general wholesale price level may be broken. The first (1790-1820) corresponds roughly with the upward trend of general prices, culminating in 1814-17. The second (1820-60) corresponds with the period of declining general prices culminating in |