countries. Our high protective tariff has probably retarded in some degree the levelling influence; but the general forces seem to have exercised their influence none the less.

Another and highly important phase of the monetary history of this period receives attention in that part of the Report which presents statistics on wages. The methods and results of this part of the investigation are to be discussed by another hand;' but some of the salient results may be here noted.

Returns of wages for 543 distinct series of laborers were obtained for the period from 1860 to 1891, and for 61 of these series returns were obtained for the whole period from 1840 to 1891. Average wages for each of 21 occupations were computed on the index number plan; and a general index number indicating the movement of wages for all the occupations was also computed. In these calculations, the wages for 1860 were again taken as the base, and the results thus made readily comparable with those for prices. The average, or index numbers, are in one sense more accurate and significant as to wages than they are as to prices. The divergence of the individual wages from the average of any one occupation is comparatively small: the average is more nearly a true average. So as to the movement of wages in one occupation, as compared with the general movement: the upward or downward fluctuations in the general average is reflected with greater faithfulness in the average wages of the several occupations than is the general movement of prices in the quotations for particular articles or groups of articles. The inevitable fictitious quality of a general index number thus calls for less constant allowance in using these results of the statistics of wages than in using the figures for prices.

The main results as to general money wages are indicated in the appended chart III, on which the general index numbers of wages are plotted. For readier comparison, the line indicating the movement of prices (calculated by simple

1 In a paper presented to the International Institute of Statistics by Colonel Carroll D. Wright, of the Department of Labor. It will appear, with the present paper, in the Bulletin of the Institute.

average) is also given. It will be seen that money wages responded with unmistakable slowness to the inflating influences of the Civil War. In 1865, when prices stood at 217 as compared with 100 in 1860, wages had only touched 143. The course of events at this time shows the truth of the common statement that in times of inflation wages rise less quickly than prices, and that the period of transition is one of hardship to the wages-receiving class. On the other hand, the sluggish movement of wages shows itself in the opposite way

in the succeeding period of falling prices. As wages rise less quickly than prices, so they fall less quickly. The upward movement of wages continued after 1865, rapidly until about 1867, thereafter more slowly. But for the activity of the speculative period preceding the crisis of 1873, it is probable that money wages would have begun to fall again as early as 1870. As it was, the advance continued slowly, until the crash of 1873 precipitated a downward movement which lasted until 1879, and corresponds to the abrupt fall of prices during

the same years. With the resumption of specie payments, and the new and more solid start which the industry of the country then took in all directions, a striking inverse movement of wages and prices took place. From 1879 wages rise; there is a slight interruption of the upward movement in the depressed years 1884-86, but otherwise the rise is steady. Prices rise in the "boom" years 1880-82, and thereafter fall unmistakably. Taking these years as a whole, we have strong testimony of that inverse movement of prices on the one hand, and of wages, and indeed all money incomes, on the other hand, which seems to have taken place in all civilized countries during the last generation. It may be fairly said that this inverse movement shows itself in the United States even before 1879. In the long period of depression that lasted from 1873 to 1879, wages fell more slowly than prices, the inevitable readjustment from a paper to a specie standard operating more incisively in the decline of prices. Wages virtually rose, as compared with prices, before the direct rise in money wages set in. All in all, the figures show that the purchasing power of money wages has been rising steadily for at least twenty years, and that the decline in prices since 1873 and especially since 1882 has been a source of prosperity and not of depression to the community at large.

From this digression in regard to wages we may return to some further consideration of the movement of prices, and, more particularly, to an interesting phase of that movement during the period of inflation: the relation between the specie premium and advance in general prices. For some purposes, it is necessary to regard the premium on specie as a mark of the depreciation of an inconvertible paper money. But economists have long been attentive to the fact that the real depreciation—the rise in prices over and above what they would have been if the currency had remained on a specie basis-is by no means measured with accuracy by the specie premium. The premium may be greater than the general rise in prices; it may be less. So far as the United States during and after the civil war is concerned, it has been usually supposed that in the first stages

of the paper emission, and especially during 1863-65, the specie premium was greater than the general rise in prices; while in later years, say in 1866-70, the reverse is supposed to have been the case, the rise in prices being greater than the gold premium. It is pertinent to inquire what light our statistics throw on this aspect of the history of prices.

Unfortunately, the fact that the index numbers were calculated on the basis of January prices makes it possible to get conclusions only of a very limited sort. On chart II the general index number of prices, as they stood in January of each year, is compared with the gold premium as it stood in January. For the earlier years of the papermoney period, however, the comparison derivable from the lines is of little value. During the first years of the period, from 1862 to 1865, prices and the gold premium fluctuated with great rapidity and with extraordinary irregularity. The paper-money issues were on a large scale, and depreciation set in very quickly. The progress and extent of the real depreciation, as indicated by the change in prices, could be ascertained only by getting statistics from month to month, or, indeed, from week to week. The gold premium reached its highest point in July, 1864, when it touched 285. Throughout 1865 it fluctuated violently. It ranged between 285 and 222 in the month of July, 1864, and during this year ranged between 245 and 151. How prices fluctuated during this feverish year, we have no means of knowing. Few individual commodities showed fluctuations as great as gold, and a general index number, month by month, would probably show a less spasmodic and irregular movement than did the gold premium. It is still possible, even probable, that the general movement of prices lagged behind the gold premium. In the first month of 1865, for which we have the high index number of prices indicated on the chart, there is no great difference between the gold premium and the index of prices; but it is not improbable that by this date prices had just attained their highest point, while the premium on specie was already on the decline. All this, however, is speculation; our statistics do not give us more than a hint of the course of events during the period

of sudden inflation. It should be said, however, that the materials for a more detailed investigation are contained in the specific quotations of prices printed in the body of the Report. There the economist will find quotations of prices for all the articles for each quarter of each year,-for January, April, July, and October. Index numbers prepared from these quarterly quotations might furnish instructive results, and present a tempting opportunity for those who may be able to give the investigation the weary labor which it would entail.

After the first years of the period of inflation, when the movement of prices became less irregular and fluctuations in the gold premium less abrupt, the January index number and the specie premium may be compared with better promise of significant results. And then the movement is certainly striking. The decline in the gold premium is more rapid than the decline in general prices. The real depreciation of the paper, in other words, was greater than its discount in terms of gold would indicate. This is more particularly the case in the years immediately after the war, -in 1866-68, and again in the years immediately before and after the great crisis of 1873,-i. e. in 1871-74. After the war the gold premium fell sharply and suddenly from a range of 200 and over to one of 140 or thereabouts; yet prices fell much more gradually and more slowly. Again, in the years after 1870 the gold premium ranged between 110 and 120, and in general tended to fall. The January quotations in this period are a fair indication of the range of the premium throughout the year. Yet in '70-'73, prices showed a tendency to rise rather than to fall, and the real depreciation of the currency was greater than the specie premium would indicate. After 1873, prices fell sharply and continually; the gold premium fell much more slowly. The country was nearer a specie basis in 1873, so far as the gold premium went, than it was so far as the general range of prices went. The final and real readjustment to a specie basis was accomplished rather by the relentless fall in prices and in wages between 1873 and 1879, than by the more obvious, but less incisive, decline in the gold premium. F. W. TAUSSIG.

Harvard University, Cambridge, Mass.