100, and the average decline in the prices of these same articles for the year 1810 was found to be 20 per cent, the index number for the year 1800 would be 100, and for the year 1810, 80.[1]
The difficulties in the way of obtaining satisfactory averages from comparisons of prices at different periods by the above or any other methods are, however, almost insuperable; so that it may well be doubted whether the determination of an average of general prices is ever within the bounds of possibility. Quotations for a given day, or month, do not necessarily show the average for the year; and, in like manner, the selection of a limited number of articles for comparison can not insure correct conclusions respecting the movement of prices in general. All methods of comparing price variations which content themselves with mere average quotations of different articles, and which do not pay due regard to the relative importance of each article in the domestic and foreign commerce of a country; which, for example, allow a change of 80 per cent in the price of an article like cochineal, of which the value sold in any one year is small, to balance a change of 2 per cent in an article, like sugar, the value of which annually sold is enormous, are also in a great degree deceptive and worthless;[2] and even when in the comparison of prices, the importance of considering relative quantities is fully recognized, the data for ascertaining these relations are extremely uncertain and questionable. The utmost of service that all such tabular comparisons of prices, even when prepared with all desirable qualifications, are capable of rendering, would, therefore, seem to be limited to the affording of important inferences respecting variations of prices, or to the showing
- ↑ For a full exhibit and discussion of these tables, reference is made to a paper prepared and laid before the British Royal Commission (third report, Appendix B, pp. 312-390, 1886), by R. H. Inglis Palgrave, F. R. S.; and also to an article in the (Harvard) "Quarterly Journal of Economics" (vol. i, No. 3, Boston, 1887), by Professor J. Laurence Laughlin, Professor of Political Economy, Harvard University.
- ↑ One of the best-known tables of this character, embracing twenty-two different articles, has been kept by the London "Economist" for many years as a constituent element of current British commercial history; and the objections inherent in the system adopted are forcibly illustrated by the following recent occurrence, to which attention has been called by the "New York Commercial Bulletin": Thus, a comparison of index numbers for Januuary and July, 1886, and for January, 1887, as deduced from the "Economist's" tables of prices, indicated a small advance for the latter month in the general level of British prices. But the first article on the "Economist's" list of prices is coffee, which advanced from July 1, 1886, to January, 1887, to a degree sufficient to alone add 50 to the index number of January; while the entire increase for the whole twenty-two articles was only 36; or, in other words, if coffee alone were omitted from the list of articles compared, the net result would show an apparent decline instead of any advance in the general level of prices. "Certainly," as the "Commercial Bulletin" remarks, "it is difficult to attach much importance to results having no better basis than this. For coffee is by no means one of the most important articles compared; it is greatly exceeded in importance by at least twelve of them. But the change in that one article happens to have been surprisingly great, and it thus outweighs far more important changes in other articles, such as iron or meats."
