Note 466 — Chapter XXI

I do not believe that we have sufficient agreement among the best students of the statistics of the precious metals to justify any statistical conclusions regarding the laws governing the industrial consumption of gold and silver. Even the facts as to the proportions of annual production of gold in recent years going to money and to the arts are in dispute. Thus, DeLaunay (The World's Gold, New York, 1908, p. 176), divides the annual output as follows: Exportation to the East, and loss, 16%; coinage, 44%; industry, 40%. The industrial employments are divided as follows: jewelry, 24% (of total annual gold production); watch cases, 10%; gold leaf, 2.25%; watch chains, 1.75%; plate, 0.75%; various uses, as pens, dentistry, chemical works, etc., 1.25%. DeLaunay's competence as an authority is attested by various writers, among them W. C. Mitchell (Business Cycles, p. 281). Mitchell, comparing DeLaunay's estimates with divergent estimates of other authorities, concludes that there is not sufficient evidence to justify definite conclusions. I do not think that anyone who has read the criticisms which Touzet has brought together (Emplois Industriels des Métaux Précieux, Paris, 1911, pp. 49-52) of the methods employed in the investigations by the Director of the United States Mint in 1879, 1881, 1884, 1886, and 1900, will have large confidence in the exactness of the results reached in those investigations. (See annual reports of the Director of the Mint for the years in question.) Touzet's careful and elaborate study employs the figures of these investigations as the best available, but with substantial misgivings. There are many indeterminate elements in the problem, as shown by both Touzet and DeLaunay, among them, the extent to which coin is melted down for industrial purposes.

The Director of the Mint would assign a much higher proportion of the annual output to coinage than would DeLaunay.

Earlier studies, by Soetbeer and Suess, seem quite out of harmony with these conclusions. (Suess, Eduard, The Future of Silver, Washington, Government Printing Office, 1893, pp. 51-53.) Suess thinks that virtually as much gold was going into the arts uses as was being produced, in 1892, and quotes Soetbeer (Litteraturnachweis, p. 285) as admitting that such a contention may not be demonstrable, but at the same time holding that it cannot be disproved.

In the face of what seems to be a really indeterminate statistical problem, I content myself with the theoretical conclusions in the text. Because I cannot find adequate grounds for confidence in the main source from which he has drawn his statistics, I refrain from a criticism of the theory and method underlying Professor J. M. Clark's ingenious effort to derive statistical laws for the elasticity of the arts demand for gold. (American Economic Review, Sept. 1913.)