• Choose a category
• All
• Classic-Fiction

By Root 311 0

gold by following the same procedure for gold as for silver, namely, estimating (a) a demand equation for the nonmonetary use of gold and (b) the hypothetical quantity of gold that would have been demanded if the United States had gone on a silver standard. However, explorations along this line have proved disappointing. First, estimating the hypothetical nonmonetary demand is even more difficult for gold than for silver.* Second, estimated demand equations for gold yielded negative, though statistically not significant, coefficients for the real price of silver instead of the positive coefficient that would be expected for a substitute for gold.† That result is inconsistent with the positive coefficient for the real price of gold in demand equations for silver, a violation of the Slutsky cross-equation condition. Eliminating that contradiction requires estimating the silver and gold demand equations simultaneously, imposing the appropriate cross-equation restriction. Given the drastic difference between the two separate equations, I doubt that the result would deserve much confidence. Finally, these problems resolved, the simultaneous solution requires solving a fourth-degree equation in the U.S. price level.

I am thus left in a quandary. I am unhappy with what I have done but even more unhappy with the most obvious alternative, a simplified general equilibrium analysis. A comprehensive general equilibrium analysis would have to include not only the determinants of gold production and silver production, which I have completely neglected, but also the determinants of the fraction of gold production and silver production that go into monetary and nonmonetary use. Construction of such an expanded general equilibrium model would be extremely laborious and would deserve little confidence. Under the circumstances, I am inclined to leave well enough alone, while at the same time acknowledging that the estimates are subject to a wide margin of error—particularly with respect to year-to-year movements.

Source Notes

(in the order in which the variables are introduced)

P 1869–1914: Friedman and Schwartz (1982, table 4.8); 1865–68: extrapolated backward from Hoover (1960, p. 142).

PS U.S. Bureau of the Census (1975, p. 606, series 270). For 1865 to 1878, original source gives price in gold dollars; adjusted to greenback price by multiplying by reciprocal of gold value of currency, from Warren and Pearson (1933, table 69, p. 351).

WMG, WNMG U.S. Commission on the Role of Gold (1982, table SC 7, p. 198).

UMG U.S. Commission on the Role of Gold (1982, table SC 9, p. 203); 1865–77 shifted from June 30 to December 31 date by two-year moving average of June 30 data for 1865–78.

EWMG = WMG—UMG.

RPG 1865–78: reciprocal of gold value of currency (from Warren and Pearson 1933, table 69, p. 351) times legal price of gold ($20.67183) divided by P; 1879–1914: legal price divided by P.

UKP 1868–1914: Friedman and Schwartz (1982, table 4.9); 1865–67: extrapolated back from 1868 by price index implicit in Deane (1968).

SPROD Warren and Pearson (1933, table 24, p. 139). 1865–75: linear interpolation between centered five-year averages in table; thereafter, annual figures in table.

EWMDS Drake (1985, table A, [>]) gives estimates for successive five-year periods based on annual reports of the director of the U.S. Mint. I simply assumed that the same amount was accumulated each year during the successive five-year periods. The numbers are small and do not vary drastically from one period to the next, so not much error is introduced by this assumption. However, I suspect that the initial estimates are subject to a large margin of error.

UMDS Purchases under the silver purchase laws of February 12, 1873, January 14, 1875, February 28, 1878, and July 14, 1890, are given in U.S. Secretary of the Treasury (1899, p. 207). For the first two purchase laws, only the total is given; I have assumed the amount purchased was the same in each month of the period for which each law was in effect. For the final two laws, figures are given for fiscal