"The theory that wages depend entirely on the efficiency of labor, or on the product of industry, is a new form of the old doctrine of the wages-fund."

Excerpts from "The Effect of an Eight Hours' Day on Wages and the Unemployed" by  Charles Beardsley, Jr. (The Quarterly Journal of Economics, Vol. 9, No. 4 (Jul., 1895), pp. 450-459):

The argument of workingmen that the general adoption of an eight hours' day would raise wages and absorb the unemployed is well known. A reduction in hours of work would be equivalent to the withdrawal from the ranks of men now employed of a certain number of laborers. The gap thus made would be filled by the unemployed. It is the competition of the fringe of unemployed or intermittently employed (comprising 10 per cent. of the working classes in England in normal years, according to Mr. Tom Mann) that keeps down the wages of the employed. If the number of the unemployed were lessened, wages might rise.

The reply which has been made to this argument by Mr. John Rae in his valuable and entertaining book, Eight hours for Work (1894), and by other writers, does not appear to be conclusive. It is said that the demand for work comes from the product of work, and that commodities constitute the demand for commodities. If the output of commodities falls off, the demand for them, and therefore for labor, must fall off also. So that (it is said), if a general reduction of hours resulted in a diminished national dividend, wages, instead of rising, would fall. In Mr. Rae's words,

The only way to increase the demand for labor all round is to increase the production of labor all round, and a general or serious diminution of production always causes a general or serious decrease in the demand for labor.… 

But, if all trades together were to restrict their output in the hope of distributing the work better, they would find they had merely less work to distribute; and, instead of making work for the unemployed, they would have unmade the work of a considerable portion of those now employed.… 

The effect of shorter hours on the general wages of labor depends entirely on their effect on production. If they lessen production generally, they will lower wages generally.

Mr. Rae's position seems perfectly clear, but it depends on a half-truth. Ceteris paribus, wages vary with the productiveness of industry, but only ceteris paribus. The theory that wages depend entirely on the efficiency of labor, or on the product of industry, is a new form of the old doctrine of the wages-fund. The characteristic feature of the classical doctrine was the assumption that the wages-fund was an inelastic quantum of the total circulating capital. The error of the theory that wages are measured by amount of product is in the implication that the proportion of wages to the total product of industry is at any given time rigidly fixed. According to the theory that wages are limited by capital, wages might rise if capital increased. According to the doctrine that wages depend on product, wages may rise if the product increases. Both theories ignore the fact that a change in the volume of the national dividend may be accompanied by a readjustment of the relative proportions of the shares in distribution which will neutralize, or more than neutralize, the effect of the change in the national dividend so far as any particular one of those shares is concerned. If the national dividend is diminished, the wages-fund will be diminished, profits will fall, interest and rent will be diminished, provided only that the relative magnitudes of wages, profits, interest, and rent remain unaltered. It does not follow that if shorter hours lessen, or tend to lessen, the national dividend, they will necessarily lessen the wages-fund. For the wages-fund is the product of two factors: it is the national dividend multiplied by a ratio.

Now, shorter hours of work would give to large numbers of laborers, at present unorganized or imperfectly organized, an opportunity which they are far from possessing. These workers are now under the tyranny of competition. They keep down their own wages by bidding against each other, or rather the casually employed keep down their own wages by bidding against each other, and the wages of the regularly employed by bidding against them, and standing ready to take their places at wages-current. To whatever degree, by a redistribution of work, this cutthroat competition could be mitigated, it would become possible to control the supply of labor, and to exact a monopoly price for it. In order to reduce the severity of this competition, or practically destroy it altogether, it would probably be good policy for the employed to divide even the present wages-fund with the unemployed. With the unemployed out of the way, effective united action on the part of laborers would be possible, and considerable advances in wages obtained, especially by the lower grades of unskilled workers.

But it has been said that, while a single trade may increase wages by regulating the supply of labor, all trades together cannot. This amounts to saying that a general rise in wages (relatively to the other shares in distribution) is impossible. It amounts, as I have already pointed out, to a doctrine of a rigid wages-fund. For, unless wages can be raised by checking competition among workingmen, they can hardly (relatively speaking) be raised at all in the present social order. There is no assurance that the constant growth of the national dividend, under a regime of unchecked competition, is accompanied by a corresponding increase in wages.