part at least, to general causes. We must, if the inference is to be at all accurate, go farther than this and assume that in the 22 articles the causes affecting the relative prices have, on the whole, balanced one another, and thus left the movement due to general causes open to measurement.1

To resume the main argument: however difficult it may be to measure the general rate of wages in the same country at different times, or in different countries at the same time, there is no doubt that we can distinguish between the causes that affect wages generally and those that affect some particular kind. The distinction was drawn and developed by Adam Smith, and has been adopted by all subsequent writers. Where the conflict of opinion arises is as to the nature and number of the general and special causes respectively.

In most economic treatises since Ricardo, by a "country" or "nation" is understood an area or population throughout which industrial competition may be considered as the

1 If it be assumed that mere changes in currency cannot ultimately disturb relative values, it follows that after time has been given for readjustment, such changes must have operated uniformly. If, then, a number of commodities be selected, the relative values of which have remained comparatively steady (that is to say, which have not suffered any considerable change in the conditions of production or demand), the movement in prices in these commodities must be ascribed to causes of a general kind. Hitherto, so far as I know, no one has attempted to select commodities with this principle definitely in view. "The only mode of eliminating these fluctuations," says Jevons, Currency and Finance, p. 26, "is to render our inquiry not more exclusive but more inclusive." The assumption is that the variations of some will compensate those of others.

If the surface of a lake or reservoir were still, the fall or rise during any period might plainly be measured at any point. If there were great waves and the measure were taken from a boat, an average of soundings must be made to determine the rise or fall in the depth of the water. Similarly the index numbers of particular commodities will furnish a measure of any general alteration in the height of prices. If relative values remained absolutely steady a movement in the price of any single commodity would measure the movement of the whole; but if relative values change a compensatory method must be adopted.

principal economic force in the distribution of wealth, and in which laws and customs are supposed to have comparatively little effect. Thus Mill: "Competition must be regarded in the present state of society as the principal regulator of wages and custom or individual character only as a modifying circumstance, and that in a comparatively slight degree." As already stated, however, in the present work, this view will be adopted as provisional only, in order to bring out by contrast the earlier stages of development. The theory of wages, then, may be divided into two parts, giving answers to two questions: 1. What are the causes which determine the general rate of wages? 2. Why are wages in some occupations and at some times and places above or below this general rate?

§ 8. The Wages-Fund Theory. With regard to the first question, Adam Smith, as in almost every important economic theory, gives an answer which combines two views which were subsequently differentiated into antagonism. "The produce of labour constitutes the natural recompense or wages of labour," is the opening sentence of his chapter on wages.2 But then he goes on to say that "this original state of things, in which the labourer enjoyed the whole produce of his own labour, could not last beyond the first introduction of the appropriation of land and the accumulation of stock." And he thus arrives at the conclusion that "the demand for those who live by wages, it is evident, cannot increase, but in proportion to the increase of the funds which are destined to the payment of wages." This is the germ of the celebrated wages-fund theory which was carried to an extreme by J. S. Mill and others; and, although Mill abandoned the theory some time before his death, he was unable to eradicate it from his systematic treatise, and to reduce it to its proper dimensions. It is important to observe that in the hands of

1 Bk. II., Ch. XI., § 1.

2 In discussing the general rate of wages I have made considerable use of my article "Wages" in the Encyclop. Brit.

Mill this theory was by no means, as was afterwards maintained by Professor Cairnes, a mere statement of the problem to be solved. According to Cairnes1 the wages-fund theory, as given in Mill's Principles,2 embraces the following statements: (1) The wages-fund is a general term used to express the aggregate of all wages at any given time in possession of the labouring population; (2) the average wage depends on the proportion of this fund to the number of people; (3) the amount of the fund is determined by the amount of general wealth applied to the direct purchase of labour. These propositions Cairnes easily reduces to mere verbal statements, and he then states that the real difficulty is to determine the causes which govern the demand and supply of labour. But the most superficial glance, as well as the most careful survey, will convince the reader of Mill's chapters on wages, that he regarded the theory, not as the statement, but as the solution of the problem. For he applies it directly to the explanation of movements in wages, to the criticism of popular remedies for low wages, and to the discovery of what he considers to be legitimate and possible remedies. In fact, it was principally on account of the application of the theory to concrete facts that it aroused so much opposition, which would have been impossible if it had been a mere statement of the problem.

The wages-fund theory, as a real attempt to solve the wages question, may be resolved into three propositions, which are very different from these verbal truisms. (1) In any country, at any time, there is a determinate amount of capital unconditionally destined for the payment of labour. This is the wages-fund. (2) There is also a determinate number of labourers who must work independently of the rate of wages, that is, whether the rate is high or low. (3) The wages-fund is distributed amongst the labourers solely by means of competition, 1 Leading Principles of Political Economy. 2 Bk. II., Ch. XI., § 1.

masters competing with one another for labour, and labourers with one another for work, and thus the average rate of wages depends on the proportion between wagecapital and population. It follows then, according to this view, that wages can only rise either owing to an increase of capital or a diminution of population, and this accounts for the exaggerated importance attached by Mill to the Malthusian theory of population. It also follows from the theory, that any restraint of competition in one direction can only cause a rise of wages by a corresponding fall in another quarter, and in this form it was the argument. most frequently urged against the action of trade unions. It is worth noting, as showing the vital connexion of the theory with Mill's principles, that it is practically the foundation of his propositions on capital in his first book, and is also the basis of the exposition in his fourth book of the effects of the progress of society on the condition of the working-classes.

It has often been remarked that, in economics as in other sciences, what eventually assumes the form of the development of, or supplement to, an old theory, at first appears as if in direct antagonism to it, and there is reason to think that the criticism of the wages-fund theory was carried to an extreme, and that the essential elements of truth, which it contains, were overlooked. In many respects the theory may be regarded as a good first approximation to the complete solution of the problem. The causes which it emphasises too exclusively are after all veræ causæ, and must always be taken into account. There can be no doubt, for example, that under certain conditions, a rapid increase in the labouring population may cause wages to fall, just as a rapid decline may make them rise. The most striking example of a great improvement in the condition of the labouring classes in English economic history is found, as already shown, immediately after the occurrence of the Black Death in the middle of the fourteenth century. The sudden and extensive thinning of the ranks of

labour was manifestly the principal cause of the great improvement in the condition of the survivors.

Again, as, regards the amount of capital competing for labour, the reality of the cause admits of no dispute, at any rate in any modern society. The force of the element is, perhaps, best seen by taking a particular case, and assuming that the general wages-fund of the country is divided into a number of smaller wages-funds. Take, for example, the wages of domestic servants, when the payment of wages is made simply for the service rendered. We may fairly assume that the richer classes of the community practically put aside so much of their revenue for the payment of the wages of their servants. The aggregate of these sums is the domestic wages-fund. Now, if owing to any cause, the amount available for this purpose falls off, whilst the number of those seeking that class of employment, remains the same, the natural result would be a fall in wages. It may, of course, happen in this, as in other cases, that the result is not so much a direct fall in the rate of wages, as a diminution of employment — but even in this case, if people employ fewer servants, they must do more work. Again, if we were to seek for the reason why the wages of governesses are so low, the essence of the answer would be found in the excessive supply of that kind of labour compared with the funds destined for its support. And similarly, through the whole range of employments in which the labour is employed in perishable services and not in material products, the wages-fund theory brings into prominence the principal causes governing the rate of wages, namely, the number of people competing, the amount of the fund competed for, and the effectiveness of the competition. This view, also, is in harmony with the general principles of demand and supply. If we regard labour as a commodity and wages as the price paid for it, then we may say that the price will be so adjusted that the quantity demanded will be made equal to the quantity offered at that price,