All the main elements of the classical wages-fund doctrine were present in Mill's Principles, including the assumption of a point-input-point-output production process. Mill assumed that the present remuneration of labor was the consequence of past applications of capital and labor, and he believed that a proportion of total output was destined for labor in advance of production. He further perceived and applied the doctrine at an aggregate level and in real terms. This constitutes his theoretical view of the doctrine.
By 1869, Mill had altered his views on the wages-fund doctrine, and there has been a great deal of controversy as to why he did so. Several explanations deal exclusively with considerations other than Mill's theoretical views. Since Mill recanted the doctrine on the occasion of his review of a book by W.T. Thornton, one reason given for the recantation is his friendship with Thornton. Another explanation given is Mill's late but deep involvement in social reform. A third reason often cited is the combined influence on Mill of his wife, Harriet Taylor, and of the philosopher Auguste Comte. While none of these influences should be discounted completely, it seems clear from a close examination of Mill's writings that by 1869 he had changed his theoretical view of the wages-fund. It is on this last matter that attention will be focused in this chapter.
The central issue in the recantation concerned the fixity of the fund earmarked for the payment of labor. The idea of a fixed fund in the short run implied that in the aggregate, workers could claim no more in wage payments than an amount that would exactly deplete the fund. Thus the doctrine of the wages-fund was frequently used to demonstrate the futility of efforts by labor unions to raise their aggregate compensation. The long run was a different matter—no classical economist argued that the fund was fixed over the long run. However, some advanced the argument that if labor unions were too aggressive in pushing their claims, profit expectations would decline, so that in the future less capital would flow into the fund, thereby reducing real wages somewhere down the road. In later life, Mill became sympathetic to labor unions, and this may have been the impetus that led him to reexamine the concept of the wages-fund, in particular the subject of its short-run fixity.
In his 1869 review of Thornton's book On Labour, Mill undid the fixity assumption. Of the aggregate amount that is spent on wages, Mill asserted only that there is some upper limit. He wrote:
.. .there is an impassable limit to the amount which can be so expended; it cannot exceed the aggregate means of employing classes. It cannot come up to those means; for the employers have also to maintain themselves and their families. But, short of this limit, it is not, in any sense of the word, a fixed amount ("Thornton on Labour and Its Claims," p. 516).
Mill's argument advanced to the point that he divided the employer-capitalist's means into two parts: his capital and his income on that capital. While the former is usually equated in classical economic nomenclature with the wages-fund, Mill argued that the capitalist could add to that amount by discretionary reductions of his income. The capitalist, in other words, might respond to exogenous variables (e.g., union pressure, different profit expectations, etc.) in such a way that he voluntarily reduced expenditures on himself and his family in order to spend more on labor. In this way, Mill apparently thought that labor unions might be able to redistribute income in favor of the workers. Unfortunately, Mill's argument did not distinguish between money wages and real wages, nor between short-run and long-run effects. Consequently, his recantation does not rest on a sound theoretical footing.
A Short-Run Wages-Fund Model
In order to expose the deficiencies of Mill's recantation, we shall place it in the context of a short-run wages-fund model that rests on the usual classical assumptions. Those assumptions are:
1 Production takes place within a point-input-point-output production process.
2 The entire output of the economy is composed of fixed capital, wage goods, and capitalist consumables. There is, moreover, no transference of demands between markets; i.e., wage earners do not transfer demands to capitalist consumables and vice versa.
3 Production in all industries is marked by a constant ratio of fixed to circulating capital.
4 Perfect competition (i.e., constant costs of production) exists everywhere.
5 The money supply is fixed for the term in question.
6 Population and productivity remain unchanged during the period in question.
Under these assumptions, the aggregate stock of goods in real terms during any period, say t1 is determined by past production and cannot be increased during t1,. In real terms, consumption and investment decisions are made at the beginning of the period (i.e., the end of t0) and the entire stock of goods is depleted by the end of the period, albeit at different rates of use. For example, consider Figure 5, where the total stock of goods at the end of t0 is represented by OY0, divided so that OM0 is equal to fixed capital (e.g., machinery), M0W0 is equal to wage goods available for purchase by workers, and W0Y0 is equal to capitalist consumables. This tripartite division conforms to Mill's representation. Under the usual assumptions of the wages-fund theory, these various stocks are used up during the production period, so that at the end of t1 each has fallen to zero.
Now let us examine the effects of a decision by the capitalist to reduce his own real income (W0Y0) in order to spend more on labor, the prospect Mill raised in his recantation. The effects of this redistribution of income are carried through in Figures 6a and 6b. The former depicts the market for goods that are bought by workers, the latter the market for goods purchased by capitalists. Under the rigid supply conditions of the wages-fund model, output in each period is fixed and determined by the previous period. Thus the supply curves in Figures 6a and 6b are vertical lines. A voluntary reduction in real income by capitalists will cause the demand for capitalist consumables to shift to the left, lowering the average price of such goods from Pc to P'c. Pari passu, an increase in workers' real income will shift the demand for wage goods to the right, thereby raising the average price of those goods from Pw to P'w.
The conclusion of this analysis is that under the assumptions of the classical wages-fund doctrine, the effects of any reallocation of funds by capitalists in favor of labor are solely upon prices in the two markets. Furthermore, given a constant money stock and velocity, the price changes in the two markets will be proportionate in opposite directions, so that the aggregate price level will not be affected. More important from the standpoint of the laboring classes, the increase in money wages occasioned by the transfer of income from the capitalist class produces a price increase in wage goods that offsets the rise in money wages. Real wages remain unaffected by the transfer. Since Mill implied that workers would be better off under such a transfer, it seems clear that he confused real wages with money wages.
The nature of the long-run adjustments that would accompany the kind of income redistribution just considered does not hold any brighter prospects for permanent increases in real wages. Given price changes in the two markets traced above, higher profits in the wage-goods industry would signal new firms to enter whereas lower profits in the capitalist-consumables market would encourage some firms to exit. These long-run changes can be envisioned by shifting the vertical supply curve to the right in Figure 6a and to the left in Figure 6b. Under constant-cost conditions, price would tend to return to Pw in Figure 6a and to Pc in Figure 6b. The adjustments in each market might be lengthy, but the tendency would be for prices to return to their former level before the income transfer Mill suggested. The point that Mill seemed to forget (or deny) in his 1869 recantation is that in the classical world, permanent increases in real wages are traceable to real factors only, such as improvements in technology or some other increase in worker productivity. Mill seems to have been victimized by the idea that aggregate welfare can be improved by taking income from one group and giving it to another, an idea that seems to die hard, as witnessed by contemporary American and British experience with government-forced attempts to redistribute income.
Cairnes's Last Stand
In 1874, J.E. Cairnes, in the face of the attacks by Longe and Thornton and of Mill's recantation, attempted to rehabilitate the wages-fund doctrine and to revive classical theory. In Some Leading Principles of Political Economy Newly Expounded, Cairnes defended the validity of the doctrine. The real issue was not the determinacy or indeterminacy of the fund, but simply what the capitalist qua capitalist did in the real world. Cairnes did not view the capitalist as setting aside a specific amount for wages, but as long as he acted as a capitalist and made investments, a wages-fund would exist.
Determinants of the Fund
Cairnes went into great detail on the determinants of the individual capitalist's investment, and he specifically mentioned the amount of his total means, the capitalist's time preference, and the opportunity for making a profit. The investment determinants for society as a whole parallel Mill's, i.e., the "effective desire of accumulation" and the extent of the field for investment. Total investment, which includes investment in fixed capital and raw materials, does not give an accurate understanding of the wages-fund, however, and so Cairnes was led to investigate the proportion of wages to total capital. Here he made some incisive statements. The proportion of wages to total capital, Cairnes maintained, is determined by three factors: (1) the nature of national industries, or the economy's production function; (2) the total capital of the country; and (3) specifically the supply of labor.
On the first point Cairnes was very clear. As he noted:
If, for example, his [the capitalist's] purpose is to engage in cotton or woollen manufacture, a very large proportion of his whole capital will assume the form of buildings, machinery, and raw wool or cotton... of fixed capital and raw material, which would leave a correspondingly small proportion available for the payment of wages (Some Leading Principles, p. 170).
Other things being equal, in agricultural nations, where labor inputs are large, one would expect a large proportion of total capital to go to labor. But given the state of the economy's production function, the price of inputs determines factor shares. Supply-and-demand conditions for inputs obviously affect input prices, and Cairnes believed that the demand for labor was inelastic, since presumably investment was predetermined by other factors. Shifts in the supply of labor, however, were inversely related to the size of the wages-fund, though Cairnes thought that such shifts were only minor influences on the size of aggregate wages.
Clearly Cairnes was into an explanation of the state of wages over time. Ignoring the discrete production period assumption, Cairnes looked to the growth or decline of aggregate capital and to the nature of national industries as prime variables in the determination of wages. In this matter he adapted to the Ricardian position that a growth in fixed capital relative to total capital would cause profits and investment to fall over time. Technology would forestall the stationary state, but the latter would arrive nonetheless. The long-run prospects for laborers were bleak unless, of course, they saved and became part of the entrepreneurial class.
Cairnes's discussion of the long run is perfectly consistent with classical positions on the wages-fund, but his defense of the doctrine broke down when he turned (again following Mill) to a discussion of the trade union issue. Unfortunately, at this juncture Cairnes fell into Mill's error of identifying the fund with a money amount. This problem is implicit in Cairnes's statement that:
If beyond the amount actually spent on wages at any given time there be an indefinite margin of wealth which workmen by judicious combination may conquer; then it is evident Trades-Unionism has a great field before it, and workmen will naturally and properly look to this agency as the principal means of improving their condition (Some Leading Principles, p. 214).
This "indefinite margin of wealth" is clearly a money amount in the hands of capitalists.
In short, wages can be higher or lower only within limits. Subsistence, of course, sets the lower limit. If union demands go beyond the upper limit, entrepreneurial investment will cease, and the fund will decline. But when profits are within these limits, union leaders can bargain successfully.
Cairnes's Impact on Classical Economics
Where does Cairnes's defense leave the wages-fund doctrine? In terms of the short run, Cairnes would have to deny that wages are fixed, since he viewed the fund as a money amount. In real-goods terms, of course, the fund is fixed as of any moment, and given the discrete time period of production assumption, the average wage is fixed over the period. Critics and defenders of the doctrine simply failed to come to grips with a correct, albeit naive, theoretical construct.
As a believer in the classical stationary state, Cairnes was very pessimistic concerning the long-run outcome of the condition of the laboring classes, especially in Great Britain. The declining rate of profit and the growth of land rent were the limits to progress. As he noted:
Against these barriers Trades-Unions must dash themselves in vain. They are not to be broken through or eluded by any combination, however, universal, for they are barriers set by nature herself (Some Leading Principles, p. 283).
However, Cairnes was not totally devoid of suggestions for labor. First, he admonished labor for squandering higher real wages on larger populations, and he urged "self-restraint," as Malthus had. Second, laborers should attempt to become part of the capitalist class by saving to augment the fund.
The laborer must learn to save, and Cairnes offered an empirical example of how this might be done. Calculating that £120 million was spent on alcoholic drinks per year, Cairnes attributed the largest portion of the expenditure to laborers. (He offered the further judgment that three-fourths of it was both physically and morally injurious to them.) Thus Cairnes concluded that half of this sum could be saved and channeled into labor cooperatives. Though such massive investment would go far toward improving labor's position in income distribution, Cairnes was not especially sanguine concerning his proposals. With typically Victorian disdain he noted:
What workmen have to overcome in order to engage effectively in cooperative industry is, first, the temptation to spend their means on indulgences generally pernicious, and which at all events may without detriment be dispensed with; and, secondly, the obstacles incident to their own ignorance and generally low moral condition (Some Leading Principles, pp. 289-290).
Indeed, by identifying the fund with a money amount, Cairnes expanded the confusion established by Mill and by the doctrine's critics. Many of his other writings on distribution (his adherence to a residual theory of wages rather than profits, for example) are similarly marred by confusion. The decline of classical economics was thus in evidence, not only because a competitive theory (marginalism) was in ascendance, as we shall soon see, but also because important members of the tradition misinterpreted some of its fundamental precepts. J.E. Cairnes, although he made advances in other areas (particularly in scientific methodology), may have undermined the theoretical tradition he fought so energetically to protect.