"Where one portion of the workman's labor consists in the exertion of mere physical force, as in weaving, and in many similar arts, it will soon occur to the manufacturer, that if that part were executed by a steam-engine, the same man might in the case of weaving, attend to two or more looms at once; and, since we already suppose that one or more operative engineers have been employed, the number of looms may be so arranged that their time shall be fully occupied in keeping the steam-engine and the looms in order.

"Pursuing the same principles, the manufactory becomes gradually so enlarged, that the expense of lighting during the night amounts to a considerable sum; and as there are already attached to the establishment persons who are up all night, and can therefore constantly attend to it, and also engineers to make and keep in repair any machinery, the addition of an apparatus for making gas to light the factory leads to a new extension, at the same time that it contributes, by diminishing the expense of lighting, and the risk of accidents from fire, to reduce the cost of manufacturing.

"Long before a factory has reached this extent, it will have been found necessary to establish an accountant's department, with clerks to pay the workmen, and to see that they arrive at their stated times; and this department must be in communication with the agents who purchase the raw produce, and with those who sell the manufactured article." It will cost these clerks and accountants little more time and trouble to pay a large number of workmen than a small number; to check the accounts of large transactions, than of small. If the business doubled itself, it would probably be necessary to increase, but certainly not to double, the number either of accountants, or of buying and selling agents. Every increase of business would enable the whole to be carried on with a proportionally smaller amount of labor.

As a general rule, the expenses of a business do not increase by any means proportionally to the quantity of business. Let us take as an example, a set of operations which we are accustomed to see carried on by one great establishment, that of the Post Office. Suppose that the business, let us say only of the London letter-post, instead of being centralized in a single concern, were divided among five or six competing companies. Each of these would be obliged to maintain almost as large an establishment as is now sufficient for the whole. Since each must arrange for receiving and delivering letters in all parts of the town, each must send letter-carriers into every street, and almost every alley, and this too as many times in the day as is now done by the Post Office, if the service is to be as well performed. Each must have an office for receiving letters in every neighborhood, with all subsidiary arrangements for collecting the letters from the different offices and re-distributing them. I say nothing of the much greater number of superior officers who would be required to check and control the subordinates, implying not only a greater cost in salaries for such responsible officers, but the necessity, perhaps, of being satisfied in many instances with an inferior standard of qualification, and so failing in the object.

Whether or not the advantages obtained by operating on a large scale, preponderate in any particular case over the more watchful attention and greater regard to minor gains and losses, usually found in small establishments, can be ascertained, in a state of free competition, by an unfailing

Wherever there are large and small establishments in the same business, that one of the two which in existing circumstances carries on the production at greatest advantage, will be able to undersell the other. The power of permanently underselling can only be derived from increased effectiveness of labor; and this, when obtained by a more extended division of employment, or by a classification

tending to a better economy of skill, always implies a greater produce from the same labor, and not merely the same produce from less labor; it increases not the surplus only, but the gross produce of industry. If an increased quantity of the particular article is not required, and a part of the laborers in consequence lose their employment, the capital which maintained and employed them is also set at liberty; and the general produce of the country is increased, by some other application of their labor.

Another of the causes of large manufactories, however, is the introduction of processes requiring expensive machinery. Expensive machinery supposes a large capital, and is not resorted to except with the intention of producing, and the hope of selling, as much of the article as comes up to the full powers of the machine. For both these reasons, wherever costly machinery is used, the large system of production is inevitable. But the power of underselling is not in this case so unerring a test as in the former, of the beneficial effect on the total production of the community. The power of underselling does not depend on the absolute increase of produce, but on its bearing an increased proportion to the expenses; which, as was shown in a former chapter,* it may do, consistently with even a diminution of the gross annual produce. By the adoption of machinery, a circulating capital, which was perpetually consumed and reproduced, has been converted into fixed capital, requiring only a small annual expense to keep it up; and a much smaller produce will suffice for merely covering that expense and replacing the remaining circulating capital of the producer. The machinery therefore might answer perfectly well to the manufacturer, and enable him to undersell his competitors, although the effect on the production of the country might be not an increase but a diminution. It is

* Supra, chap. vi. p. 114.

true, the article will be sold cheaper, and therefore, of that single article, there will probably be not a smaller, but a greater quantity sold; since the loss to the community collectively has fallen upon the work-people, and they are not the principal customers, if customers at all, of most branches of manufacture. But though that particular branch of industry may extend itself, it will be by replenishing its diminished circulating capital from that of the community generally; and if the laborers employed in that department escape loss of employment, it is because the loss will spread itself over the laboring people at large. If any of them are reduced to the condition of unproductive laborers, supported by voluntary or legal charity, the gross produce of the country is to that extent permanently diminished, until the ordinary progress of accumulation makes it up; but if the condition of the laboring classes enables them to bear a temporary reduction of wages, and the superseded laborers become absorbed in other employments, their labor is still productive, and the breach in the gross produce of the community is repaired, though not the detriment to the laborers. I have restated this exposition, which has already been made in a former place, to impress more strongly the truth, that a mode of production does not of necessity increase the productive effect of the collective labor of a community, because it enables a particular commodity to be sold cheaper. The one consequence generally accompanies the other, but not necessarily. I will not here repeat the reasons I formerly gave, nor anticipate those which will be given more fully hereafter, for deeming the excep tion to be rather a case abstractly possible, than one which is frequently realized in fact.

A considerable part of the saving of labor effected by substituting the large system of production for the small, is the saving in the labor of the capitalists themselves. If a hundred producers with small capitals carry on separately

the same business, the superintendence of each concern will probably require the whole attention of the person conducting it, sufficiently at least to hinder his time or thoughts from being disposable for anything else; while a single manufacturer, possessing a capital equal to the sum of theirs, with ten or a dozen clerks, could conduct the whole of their amount of business, and have leisure too for other occupations. The small capitalist, it is true, generally combines with the business of direction some portion of the details, which the other leaves to his subordinates; the small farmer follows his own plough, the small tradesman serves in his own shop, the small weaver plies his own loom. But in this very union of functions there is in a great proportion of cases a want of economy. The principal in the concern is either wasting, in the routine of a business, qualities suitable for the direction of it, or he is only fit for the former, and then the latter will be ill done. I must observe, however, that I do not attach, to this saving of labor, the importance often ascribed to it. There is undoubtedly much more labor expended in the superintendence of many small capitals than in that of one large capital. For this labor, however, the small producers have generally a full compensation, in the feeling of being their own masters, and not servants of an employer. It may be said, that if they value their independence they will submit to pay a price for it, and to sell at the reduced rates occasioned by the competition of the great dealer or manufacturer. But they cannot always do this and continue to gain a living. They thus gradually disappear from society. After having consumed their little capital in prolonging the unsuccessful struggle, they either sink into the condition of hired laborers, or become dependent on others for support,

§ 2. Production on a large scale is greatly promoted by the practice of forming a large capital by the combination of