If, however, more than one commodity, or more than one variety of the same commodity, is manu factured, this method of unit costs is clearly inaccurate, unless each unit of each line of product makes equal use of all factory facilities. Furthermore, if the cost of the material is used as a basis, it is very clear that a line of goods involving expensive material will be burdened more than one in which a less valuable ma terial is used. Thus, two pieces of jewelry may be manufactured, alike in every respect, and employing the facilities of the factory to exactly the same extent, one of them being made of gold, and one of some less valuable material—clearly, the product made of the more valuable material will be compelled, under this method, to carry rnor'e than its just share of expense.
It would be logical, again, to charge more of the foundry expense against a large casting than against a small one; but in the machine shop the large casting might have a very small amount of work done upon it, while the small one might require many hours of labor and machine work. In this case, therefore, the small casting might be entitled to a much heavier ma chine-shop expense than the larger one. Manufac turers using this method of distributing expense will experience difficulty in competing, in jobs involving much material and little work, against rival manufac turers using other and more logical methods.
Despite these obvious defects, however, the method is often useful and sufficiently accurate in handling such departments as foundries, and porcelain works attached to factories of the intermittent types, es pecially if the product of these departments is fairly uniform in character and size, and is passed thru in large lots for stock orders. The limitations of this system are most apparent where the product is varied in quantity and size.
4. Distribution by percentage on labor cost.—In distributing the expense by percentage on labor cost it is assumed that the burden varies directly with the direct labor expended upon the job. Thus, with the data assumed in Section 2 of this chapter, suppose that the total direct labor for the given period is $5,000. Then, in the preceding exam ple the expense which every dollar of direct labor should carry into the cost of the machine will be $5,000 = $1.00. The expense which the job in ques $5,000 tion must bear is therefore $300 X 1.00 = $300, and the total factory cost will be: This result, as might be expected, differs 'materially from the cost obtained with material as the basis, since there is no connection between the relative amount of material and labor that enter into the product. The amounts assumed above for these items, while taken at random, are entirely probable.
5. Advantages and defects.—In shops where the products are closely similar in kind and do not differ greatly in size this method may, in many cases, give results accurate enough to justify its use. In jus
tification of the method it is sometimes urged that the indirect labor varies quite closely with the direct labor, and that, therefore, the direct labor is a measure of expense. Granting that this may be true, tho it is not always so, it should be remembered that expense is composed of several other items besides indirect labor, and that these items neither are connected in any way with direct labor, nor vary directly with it. Because of its simplicity and because wages are such an important and evident part of the cost of produc tion, this method has been, and still is, in extensive use. It has, however, serious limitations.
First, this method does not discriminate between the cost of work done by a rapid workman and the cost of work done by one less rapid. Thus, a job which takes a rapid man, receiving sixty cents per hour, four hours to complete, is burdened with the same expense as a job done by a slower man, earning forty cents, and requiring six hours of his time. This method of distributing the expense, therefore, does not differentiate between these two men in the final shop costs. Yet the slower man has employed to a greater degree the shop facilities—power, heat, lighting, floor space, tools, etc.—and, very clearly, has caused the work to cost the firm much more than it did in the case of the more rapid workman. Then, too, because of the decreased rate of output due to the slower man, there is less total product over which to spread the expense, and each part must bear a greater proportion as the rate of production falls, since, as already explained, expense does not vary directly with product, but increases relatively as pro duction decreases.
This method also fails entirely to differentiate be tween the cost of large work and the cost of small work, and the error involved in its use may become large where the size of product and the size of the machinery required, vary widely. Thus, the expense charged to a job involving labor worth $10 by a mechanic using a file only, is exactly the same under this method as that laid upon a job involving the same labor charge, but done by a very large boring mill requiring the use of much greater floor-space, beat ing, lighting and power, as well, perhaps, as the serv ice of a large overhead crane. The work done on a very costly automatic machine will, under this method, bear practically no expense burden, tho it is evident that the interest charge, alone, is much greater for the automatic machine than for hand work, which may bear heavy expense charges. In brief, this method of distributing expense is an averaging method, and fails wherever the indirect expenses vary, since it does not give accurate results in proportion to the varia tion in the components of the expense to be dis tributed.