The method of percentage on man-hours would seem to be more logical, in general, than that known as percentage on labor for other reasons. A great many of the principal items of indirect cost as, for instance, heat, light, depreciation, rent, insur ance, taxes, interest, etc., are more fairly propor tional to time than to wages or labor costs. There is no reason for assuming that the proportion of these items will be high where wages are high, and low where wages are low; in fact, such items as super vision are likely to be highest where labor costs are lowest. The interest on investment may be very high where low-priced labor is employed in operating auto matic and semi-automatic machines.
On the other hand, this method is no better than that of percentage on labor, so far as differentiation between the use of equipment of different value and size is concerned. The same hourly rate is applied to a hand worker as is applied to a $20,000 boring mill. This results in overcharging small, cheap product, and undercharging the larger and more expensive work. When applied over an entire factory which is engaged in making a wide range of work, competi tion with the small articles of product becomes diffi cult if such competition is against other manufactur ers specializing in these small articles.
It will be evident that this method can be applied with accuracy only when the work is of fairly uniform size and character, but it should also be noted that the errors of this method, like those of the percentage on-labor plan, can be obviated to some extent by care ful departmentization. Where tools and processes can be grouped so as to obtain equal conditions for similar types of machinery and similar processes, this plan can often be used with success. Like the pre ceding systems it has the virtue of simplicity and in many cases this is no small advantage.
10. Inadequacy of foregoing methods.—It will be clear that under any of the plans of distributing ex pense that have been discussed an averaged result is ultimately obtained. It is clear, also, that the total expense can be distributed with certainty by any of these methods. None of them, however, takes ac count of the fact that expense does not weigh equally upon all parts of the productive activities. It is for this reason that these methods apply with some semblance of accuracy only where the conditions are more or less uniform, where the wages paid do not vary greatly, where the size and character of the equip ment do not differ, and where the lines of product are not diverse in character. In other words, where
the conditions are simple, only simple methods are required. It is for this reason, also, that any of the methods that have just been discussed apply more closely when the factory can be departmentized so as to group machines and processes of similar kind and equal size together. Where this can be done, and each department is accurately charged with its share of expense, almost any method of dividing the depart mental expense will give fairly satisfactory results, since each man or tool in each department is, gener ally speaking, equally taxable. Minute department ization of this character is not usually possible, how ever, and where these simple conditions do not obtain, and accurate costs are desired, the averaging methods so far described are not applicable with any degree of accuracy.
The reasons explanatory of this fact are funda mental. When factories were smaller, and machine processes did not constitute so large a factor in manu facturing, wages or time did, no doubt, serve as a satisfactory measure of expense, especially where only rough costs were required. But neither wages nor labor hours can be taken as a measure of expense in modern factories where the product varies as to both size and character. This is markedly true where machine processes constitute a large portion of the cost of production. The elements of expense that attach themselves to a handworker may be, and gen erally are, very different in value from those connected with an expensive automatic machine. The range of processes found in some modern factories is almost bewildering; so much so, in fact, that any distribu tion of expense by averaging methods may be little short of a wild guess, so far as accurate allocation is concerned. This complexity is still further increased when machines and processes are used intermittently in ever changing lines of product. It is true, of course, that in all complex cases a certain amount of averaging must be done, but this can be minimized and the results made more accurate than those ob tained by the methods heretofore outlined.