PREDETERMINATION OF COSTS—MATERIALS AND LABOR 1. General.—The functions of a cost system as a means of finding and recording costs, and also as a basis of managerial reports, have been discussed in the preceding sections. It remains to consider the third, and in some respects the most important, use of a cost system, namely, the predetermination and control of costs. It has been shown (see Section 15, Chapter XVIII) how the cost of an Article already produced may, perhaps, be reduced after studying the cost rec ords; and it will be clear that, in general, accurate costs are a powerful means of studying cost reduc tion. If the costs show a certain record of perform ance, there would seem to be no reason why this performance cannot be repeated. Experience shows, however, that such is not always the case; in fact, it is a rare thing to have two lots of product come out at the same cost, even tho produced on the same ma chines and by the same men, unless special care is taken to insure this result ( see Figure 24, page 280) .
Yet the ability to predict performance and costs be comes more important daily as competition becomes more keen and strenuous. A few years ago it was not uncommon practice in the case of construction jobs, to trust to the honor and ability of a contractor, and to reward him by paying him a percentage on costs. This form of contracting still survives to a limited extent tho, generally, the contractor is called upon to state a definite maximum price. In most cases, however, the contractors, builders or merchants who are supplying goods are required to bid on the contract, the lowest bidder securing the work. From the grocer who sells a pound of sugar, to the man who bids on the construction of a canal or a battleship, a guarantee of maximum price is usually demanded be fore any business is transacted; and the bargain is not infrequently cemented by legal agreements and bonds, or other financial guarantees of faithful performance.
2. Difficulties in predicting perf ormance.—When the contractor or merchant turns his eye inward on his enterprise, with a view to computing the price at which he can compete, he is, in general, confronted with the difficulties that have been discussed in the preceding chapters; and he must, in most cases, use extreme care in answering the momentous question—What is the cost of production? Material values he can usually compute with fair accuracy, but the case of labor is different. He finds that, in general, the price of
labor per diem is fairly definitely fixed; but he may find no relation between that price and the output per diem, and the vexing problem of expense is ever with him. If the work is similar to some he has already done, good cost records will give him an adequate idea of his limitations, provided he can insure a repetition of the performance. If the work is ,new in character, his costs may still be of great value, tho not based on this new class of work. The wide variation in almost any set of bids, even where the article is closely speci fied, is sufficient evidence of the difficulties that are to be met predicting costs of production. For these reasons, and for others, which have been discussed in the foregoing chapters, there is a rapidly growing de mand on the part of managers for more accurate methods of cost prediction than are generally avail able.
3. Estimating costs of production.—Costs of pro duction may be estimated for one of two purposes— namely, as a basis for bidding on work in the open market, or for fixing the price of production in the factory in advance of actual work. Estimating for the first purpose may here be considered, so far as cost finding bears upon this problem.
Estimating, of any kind, without the use of cost data can be little more than skilful guessing; unfor tunately, much of the estimating that is done is of this character. Where large margins of safety can be had in the way of good profits, these simple methods may suffice; but wherever competition is a factor, suc cessful bidding must rest on cost data of some kind. In many callings and industries, considerable data of this kind have been compiled, and may be found in books on estimating. It should be noted, however, that such data must be considered as representing average results only; a keen estimator will not be sat isfied with accuracy of his predictions until he pos sesses data representing the best practice of his own enterprise. In general, therefore, each enterprise must collect its own data in accordance with the prin ciples of cost finding which have been discussed in the preceding chapters.