This section is from the "Practical Banking" book, by Albert S. Bolles.
Another guard might be established. The depositor might be required to make two tickets describing his deposits, leaving one with the discount clerk or other designated official, and the money and second ticket with the receiving teller. Then the general bookkeeper might post from one set of tickets and the individual bookkeepers from the second set, and thus a good system of checks be established.
Another thing can be done, namely, after counting the money of the depositor, to return it and have him deposit it in a proper receptacle, and which the receiving teller cannot open. If this were done he would have no money whatever in his possession, save during the count. But then it would be necessary for somebody to handle it, and when this was done the risk of loss would be incurred. Many of the recent defalcations have been by receiving tellers, who have made false entries of the amount received. Some of them would have been sooner detected by writing up the books of depositors more frequently, or by having them compared with the books of the bank by other officials than those who made the entries. The practice of writing up the books of depositors with regularity is falling into disuse; a reform in this regard ought to be made at once. Moreover, why could not a rule be made, or statute enacted, if necessary, requiring depositors to present their books within stated periods—thirty, sixty, or other days—and, if not ob served, absolving the bank for all losses from error that may have occurred? It may be fairly presumed that errors could be more easily corrected soon after their happening than they can be at a much later time. Surely a bank ought not to suffer for the contributory negligence of its depositors.
Now, let us consider the paying teller. He must be entrusted with money for making payments. But, why could not a clerk or other person be detailed every day for counting the paying teller's cash in the morning before beginning business, and after his proof is made at the close of the day, for the purpose of finding out whether any error or fraud has been committed? What is the present system? The paying teller makes out his proof at the close of the day and submits it to his superior officer; but whether his cash corresponds with his statement, who knows? When Scott, the paying teller of the Manhattan Bank of New York, ran away with a large amount, a few months ago, it was assumed that he took it during his last day of service, but on what reason is the assumption founded? May he not have abstracted money from time to time for a long period—from the time, in fact, since the last count had been made by others? Well knowing that his cash was recounted only at considerable intervals, he was quite safe in helping himself if he desired. This judgment, therefore, of a sudden robbery is not tenable, so long as the present system is continued; whether the paying teller's proof is correct or not is purely a matter of faith. Why is it not practicable and also desirable to designate a man every day to go over the paying teller's cash after him, and thus make more certain the truth of his statement? Of course, if the examiner and paying teller should combine to commit fraud, it would be just as easy as before. We would suggest the designation of a different person from day to day: sometimes the president doing it, sometimes the cashier, and, again, a director might be willing to make an examination of that kind; and so, by varying the examination constantly, might we not with reason hope that the paying teller would exercise greater care in making up his statement than he does at present.
In suggesting such an examination to a bank officer, not long since, he remarked that no paying teller would submit to it; but another bank officer of longer experience remarked that a paying teller who would not submit was quite unfit for his place, that no honest officer objected to any investigation, however searching or frequent. If it were general, and not spasmodic or casual, and applied to all paying tellers without distinction, it is believed they would not object to the examination. We all know that examinations are constantly made, of one kind or another; moreover, it should be remembered that all checks and investigations are for a twofold purpose—to discover errors as well as frauds, for it is inevitable that, in conducting a business with the utmost care, errors will occur, and investigations are always in order to prevent and discover them.
A similar examination might be daily conducted of the receiving teller's cash. By counting the amount in the morning and the first set of depositors' tickets above mentioned, and his cash at the close of the day, it would seem to be impossible for him to commit fraud, except through collusion.
Another precaution that may be mentioned in this connection, is that of keeping the largest possible amount of money under double lock and key, so that conjoint action will be necessary to get it.
Some banks keep their reserve in this way, but why not extend the precaution further? Each bank, of course, must determine for itself how far it can go in this direction. Probably a little thought bestowed on this matter would be quite enough to show the practicability of diminishing very largely the amount of ready money in the absolute possession of the receiving or paying teller. Why would it not be practicable, whenever a given sum was received, say, $15,000, $20,000, or $25,000 or more, to have it counted, and, perhaps, recounted by another, and labeled, and put away under double lock and key, and on the other hand, of giving to the paying teller a comparatively small sum from time to time as the occasion required.