United States Fidelity & Guaranty Co. v. Federal Reserve Bank of New York

HAIGHT, District Judge: Plaintiff Union Trust Company of Maryland (“Union Trust”) and its insurer, United States Fidelity and Guaranty Company, brought this action to recover damages caused by a clever check fraud perpetrated upon Union Trust by a nonparty. Following denial of its motion to dismiss, in an opinion reported at 590 F.Supp. 486 (hereafter cited as “USF & G /”), defendant Federal Reserve Bank of New York (“New York Fed”) impleaded third-party defendants State Bank of Albany (“Albany State”), 1 Philadelphia National Bank (“PNB”), and First Pennsylvania Bank (“First Penn”). Soon after filing of the third-party complaint, plaintiffs amended their complaint to assert claims against Albany State. It was unnecessary for them to assert claims against the remaining third-party defendants because the Court accepted for transfer and consolida *363 tion a suit which plaintiffs had previously-filed against those two Pennsylvania banks in the Eastern District of Pennsylvania, United States Fidelity and Guaranty Co. v. Philadelphia National Bank, No. 83-1304 (E.D.Pa.). That action was redesigna-ted civil action no. 84 Civ. 6950 (CSH) following the transfer. Since decision of the motion to dismiss, all parties have engaged in extensive and revealing discovery. Following the completion of discovery, defendant New York Fed initiated the current round of motions by moving for summary judgment on plaintiffs' claims. Albany State filed a similar motion soon after. Plaintiffs opposed both motions and moved to amend their complaint against New York Fed. See note 9, infra. Also before the Court are three motions for summary judgment which were pending in Pennsylvania district court at the time of transfer of the related case. These are cross-motions by plaintiffs and First Penn and a motion on plaintiffs’ claims by PNB. Because the facts are for the most part undisputed, these motions together present most of the legal issues which would be encountered at trial, if the complaints survive the present motions. I begin with a review of the facts revealed by discovery. I. As described more fully in USF & G I, 590 F.Supp. at 489-91, Union Trust was fraudulently induced to permit a depositor to withdraw funds against a worthless check. In April 1980, a man who called himself Marvin Goldstein established a checking account with Union Trust. Soon after, he deposited a check for over $880,-000 in the account. The account upon which the check purported to be drawn did not exist, but a clever manipulation of the numerals on the face of the check caused it to be routed among a number of New York and Pennsylvania banks before being returned to Union Trust as uncollectible. In the meantime, Union Trust had permitted Goldstein to withdraw a substantial amount of cash from his account, having assumed from the lapse of time that the check had been paid. 2 The foregoing information was pleaded in the original complaint, and discovery has confirmed its accuracy. The interesting details unearthed in discovery primarily concern not the behavior of the defendant banks in routing the bogus check but that of Union Trust’s employees in accepting it and releasing fundá against it. A summary of that new information follows. On April 16, 1980, Goldstein walked into a Baltimore branch of Union Trust. He told the branch manager, John Gemmill, that he and his father were precious metals dealers and that he planned to establish a Baltimore office of his father’s New York business, Goldstein Precious Metals and Stones. In preparation, he sought to open a checking account with Union Trust in the name of the business. In opening the account, Goldstein produced one piece of personal identification, a New Jersey driver’s license, and a New York certificate of business proprietorship, and supplied a New York bank reference. Gemmill recorded this and other information on a “New Account Information Form.” Goldstein opened the new account with a cash deposit of $15,000. Gemmill then turned the new account form over to assistant branch manager John Clement with instructions to prepare two signature cards. Clement did so, but he unaccountably neglected to transcribe the bank reference from the new account form to the signature cards. The form and one card were then sent to central Union Trust files, while one card was retained at the local branch. One week later, Goldstein returned to the branch and withdrew $14,000 from his account, reducing his balance to $1,000. *364 Little was heard from him until May 6, when he deposited a check for $880,000 at a second Union Trust branch located a few blocks from the branch with which he had opened the account. Deposit of such a large check ordinarily triggers self-protective internal alerts at a bank, and Union Trust was no exception. Tellers who received for deposit checks over $100,000 were supposed to notify the branch manager, according to written Union Trust procedural guidelines. The manager was then to decide whether a “hold” should be placed on the check — that is, whether the depositor should be denied access to the deposited funds for an extended period of time in order to permit the bank to confirm the collectibility of the check. In the absence of such a hold, funds ordinarily become available to the depositor within one to two days of deposit of the check. The teller who accepted Goldstein’s check, however, neither notified bank officers of the large deposit nor placed a hold on it. The teller, however, was merely the bank’s first line of defense against fraud. During the next few days, Gemmill, the branch manager, was reviewing a document known as a “balance fluctuation report,” designed to alert bank officers to unusually large balance changes in the accounts under their supervision. The leap in the Goldstein Precious Metals account balance from one thousand to nearly a million dollars naturally caught his attention, and he decided to investigate. Gemmill first requested, or had Clement request, credit reports on the Goldstein business from two national credit reporting services. Both services reported that they had no record of Goldstein Precious Metals and Stones. Upon receiving this information, Gemmill had Clement retrieve the Goldstein signature card in order to pursue the credit references which would ordinarily be listed on it. Because Clement had neglected to transfer the reference from the account form, however, the signature card was no help, and neither bank employee pursued this further. 3 Wisely, Gemmill instructed Clement to look into the $880,000 check. The check was drawn on an account at First Penn of a company called Metropolitan Investment Corporation. On instructions from Clement, an employee of Union Trust’s credit department called officials at First Penn and was told that no such account existed. First Penn had no banking relationship with Metropolitan Investment Corporation. This information was relayed to Clement, who told Gemmill. All of the foregoing occurred on or before Friday, May 9, within four days of the check’s May 6 deposit. According to his deposition testimony, Gemmill, who was leaving for vacation on May 9, told Clement to report to senior bank officers any activity in the Goldstein account, perhaps recognizing that the First Penn report made it unlikely that the Gold-stein check would be honored. Clement remembers no such instructions. Either way, no other action was taken at that time to prevent Goldstein from withdrawing funds against the check. On Monday, May 12, Gemmill was temporarily replaced by another bank officer. The next day, Goldstein telephoned Clement to ask instructions for making a “wire transfer” of funds, that is, for arranging the automatic transfer of funds from his account to an account in another bank. When Goldstein arrived at the branch to arrange for the transfer, he presented Clement with a bottle of expensive champagne. They discussed arrangements for the transfer, and Clement told him that the bank could only wire “collected” balances, that is, could only wire funds from the check after a sufficient time had passed to permit the bank to conclude that the check had been paid. 4 Apparently Goldstein was *365 willing to wait, No further action was taken on Tuesday. Two days later, on Thursday, May 15, seven business days after deposit of the check, Goldstein returned and sought both to effect a wire transfer of $660,000 and to withdraw $95,000 in cash. The wire transfer was to be made to the account of a Maryland coin dealer. Apparently concluding that sufficient time had passed, Clement undertook to make the wire transfer. His first step was to check the Union Trust computer to find out whether sufficient funds were available in Goldstein’s account to satisfy the transfer request. Because this act was crucial to the success of the fraud, it must be examined in some detail. As explained in note 4, supra, banks are never notified when checks deposited with them are paid by the payor bank. The large volume of checks in the banking system would make any such notification system expensive and unwieldy. If payment is refused, however, the payor bank must notify the depositary bank of the refusal within roughly twenty-four hours of its receipt of the check. See N.Y.U.C.C. § 4-802. In order to protect themselves against uncollectible checks, banks commonly guess at the amount of time the check is likely to spend in the collection system before reaching the payor bank and place a hold on the deposited check for at least that amount of time. Union Trust’s computer was charged with keeping track of such holds. The Union Trust computer apparently registered two types of holds. When a check was deposited into a checking account such as Goldstein’s, the Union Trust computer automatically placed a one- or two-day hold on the deposited check. These holds were keyed to the Federal Reserve clearing system. Member banks of the Federal Reserve system keep accounts with their local Federal Reserve bank. See USF & G I, 590 F.Supp. at 490 n. 7. One of the functions of these accounts is to permit the Reserve bank to credit and debit member banks for checks cleared through the Reserve bank. Banks are ordinarily not given immediate credit for or access to the funds represented by checks such as that deposited with Union Trust by Goldstein. 5 In other words, regardless of whether a depository bank gives its depositor immediate access to the funds represented by a deposited check, the Federal Reserve will not, in most instances, give the bank immediate credit for the check in the account which the bank maintains with the Reserve bank. Instead, the Reserve bank credits the check to a one-or two-day deferred account, equivalent to placing a hold on the check, with the length of the deferral depending primarily upon the distance the check must travel to reach the payor bank. Once the deferral period elapses, the Reserve bank grants the depositary bank a provisional credit for the amount of the check. In reality, however, checks frequently do not clear in the one- or two-day time period allotted for this purpose by the Federal Reserve. The provisional status of the credit given by the Reserve bank signifies that although a credit has been given to the account of the depositary bank, no corresponding debit has been made in the account of the payor bank. Once the payor bank receives the check, it must notify the Reserve bank within twenty-four hours whether it will honor the check. If the Reserve bank is notified that the check has been honored, it debits the account of the payor bank and removes the provisional status of the credit given the depositary bank. 6 *366 The Union Trust computer automatically placed the equivalent of a hold on all deposited checks corresponding to the length of the deferral imposed by the Federal Reserve. The duration of Reserve system deferrals were never intended, however, to estimate or correspond precisely to the actual clearing time of a check. On the contrary, they intentionally underestimate this time. For that reason, as suggested above, banks frequently place their own extended hold on checks which are not likely to clear within one or two days. This prevents the depositor from withdrawing funds until the bank has satisfied itself that no word of dishonor has arrived or, in theory, will arrive. Such holds could also be recorded on the Union Trust computer, but they were not imposed automatically. As noted above, no such hold was placed on the Goldstein check. The only hold in the Union Trust computer was a two-day Federal Reserve hold, which expired several days before May 15. Therefore, when Clement checked, he found no hold on the Goldstein funds. This, of course, did not indicate that the Goldstein check had been paid. It simply indicated that whatever precautions Union Trust had taken against premature withdrawal had expired, and the check had not been returned. Clement called the bookkeeping department of the bank, which confirmed that no hold was in effect — in other words, that, according to the bank’s computer, the funds from the check were available for withdrawal. Apparently misunderstanding the nature of the information supplied by the computer, Clement, according to his deposition testimony, concluded from his computerized inquiry not only that all holds had expired but that the check had actually cleared. Although he was aware of the First Penn report that no account existed against which the check could possibly have been drawn, he concluded that the report must have been incorrect. Without checking once again with First Penn, he authorized the wire transfer and the cash withdrawal. Goldstein picked up the cash in person on May 15. The wire transfer was made to the bank of the Maryland coin dealer the same day. The dealer, however, had insufficient coins on hand to satisfy Goldstein’s order. The dealer gave Goldstein the coins available on May 15; the remainder were delivered to him in mid-afternoon the following day. Goldstein thereafter disappeared with the cash and coins. At about the same time Goldstein was receiving his coins on May 16, a Union Trust Vice President was informed that First Penn was returning the $880,000 check. Only later was it discovered that the Goldstein check was a fraud. The numbers printed in magnetic ink at the bottom of the check and the routing number, printed in ordinary ink in the top right hand corner, did not match, a circumstance which caused the check to be routed to several banks over several days before arriving at First Penn. 7 It is now possible to describe the circumstances surrounding the routing of the check in more detail. Union Trust’s central check processing facility attempted to computer-process the check on May 6, the day of deposit. This machine rejected the check because the magnetic ink character recognition number (“MICR number”) was not printed in magnetic ink and was the wrong size, necessitating hand processing. 8 At that time all *367 of Union Trust’s non-local checks which required hand processing were sent to PNB for processing and collection. The Goldstein check was accordingly dispatched to PNB in the early morning of May 7; because of its great value, it was sent by courier. Based on the check’s routing number, it was sent by PNB to a New York Fed processing center in Utica, New York, for forwarding to Albany State. Albany State received the check on the morning of May 9 and returned it to the Utica center of New York Fed, stamped “Sent in Error,” on May 12. 9 The next day the Utica center sent the check to New York Fed’s New York City office. Despite irregularities in the check described as “glaring” by a New York Fed chief of operations, that office failed to detect the fraudulent nature of the check. Instead, on May 14, New York Fed sent the check to the Federal Reserve Bank of Philadelphia. A document accompanying the cheek indicated that New York Fed intended the check to be presented for collection to First Penn rather than returned to PNB as an unpaid item. 10 The Philadelphia Reserve bank presented the check to First Penn at 9:00 a.m. on the same day it received it, May 14. At this time, it must be remembered, Goldstein had not yet been permitted to withdraw any funds against the check by Union Trust. If First Penn was going to dishonor the Goldstein check, it was required by the U.C.C. to notify PNB of this fact by midnight of May 15. It did not. 11 Instead, processing of the check was not completed until 9:45 a.m. on the morning of May 16, when First Penn finally notified PNB of the dishonor. At this point it was still not too late to recover the fraudulently gained funds, for the second shipment of gold coins had not yet been made to Goldstein. Quick action might have caught him. Quick action, however, was not forthcoming. PNB did not pass word of the dishon- or to Union Trust until mid-afternoon of May 16, too late to prevent Goldstein from slipping away. He has never been located. II. Albany State and the three defendant banks all move for summary judgment on plaintiffs’ claims. Each is in a somewhat different factual posture, and each accordingly raises somewhat different legal issues. Overhanging all of the arguments, however, is the issue of the effect of Union Trust’s behavior in its dealings with Gold-stein. It cannot be contended that Union Trust’s behavior was anything but reckless. Gemmill opened an account with a man who presented only one form of personal identification, an out-of-state driver’s license. Whether or not this alone was poor commercial practice, it placed particu *368 lar importance on the bank reference which he provided, for the reference was Union Trust’s only independent means of confirming Goldstein’s identity. Yet Clement left the reference off the signature card, and neither man made an effort to track it down when it was needed. In addition, the teller who accepted the check for deposit did so without presenting it to a branch manager, in breach of bank regulations. 12 This failure was particularly serious, for it deprived the banking system of its one opportunity for the check to be examined by an experienced professional who was unhurried by time pressures and simultaneously had access to information about the depositor. The most astounding act of carelessness, however, was the bank’s release of funds against a check which it had been told was drawn on a non-existent account. Such an act was more than mere negligence, for it entailed acting in the face of a known and obvious risk. Viewed in its entirety, Union Trust’s conduct was breathtakingly foolhardy; nay, commercially suicidal. 13 New York Fed argues, with backing from Albany State, that such almost incomprehensible conduct should act as a bar to any recovery based on the dramatically lesser negligence of other banks in the check’s chain of collection. A. In understanding this claim, it is helpful to review the basis of New York Fed’s alleged liability. Section 4-202(1) of the U.C.C. imposes upon all banks which handle checks sent for collection a duty of ordinary care both in forwarding checks for collection and in returning checks deemed uncollectible. Care must be exercised both in the promptness of action and in the choice of action. Timing is handled by the “midnight deadline” rule; banks exercise care in the timing of an action by responding, generally, before midnight of the business day following the day on which the check is received. N.Y.U.C.C. §§ 4-104 and 4-202(2). The proper choice of action cannot be so easily codified and is governed by a general rule of reasonableness borrowed from tort law. USF & G I, 590 F.Supp. at 491-92. This is the duty New York Fed is accused of failing to satisfy. At the time of USF & G I, the lack of discovery made it impossible to define precisely New York Fed’s alleged breach of care, since its exact actions were then unknown. It now appears that plaintiffs' claim is twofold: that New York Fed should have, in the exercise of ordinary care, either 1) recognized the risk of fraud inherent in the check and sent it for return to PNB rather than collection from First Penn or 2) recognized that the routing of the check to Albany State caused a significant risk that Union Trust would release funds before the check cleared and sent wire notice to Union Trust or PNB that the check had taken the long route to First Penn. See USF & G I, 590 F.Supp. at 494-99. Union Trust argues that these actions would have permitted it either to prevent the fraud or to catch Goldstein before he absconded with the cash and coins. It is by no means a foregone conclusion that New York Fed’s failure to take these actions, given the realities of modern check *369 processing, was negligent. USF & G I, 590 F.Supp. at 499. On the other hand, as discussed above, Union Trust’s recklessness was gross and obvious. New York Fed argues, on several grounds, that the tenuous nature of its alleged negligence should not serve as a basis for recovery in the face of Union Trust’s conduct. The primary difficulty with defendant’s theory is that it is not expressly sanctioned by the UCC. Plaintiffs’ suit is for violation of a duty imposed in the first instance by the UCC, and it is to that body of law which resort must first be had to determine the rights and liabilities of the parties. Although the UCC in some specific circumstances apportions liability on the basis of relative fault—most notably in the law of forged signatures and endorsements, see, generally, Five Towns College v. Citibank, N.A., 108 A.D.2d 420, 489 N.Y.S.2d 338, 342-43 (1985), it incorporates no general rule of comparative or contributory negligence. In the area with which we are concerned, check collection, the Code imposes upon collecting banks a duty of due care and subjects them to “but for” liability for violations of the duty. N.Y.U.C.C. §§ 4-103(5) and 4-202(1); Northpark National Bank v. Bankers Trust Co., 572 F.Supp. 524, 531 (S.D.N.Y.1983) (hereafter “Northpark”). There is no express requirement that the plaintiff demonstrate its own due care as a prerequisite to recovery, nor is there any mention of comparative negligence. 14 There is simply no mention of the effect, if any, of a plaintiff’s negligence on its recovery under § 4-202(1). The question cannot be permitted to end there, however. Article 4 of the Code, governing check collection, was developed in the 1950s, prior to the advent of large scale fraud upon the check collection system. See Northpark, 572 F.Supp. at 533. The rules of liability which govern the allocation of losses arising from check transit were not—could not have been—designed with such fraud in mind. In these circumstances, to adhere blindly to the limitations imposed by those rules, if to do so would violate the policies which the UCC otherwise seeks to promote, would be unwise and unjust. Nor does the Code demand such adherence. As Judge Knapp noted in his seminal Northpark decision, “the history of the UCC makes it abundantly clear that, especially in the context of those provisions which impose a duty of care, the Code’s watchword is ‘flexibility.’ ” 572 F.Supp. at 533. I do not, therefore, find the lack of a rule of contributory or comparative negligence in Article 4 to be an insuperable barrier to defendants’ claim that such a rule should be imposed. Rather, the decision turns upon whether to do so would, on the one hand, be consistent with the aims of § 4-202(1) and Article 4 as a whole and, on the other, would promote the policies which animate the rules of liability found elsewhere in the Code. B. In making its claim, New York Fed argues primarily by analogy to tort law. Because, the argument goes, the ordinary care standard of § 4-202(1) is borrowed from tort law, tort principles of comparative or contributory negligence should apply. As plaintiffs point out, however, this is not a tort case. The standard of care, though reliant on a concept borrowed from tort law, is imposed by the UCC. In allegedly violating § 4-202(1), defendants did not breach a duty imposed by tort law; they breached a duty imposed by statute, the UCC. It is the policies served by the UCC, therefore, rather than those served by tort law, which must determine whether plaintiff’s negligence should affect its recovery under § 4-202(1). Because, as discussed below, the aims of tort law are not identical with those of the UCC, I decline to borrow the tort rule. *370 Tort law is designed primarily to apportion loss. Because it is typically imposed upon lay parties with little or no appreciation of its finer points, only secondarily can it hope to guide behavior so as to minimize harm. Rather, it most often becomes relevant only after the fact; the courts are asked to decide who must bear or respond for a loss previously incurred, and to what degree. As such, its guiding principle is fairness. In tort law courts have equated fairness with fault. The rule of comparative negligence is a perfect expression of this principle. The UCC, however, was designed to facilitate commerce primarily by guiding and making predictable the consequences of behavior. It is imposed primarily upon a comparatively sophisticated group, businessmen and bankers, who look to its provisions to direct their business transactions. This is not to overlook its loss allocation function. Rules designed to guide behavior are inevitably turned against those who fail to follow them, and even the most carefully planned transactions may end in dispute. This function, however, is secondary to the creation of a system of rules to bring order and predictability to commercial transactions. Interestingly, this fact is forcefully demonstrated by a series of provisions in the Code which were designed, unlike the bulk of the Code, to distribute loss. Several provisions of Article 3 determine which party must bear loss which results from the use of a forged signature or endorsement on a negotiable instrument. Tort law would ordinarily distribute such a loss on the basis of fault: that party or parties whose carelessness resulted in the forgery would bear the consequences of it. The UCC, however, for the most part does not look at actual fault. 15 Instead, it places responsibility on the party which ordinarily would be in the best position to prevent the loss. 16 See Northpark, supra, 572 F.Supp. at 535 and n. 26; Underpinning & Foundation Constructors, Inc. v. Chase Manhattan Bank, N.A., 46 N.Y.2d 459, 468, 414 N.Y.S.2d 298, 302, 386 N.E.2d 1319 (1979). Such a result accomplishes two purposes: first, it increases the efficiency and fraud-resistance of the banking system by placing upon those best able to guard against it the responsibility for preventing fraud, 17 and, second, it speeds the resolution of disputes by establishing clear rules of liability which do not depend heavily upon the specific facts of individual instances of fraud. See Perini Corp. v. First National Bank, 553 F.2d 398, 405-06 (5th Cir.1977) (describing this policy as promoting “finality”). Therefore, while the purpose of these rules is, as with tort law, to apportion loss, they are guided not by the policy which guides tort loss apportionment — fairness in the circumstances — but by the policies which shape the UCC — the promotion of efficient and predictable commerce. Because tort law and the UCC are designed to serve different ends, I find it inappropriate to borrow a rule of comparative or contributory negligence from tort law for use in the UCC. Nevertheless, the question remains open whether such a rule should be implied from within the UCC itself. *371 C. The policies which underlie the UCC were explored above. As relevant here, they were usefully summed up by Judge Knapp in Northpark: “[i]f there is a policy implicit in the UCC’s rules for the allocation of losses due to fraud, it is surely that the loss be placed on the party in the best position to prevent it.” 572 F.Supp. at 535. See similarly, Leigh Co. v. Bank of New York, 617 F.Supp. 147 (S.D.N.Y.1985), at 151. These rules are most elaborately developed in connection with the law of forged endorsements and signatures. Liabilities for forgeries are assigned in the first instance without regard to fault. Instead, as noted above, the drafters placed liability upon the party which they determined would typically be best situated to prevent a particular type of forgery. Creation of this type of rule is essentially a legislative task, outside the scope of the proper exercise of judicial power. Even assuming that courts are competent to determine which party most commonly can prevent a particular type of forgery, the decision whether to assign liability to that party in all circumstances, without regard to fault, is best left to the legislature. This is particularly true where, as here, the legislature has already undertaken to create a complex network of laws in the area. For that reason, I rejected in USF & G I New York Fed’s request that I create such a rule, on grounds of public policy, assigning strict liability for MICR fraud to depositary banks. 590 F.Supp. at 500 and n. 23. Nor is it particularly clear that depositary banks are best situated in all, or even most, cases to detect MICR fraud. If such fraud is carried out with more sophistication than demonstrated by Goldstein, the deposit and collection of funds may escape the notice of even a careful bank. In such cases, the bank best situated to detect the fraud might be the first bank to refuse payment, Albany State in this case. That bank would be the first with concrete knowledge that the check is flawed. 18 On the other hand, the fraud could not be confirmed until the check has been examined by the purported payor. Which of these banks should be assigned liability under all circumstances, if any of them should, is a choice I declined and decline to make. If such legislative assignments of strict liability comprised the whole of the UCC’s system for distributing loss due to forgeries, it might properly be concluded that the concept of contributory or comparative negligence had no role to play within the UCC. However, all rules governing liability for forged signatures are subject to the following provision, which incorporate what is, in effect, a rule of contributory negligence: Any person who by his negligence substantially contributes to a material alteration of the instrument or to the making of an unauthorized signature is precluded from asserting the alteration or lack of authority against a holder in due course or against a drawee or other pay- or who pays the instrument in good faith and in accordance with the reasonable *372 commercial standards of the drawee’s or payor’s business. N.Y.U.C.C. § 3-406. 19 Section 3-406 is consistent with the remainder of the Code. It places a burden of due care on a person well situated to prevent a forgery: the drawer of the check. Drawers are forced to draw their checks so as to prevent alterations and to safeguard their blank checks so as to prevent their falling into unauthorized hands. 20 While § 3-406 does not directly apply to the case at bar, I am persuaded that it is appropriate to apply its spirit by analogy to apportionment of loss due to MICR fraud. The depositary bank, like the drawer of the check, is well situated to protect the system against MICR fraud. The depositary bank has an opportunity to examine the check free of the time pressures which prevent collecting banks from giving checks more than a cursory glance. Perhaps more important, the depositary bank is in the unique position of being able to examine both the depositor and the check. No other bank in the collecting chain can examine the depositor, a crucial disadvantage given the seeming difficulty of detecting this type of fraud. In many cases examination of the depositor and check might well reveal nothing unusual or alarming, 21 in which event the collecting banks will be obliged, to avoid liability, to demonstrate their compliance with § 4-202. As this case proves, however, a careful examination of either depositor or check, or both, might well reveal the fraud and protect the banking system. Because the depositary bank is uniquely situated to perform such an examination, it is entirely consistent with the policies served by the UCC to place upon it the duty to do so. Conversely, to refuse to imply such a duty would be equally inconsistent with those policies. It would permit the individual or entity in the best position, at least in those circumstances, to prevent the fraud to evade any duty to do so. Not only would such an approach be inefficient, it could well be, as this case demonstrates, entirely unjust. Moreover, because only the depositary bank possesses this ability to examine both the depositor and the check, it is appropriate to place upon it the initial burden of care. In the banking system, the depositary bank is the first line of defense against MICR fraud, and the most efficient point at which to take precautions against it. It may properly be prevented from holding other banks liable if it has not adequately fulfilled this role. 22 *373 I hold that a depositary bank which is a victim of MIOR fraud may be precluded from recovering damages from collecting banks under UCO § 4-202(1) if those banks can demonstrate that the negligence of the depositary bank played a substantial role in the success of the fraud. 23 I have no hesitation in holding that, for the reasons stated previously, it has been demonstrated as a matter of law that the recklessness of Union Trust played a substantial-nay, indispensable-role in the success of this fraud. I grant summary judgment in favor of defendants Albany State, New York Fed, and PNB. III. First Penn stands on a somewhat different footing, since it is alleged to be liable not under § 4-202(1) but § 4-302. Section 4-302 makes a payor bank which fails to return a dishonored check within its midnight deadline liable for resulting damages. It has been repeatedly recognized, however, that § 4-302 does not shift the burden of loss to a payor bank which misses its deadline if the payee was already aware when presenting the check that it would not be accepted or paid except by mistake. See N.Y.U.C.C. § 3-511(2)(b); Bank Leumi Trust Co. v. Bally's Park Place, 528 F.Supp. 349 (S.D.N.Y.1981); Leaderbrand v. Central State Bank of Wichita, 202 Kan. 450, 450 P.2d 1, 9 (1969); cf. Continental National Bank v. Sanders, 581 S.W.2d 293, 296 (Tex.Civ.App.1979). Although Union Trust did not know that the Goldstein check would not be paid when it forwarded the check for collection, it learned long before releasing funds to Goldstein that First Penn had no account from which to pay the check. The same policy reasons which preclude recovery under § 4-302 by those who forward checks which they have reason to know are uncol-lectible also argue in favor of precluding Union Trust from recovering. It had reason to know of the uncollectibility of Gold-stein's check. Therefore, it can and should be estopped from claiming that at the time it suffered the loss it was not aware that First Penn would refuse the check. To hold otherwise would be inequitable to First Penn. First Penn informed Union Trust that it had no account relationship with the purported drawer of the check. It should not now be held liable because Union Trust chose to ignore that information and subsequently release funds against the check. Also relevant is National Savings and Trust Co. v. Park Corp., 722 F.2d 1303, 1304 (6th Cir.1983), cert. denied, 466 U.S. 939, 104 5.Ct. 1916, 80 L.Ed.2d 464 (1984), in which it was held that a bank may recover funds paid by mistake on a bad check unless the plaintiff has changed its position in reliance on the payment. Because Union Trust had been informed before allowing Goldstein to withdraw funds that First Penn would not pay the check, Union Trust may be estopped from claiming that it permitted withdrawal in reliance on First Penn's deemed payment at the expiration of its midnight deadline. Under both lines of authority, First Penn is entitled to summary judgment on plaintiffs' claims. VI. For the reasons stated above, all defendants are entitled to summary judgment on plaintiffs' claims. The third-party claims are therefore moot. The Clerk is directed to dismiss both complaints with prejudice. It is SO ORDERED. Footnotes [1] Since the occurrence of the events which form the basis of this lawsuit, Albany State has changed its name to Norstar Bank of Upstate New York. I will continue to refer to it by its prior name, as do the parties. [2] As explained in USF & G I, 590 F.Supp. at 489, familiarity with which is assumed throughout this decision, this would ordinarily have been an entirely legitimate assumption; indeed, it is an assumption made thousands of times daily by every bank in the nation without adverse consequences. [3] According to an affidavit filed by New York Fed, the New York bank reference, if contacted, would have reported no history of dealings with the Goldstein company. [4] As explained in USF & G I, 590 F.Supp. at 489, banks are never actually notified when a check which has been deposited with them is paid by, or “clears," the payor bank. Instead, they are only told if it fails to clear. Therefore, before releasing funds against a suspect check, banks *365 commonly hold the check for the number of days which experience has taught them will be required for the check to reach the payor bank and for that bank to inform them if the check is uncollectable. If no word arrives in this time, it is assumed that the check has been paid. [5] There are a few exceptions to this rule of delayed availability — checks drawn on the United States Treasury, for example, are immediately credited to the depositary bank — but they are not relevant here. [6] The foregoing simplified description of the operation of this aspect of the Federal Reserve system is derived primarily from Eric N. Compton, Principles of Banking, 154-56 (2d Ed.1983). [7] For a fuller explanation of the operation of the fraud, see USF & G I, 590 F.Supp. at 490. [8] It is interesting to note that Goldstein's technique of printing an improper MICR number has been used in other types of fraud. According to published reports, the scheme which resulted in the well-publicized May, 1985, conviction for fraud of the brokerage firm of E.F. Hutton & Co. also used improperly printed checks. Hutton had checks printed with incorrect MICR numbers to foil its banks’ computer-aided check-processing machines, thereby lengthening the time needed for processing the checks and increasing Hutton’s "float,’’ or amount of funds on deposit by checks which have not yet been collected from the payor bank. Hutton deposited the flawed checks drawn on its own accounts at various banks in its accounts at other banks. Until the checks were paid by the former banks, Hutton received payment for use of the funds by both banks at once. The longer the time required to process the check, the greater the “double” payments *367 Hutton received. This was where the incorrect MICR numbers came in handy. See Bleakley, How Hutton Scheme Worked, N.Y. Times, May 17, 1985, at D4, cols. 2-4. [9] In USF & G I one issue was whether New York Fed had a regulatory duty to "wire advice" to Union Trust of this return. 590 F.Supp. at 492-94. I concluded that New York Fed had no duty to originate such advice but only to forward wire advice given to it. 590 F.Supp. at 493-94. Albany State now claims that it did wire advice. On the basis of this allegation, which New York Fed denies, plaintiffs move to amend to assert New York Fed’s failure to forward the advice. See USF & G I, 590 F.Supp. at 494. I deny the motion. Plaintiffs have adduced no evidence in support of their claim that Albany State wired advice. There is only the bank’s bare allegation. Therefore, the condition set out for amendment in USF & G I, 590 F.Supp. at 494, that “discovery [has] reveal[ed] that New York Fed did indeed receive wire advice” has not been fulfilled. Discovery revealed no evidence one way or the other as to wire advice. [10] Because both PNB, to which the check would have been returned, and First Penn, to which it was sent for collection, are in the same Federal Reserve district, it was impossible to determine on the facts pleaded in the complaint whether New York Fed had returned the check or sent it for collection. See USF & G I, 590 F.Supp. at 497. It now seems clear, based on discovered materials, that New York Fed did not return the item but sent it for collection. [11] On this point, the statement of facts in USF & GI is in error, since the initial pleadings seemed to indicate that each bank in the chain met its midnight deadline. 590 F.Supp. at 490. [12] In retrospect, the teller’s failure to place a hold on the check, while poor practice, was of less consequence. Union Trust’s standard, and apparently proper, hold for a check drawn on a Pennsylvania bank was five to six days. Since Goldstein withdrew funds on the seventh day, the hold would not have prevented execution of his scheme. [13] This is not the whole of the bank’s carelessness. First, discovery revealed that after being rejected by the automatic sorting machinery the check was examined by a Union Trust supervisor, but this supervisor failed to detect the "glaring” facial irregularities in the check and instead forwarded it to PNB. Second, Clement’s wire transfer of $660,000 exceeded the amount of money he was authorized by bank regulations to transfer. Third, several of Goldstein’s actions known to Clement labelled him as "shady,” including his transparent attempt at a bribe and his request that the money be wired to a coin dealer, who presumably would convert the funds into highly liquid assets. [14] The latter is not surprising. The UCC predated the widespread acceptance of comparative negligence of the 1970s. More important, the concept of comparative negligence is anathema to a code which had as a primary aim the fostering of swift resolution of disputes through the clear delineation of commercial liabilities. See Perini Corp. v. First National Bank of Habersham County, 553 F.2d 398, 405 (5th Cir.1977). [15] The exception, discussed infra, is UCC § 3-406. [16] In most cases, loss falls on the forger's immediate transferee. N.Y.U.C.C. §§ 3-417(2)(a), 4-207(2)(a), 3-414(1), and 4-212(1). However, when a check is paid over a forged signature, loss is placed upon the drawee bank, which in theory could have checked its client’s signature. N.Y.U.C.C. § 3-417, comment 3. See generally, Northpark, supra, 572 F.Supp. at 535 n. 26. [17] Because such parties are generally commercial banks, they are presumed to be aware of the UCC’s rules of liability and to take precautions to guard against forgeries. The rule of absolute liability regardless of fault would, one would think, tend to increase their efforts at protecting against forgeries, since to avoid liability they would be forced to guard against not only their own carelessness but the carelessness and deviousness of others. In theory, by placing this burden upon the shoulders of those best able to guard against fraud the UCC achieves such fraud resistance in the most efficient manner. [18] It appears that the first banks in the collection chain, Union Trust, PNB, and New York Fed here, would not automatically be suspicious of a check bearing a routing number inconsistent with its listed name of the payor bank. According to the deposition testimony of a Union Trust employee, such checks may be intentionally and properly circulated. They are used, when one bank maintains an account at a second bank. Union Trust, in fact, participated in such an arrangement. On these checks, Union Trust's routing number appeared in the upper right corner of the checks, and the checks were drawn on an account held by Union Trust, but the name of a second bank, Union Trust's customer bank which maintained the account at Union Trust, appeared where one would otherwise have expected Union Trust’s name. The employee called such checks “due-from” checks. The Goldstein check, of course, had other irregularities which might have attracted attention: the MICR numbers were neither in magnetic ink nor of the proper size and style, and they were nonsensical. The discrepancy between the Albany State and First Penn designations alone, however, would not automatically have suggested fraud to a collecting bank. [19] A more complex and specific provision, N.Y. U.C.C. § 4-406, establishes non-exclusive circumstances which constitute per se negligence by checking account customers who are victims of forged signatures. [20] According to the Code, “[t]he most obvious case is that of the drawer who makes use of a signature stamp or other automatic signing device and is negligent in looking after it.” N.Y.U. C.C. § 3-406, comment 7. Again, this shows the commercial orientation of the Code; businessmen are warned to watch those with access to signing machines. [21] See note 18, supra. [22] To this extent, this imposed rule differs from § 3-406, which requires an otherwise liable bank seeking to evade liability to demonstrate not only the drawer’s lack of due care but its own exercise of due care. The rules of liability of which § 3-406 is a part are quite different, however, from § 4-202. "Otherwise liable” banks are not liable, as is the case with banks liable under § 4-202, because of their negligence. They are strictly liable. It is arguably appropriate to require them to demonstrate their own exercise of care before evading the severe standard of strict liability. Further, drawers, who may be made liable under § 3-406, stand on a much different footing than the banks which would be liable. They are generally individuals and corporations, and the forgeries are presumably carried out without their awareness. Because of this, they do not possess the tools to protect against fraud — for example, the chance to ask a customer for identification or to check a signature card — which the strictly liable banks do. Indeed, it is these tools to protect against forgery which the UCC uses to justify holding those particular banks liable. See Northpark, supra, 572 F.Supp. at n. 26. By contrast, depositary banks under § 4-202(1) differ from the collecting banks in their ability not only to examine the check but the depositor. They thus have advantage over the collecting banks. It is therefore appropriate to place liability on their shoulders should they fail to exercise due care, even if the collecting banks have also failed to do so. [23] It might be contended that application of comparative negligence would be more fair. The UCC, however, contains no provision for comparative negligence. See Northpark, supra, 572 F.Supp. at n. 28. Even in New York the legislature has left the UCC's all-or-nothing rules alone while statutorily adopting comparative negligence in other areas. See N.Y.Civ.P.L.R. § 1411. Contributory negligence serves the UCC's aims of efficiency and finality by definitively assigning liability. See note 14, infra. It would be improper to impose the more tort-oriented comparative negligence in the absence of legislative approval of such a step.