Miguel Villafane-Neriz, Insurance Commissioner of Puerto Rico v. Federal Deposit Insurance Corporation
75 F.3d 727
1st Cir.
1996
Check Treatment
TORRUELLA, Chief Judge.

This appeal seeks review of a decision of the United States District Court for the District of Puerto Rico, which entered summary judgment on behalf of appellee the Federal *729 Deposit Insurance Corporation (“FDIC”), in its corporate capacity. Appellant Miguel Villafañe-Neriz, Insurance Commissioner of Puerto Rico (the “Commissioner”) seeks to recover FDIC deposit insurance for the $50,-000 value of a certificate of deposit (the “Certificate” or the “CD”) purchased by the Guaranty Insurance Company (“Guaranty”), which was assigned to the Commissioner simultaneously with its purchase. The district court held that the FDIC properly relied on the books and records of an insolvent institution in making its determination that the Commissioner was not entitled to deposit insurance. The sole issue before us is whether the district court erred in granting summary judgment against the Commissioner in his action against the FDIC in its corporate capacity. 1 For the reasons stated herein, we affirm.

BACKGROUND

The facts of this case are undisputed. On July 20, 1983, in compliance with the Puerto Rico Insurance Code’s statutory deposit requirement, 26 L.P.R.A. §§ 801-809 (1976), Guaranty purchased the six-month CD from the Girod Trust Company (“Girod” or the “Bank”) in the principal amount of $50,000. On the same day Guaranty assigned and conveyed its interest in the Certificate to the Commissioner. Girod was not a party to the assignment. Another document was executed on the same date, entitled “Requisition to the Bank.” This document stated, inter alia, that Girod would not release the funds represented by the CD, “whether the principal value or income thereof,” without the Commissioner’s authorization. The Certificate was itself given to, and remains with, the Commissioner.

Less than three months after purchasing the Certificate from Girod, Guaranty executed a loan agreement, unrelated to the CD, pursuant to which it borrowed $600,000 from Girod. A promissory note for that amount, payable to Girod, evidenced the loan, and was due on April 26,1984. On January 17, 1984, the CD became due, and was “rolled over”— extended for a term of six additional months — at Guaranty’s request. In the meantime, Guaranty had fallen behind on payments due to the Bank under the $600,-000 loan agreement. On July 16, 1984, the CD came due again. Two days after its maturity, on July 18, $50,000 in proceeds from the Certificate was credited toward Guaranty’s outstanding indebtedness under the $600,000 loan agreement.

On August 16, 1984, Girod was declared insolvent and the FDIC was appointed as receiver. Four months later, on December 19, 1984, Guaranty also became insolvent, and the Commissioner was appointed its receiver in turn. As such, on August 25, 1986, the Commissioner filed a proof of claim with the FDIC, seeking payment on the CD. Having received no payment on the claim, the Commissioner filed a complaint against the FDIC in the Superior Court of Puerto Rico on May 22,1991, seeking to recover the proceeds of the CD. The FDIC removed the action to federal court pursuant to 12 U.S.C. § 1819(b), and the parties filed cross-motions for summary judgment. Without ruling on the motions, the district court requested submission of briefs on the application of 12 U.S.C. § 1823(e). The court then held that that section barred the Commissioner’s reliance upon either the Assignment or the Requisition,. and ordered summary judgment in favor of the FDIC. On appeal in Villafañe-Neriz v. FDIC, 20 F.3d 35 (1st Cir.1994), this Court reversed the judgment of the lower court and remanded the case for further proceedings consistent with its opinion. On February 7, 1995, the district court entered summary judgment dismissing the complaint. It is undisputed that the CD no longer appeared on the bank’s books and records at the time the bank failed and that the Certificate itself remains in the Commissioner’s possession.

*730 DISCUSSION

A. Standard of Review

This case centers on whether the FDIC, in its corporate capacity, was correct in determining there was no insured deposit. As the essential facts are not in dispute, and all that is before us is a question of law, our review of the district court’s decision is de novo. See, e.g., FDIC v. Keating, 12 F.3d 314, 316 (1st Cir.1993). This Circuit has not yet decided which standard a district court should use when reviewing FDIC insurance claim determinations.

There is a dispute among the circuits as to the underlying standard that should apply to the review of an FDIC insurance claim determination. The majority of circuits which have addressed the issue apply the deferential standard set out in Section 706 of the Administrative Procedure Act (“APA”), 5 U.S.C. §§ 701-706 (1994). See, e.g., Metro County Title, Inc. v. FDIC, 13 F.3d 883, 886 (5th Cir.1994) (direct petition to court of appeals for review of FDIC determination); Nimon v. RTC, 975 F.2d 240 (1992) (direct petition to court of appeals for review of Resolution Trust Corporation determination); In re Collins Sec. Corp., 998 F.2d 551, 553 (8th Cir.1993) (review of district court decision); Fletcher Village Condominium, Ass’n. v. FDIC, 864 F.Supp. 259, 263 (D.Mass.1994). The APA mandates that reviewing courts set aside agency findings that are “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). Under this deferential standard a court would “review the evidence anew and determine whether the administrative action was arbitrary and capricious.” First Nat’l Bank of Fayetteville v. Smith, 508 F.2d 1371, 1374 (8th Cir.1974), cert. denied, 421 U.S. 930, 95 S.Ct. 1655, 44 L.Ed.2d 86 (1975); see, e.g., Hymel v. FDIC, 925 F.2d 881, 883 (5th Cir.1991).

However, a recent decision by the D.C. Circuit creates a second option, holding that review of FDIC determinations, to be undertaken at the district court level, should be de novo rather than under the deferential APA standard. See Callejo v. RTC, 17 F.3d 1497 (D.C.Cir.1994). The Callejo court based its rejection of the APA on its reading of 12 U.S.C. § 1821(f) (1988 & Supp.1991), which provides for judicial review of disputed deposit insurance claims, and its revision under the Financial Institutions Reform, Recovery, and Enforcement Act of 1989. Callejo, 17 F.3d at 1501 (concluding that § 1821(f)(3) “supplants the APA and sets up a different relationship between the agencies and the courts”); see Pub.L. No. 101-73, 103 Stat. 183 (1989); cf. Pennsylvania v. FDIC, 881 F.Supp. 979, 983 (E.D.Penn.1995) (rejecting Callejo’s logic but nonetheless applying de novo standard of review on other grounds). This Circuit has expressly adopted the aspect of the Callejo decision which holds that initial jurisdiction to review claims for insurance benefits lies in the district court rather than in the court of appeals. Massachusetts v. FDIC, 47 F.3d 456, 458 (1st Cir.1995). However, the decision in that case was limited to Callejo’s jurisdictional holding. Id. at 460. Thus, Massachusetts v. FDIC does not determine the district court’s standard of review in the present case, and our decision to postpone the discussion does not clash with our earlier decision.

The district court did not explicitly state which standard of review it was applying, although it did make an isolated reference, midway through its opinion, to the “arbitrary, capricious and contrary to law” standard. We need not determine which standard the district court should have applied at this time, since we agree with the FDIC that under either the APA “arbitrary and capricious” or the Callejo de novo standard, the district court’s decision is correct. Thus we postpone discussion regarding the applicable standard of review in light of Callejo for another day.

B. Was this an insured deposit?

At the core of the parties’ dispute is whether the Commissioner was entitled to deposit insurance. That issue depends on whether there was an insured deposit at the time of Girod’s failure, a question which in turn hinges on whether and when erroneous bank records are conclusive. It is undisputed that the Bank’s account records did not disclose the existence of an account on the *731 date Girod failed, and that the original Certificate is in the possession of the Commissioner, and has been since July 1983. The FDIC argues that under its regulations and the applicable ease law, it is justified in relying solely on the failed Bank’s account records, so that its refusal to provide insurance was proper. The Commissioner counters that under Puerto Rico law the FDIC should have known the certificate had not been properly cancelled because the original was not in the bank’s possession. Because we agree with the FDIC, we affirm the decision of the court below.

In the Federal Deposit Insurance Act, 12 U.S.C. §§ 1811-1831d (1982) (as amended), Congress defined “deposit” to mean “the unpaid balance of money or its equivalent received or held by a bank or savings association in the usual course of business and for which it has given or is obligated to give credit, either conditionally or unconditionally, ... or which is evidenced by its certificate of deposit....” 12 U.S.C. § 1813(0(1) (Supp. 1995). “Insured deposit” is defined in turn as “the net amount due to any depositor for deposits in an insured depository institution as determined under sections 1817(i) and 1821(a) of this title.” 12 U.S.C. § 1813(m)(l) (1988 & Supp.1991).

The FDIC contends that it is entitled to rely exclusively on the account records of the failed institution — and so it did not have to look further afield to track down the Certificate. Our analysis of the FDIC regulations, the body of ease law, and the policy concerns underlying these regulations leads us to agree. First, the FDIC regulations, promulgated under congressional authorization, Abdulla Fouad & Sons v. FDIC, 898 F.2d 482, 484 (5th Cir.1990), themselves provide that the amount of an insured deposit at the closing of a failed bank shall be “the balance of principal and interest unconditionally credited to the deposit account as of the date of default of the insured depository institution.” 12 C.F.R. § 330.3(i)(l) (1995). Indeed, the regulations specify that, while ownership under state law is one prerequisite for insurance coverage, the deposit account records are controlling:

Deposit insurance coverage is also a function of the deposit account records of the insured depository institution, of record-keeping requirements, and of other provisions of this part, which, in the interest of uniform national rules for deposit insurance coverage, are controlling for purposes of determining deposit insurance coverage.

12 C.F.R. § 330.3(h) (including regulatory exceptions not relevant here). Reviewing courts have treated these regulations implementing and interpreting the statutory provisions dealing with deposit insurance with some deference. 2 See FDIC v. Philadelphia Gear Corp., 476 U.S. 426, 437-38, 106 S.Ct. 1931, 1937-38, 90 L.Ed.2d 428 (1986); see, e.g., Raine v. Reed, 14 F.3d 280, 283 (5th Cir.1994); Collins, 998 F.2d at 555; cf. Chevron U.S.A Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 842-44, 104 S.Ct. 2778, 2781-82, 81 L.Ed.2d 694 (establishing doctrine treating agency’s view of a statute with deference when the statute is ambiguous), reh’g denied, 468 U.S., 1227, 105 S.Ct. 28, 82 L.Ed.2d 921 (1984).

*732 Second, a series of policy considerations underlie the FDIC’s practice of relying on the books and records in making deposit insurance determinations. In purchase and assumption transactions, 3 the FDIC often must make its determinations overnight. See Raine, 14 F.3d at 283 (“We will not undermine the speed and efficiency of bank takeovers by imposing a requirement upon the FDIC to locate and evaluate every possible avenue of disputed liability in implementing the takeover of a failed bank.”); McCloud v. FDIC, 853 F.Supp. 556, 559 (D.Mass.1994). Making quick determinations both facilitates the public’s access to its savings, Abdulla Fouad, 898 F.2d at 485, and maintains the going concern value of the failed bank, Raine, 14 F.3d at 283. Finally, the regulations also avoid fraudulent increases in insurance coverage by preventing the creation of separate trust accounts after default has occurred. See Baskes v. FSLIC, 649 F.Supp. 1358, 1360 (N.D.Ill.1986).

Third, there is “a well-grounded history of permitting the FDIC to rely exclusively on the books and records of an insolvent institution in effectuating the takeover of banks and in making the many deposit insurance determinations which are necessary to that task;” Raine, 14 F.3d at 283; see McCloud, 853 F.Supp. at 559 (describing the “seemingly solid phalanx of law establishing the conclusiveness of bank account records”); see also Abdulla Fouad, 898 F.2d at 484 (providing statutory and regulatory basis for FDIC reliance on deposit account records). This reliance on the books and records draws on the FDIC regulations: the “FDIC’s longstanding practice of looking primarily at the failed bank’s deposit account records in determining insurance claims is clearly a permissible interpretation of [its] statutory mandates.” Collins, 998 F.2d at 554. Indeed, the case law the FDIC cites states that a bank’s closing is “the seminal point” of the FDIC’s determination. “That event not only trigger[s] the liquidation process, but it also cast[s] in stone the relationship of [plaintiff] to the bank.” FDIC v. McKnight, 769 F.2d 658, 661 (1985), cert. denied sub nom., All Souls Episcopal Church v. FDIC, 475 U.S. 1010, 106 S.Ct. 1184, 89 L.Ed.2d 300 (1986).

In fact, the case law supports the FDIC’s dependence on the books and records of the Bank at the time of failure even though the balance was a result of alleged unauthorized activity. 4 See Abdulla Fouad, 898 F.2d at 484-85 (finding that plaintiffs position that the FDIC should have searched bank credit files and other records before denying a claim goes against statutory and regulatory authority); Fletcher Village, 864 F.Supp. at 265 (“To hold the FDIC liable for the errors and omissions inherent in almost any routine banking transaction would divert the FDIC from its core mission of protecting the banking system from an ultimate catastrophe.”). In their analysis, “[t]hese cases reflect the severe tension between two values: the legitimate expectations of the depositor and the regulator’s desire to rely upon existing records to expedite the handling of bank emergencies.” Fedders Air Conditioning, 35 F.3d at 23.

The Eighth Circuit’s analysis in the factually similar In re Collins proves illustrative. In that ease, as here, the purchaser of a certificate assigned the proceeds to a third party, Collins. The proceeds, however, were paid to the purchaser’s account, and the CD *733 account was reflected on the institution’s account records as closed. Collins, 998 F.2d at 552-53. The trustee for the bankrupt Collins sought to recover from the institution for paying out the CD account despite the assignment, alleging negligence and breach of contract. Following the institution’s insolvency, the FDIC in its corporate capacity denied deposit insurance, and the trustee sought judicial review of its denial. Id. at 553. The court of appeals held that the FDIC properly relied on the books and records of the failed institution to deny deposit insurance, noting that the institution’s “mistaken payment may not have affected Collins’s rights against the bank, but it did extinguish the insured amount.” Id. at 554. In short, it reasoned that “[djeposit insurance protects depositors from loss due to the bank’s insolvency, not loss from the bank’s pre-insolvency mistakes.” Id. at 555. We find the Eighth Circuit’s reasoning convincing in the present case as well.

The Commissioner seeks to differentiate Collins on several bases. First, he argues that the mistake in Collins was a simple bank error, id. at 552-53, while the “cancellation” in the current case is not a simple mistake, but rather was illegal on its face. We do not find the distinction relevant, In both cases, the account was cancelled without regard to the CD’s assignment, and the bank’s records had not reflected the assignment. Second, he finds it significant that the decision in Collins noted not only that the account in controversy been closed at least a full year before the bank was declared insolvent, but also that the insolvent bank had not paid deposit insurance premiums for the account. See id. at 553-54. Here, the Commissioner contests, Girod paid deposit insurance premiums on the CD account, which was cancelled less than a month before the bank was taken over and a few days after the Treasury Department of Puerto Rico conducted the investigation that led to the bank’s closing. Again, we are not convinced of the distinction. In Collins, the court’s determination was based on the fact that the records of a failed bank indicated the amounts that were insured, and the accounts for which the FDIC collected deposit insurance premiums. The length of time that the account was closed, or the insurance premiums unpaid, was not the key: the crucial factor was the reasonableness of the FDIC reliance on the records. See id. at 554. Collins noted that the CD account was not an insured deposit for which premiums were paid, and found that Collins’ trustee had “confus[ed] the right to recover from the bank with the right to withdraw from an insured account.” Id.

Raine v. Reed offers another example of an analysis upholding the FDIC’s exclusive reliance on the books and records of a failed institution. Raine was a victim of unauthorized withdrawals from automatic teller machines, who notified her bank of the withdrawals and sought to have her account re-credited pursuant to the Electronic Fund Transfer Act (“EFTA”), 15 U.S.C. §§ 1693-1693r. Raine, 14 F.3d at 281-82. At the time of its failure, the bank had not provisionally re-credited the account pending final resolution of her request, as required by regulations implementing the EFTA. Raine contended she was entitled to deposit insurance on the basis that she should not suffer because of the bank’s mistakes. Id. at 282. However, the Fifth Circuit found that

[t]he disputed amount was simply not credited to her account at all, conditionally or otherwise. Thus, the account cannot be covered by deposit insurance because no credit for the amounts withdrawn was entered on the bank’s books at the time of failure.

Id. at 283. The court relied on the “well-grounded history” of allowing the FDIC to rely exclusively on an insolvent institution’s books and records, even where the bank itself has committed a mistake, as well as the policy rationales discussed above, in upholding the FDIC’s use of the books and records. Id. According to the court, “[t]he regulations are clear and simple, either the amount is credited to the account, in which case it is covered by deposit insurance, or the amount is not on the books, in which ease it becomes a general liability of the bank.” Id. at 284.

The Commissioner offers no authority contradicting our analysis of the FDIC regulations, policy considerations, and case law supporting the use of the failed bank’s records. Instead, the Commissioner counters with two arguments. First, he contends that even if the FDIC relied on Girod’s existing records *734 at the time of its failure in 1984, it should have concluded that the Certificate had not been properly cancelled. The Commissioner relies on 7 L.P.R.A. § 3 (1981), which states, in pertinent part, that

[t]he term “deposit certificate” shall mean any deposit which has been evidenced by a receipt or written agreement containing the term for which such deposit has been made and which also requires presentation at the bank for its collection.

(emphasis added). He concludes from this language that since Puerto Rico law mandates that the original of a certificate of deposit be presented to the bank for collection, an FDIC official reviewing Girod’s records regarding the CD’s “cancellation” had to be alerted that the Certificate was still valid by the simple fact that the original was not contained in the customer profile. By failing to do so, the Commissioner contends, the FDIC did not give the proper weight to these Puerto Rican recordkeeping requirements.

We do not find this argument convincing. On its face, the statute sets out the requirements for presentation for collection, which is not at issue here. We are concerned with what records should remain in the bank after a setoff, 5 and the language is silent on this point. The sole case the Commissioner cites to support its argument that the statute should be read to require that the original of a certificate must be presented for setoff, Walla Corp. v. Banco Commercial de Mayaguez, 114 D.P.R. 216 (1983) (holding that where bank did not have original certificates but was in rightful position to perform setoff it could publish notices in newspapers of general circulation), does not mention the statute. We could find no case law on point. Given the plain meaning of the statute, and the absence of evidence to support the Commissioner’s reading of the statute, we reject his position..

The Commissioner’s second argument relies on an exception to the general rule that the records of a failed Bank are conclusive. That exception states that “records that would otherwise be conclusive evidence may be attacked as fraudulent.” Collins, 998 F.2d at 555; see, e.g., Jones v. FDIC, 748 F.2d 1400, 1405 (10th Cir.1984); McCloud, 853 F.Supp. at 559. The Commissioner argues to this court that there was fraud in the transaction assigning the Certificate, and therefore he is entitled to attack the conclusiveness of Girod’s records.

However, we refuse to consider the Commissioner’s argument, since he raises it for the first time on appeal. 6 It is well established that this court will not consider an argument presented for the first time on appeal. See Clauson v. Smith, 823 F.2d 660, 666 (1st Cir.1987) (collecting cases). While exceptions to this rule exist, they apply only “ ‘in horrendous cases where a gross miscarriage of justice would occur,’ ” Johnston v. Holiday Inns, Inc., 595 F.2d 890, 894 (1st Cir.1979) (quoting Newark Morning Ledger Co. v. United States, 539 F.2d 929, 932 (3d Cir.1976)), and this is not such a case. That the Commissioner’s argument involves no new facts, just a new theory, is irrelevant. See Ondine Shipping Corp. v. Cataldo, 24 F.3d 353, 355 (1st Cir.1994) (“This assertion is neither original nor persuasive.”).

Therefore, since we find that under either the arbitrary and capricious standard or a more demanding de novo review the FDIC *735 was correct in relying solely on Girod’s records, and reject the Commissioner’s arguments based on Puerto Rico banking law and fraud, we affirm the district court’s holding on these issues.

C. Application of the McCarranFerguson Act

The Commissioner raises one final argument against the FDIC’s insurance determinations, based on Section 2(b) of the McCarran-Ferguson Act. 59 Stat. 33, 34 (1945), as amended, 15 U.S.C. § 1012(b). Section 2(b) states, in pertinent part:

No Act of Congress shall be construed to invalidate, impair, or supersede any law enacted by any State for the purpose of regulating the business of insurance, or which imposes a fee or tax upon such business, unless such Act specifically relates to the business of insurance____

15 U.S.C. § 1012(b) (1994). This statute creates “a form of inverse preemption, letting state law prevail over general federal rules— those that do not ‘specifically relate[ ] to the business of insurance.’ ” NAACP v. American Family Mut. Ins. Co., 978 F.2d 287, 293 (7th Cir.1992), cert. denied, 508 U.S. 907, 113 S.Ct. 2335, 124 L.Ed.2d 247 (1993); see United States Dep’t of the Treasury v. Fabe, 508 U.S. 491, 506-08, 113 S.Ct. 2202, 2211, 124 L.Ed.2d 449 (1993) (noting that the Act “transformed the legal landscape by overturning the normal rules of preemption.”). As the Supreme Court has explained, “ ‘Congress’ purpose was broadly to give support to the existing and future state systems for regulating and taxing the business of insurance.’ ” Fabe, 508 U.S. at 500, 113 S.Ct. at 2207 (quoting Prudential Ins. Co. v. Benjamin, 328 U.S. 408, 429, 66 S.Ct. 1142, 1154, 90 L.Ed. 1342 (1946)). Indeed, the quoted language of Section 2(b) “impos[es] what is, in effect, a clear-statement rule, a rule that state laws enacted ‘for the purpose of regulating the business of insurance’ do not yield to conflicting federal statutes unless a federal statute specifically requires otherwise.” Id. at 507, 113 S.Ct. at 2211.

The Commissioner seizes on Section 2(b) to contend that the district court’s decision “renders meaningless” the Puerto Rico Insurance Code provisions requiring that insurance companies make statutory deposits. See 26 L.P.R.A. §§ 801-809. Because Guaranty assigned the CD to the Commissioner in order to comply with this statute, the Commissioner concludes that the district court’s decision upholding the FDIC’s refusal of deposit insurance impairs the Commissioner’s ability to regulate the business of insurance in Puerto Rico, and therefore violates Section 2(b). The decision, the Commissioner contends, particularly impairs “obtaining eligible deposits from insurance companies to comply with the statutory deposit requirement of the Insurance Code, whose ultimate aim is to protect policyholders in case of the insurer’s insolvency.” (Appellant’s Brief, at 20).

The Supreme Court has set out the factors required for a federal statute to fall within the McCarran-Ferguson Act. First, the federal statute must “invalidate, impair, or supersede” the state act. Second, the federal statute must not “specifically relat[e] to the business of insurance.” Finally, the state law must have been enacted “for the purpose of regulating the business of insurance.” Fabe, 508 U.S. at 507, 113 S.Ct. at 2208. We need go no further than the first factor in our analysis, as the Commissioner’s argument that the application of the Puerto Rico Insurance Code is impaired fails.

The Supreme Court faced the question of when an “impairment” occurs in SEC v. National Sec., Inc., 393 U.S. 453, 89 S.Ct. 564, 21 L.Ed.2d 668 (1969). In that case, the SEC sought to unwind the merger of two insurance companies which had been approved by the state of Arizona, on the basis that the merger was obtained through use of fraudulent misrepresentations. Arizona argued that the SEC’s action would violate the McCarran-Ferguson Act. The Supreme Court, finding that the essential question was “whether the McCarran-Ferguson Act bars a federal remedy which affects a matter subject to state insurance regulation,” id. at 462, 89 S.Ct. at 569, disagreed.

The gravamen of the complaint was the misrepresentation, not the merger____ Nevertheless, [the state] contend[s] that any attempt to interfere with a merger approved by state insurance officials would “invalidate, impair, or supersede” the state *736 insurance laws____ We cannot accept this overly broad restriction on federal power.
It is clear that any “impairment” in this case is a most indirect one.

Id,, at 462-63, 89 S.Ct. at 569-70. The courts have relied on this logic to conclude that “application of a federal law [will] be precluded only where the federal law expressly prohibits] acts permitted by state law, or vice versa.” Merchants Home Delivery Serv., Inc. v. Frank B. Hall & Co., 50 F.3d 1486, 1492 (9th Cir.1995) (holding that application of a federal law which prohibited acts also prohibited by state insurance law did not invalidate, impair, or supersede state law, despite their differing remedies), cert. denied sub nom., Prometheus Funding Corp. v. Merchants Home Delivery Serv., Inc., — U.S. -, 116 S.Ct. 418, 133 L.Ed.2d 335 (1995); see American Family Mut. Ins. Co., 978 F.2d at 296-97 (drawing on analogies to the principles governing federal preemption of state law).

Application of this “direct conflict” test quickly defeats the Commissioner’s argument. In short, nothing in the district court opinion — or the FDIC regulations — impairs the Commissioner’s authority or ability to obtain deposits from insurance companies to comply with the statutory deposit requirement. The opinion and regulations merely set out what records the FDIC may rely on in making insurance determinations. The Commissioner’s loss is the product of events, not a conflict between federal and Commonwealth statutes. In the absence of a direct prohibition, we refuse to hold that there has been an impairment merely because in this circumstance the Commissioner suffered a loss. Cf. Merchants Home Delivery, 50 F.3d at 1492 (“The language of § 2(b) is inconsistent with a congressional intent to allow states to preempt the field of insurance regulation.”).

CONCLUSION

For the reasons stated above, we affirm.

Notes

1

. In its corporate capacity, the FDIC functions as a bank regulator and insurer of bank deposits. 12 U.S.C. §§ 1818, 1821(a) (1988 & Supp.l99Í). The Commissioner does not seek review of that part of the district court decision that dismissed the complaint as against the FDIC as receiver of the former Girod Trust Company.

2

. The Commissioner asks us to note that "deposit account records" are defined to include certificates of deposits and "other books and records of the insured depository institution, ... which relate to the insured depository institution’s deposit taking function.” 12 C.F.R. § 330.1(d) (1995). This language proves unhelpful, however, since it is undisputed that there was no Certificate among the Bank’s records at the time of failure. Had the Certificate remained in the records, this case would likely not have arisen.

At oral argument, the Commissioner raised the application of 12 C.F.R. § 330.4(b)(4), which states, in pertinent part:

If any deposit obligation of an insured depository institution is evidenced by a negotiable certificate of deposit ... the FDIC will recognize the owner of such deposit obligation for all purposes of claim for insured deposits to the same extent as if his or her name and interest were disclosed on the records of the insured depository institution ....

The Commissioner relies on this to argue that he should be recognized as the owner for insurance claims purposes. We are inclined to consider the argument waived, as the Commissioner did not argue on the basis of section 330.4 below or, indeed, in his brief to us. Even if we were to address his argument, however, this section would not aid his cause since, assuming the Certificate was a negotiable instrument, it merely sets him up as the "owner of such deposit obligation.” Establishing that the FDIC should recognize him as the owner of the obligation— which it argues it has — does not save him from the language of section 330.3(h) and the fact that the bank records did not disclose the existence of an account at the time of failure.

3

. A purchase and assumption transaction occurs when, in its capacity as receiver, the FDIC sells a failed bank’s healthy assets to a purchasing bank in exchange for that bank’s promise to pay the depositors of the failed bank. FDIC-receiver next sells the ‘bad’ assets to itself as FDIC-cotporate. FDIC-receiver uses the money received to pay the purchasing bank to make up the difference between what the purchasing bank was willing to pay for the good assets and what it must pay out to the failed bank’s depositors. FDIC-corporate then tries to collect on the bad assets. This purchase and assumption generally needs to be executed with speed, often overnight. Timberland Design, Inc. v. First Serv. Bank For Sav., 932 F.2d 46, 48 (1st Cir.1991).

4

. This circuit has not previously reached the issue of whether a bank’s correctly recorded but unauthorized activity precludes an insurance claim. See FDIC v. Fedders Air Conditioning, 35 F.3d 18, 23 (1st Cir.1994). Indeed, in the present case's first appearance before this court, we noted we had not decided “whether or to what extent we would be willing to follow the Eighth Circuit's holding in In re Collins." Villafañe-Neriz, 20 F.3d at 40 n. 6. In affirming the district court’s decision today, we adopt much of the logic of the Collins decision, as well as the decisions of the district court for Massachusetts which have decided similar cases. See Fletcher Village, 864 F.Supp. at 265 (adopting the reasoning of Collins)', McCloud, 853 F.Supp. at 559.

5

. It is irrelevant to this discussion that this Court, in our earlier panel opinion, see Villafañe-Neriz, 20 F.3d at 41, stated that "Girod’s crediting of the certificate of deposit towards Guaranty’s $600,000 loan and the resulting 'relationship' between the certificate of deposit and the loan, cannot be characterized as anything other than ineffective." Our holding here is not based on the premise that a setoff took place, but on the fact that the records indicated that the account no longer existed.

6

. We note in passing that the Commissioner argues on the basis of fraud in the transaction underlying the records, since there is no question that the records themselves were accurate. However, the cases the Commissioner seeks to rely on in his argument — McCloud, Abdulla Fouad, and Collins — all discuss an exemption based on fraud in a bank's recordkeeping. See Collins, 998 F.2d at 556 (noting that there was "no allegation of fraud” where the bank’s records "accurately reflected payout of $100,000 to the wrong party.”); Abdulla Fouad, 898 F.2d at 485-86 (refusing to extend exception to include proof directed to the deposit account recording); McCloud, 853 F.Supp. at 560 (concluding that, where there was evidence that the records of deposits were altered during the course of fraudulent conduct by the bank president, "it was arbitrary, capricious and contrary to law for. the agency to consider the customer profile as of the day of the bank’s default as conclusive” (footnote omitted)). Thus we question whether the Commissioner would have met with success, even if he had both raised this argument in the court below and demonstrated that there was fraud in the transaction.