FEDERAL DEPOSIT INSURANCE CORPORATION, аs Receiver of AmTrust Bank v. AMFIN FINANCIAL CORPORATION; AMFIN REAL ESTATE INVESTMENTS, INC.; AMFIN INSURANCE AGENCY, INC.; AMFIN INVESTMENTS, INC.; AMFIN PROPERTIES, INC.; AMFIN MANAGEMENT, INC.
No. 13-3669
UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT
July 8, 2014
RECOMMENDED FOR FULL-TEXT PUBLICATION
Pursuant to Sixth Circuit I.O.P. 32.1(b)
File Name: 14a0142p.06
No. 1:11-cv-02574—John R. Adams, District Judge.
Argued: March 19, 2014
Decided and Filed: July 8, 2014
Before: GILMAN, COOK, and McKEAGUE, Circuit Judges.
COUNSEL
ARGUED: Joseph Brooks, FEDERAL DEPOSIT INSURANCE CORPORATION, Arlington, Virginia, for Appellant. Pierre H. Bergeron, SQUIRE SANDERS (US) LLP, Cincinnati, Ohio, for Appellees. ON BRIEF: Joseph Brooks, FEDERAL DEPOSIT INSURANCE CORPORATION, Arlington, Virginia, for Appellant. Pierre H. Bergeron, Lauren S. Kuley, SQUIRE SANDERS LLP, Cincinnati, Ohio, Philip M. Oliss, SQUIRE SANDERS LLP, Cleveland, Ohio for Appellees.
COOK, J., delivered the opinion of the court, in which McKEAGUE, J., joined and GILMAN, J., joined except as to Part II.B. GILMAN, J. (pp. 12-14), delivered a separate concurring opinion.
COOK, Circuit Judge. This case presents the question of who owns a $170 million tax refund: a parent corporation that filed a consolidated tax return on behalf of its subsidiaries and to whom the IRS issued the refund, or the subsidiary whose net operating loss generated the refund. The answer depends on whether the parties had a debtor-creditor relationship—in which case the parent corporation owns the refund but with an obligation to repay the subsidiary—or an agency/trust relationship—in which case the subsidiary owns the refund, with the parent acting merely as an agent or trustee.
Appellee AmFin Financial Corporation (“AFC“), the parent of a group of banks that included AmTrust Bank (“AmTrust“), insists that a tax-sharing agreement (“TSA“) mandates that a $170 million tax refund (“Refund“) generated by AmTrust‘s net losses belongs to AFC‘s bankruptcy estate, and that AmTrust is merely another creditor of the estate. The district court agreed, holding that the TSA unambiguously allocated the Refund to thе now-bankrupt AFC. Finding the TSA silent on this issue, we reverse and remand with instructions that the district court consider extrinsic evidence concerning the parties’ intent in light of Ohio agency and trust law.
I.
In 2006, AFC and its affiliates (“Affiliated Group“), including AmTrust, entered into the current TSA for the purpose of allocating tax liability. In November 2009, AFC filed for Chapter 11 bankruptcy protection, and, as part of that reorganization, the federal Office of Thrift Supervision closed AmTrust and placed it into FDIC receivership. See In re AmTrust Fin. Corp., 694 F.3d 741, 745–49 (6th Cir. 2012). AFC later filed a consolidated 2008 tax return on behalf of the Affiliated Group showing a total net operating loss of $805 million, with AmTrust‘s losses accounting for $767 million of the total. After AFC took the position that any refund would belong to its bankruptcy estate, the parties agreed by stipulation to deposit refunds in a segregated account pending full adjudication of the parties’ respective ownership claims. When
The FDIC filed a complaint seeking a declaratory judgment that AmTrust owns the Refund. The FDIC later moved to amend its complaint after uncovering new evidence regarding the parties’ intent as to the ownership of refunds. The district court denied this mоtion, granting AFC judgment on the pleadings instead. Borrowing another court‘s analysis of a different TSA, the district court reasoned that the TSA‘s use of terms such as “reimbursement” and “payment” definitively established a debtor-creditor relationship between AFC and its subsidiaries as to tax refunds, thereby justifying the court‘s awarding the Refund to AFC‘s bankruptcy estate. Though the FDIC proffered extrinsic evidence showing that the parties intended to create an agency or trust relationship under Ohio law with respеct to tax refunds, the district court declined to consider that evidence and rejected the FDIC‘s trust and agency arguments without further analysis. The FDIC appeals.
II.
We review a district court‘s judgment on the pleadings “using the same de novo standard of review employed for a motion to dismiss under
Because the district court concluded that the TSA unambiguously allocated the Refund to AFC‘s bankruptcy estate, we first consider the effect of that document.
Finding the TSA integrated and unambiguous, the district court concluded as a matter of law that it created a debtor-creditоr relationship with respect to tax refunds between AFC and its affiliates, including AmTrust, and thus the Refund belonged to AFC‘s bankruptcy estate. We review de novo “a district court‘s conclusions regarding ambiguity in contract language.” Schachner v. Blue Cross & Blue Shield of Ohio, 77 F.3d 889, 893 (6th Cir. 1996).
1. The Text of the TSA
The FDIC first argues that the TSA says nothing about who owns tax refunds issued to AFC in its role as filer of consolidated tax returns on behalf of the Affiliated Group.1 The stated purpose of the TSA is to “specify the manner in which [the Affiliated Group] will share the Consolidated Tax Liability . . . and the mаnner in which certain Tax Attributes . . . are to be treated among the members of the Affiliated Group.” To that end, the TSA directs that “[e]ach Affiliate shall be responsible for and shall reimburse the Paying Entity for its share of the Affiliated Group‘s Consolidated Tax liability.” Importantly, none of the TSA‘s provisions addresses the disposition or ownership of a refund issued by the IRS to AFC in its capacity as parent of the Affiliated Group. Rather, in keeping with the TSA‘s purpose, section two of the TSA allocates tax liability and requires members to timely pay their shares, and section three of the TSA allows members of the Affiliated Group to use other members’ net operating losses to reduce their tax liability.
AFC claims that section four of the TSA applies here “[b]y requiring that members promptly ‘settle any amounts’ calculated by reference to the refund.” The relevant text reads:
4. Adjustment. In the event of any adjustment of the Consolidated Tax Liability of the Affiliated Group for any Consolidated Return year by reason of the filing of [a] . . . loss carryback refund application, . . . the respective liabilities of Common Parent and each Affiliate shall be redetermined hereunder after fully giving effect to any such adjustment, as if such adjustment had been part of the original computation; and the parties shall promptly settle any amounts owing among them.
2. The District Court‘s Analysis
For its part, the district court pointed to no particular provision to support its conclusion that the TSA unambiguously allocated the Refund to AFC, relying on another court‘s analysis of an altogether diffеrent TSA instead. See In re IndyMac Bancorp, Inc., No. 2:08-bk-21752-BB, 2012 WL 1037481 (Bankr. C.D. Cal. Mar. 29, 2012) (report and recommendation), aff‘d, 554 F. App‘x 668 (9th Cir. 2014). That TSA “expressly authorize[d] [the parent corporation], in its sole discretion, to determine whether any tax refunds to which the consolidated group is entitled will be paid or credited against future tax liabilities of the consolidated group.” Id. at *14 (internal quotation marks omitted). That agreement also “provide[d] for ‘payment’ or ‘reimbursement’ from [the parent corporation] to the [affiliate bank] under certain conditions if [the аffiliate bank] suffer[ed] losses that would have entitled it to a refund had it filed separate tax returns.” Id. The TSA in IndyMac thus expressly stated the circumstances under which the parent corporation would disburse refunds to the group and gave the parent corporation discretion as to whether to distribute refunds at all. And in affirming, the Ninth Circuit explicitly referenced the fact that the TSA there “describes the process by which [the parent corporation] will allocate . . . tax refunds” and “givеs [the parent corporation] ‘sole discretion’ to determine the ‘means and manner’ of . . . paying refunds.” IndyMac, 554 F. App‘x at 668.
Likewise, the TSAs in other cases cited by AFC include language directly addressing the distribution of refunds. See, e.g., In re Imperial Capital Bancorp, Inc., 492 B.R. 25, 30 (Bankr. S.D. Cal. 2013); In re First Cent. Fin. Corp., 269 B.R. 481, 490 (Bankr. E.D.N.Y. 2001). As no similar language appears in the TSA here, these cases offer AFC no support.
As pressed by the FDIC, the Eleventh Circuit persuasively rejected AFC‘s refund-ownership position in a similar case, In re BankUnited Financial Corp., 727 F.3d 1100 (11th Cir. 2013). Like the district court here, the BankUnited bankruptcy court held that the TSA‘s use of terms such as “payables” and “receivables” evidenced the parties’ unambiguous intent that a bank‘s parent company would retain tax refunds generated by the bank with only a debtor‘s obligation to repay the bank. In re BankUnited Fin. Corp., 462 B.R. 885, 899–900 (Bankr. S.D. Fla. 2011). The Eleventh Circuit reversed, rejecting the bankruptcy court‘s terminology-based rationale and holding that “[w]e find no words in the TSA from which it could reasonably be inferred that the parties agreed that the [parent company] would retain the tax refunds . . . and . . . be indebted to the [b]ank.” BankUnited, 727 F.3d at 1108. The court furthеr explained that “[i]f . . . the parties created a debtor-creditor relationship, we would expect to find . . . protection[s] for the creditor [bank] that would help guarantee the debtor [parent company‘s] obligation, such as a fixed interest rate, a fixed maturity date, or the ability to accelerate payment upon default.” Id.
Just so here: The TSA says nothing about tax refunds received by AFC on behalf of the group and includes no protections for the putative creditor, as one would expect if the parties intended a debtor-creditor relationship. And, just like BankUnited, AFC‘s straining to imbue the commonplace terms “payment” and “reimbursement” with specialized and unambiguous meaning falls flat. See Alexander v. Buckeye Pipe Line Co., 374 N.E.2d 146, 150 (Ohio 1978) (“[C]ommon words appearing in a written instrument are to be given their plain and ordinary meaning unless manifest absurdity results or unless some other meaning is clearly intended from the face or overall contents of the instrument.“).2
4. Conclusion
In sum, nothing in the TSA evidencеs an unambiguous intent to create a debtor-creditor relationship and thereby allocate the refund to AFC. Moreover, persuasive case law supports our conclusion that the use of terms such as “reimbursement” and “payment” need not create a
B. The Propriety of the Bob Richards Court‘s Analysis
Finding the TSA ambiguous, we next consider the FDIC‘s contention that we should apply the principle first enunciated in In re Bob Richards Chrysler-Plymouth Corp., 473 F.2d 262, 265 (9th Cir. 1973), that “[a]bsent any differing agreement[,] . . . a tax refund resulting solely from offsetting the losses of one member of a consolidated filing group against the income of that same member in a prior or subsequent year should inure to the benefit of that member.” According to the FDIC, the Refund belоngs to AmTrust under this analysis.
But this court-created “rule” is a creature of federal common law, see In re NetBank, Inc., 729 F.3d 1344, 1347 n.3 (11th Cir. 2013), and, as AFC points out, federal common law constitutes “an unusual exercise of lawmaking which should be indulged only in a few restricted instances.” Cent. States, Se. & Sw. Areas Pension Fund v. Mahoning Nat‘l Bank, 112 F.3d 252, 256 (6th Cir. 1997) (internal quotation marks omitted) (quoting Milwaukee v. Illinois, 451 U.S. 304, 313 (1981)). In particular, the Supreme Court instructs courts to invoke federal common law only when “there is a significant conflict between some federal policy or interest and the use of state law.” O‘Melveny & Myers v. FDIC, 512 U.S. 79, 87 (1994).
This case does not call for fеderal common law because our precedent recognizes that “[s]tate law determines whether [property is] excluded from the debtor‘s estate.” In re Cannon, 277 F.3d 838, 849 (6th Cir. 2002) (citing Barnhill v. Johnson, 503 U.S. 393, 398 (1992)); see also Butner v. United States, 440 U.S. 48, 54 (1979) (“Congress has generally left the determination of property rights in the assets of a bankrupt‘s estate to state law.“). If Congress generally allows state law to determine the property interests subject to bankruptcy, we cannot say that this situation constitutes one of the “few restricted instances” where fеderal common law “should be indulged.” See Cent. States, 112 F.3d at 256; see also United States v. Kimbell Foods, Inc., 440 U.S. 715, 729 (1979) (rejecting “generalized pleas for uniformity” absent “concrete evidence that adopting state law would adversely affect” federal interests).
C. Extrinsic Evidence of a Resulting Trust or Agency Relationship Under Ohio Law
Because the TSA says nothing about the ownership of refunds and we decline to apply federal common law, we agree with the FDIC‘s alternative argument that the district court must look to the evidence of the parties’ intent unearthed during discovery. The FDIC contends that this evidence shows either a trust or agency relationship with respect to tax refunds, such that the Refund properly belongs to AmTrust.3 AFC lodges a number of objections to the consideration of this evidence. None has merit.
1. Resulting Trust
Under Ohio law, “a resulting trust is based on the parties’ intentions.” Brate v. Hurt, 880 N.E.2d 980, 985 (Ohio Ct. App. 2007). “A resulting trust has been defined as one . . . where the legal estate in property is transferred or acquired by one under facts and circumstances which indicate that the beneficial interest is not intended to be enjoyed by the holder of the legal title.” First Nat‘l Bank of Cincinnati v. Tenney, 138 N.E.2d 15, 17 (Ohio 1956) (internal quotаtion marks omitted). According to the FDIC, AFC and its affiliates intended that AFC distribute tax refunds received on behalf of the Affiliated Group as though the companies had filed separate
First, AFC makes much of the fact that the TSA contains no language evincing a trust relationship. But the FDIC never argued that the TSA created an express trust; rather, the FDIC urges the court to find an implied resulting trust.
Second, AFC cites In re Omegas Group, Inc. for the proposition that the Sixth Circuit recognizes no implied trusts in bankruptcy. 16 F.3d 1443 (6th Cir. 1994). The Omegas Group court held that a constructive trust could not exclude property from a bankruptcy estate. Id. at 1451. The court reasoned that “[i]mposition of a constructive trust clearly thwarts the policy of ratable distribution” because “a constructive trust . . . is a remedy [and] does not exist until a plaintiff obtains a judicial decision finding him to be entitled to a judgment ‘impressing’ defendant‘s property.” Id. (internal citations omitted). AFC maintains that this reasoning appliеs with equal force to resulting trusts.
But resulting trusts, unlike constructive trusts, do not subvert the policy of ratable distribution. As explained by the Ninth Circuit Bankruptcy Appellate Panel:
[T]he bankruptcy policy of ratable distribution [is implicated] only in determining whether to impose a constructive trust, not in applying state law governing resulting trusts. Enforcement of a resulting trust differs from postpetition imposition of a constructive trust in two important respects. First, a resulting trust gives effect to the intent of the parties. As such, the circumstances in which it is effective arе governed by the usual standards for establishing intent, and not by the “relatively undefined” equitable considerations that govern constructive trusts. Second, because a resulting trust merely gives effect to the original will of the parties, it is effective from the date of the original transfer, and does not undermine ratable distribution among creditors who possess similar legal rights as of the petition date.
In re Sale Guar. Corp., 220 B.R. 660, 667–68 (B.A.P. 9th Cir. 1998) (internal citations omitted); see also In re Cedar Funding, Inc., 408 B.R. 299, 314–15 (Bankr. N.D. Cal. 2009) (“Because a resulting trust is a judicial recognition that the parties’ relationship is and was always intended to be that of trustee and beneficiary, not debtor and creditor, the trust property never belongs to the debtor and does not become part of the bankrupt‘s estate.“).
Last, AFC notes that Ohio law “generally” recognizes resulting trusts in three situations: “(1) purchase-money trusts, (2) instances where an express trust does not exhaust the res given to the trustee, and (3) where express trusts fail, in whole or in part.” Univ. Hosps. of Cleveland, Inc. v. Lynch, 772 N.E.2d 105, 117 (Ohio 2002). Because none of these situations presents itself, AFC insists that the FDIC‘s trust theory cannot succeed. But the court in Lynch did not purport to set forth an exclusive list of recognizable resulting trusts. Indeed, Ohio courts frequently recognize resulting trusts in other situations. See, e.g., Woodward v. Kleese, No. 2007-T-0002, 2007 WL 2822753, at *5 (Ohio Ct. App. Sept. 28, 2007) (affirming imposition of resulting trust when circumstances showed transferred money was to be returned to the transferor upon request); Bell v. Straight, Inc., 707 F. Supp. 325, 329 (S.D. Ohio 1989) (recognizing viability of resulting-trust theory when plaintiffs alleged they contributed money to a charity for a specific purpose but the charity failed to use money for that purpose). These cases underscore the key resulting-trust inquiry: whether the parties intended the rеcipient of the property to also hold the beneficial interest.
2. Agency Relationship
The FDIC‘s evidence might also establish an agency relationship. An agency-principal relationship exists in Ohio “when one party exercises the right of control over the actions of another, and those actions are directed toward the attainment of an objective which the former seeks.” Hanson v. Kynast, 494 N.E.2d 1091, 1094 (Ohio 1986). AFC repeats its mantra that the court may not consider any extrinsic evidence bеcause the TSA “governs tax matters[] and . . . is unambiguous.” But as discussed above, the TSA says nothing concerning the disposition and ownership of Affiliated Group members’ shares of tax refunds. And Ohio law instructs that courts may consider extrinsic evidence if “the language of a contract is unclear or ambiguous, or
III.
For these reasons, we REVERSE and REMAND for proceedings consistent with this opinion.
RONALD LEE GILMAN, Circuit Judge, concurring. I agree with the result reached by the lead opinion and with all of its analysis other than in Part II.B., which is titled “The Propriety of the Bob Richards Court‘s Analysis.” The lead opinion in Part II.B. rejects what it characterizes as the federal common-law rule set forth in Western Dealer Management, Inc. v. England (“In re Bob Richards Chrysler—Plymouth Corp.“), 473 F.2d 262 (9th Cir. 1973), and instead instructs the district court to look solely to state law in deciding whether AFC (the bank holding company) or AmTrust (the bank) owns the tax refund in question.
In contrast to the lead opinion, I do not read the Bob Richards decision as requiring an “either-or” choice between federal and state law. The key holding of that case is as follows:
[W]here there is an explicit agreement, or where an agreement can fairly be implied, as a matter of state corporation law the parties are free to adjust among themselves the ultimate tax liability. But in the instant case the parties made no agreement concerning the ultimate disposition of the tax refund. Absent any differing agreement[,] we feel that a tax refund resulting solely from offsetting the losses of one member of a consolidated filing group against the income of that same member in a prior or subsequent year should inure to the benefit of that member.
Id. at 264–65. That is, Bob Richards held that courts should first look to state corporate and contract lаw to resolve doubts in situations such as this and, only if the court comes up empty handed, should the court then allocate the refund to the entity that generated the refund because federal tax law does not change the ownership of the refund.
This view appears to have gained general acceptance since the Ninth Circuit issued its decision in 1973. In In re Prudential Lines Inc., 928 F.2d 565 (2d Cir. 1991), for example, the Second Circuit considered whether a parent or a subsidiary owned a net operating loss that had been accrued by the subsidiary prior to the subsidiary entering bankruptcy. The Second Circuit first acknowledged that “where there is an explicit agreement, or where an agreement can fairly be implied, as a matter of state corporation law the parties are free to adjust among themselves
The Fifth Circuit followed a similar approach in Capital Bancshares, Inc. v. Federal Deposit Insurance Corp., 957 F.2d 203 (5th Cir. 1992). Considering whether a parent or a bankrupt subsidiary owned a tax refund paid to the parent, the court reasoned: “Following the In re Bob Richards reasoning, the refund is the property of the Bank in the absence of a contrary agreement.” Id. at 208.
This brings us to two more recent decisions in the Eleventh Circuit. The first is In re BankUnited Financial Corp, 727 F.3d 1100 (11th Cir. 2013), cert. denied, 134 S. Ct. 1505 (2014), where the Eleventh Circuit considered whether a tax refund generated by an insolvent bank was owned by the bank or by the bank‘s holding company that filed the consolidated tax return. It framed the dispositive question as “a matter of contract interpretation,” id. at 1104, and ultimately concluded, based on an analysis of the Tax Sharing Agreement, that the tax refund should be distributed to the bank, id. at 1109.
The Eleventh Circuit quickly returned to the same issue in In re NetBank, Inc., 729 F.3d 1344 (11th Cir. 2013). Again considering the tax-sharing agreement before it, the Eleventh Circuit concluded as follows:
Following the BankUnited decision, and implementing the express provision in the instant TSA, we apply state contract law—i.e., Georgia contract law. We notе, however, that the outcome of the instant case would not be different if the ”Bob Richards rule” were applied. We conclude that the intent of the parties expressed in the TSA—the controlling factor under either Georgia contract law or the federal common law as articulated in the ”Bob Richards rule“—created an agency relationship.
I believe that all of these cases can be reconciled with the following protocol: First look to the tax-sharing agreement (TSA) to settle the refund issue. But if the TSA is ambiguous, then determine whether an agreement between the parties can be implied under state law. And if that inquiry also proves inconclusive, then the loss belongs to the entity that generated the tax refund because federal tax law does not change the ownership of the refund.
This protocol is based on the foundational case of Bob Richards and has essentially beеn followed by all of our sister circuits that have weighed in on the issue. I thus see no need for the lead opinion to opine on what is sees (wrongly in my view) as a purported conflict between state law and federal common law.