ORDER REVERSING BANKRUPTCY COURT’S JUDGMENT
William J. Martinez, United States District JudgeThe Federal Deposit Insurance Corporation (“FDIC”), acting as receiver for the defunct United Western Bank (“Bank”), appeals the Bankruptcy Court’s determination that a tax refund generated on account of the Bank’s losses should remain a part of the bankruptcy estate of the Bank’s parent company, United Western Bancorp, Inc. (“Holding Company”). See In re United W. Bancorp, Inc., 558 B.R. 409 (Bankr. D. Colo. 2016) (“UWBI ”). For the reasons explained below, the Court finds that the relevant contract between the Bank and the Holding Company is ambiguous regarding whether the Holding Company may keep the tax refund in the present circumstances. That contract further requires that any ambiguity be construed in favor of the Bank. Accordingly, the tax refund is not part of the Holding Company’s bankruptcy estate and must be remitted to the Bank. The judgment of the Bankruptcy Court is reversed.1
I. STANDARD OF REVIEW
In reviewing a bankruptcy court’s decision, the district court normally functions as an appellate court, reviewing the bankruptcy court’s legal conclusions de novo and its factual findings for clear error. 28 U.S.C. § 158(a); In re Warren, 512 F.3d 1241, 1248 (10th Cir. 2008). The Bankruptcy Court’s judgment rested on a *879contract interpretation made as a matter of law, so this Court’s review is de novo. In re Universal Serv. Fund Tel. Billing Practice Litig., 619 F.3d 1188, 1203 (10th Cir. 2010).
II. BACKGROUND & PROCEDURAL HISTORY2
A. The Holding Company and the Bank
The Holding Company owned thirteen subsidiaries. (App. 41, 45-46.) One of those subsidiaries was the Bank, which the Holding Company wholly owned, and which was the Holding Company’s principal asset. UWBI, 558 B.R. at 416. The Bank operated eight branches and a loan servicing office in Colorado. Id.
B. The Tax Allocation Agreement (TAA)
The Internal Revenue Code permits an “affiliated group” of corporations (those with a common parent and a chain of sufficient stock ownership) to file a “consolidated [tax] return” that aggregates the gains and losses of all of them as if one corporation. See 26 U.S.C. §§ 1501-04. To facilitate such consolidated filing, eligible affiliated groups often enter into a written agreement amongst themselves known as a tax sharing agreement or a tax allocation agreement. Here, the Holding Company and its subsidiaries entered into a “Tax Allocation Agreement” (“TAA”).3
The TAA is dated January 1, 2008, and was signed by representatives of the Holding Company and its thirteen subsidiaries, including the Bank. (App. 41, 45-46.) It refers to all of the subsidiaries combined as “the Group,” and also sometimes as “the Affiliates.” (App. 41.) The TAA’s recitals announce its purpose as follows: “to establish a method for (i) allocating the consolidated tax.liability of the Group among its members, (ii) reimbursing [the Holding Company] for the payment of such tax liability, and (iii) compensating each member of the Group for the use of its losses by any other member of the Group.” (Id.)
To accomplish this purpose, the TAA first proclaims the following “General Rule” for federal tax filings:
Except as specifically set forth herein to the contrary, each Affiliate shall pay [the Holding Company] an amount equal to the federal income tax liability such Affiliate would have incurred were it to file a separate return (or, if appropriate, a consolidated return with its subsidiary affiliates). If [the Bank] incurs a net operating loss or excess tax credits, the [Bank] is entitled to a refund [from the Holding Company] equal to the amount that it would have been entitled to receive had it not joined in the filing of a consolidated return with [the Holding Company]. Similar treatment is optional at [the Holding Company’s] discretion for [other] Affiliates. Any refund shall generally not exceed the amount claimed or received as a refund resulting from a carryback claim filed by [the Holding Company], However, this shall not prevent [the Holding Company] from the ability to make a refund over the amount received or claimed as a refund or carryback, if in its sole discretion it *880believes such payment is in its best interest.
(TAA § A.l.) Having proclaimed this general rule, the TAA then goes on to re-proclaim its purpose, although with a different focus than that evident in the recitals: “In essence, this Agreement requires that each [Affiliate] be treated as a separate taxpayer with [the Holding Company] merely being an intermediary between an Affiliate and the Internal Revenue Service (TRS’).” (Id. § A.2.)
The details of actual cash flow to and from the Holding Company and the Affiliates are addressed later in the TAA, and here the TAA starts to become somewhat convoluted. As best the Court can discern, each Affiliate was required to pay to the Holding Company the Affiliate’s “hypothetical estimated income tax liability” on a quarterly basis at around the same time that the Holding Company was required to make estimated quarterly payments to the IRS. (Id. §§ F.l, F.2.) However, “[payments [from the Holding Company] to an Affiliate for net operating losses or similar items shall not to be made under this [quarterly] provision, but rather on an annual basis pursuant to Section A.” (Id, § F.3.)
While the cross-referenced “Section A” certainly discusses refunds from the Holding Company for an Affiliate’s net operating losses, it actually says nothing about the timing of those refunds, e.g., on an annual basis or otherwise. Rather, that seems to come from Section E. That section first instructs Affiliates that they must make “preliminary tax settlement payments ... on or before March 15 following the end of the appropriate taxable year.” (Id, § E.l.) The Court presumes this refers to any amounts over those already paid on a quarterly estimated basis during the previous year. In any event, the various parties’ obligations are trued-up towards the end of each year: “Final tax settlement payments or refunds are due on or before November 15.” (Id. § E.2.) This appears to be the “annual basis” referred to for refunds based on quarterly net losses.
The TAA actually contains three distinct refund provisions. One such provision is that just discussed, ie., the “[f]inal tax settlement ... refundí ]” that is “due on or before November 15” of each year, with reference to the previous taxable year. This refers to a payment from the Holding Company to the Affiliate, likely from monies received from other Affiliates that owed taxes. The second refund provision is simply an accelerated process to obtain the same payment: “an Affiliate with a taxable loss for the year may recover [from the Holding Company] estimated taxes paid for that year before final settlement if an ‘expedited refund’ claim is filed with [the Holding Company] by February 15 following the end of the tax year.” (Id. § E.l.)
The third refund provision is of the most interest here, as it refers to refunds received by the Holding Company from the IRS, not any sort of refund of amounts paid by the Affiliates to the Holding Company. It establishes a 10-business-day deadline for distributing such refunds:
In the event of any adjustment to the tax returns of the Group as filed (by reason of an amended return, claim for refund, or an audit by a taxing authority), the' liability of the parties to this Agreement shall be re-determined to give effect to any such adjustment as if it had been made as part of the original computation of tax liability, and payments between the appropriate parties shall be made within 10 business days after any such payments are made [to the IRS] or refunds are received [from the IRS], or, in the case of contested *881proceedings, within 10 business days after a final determination of the contest.
(Id. § H.1.)4 In other words, although an Affiliate with net losses may make a claim on the Holding Company for a refund of estimated tax payments, it appears the Holding Company has a self-executing duty to distribute to the Affiliates any actual refund received from the IRS.
Three other provisions of the TAA are notable. First, it “shall be governed by and construed in accordance with the laws of the State of Colorado and the applicable laws of the United States of America.” (Id. § H.6.) Second, through the TAA,
[e]ach Affiliate hereby appoints [the Holding Company] as its agent ... for the purpose of filing such consolidated Federal Income tax returns for the [Group] as [the Holding Company] may elect to file and making any election, application or taking any action in connection therewith on behalf of the Affiliates. Each such Affiliate hereby consents to the filing of any such returns and the making of any such elections and applications.
(Id. § G.1.) Third, the TAA contains yet another statement of its purpose (ie., in addition to those statements found in its recitals and in § A.2), followed by an “ambiguity favors the Bank” clause: “The intent of this Agreement is to provide an equitable allocation of the tax liability of the Group among [the Holding Company] and the Affiliates. Any ambiguity in the interpretation hereof shall be resolved, with a view to effectuating such intent, in favor of any insured depository institution.” (Id. § H.4.)
C. The Origin of This Dispute
In January 2011, the Office of Thrift Supervision closed the Bank and appointed the FDIC as its receiver. UWBI, 558 B.R. at 416. Thus, the FDIC assumed the role of marshaling the Bank’s assets as best as possible to pay the Bank’s obligations.
Later in 2011, apparently, the Holding Company filed its consolidated 2010 tax return on behalf of the Affiliates, including the Bank. In fact, the Bank was particularly important to this tax return. Whereas the Bank had generated taxable income in 2008 (on which the Holding Company paid taxes), the Bank generated an even larger taxable loss in 2010. Id. at 417.5 The Internal Revenue Code permits corporations to “carryback” net operating losses for up to two taxable years. See 26 U.S.C. § 172. Thus, the Holding Company was permitted to carryback the Bank’s 2010 losses to offset the taxes paid in 2008. It therefore claimed a refund on its 2010 tax return of about $4.8 million. UWBI, 558 B.R. at 417. There is no dispute that, to whatever extent a refund was due, it was entirely the result of revenue generated by the Bank in 2008 and losses incurred by the Bank in 2010—or in other words, neither the Holding Company itself nor any Affiliate generated any gains or losses relevant to the requested refund.
While that refund claim was still pending, the Holding Company found itself insolvent—because the Bank was its only real source of operating income—and so the Holding Company filed for Chapter 11 reorganization in March 2012. (App. 17.) About a year later, the Bankruptcy Court converted the proceeding to a Chapter 7 liquidation. (Id.) Thus, the Trustee was appointed to perform essentially the same *882role for the Holding Company that the FDIC was performing for the Bank: to realize as much value as possible from the Holding Company’s assets so that creditors could receive at least some compensation.
D. The Adversary Proceeding
“After learning of the anticipated Tax Refund, the Trustee (acting on behalf of the [Holding Company’s bankruptcy] estate) filed [an] adversary proceeding against the FDIC (acting as receiver of the Bank)” to settle the question of whether the refund would belong to the Holding Company (the Trustee’s position) or the Bank (the FDIC’s position). UWBI, 558 B.R. at 412. The parties agreed “that the underlying facts [were] undisputed and the contest [could] be decided as a matter of law.” Id. They accordingly filed cross-motions for summary judgment.
The crux of the dispute was whether the anticipated refund would be considered property of the Holding Company’s bankruptcy estate. If so, the Bank could make “a general unsecured claim against [the Holding Company’s] bankruptcy estate for some or all of the Tax Refund, which should share pari passu with other general unsecured claims against [the estate].” UWBI, 558 B.R. at 415. On the other hand, the Bankruptcy Code recognizes that a debtor might possess “only legal title and not an equitable interest” in certain property. 11 U.S.C. § 541(d). If this is the case, the property in question “becomes property of the estate .,. only to the extent of the debtor’s legal title to such property, but not to the extent of any equitable interest in such property that the debtor does not hold.” Id. Naturally, the Trustee argued that the Holding Company would possess both legal title and the equitable interest in the anticipated refund, while the FDIC argued that the Holding Company would lack at least an equitable interest, if not legal title as well.
While the parties’ cross-motions were pending, the IRS issued the refund in the adjusted amount of approximately $4.1 million (“Tax Refund” or “Refund”). It deposited that sum into the Bankruptcy Court’s registry pending resolution of the adversary proceeding. UWBI, 558 B.R. at 412, 417 n.21.
E. The Bankruptcy Court’s Decision
On September 16, 2016, the Bankruptcy Court resolved the parties’ cross-motions in a lengthy, thorough, and thoughtful opinion. Various portions of the Bankruptcy Court’s analysis will be examined in detail below. For present purposes, it is enough to state that the Bankruptcy Court held:
• the Holding Company “has at least bare legal title to the Tax Refund,” id. at 428 (emphasis in original), a matter that the FDIC does not challenge on appeal and therefore will not be discussed further;
• the TAA is unambiguous, id. at 424 & n.26; and
• the TAA’s unambiguous terms establish that the relationship between the Holding Company and the Bank was that of debtor and creditor, not that of agent and principal or trustee and beneficiary, meaning that the Holding Company possesses an equitable interest in the Refund in addition to legal title, id. at 424-36.
Thus, the Bankruptcy Court concluded, the Refund was a part of the Holding Company’s bankruptcy estate and the Bank could only seek it through a general unsecured claim. Id. at 436-38.
The Bankruptcy Court entered final judgment based on this order, and the FDIC timely appealed. (App. 406.) This *883Court has jurisdiction to hear the appeal under 28 U.S.C. § 158(a)(1).
III. ANALYSIS
A. Preliminary Observation
There is an air of unreality about this litigation. Under normal circumstances, it would never have been brought. If the directors of a wholly-owned subsidiary elected to sue the parent company for a refund wrongly withheld under a tax allocation agreement, it is inconceivable that the parent would do anything other than replace the subsidiary’s directors with those who would cause the subsidiary to withdraw the lawsuit. Only in situations where an independent fiduciary takes control of the parent or the subsidiary—or where separate fiduciaries take over both, as in this case—is such a lawsuit likely to exist. Not surprisingly, then, every case cited by - either party in which a court addresses ownership of a tax refund under a tax sharing or allocation agreement involved either a bankruptcy trustee or the FDIC as a failed bank’s receiver.
Yet the parties here agree (as do the various cases they cite) that the question of refund allocation is ultimately a matter of contractual intent. But in what' sense can a court analyze contractual intent in these circumstances? How can a court say there was a “meeting of the minds” in any true sense between a parent company and a wholly-owned subsidiary? How does the subsidiary have any intent apart from that of its parent? Or to put it in concrete terms applicable to this case, imagine a parent company’s officers considering a draft tax allocation agreement and asking themselves, “Should this company or one of its subsidiaries end up in bankruptcy or receivership, and therefore pass out of our control, would we want any outstanding tax refund to remain the property of the parent, or to be distributed to the subsidiaries as usual?” It is difficult to imagine the parent company’s officers electing the latter course—and therefore difficult to imagine, in the present circumstances, how the TAA could require the refund to flow to the Bank in violation of the Holding Company’s near-certain contrary intent otherwise.
But these musings prove too much. The corporate fiction is deeply ingrained in American law. Formal agreements between parents and subsidiaries, and between subsidiaries themselves, are routine. Abiding by such formalities is generally a legal requirement, even if subsidiaries are really only carrying out the will of the parent. Thus, the analysis below proceeds as if the parties have always had their own respective purposes and interests to protect. But, as will become clear, the analysis below’ also will not stray far from the reality that these potentially conflicting purposes and interests will likely manifest themselves only in proceedings related to bankruptcy and receivership.6
B. General Principles
“The commencement of a [bankruptcy action] creates an estate.” 11 U.S.C. § 541(a). This estate comprises, among many other things, “all legal or equitable interests of the debtor in property as of the commencement of the case.” Id. § 541(b)(1). But, as noted above, the estate does not include property in which the debtor possesses “only legal title and not an equitable interest.” Id. § 541(d).
*884“In the absence of any controlling federal law, ‘property’ and ‘interests in property are creatures of state law.” Barnhill v. Johnson, 503 U.S. 393, 398, 112 S.Ct. 1386, 118 L.Ed.2d 39 (1992). Thus, whether the Holding Company possesses both a legal and an equitable interest in the Refund (thus making it a part of the bankruptcy estate) or only a legal interest (thus excluding it from the bankruptcy estate) should be a question of what sort of property interest.the TAA created under Colorado law, which governs that agreement.
C. The Bob Richards Rule
The Court says “should be a question ... under Colorado law” for good reason. Whether Colorado law actually applies (as opposed to some sort of federal common law) has been complicated by the “Bob Richards rule,” on which the FDIC relied heavily in the Bankruptcy Court and which it continues to press here,
1. Bob Richards and Barnes v. Harris
The Bob Richards rule comes from In re Bob Richards Chrysler-Plymouth Corp., Inc., 473 F.2d 262 (9th Cir. 1973). That case, like this one, involved a dispute over whether a parent or a subsidiary was owed a tax refund resulting from the consolidated tax filing. Id. at 263. In that case, like this one, it was acknowledged that the refund was due solely to losses incurred by the subsidiary. Id. In that case, however, the subsidiary was the party in bankruptcy, and its trustee was suing the parent to obtain the refund for-the bankruptcy estate—in contrast to the present case, where the Trustee speaks for the parent and is suing to prevent the Refund from leaving the bankruptcy estate.
But Bob Richards nonetheless presented the same basic question: to whom does the tax refund belong? The Ninth Circuit observed that “as a matter of state corporation law the parties are free to adjust among themselves the ultimate tax liability” through “an explicit agreement, or where an agreement can fairly be implied.” Id. at 264. “But in the instant case,” however,
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. Allowing the parent to keep any refunds arising’solely from a subsidiary’s losses simply because the parent and subsidiary chose a procedural device to facilitate their income tax reporting unjustly enriches the parent.
Id. at 265. This is what the parties here refer to as the Bob Richards rule.
The Ninth Circuit cited no authority for this proposition. It recognized that “state corporation law” permitted the affiliated corporations to explicitly allocate tax matters amongst themselves, but when it came to the lack of such ah agreement, the Ninth Circuit did not say whether the unjust enrichment rule it announced flowed from state law, federal law, or something else. This is significant because, as noted, property interests included within a debtor’s bankruptcy estate ordinarily “are creatures of state law.” Barnhill, 503 U.S. at 398, 112 S.Ct. 1386.
This has led the Sixth and Eleventh Circuits to conclude that the Bob Richards rule could only be an announcement of federal common law. FDIC v. AmFin Fin. Corp., 757 F.3d 530, 535 (6th Cir. 2014) (“AmFin ”); In re NetBank, Inc., 729 F.3d 1344, 1352 n.3 (11th Cir. 2013). And the Sixth Circuit has rejected Bob Richards on *885this basis, finding it an unnecessary exercise of federal common law authority. Am-Fin, 757 F.3d at 535-36.
If writing on a clean slate, this Court would be inclined to agree with the Sixth Circuit—the Bob Richards rule can only be grounded, if anywhere, in federal common law, and yet Bob Richards does not explain the need for a federal common law rule with respect to ownership of tax refunds, as compared to other sums of money whose ownership has been effectively analyzed under state law. See, e.g., Lubin v. Cincinnati Ins. Co., 677 F.3d 1039, 1041-42 (11th Cir. 2012) (analyzing under applicable state law whether an insurance payout was property of the bankrupt holding company or of a subsidiary failed bank). Thus, at a minimum, Bob Richards should not be reflexively applied. At oral argument before the Bankruptcy Court, the bankruptcy judge displayed similar concern. (See, e.g., App. at 275-76, 292.) Cf. Danforth v. Minnesota, 552 U.S. 264, 289-90, 128 S.Ct. 1029, 169 L.Ed.2d 859 (2008) (“while there are federal interests that occasionally justify ... development of common-law rules of federal law, our normal role is to interpret law created by others and not to prescribe what it shall be” (internal quotation marks and footnote omitted)); United States v. City of Las Cruces, 289 F.3d 1170, 1186 (10th Cir. 2002) (“The reluctance to create common law is a core feature of federal court jurisprudence. Federal courts should only fashion common law in a few and restricted circumstances.” (internal quotation marks and citations omitted)).
But this Court does not write on a clean slate, due to the Tenth Circuit’s recent decision in Barnes v. Harris, 783 F.3d 1185 (10th Cir. 2015), Barnes was a shareholder derivative action against a failed bank’s holding company. Id. at 1188. Among the plaintiffs’ theories of liability against the holding company’s directors was that the holding company should have held onto at least a portion of a $9 million tax refund received on behalf of the holding company and its subsidiaries. Id. at 1189, 1195. The district court dismissed this theory of derivative liability and the Tenth Circuit affirmed, relying on Bob Richards-.
As the district court explained, a tax refund due from a joint return generally belongs to the company responsible for the losses that form the basis of the refund. See [Bob Richards ]. Plaintiffs did not allege that the Holding Company possessed any business- interests other than the Bank that might have generated losses....
Plaintiffs counter that companies may agree to alter the default allocation rule by agreement. See Bob Richards, 473 F.2d at 265 .... Yet plaintiffs have not alleged the existence of any agreement to allocate the refund....
Id. at 1195-96. The Tenth Circuit therefore found the plaintiffs’ claim inadequately pleaded in this regard.
The Churt cannot deem Barnes’s adoption of Bob Richards to be dicta. “Dicta are statements and comments in an opinion concerning some rule of law or legal proposition not necessarily involved nor essential to determination of the ease in hand.” Thompson v. Weyerhaeuser Co., 582 F.3d 1125, 1129 (10th Cir. 2009) (internal quotation marks omitted). The Bob Richards rule was essential to the Tenth Circuit’s decision to affirm the district court. Moreover, the Tenth Circuit was explicitly presented with the argument— albeit for the first time in the appellant’s reply brief—that the Bob Richards rule was unnecessary federal common law. See 2014 WL 3795344, at *18-19. Thus, this Court cannot say that the Tenth Circuit was unaware of Bob Richards’s questiona*886ble status. For whatever reason, it chose to say nothing about the federal common law argument, but the argument was certainly before the court, the court did not announce that the argument had been forfeited (e.g.) as untimely),7 and the court applied the Bob Richards rule as if beyond question.
2. Effect of Bob Richards Here
But what, precisely, is the scope of the Bob Richards rule? The FDIC argued in the Bankruptcy Court, and continues to argue here, that Bob Richards mandates its default presumption unless there exists an inter-corporate agreement that unambiguously allocates the refund away from the party incurring the relevant losses. (App. 91-92, 354; ECF No. 10 at 27-28, 37, 39.)8 The Trustee argued in the Bankruptcy Court, and continues to argue here, that Bob Richards only applies in cases where there is no tax allocation agreement of any kind, which is not the case here. (App. 238-39; ECF No. 11 at 15-16.)
The Bankruptcy Court interpreted Bob Richards to fall essentially halfway between these two positions. See UWBI, 558 B.R. at 432-34. The Bankruptcy Court did not endorse the Trustee’s claim that any tax allocation agreement of any kind overcomes the Bob Richards presumption, whether the agreement addresses refunds or not. But the Bankruptcy Court also rejected the Trustee’s argument that the agreement in question must clearly allocate the refund away from the party incurring the losses. Rather, the Bankruptcy Court strictly interpreted the language from Bob Richards where the Ninth Circuit announced that “the parties made no agreement concerning the ultimate disposition of the tax refund.” Bob Richards, 473 F.2d at 265. In this case, said the Bankruptcy Court, “the TAA is an agreement ‘concerning ultimate disposition of the tax refund’—the exact type of agreement that was absent in Bob Richards. Since such an agreement is present, the Bob Richards default rule is facially inapplicable.” UWBI, 558 B.R. at 433. The Bankruptcy Court therefore stood by its analysis of what this Court will call the “IndyMac factors” (discussed further below), which led the Bankruptcy Court to conclude—as an IndyMac analysis essentially always does—that the agreement in question created nothing more than a debtor-creditor relationship, thus leaving both legal and equitable title to the Refund in the Holding Company’s hands. Id. at 424-28.
This Court agrees with the Bankruptcy Court that, under any reasonable definition of “concerning,” the TAA is an agreement concerning ultimate disposition of tax refunds. Ironically, however, interpreting the Bob Richards rule in this manner will usually put the subsidiary in a worse position than if no tax allocation agreement ever existed. That is aptly illustrated here. When the Holding Company receives a refund from the IRS, the TAA imposes upon the Holding Company a self-executing duty to “re-determine[ ]” the Affiliates’ *887tax liability “as if [the refund] have been made as part of the original computation,” and then the Holding Company must distribute that refund to the Affiliates within ten business days. (TAA § H.l.) So, yes, the TAA “concerns” the “ultimate disposition” of the refund, and in fact declares that the refund belongs to the Affiliate that incurred the relevant losses. But by so declaring, it turns out that the Affiliates have actually weakened their claim to any refund. They have now given a Bankruptcy Court—one of the only venues in which the TAA might be litigated—an opening to disregard Bob Richards, apply the Indy-Mac factors, and hold that the refund does not belong to the Affiliates, at least not any more than any other creditor can claim that money in debtor’s possession belongs to it.
Concerned as it was for the problem of unjust enrichment, it is almost inconceivable that the Ninth Circuit meant the Bob Richards rule to apply this way. See 473 F.2d at 262 (“Allowing the parent to keep any refunds arising solely from a subsidiary’s losses simply because the parent and subsidiary chose a procedural device to facilitate their income tax reporting unjustly enriches the parent.”). This Court is convinced that if the scenario at issue here had been presented to the Ninth Circuit in the Bob Richards appeal, that court would have phrased its ruling in a manner consistent with the FDIC’s position here, i.e., that the tax allocation agreement must contradict the default rule. Indeed, the Ninth Circuit’s language already suggests that it had this very idea in mind. Immediately after the “agreement concerning the ultimate disposition of the tax refund” sentence on which the Bankruptcy Court relied, the Ninth Circuit said the following: “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. (emphasis added). “Differing” in this context can only sensibly refer to the general rule announced later in the sentence.
If Bob Richards created a federal common law rule, then this Court may conceivably invoke its own federal common law authority to clarify the meaning of Bob Richards within the District of Colorado, or even to refine the rule if needed. As noted above, however, there has been no analysis either from the Ninth Circuit or the Tenth Circuit regarding the need for, or federal interests served by, Bob Richards as a rule of federal common law. Cf. Resolution Trust Corp. v. Heiserman, 856 F.Supp. 578, 581 (D. Colo. 1994) (courts considering creation of federal common law must evaluate “whether the federal program, which by its nature is and must be uniform throughout the nation, necessitates formulation of controlling federal rules,” “whether application of state law would frustrate specific objectives of the federal program,” and whether “application of a federal rule would disrupt commercial relationships predicated on state law”). Therefore it would be difficult for this Court to declare with confidence that any clarification or extension of Bob Richards fits within those uniquely federal purposes.
As it turns out, the Court thankfully need not engage in any such inquiry. As explained below, even if the Bankruptcy Court applied Bob Richards correctly, both in letter and in spirit, the ensuing analysis of the property interest created by the TAA ultimately favors the FDIC. The Court therefore turns to that analysis.
D. The TAA as Construed Under Colorado Law
Given its interpretation of Bob Richards, the Bankruptcy Court announced that “the *888unambiguous terms of the TAA as construed under Colorado law govern the rights and obligations of [the Holding Company] and the Bank and also dictate the ultimate entitlement to the Tax Refund on a beneficial basis.” UWBI, 558 B.R. at 424 (footnote omitted). Whether those terms are unambiguous is this Court’s ultimate disagreement with the Bankruptcy Court, but this Court, like the Bankruptcy Court, must first work through the terms of the TAA to reach a conclusion about ambiguity.
1. IndyMac
The Bankruptcy Court approached this analysis through what this Court has dubbed the IndyMac factors, given their origin as an analytical test in In re Indy-Mac Bancorp, Inc., 2012 WL 1037481 (Bankr. C.D. Cal. Mar. 29, 2012).9 Just like the present dispute, IndyMac involved a subsidiary bank in FDIC receivership demanding tax refund proceeds possessed by its bankrupt holding company. Id. at *1-2. The court explained its analytical approach to the problem as follows:
First, the Court has looked to a set of cases involving the very issue presented here: a dispute about the ownership of tax refunds in bankruptcy when a pre-bankruptcy tax sharing agreement existed. Second, the Court has looked to a separate set of cases involving the interpretation of parties’ legal relationships— both inside and outside of bankruptcy— under California law.,..
* * *
The Court’s analysis of the applicable case law indicates that three key factors are examined when considering whether a particular document or transaction establishes a debtor-creditor relationship, on the one hand, or a different sort of relationship (such as a trust, mere agency, or bailment relationship), on the other hand.
Id. at *13 (footnotes omitted). The court then proceeded to discuss these three factors.
The first factor was whether the tax sharing agreement (as it was called in that case) “create[d] fungible payment obligations”—or in other words, whether the agreement created a right to the refund itself or to “amounts [paid by the holding company] equal to what the subsidiary would have received if it hypothetically were a standalone tax filer.” Id. at *13-15. The IndyMac court derived this rule from other bankruptcy court decisions (some of which have since been overruled) but also stated that it “fully accords with the Ninth Circuit’s application of California law in the bankruptcy context.” Id. at *13. With respect to the tax sharing agreement under consideration, the court found that it frequently contained the words “reimbursement” and “payment,” which “are indicative of a debtor-creditor relationship and, in comparison, are completely inconsistent with the existence of a trust or agency relationship.” Id. at *14.
The second factor was whether the tax sharing agreement contained “provisions requiring the parent to segregate or escrow any tax refunds” or “restrictions on the parent’s use of the funds while in the parent’s possession.” Id. at *15. It is not clear why IndyMac treated this as just one factor to consider, given that the case law the court cited in support was essentially unequivocal regarding the result:
*889The key principle emerging from these cases was summarized in In re Black & Geddes, Inc.: “It is a firmly established principle that if a recipient of funds is not prohibited from using them as his own and commingling them with his own monies, a debtor-creditor, not a trust, relationship exists.” 35 B.R. 830, 836 (Bankr. S.D.N.Y. 1984), These precise words have been quoted and applied by the Ninth Circuit Court of Appeals and by California’s state appellate courts. [Citing cases.]
Id. In any event, the court found that no such restrictions existed in the tax sharing agreement. Id. at *16.
The third and final factor was whether the agreement gave the parent company “sole discretion to prepare and file consolidated tax returns and to elect whether or not to receive a refund.” Id. In IndyMac, it did. Id. Thus, all three factors favored finding that the tax sharing agreement created a debtor-creditor relationship, and so the tax refunds in question remained property of the holding company’s bankruptcy estate. Id. at *17-20.
2. The Bankruptcy Court’s Application of IndyMac
In this case, the Bankruptcy Court announced at the outset that it would follow the three IndyMac factors. UWBI, 558 B.R. at 424-25.
As to the first factor (fungibility), the Bankruptcy court found that “the TAA is peppered throughout with terminology evidencing a debtor-creditor relationship including: ‘allocating,’ ‘reimbursing,’ ‘compensating,’ ‘pay,’ ‘refund,’ ‘liability,’ ‘reimburse,’ ‘liable,’ ‘payments,’ ‘refunded,’ and ‘liability,’” Id. at 425. As to the portion of the TAA specifically addressing tax refund payments received from the IRS, the Bankruptcy Court emphasized that it required “‘payments between the appropriate parties’ ” within ten business days after “ ‘re-determination’ ” of the parties’ tax liability, and therefore nothing “suggested] that the Bank had a direct interest in any IRS tax refunds.” Id. at 426 (quoting TAA § H.l).
As to the second factor (escrow, segregation of funds, etc.), the Bankruptcy Court stated that “[o]ne could search the TAA in vain for days trying to locate any express, or even implied, requirement for [the Holding Company] to escrow or segregate any funds that it might receive as a tax refund from the IRS.” Id. at 427.
As to the third factor (delegation of decision-making authority to the parent), the Bankruptcy Court emphasized TAA § G.l, by which the Affiliates appointed the Holding Company as their “agent” empowered to “elect to file and mak[e] any election, application or tak[e] any action in connection therewith on behalf of the Affiliates.” Id.
“In sum,” said the Bankruptcy Court, “under the terms of the TAA, [the Holding Company] is the beneficial owner of the Tax Refunds.” Id. at 427-28.
3. Whether IndyMac Provides Appropriate Guidance
Many bankruptcy courts since IndyMac have adopted its analysis. See, e.g., In re Downey Fin. Corp., 499 B.R. 439, 455 (Bankr. D. Del. 2013), aff'd, 593 Fed.Appx. 123 (3d Cir. 2015); In re Imperial Capital Bancorp, Inc., 492 B.R. 25, 29-30 (S.D. Cal. 2013); FDIC v. AmFin Fin. Corp., 490 B.R. 548, 554 (N.D. Ohio 2013), rev’d and remanded, 757 F.3d 530 (6th Cir. 2014). Its appeal is obvious: it is easy to apply to an otherwise complicated situation—a highly salutary feature in our perpetually congested court system. But the Court is troubled with two aspects of In-dyMac.
*890First, the idea that there are three factors to weigh is illusory. As already noted, if lack of escrow or segregation provisions automatically dictates a debtor-creditor relationship, then the Court sees little sense in treating the second factor as just one more data point to consider. Moreover, whether the parent company has complete control over the tax filing (the third factor) will likely favor the parent in every case because it is essentially mandated by an IRS regulation concerning consolidated tax filings. See 26 C.F.R. § 1.1502-77(a)(1) (“one entity (the agent) is the sole agent that is authorized to act in its own name regarding all matters relating to the federal income tax liability for the consolidated return year for each member of the group”). A provision affirming as much may not appear in every tax allocation agreement, but surely any attorney representing a parent company would, in such a situation, fill that gap by citation to this regulation.
Second, IndyMac is not a one-size-fits-all test. The IndyMac bankruptcy court itself recognized that its task was to determine under state law “whether a particular document or transaction establishes a debtor-creditor relationship, on the one hand, or a different sort of relationship (such as a trust, mere agency, or bailment relationship), on the other hand.” Id. at *13. And it attempted, in some instances more successfully than others, to ground its three factors in the law of the applicable state (California). But subsequent courts, including the Bankruptcy Court here, have applied IndyMac as if its three factors represent a federal common law of tax allocation agreements—or in other words, they apply IndyMac without reference to whether the applicable state’s law supports the factor in question. To be sure, it seems unlikely that California law on these issues differs significantly from any other state’s laws. Nonetheless, the fact that subsequent bankruptcy courts have not tied the IndyMac factors into their own states’ laws raises the possibility, with some justification, that IndyMac is being adopted for its own sake, not because it actually satisfies a court’s obligation to determine the property status of the refund under state law.
Despite all this, IndyMac certainly starts by asking the right question: does the agreement in question “establish[] a debtor-creditor relationship, on the one hand, or a different sort of relationship (such as a trust, mere agency, or bailment relationship), on the other hand.” Id. This Court would phrase it in somewhat the reverse fashion: Does the agreement create a trust relationship, agency relationship, or some other relationship under state law that conveys only legal title, not equitable title, to property? If not, it is a typical commercial (debtor-creditor) contract, meaning the party that receives property under the contract holds both legal and equitable title to that property. The Court accordingly turns to these questions.
4. The Importance of TAA § A.2
“The primary goal of [contract] interpretation is to determine and give effect to the intention of the parties.” Cache Nat’l Bank v. Lusher, 882 P.2d 952, 957 (Colo. 1994). Whether the parties intended the TAA to create a trust relationship, agency relationship, or anything like either of these cannot be analyzed without keeping in mind TAA § A.2: “In essence, this Agreement [i.e., the TAA] requires that each [Affiliate] be treated as a separate taxpayer with [the Holding Company] merely being an intermediary between an Affiliate and the [IRS].” Nothing in the TAA gives the key words here—“merely being an intermediary”—any specialized meaning. The Court therefore must read *891them according to their plain and ordinary meaning. Cache Nat’l Bank, 882 P.2d at 957.
“Intermediary” is generally defined as “a mediator or go-between; a third-party negotiator.” Black’s Law Dictionary, s.v. “intermediary” (10th ed. 2014); see also Merriam-Webster Online, s.v. “intermediary” (offering similar definitions), at https://www.merriam-webster.com/ dictionary/intermediary (last accessed July 3, 2017). “Merely” is the adverbial form of “mere,” which is generally defined as “being nothing more than.” Id., s.v. “merely, at https://www.merriam-webster.com/ dictionary/merely (last accessed July 3, 2017). Thus, TAA § A.2 declares that the purpose of the TAA is to set up an arrangement in which the Holding Company acts as nothing more than a go-between, as between the subsidiaries and the IRS. This purpose must be kept in mind in the ensuing analysis.
5. Trust
The Bankruptcy Court noted that “what the FDIC seems to be attempting is to suggest an argument sounding primarily in trust rather than agency.” UWBI, 558 B.R. at 431. But the Bankruptcy Court was “quite confused as to the FDIC’s position regarding possible trust issues.” UWBI, 558 B.R. at 431-32. The confusion is understandable. The FDIC’s summary judgment briefing below focused on establishing that an express trust is unnecessary if the Holding Company can be deemed a “conduit,” which the FDIC equated to the “intermediary” language already in the TAA. (See, e.g., App. 86, 254-55.) In other words, the FDIC appeared to be seeking recognition of some sort of generic trust-like relationship, without really identifying what that relationship was. This did not comport with the three types -of trusts that the Bankruptcy Court found to exist under Colorado law (express trusts, constructive trusts, and resulting trusts), and the Bankruptcy Court therefore rejected the possibility of a trust relationship. UWBI, 558 B.R. at 432.
In this appeal, the FDIC has not specifically argued that the Bankruptcy Court’s analysis was error. Rather, in a footnote, the FDIC argues that the Bankruptcy Court’s “discussion of trust law ... was another diversion.” (ECF No. 10 at 35 n.13.) This footnote goes on to claim that “[t]he circumstances in which Colorado law will impose a resulting trust are far broader than the bankruptcy court suggested” (id.), but it does not argue that a resulting trust would cover the situation at hand, or that the Bankruptcy Court’s analysis should be overturned. Accordingly, the FDIC has forfeited any argument sounding in trust law. See, e.g., United States v. Hunter, 739 F.3d 492, 495 (10th Cir. 2013) (arguments inadequately developed in the opening brief are forfeited).10
6. Agency
Unlike its trust argument, the FDIC certainly continues to press an agency argument. (See ECF No. 10 at 29-36.) The FDIC’s main argument in favor of an agency relationship is that TAA § G.l explicitly declares one:
*892[e]ach Affiliate hereby appoints [the Holding Company] as its agent ... for the purpose of filing such consolidated Federal Income tax returns for the [Group] as [the Holding Company] may elect to file and making any election, application or taking any action in connection therewith on behalf of the Affiliates.
Building on this, the FDIC notes that “[a]n agent is duty-bound to protect and turn over any property the agent receives for its principal.” (ECF No. 10 at 30.) See also Restatement (Third) of Agency § 8.12 cmt. b (2006) (“If the agent receives property for the principal, the agent’s duty is to use due care to safeguard it pending delivery to the principal.”); cf. Moore & Co. v. T-A-L-L, Inc., 792 P.2d 794, 798 (Colo. 1990) (declaring a real estate broker to be an “agent,” and as such, owing to the seller a duty to “account ... for all money and property received”). The Bankruptcy Court had two responses to this argument.
a. “Limited and Procedural” Agency
The Bankruptcy Court first concluded that this grant of agency “is limited and procedural only,” meaning that it only extends to precisely what it says—and it says nothing about the Holding Company “being an agent for holding any tax refunds for the Bank’s benefit.” UWBI, 558 B.R. at 430-31. The Bankruptcy Court’s notion of a “limited and procedural agency” comes from other bankruptcy cases interpreting 26 C.F.R. § 1.1502-77(a)(1), the IRS regulation discussed above which declares that one entity among a consoli-’ dated filing group (usually the parent) must be the agent for the whole vis-á-vis the IRS: “one entity (the agent) is the sole agent that is authorized to act in its own name regarding all matters relating to the federal income tax liability for the consolidated return year for each member of the group.” Other bankruptcy courts have faced an argument that the IRS regulation itself (not some provision of the relevant tax allocation agreement) creates an agency relationship that entitles the subsidiary to any refund. These courts have, not surprisingly, rejected this argument, considering that whatever “agency” this regulation establishes is for, the convenience of the IRS and has nothing necessarily to do with a consolidated filing group’s internal affairs. Thus, courts have deemed such agency to be “procedural.” See, e.g., Bob Richards, 473 F.2d at 265 (“[T]he refund is made payable to the parent and the acceptance of the refund by the parent discharges any liability of the government to any subsidiary. But these regulations are basically procedural in purpose and were adopted solely for the convenience and protection of the federal government.”); In re First Cent. Fin. Corp., 269 B.R. 481, 489 (Bankr. E.D.N.Y. 2001) (“this agency is purely procedural in nature, and does not affect the entitlement as among the members of the Group to any refund paid by the I.R.S.”); see also Jump v. Manchester Life & Cas. Mgmt. Corp., 579 F.2d 449, 452 (8th Cir. 1978) (“Though IRS regulations provide that the parent corporation is the agent for each subsidiary in the affiliated group, this agency relationship is for the convenience and protection of IRS only and does not extend further.” (citation omitted)).
Again, these cases involved an argument in which the regulation itself was deemed relevant to determining whether an agency relationship had been created. Here, the FDIC argues from the explicit language of TAA § G.l. The Bankruptcy Court’s rejection of the FDIC’s argument hinged on the belief that TAA § G.l was intended as nothing more than a contractual recognition of the relevant IRS regulations. See UWBI, 558 B.R. at 418 (describing § G.l as “mimicking the requirements of [the] *893IRS regulation”); id, at 430 (referring to “the agency referenced in the TAA (and included in the IRS regulations)”); id. at 431 (“While the TAA did identify [the Holding Company] as an agent for the Affiliated Group for purposes of filing consolidated federal income tax returns, it did so using language very similar to the IRS regulations.”).
The Bankruptcy Court’s view is reasonable, but the Bankruptcy Court cited no evidence to support such an intent. The language of TAA § G.l (“agent ... for the purpose of filing such consolidated Federal Income tax returns for the [Group] as [the Holding Company] may elect to file and making any election, application or taking any action in connection therewith on behalf of the Affiliates”) is not the same as the language of the IRS regulation (“agent ... authorized to act in its own name regarding all matters relating to the federal income tax liability for the consolidated return year for each member of the group”). If the Holding Company and the Affiliates truly wished to do nothing more than import the regulatory requirement into the TAA, one might expect that they would adopt the regulation’s language verbatim. They did not. Thus, it is equally reasonable that the parties meant something more than a basic affirmance of IRS regulatory requirements. And if so, “taking any action in connection therewith on behalf of the Affiliates” is broad enough to encompass accepting a refund on behalf of the Affiliates who might be entitled to it (or to a portion of it) if treated as separate taxpayers—as TAA § A.2 commands that they be treated vis-a-vis each other and the Holding Company, which acts “merely” as an “intermediary” with the IRS.
On the other hand, the Bankruptcy Court was correct to observe that the TAA contains none of the usual indicia of a relationship beyond a typical commercial transaction, such as restrictions on comin-gling of funds. UWBI, 558 B.R. at 427. The Sixth and Eleventh Circuits have noted, however, that such an observation may result in a wash when the tax agreement at issue similarly lacks indicia of a normal debtor-creditor relationship, such as an interest rate or collateral. AmFin, 757 F.3d at 535; NetBank, 729 F.3d at 1351. This argument may not go as far as the Sixth and Eleventh Circuits seem to take it, given that typical commercial contracts are the default and may be governed by numerous default rules in the absence of expected terms, whereas trust and agency relationships normally require some evidence of intent to create such a relationship. Nonetheless, those courts’ reasoning is well taken here when compared to the language of the TAA, which lacks any indi-cia of a debtor-creditor relationship and, in fact, affirmatively characterizes the Holding Company as a “mere[ ] ... intermediary.”
Given all this, there are at least two reasonable interpretations of TAA § G.l.
b. Subsidiary as Principal,. Parent as Agent
The Bankruptcy Court’s second reason for rejecting the FDIC’s interpretation of TAA § G.l is a conclusion that a subsidiary can never appoint a parent as its agent:
.,, the agency referenced in the TAA is not consistent with Colorado common law agency. In Colorado, there can be no agency relationship where the alleged agent is not subject to the control of the alleged principal. The FDIC is turning agency on its head because the Bank did not control [the Holding Company] (or at least the FDIC produced no evidence that the Bank controlled [the Holding Company]). Subsidiaries generally do *894not control their parents. So, the agency argument does not work.
UWBI, 558 B.R. at 431 (citations omitted).
As stated in the Court’s preliminary observation above (Part III.A), reasoning such as this proves too much. Although it may be true in a practical sense that a wholly owned subsidiary-principal could never direct a parent-agent contrary to the parent’s wishes, that does not mean that the subsidiary and the parent are not separate legal entities with at least nominally separate directors and management. When corporate formalities are properly observed, courts nearly always respect them, and thus there seems to be no reason that a subsidiary cannot efficaciously designate its parent as its agent. Accordingly, there remain at least two reasonable interpretations of TAA § G.l regarding the scope of the Holding Company’s agency on behalf of the Bank.
E. Ambiguity
In Colorado, “ambiguity of a contract ... is a question of law.” Cheyenne Mountain Sch. Dist. No. 12 v. Thompson, 861 P.2d 711, 715 (Colo. 1993). “In determining whether an ambiguity exists, [the court] must ask whether the disputed provision is reasonably susceptible on its face to more than one interpretation.” Allen v. Pacheco, 71 P.3d 375, 378 (Colo. 2003). The Court has concluded that the TAA could be reasonably interpreted both to create an agency relationship (in which case the Holding Company was required to act toward the Refund as a fiduciary for the Bank) or a standard commercial relationship (in which case the Holding Company has no greater obligation to the Bank than it does to any other creditor).
The TAA, by its express terms, breaks the tie in favor of the Bank: “The intent of this Agreement is to provide an equitable allocation of the tax liability of the Group among [the Holding Company] and the Affiliates. Any ambiguity in the interpretation hereof shall be resolved, with a view to effectuating such intent, in favor of any insured depository institution.” (TAA § H.4.) There can be no question that the “equitable allocation” in this matter is to remit the Refund to the Bank. At oral argument below, the Trustee’s counsel commendably acknowledged, “Absent bankruptcy ... the parent wouldn’t have the right to keep this refund.” (App. 348.) Because the TAA can reasonably be interpreted to require the Holding Company to act as agent on behalf of the Bank in obtaining and remitting the refund, the TAA requires that this Court so construe it.
The Bankruptcy Court partially sidestepped the ambiguity argument by announcing that the parties both agreed that the TAA was unambiguous. UWBI, 558 B.R. at 424 n.26. The FDIC indeed asserted below that the TAA’s terms unambiguously favored the FDIC’s position, but the FDIC also argued in the alternative based on TAA § H.4. (App. 251, 331-36.) Thus, the FDIC preserved the argument, and continues to urge it here in the alternative. (See ECF No. 10 at 22, 40, 41.)
The Bankruptcy Court also determined, of its own accord, that the TAA was unambiguous, but it did so because the rule of IndyMac is that an agreement with supposedly fungible obligations, a lack of co-mingling restrictions, and a full delegation of tax-related authority to the parent is, by definition, “unambiguously” a typical commercial contract, not a contract creating any heightened relationship. The Court need say nothing further about the validity of that rule as a general matter. It appears that no other case applying the IndyMac approach has faced a tax allocation agreement with language such as that contained in TAA § A.2: “In essence, this Agreement requires that each [Affiliate] be treated as a separate taxpayer with [the *895Holding Company] merely being an intermediary between an Affiliate and the [IRS].” This language must be weighed against any inferences drawn out of the IndyMac analysis, and it at least creates an ambiguity—thus triggering TAA § H.4.
Accordingly, the Court concludes that the Holding Company, construed as an agent under TAA §§ A.2, G.l, and H.4, held no more than legal title to the Refund, while the Bank held equitable title. The Refund is not part of the Holding Company’s bankruptcy estate. See 11 U.S.C. § 541(d).
IV. CONCLUSION
For the reasons set forth above, the Court ORDERS as follows:
1. The judgment of the Bankruptcy Court is REVERSED and this matter is REMANDED to the Bankruptcy Court for further proceedings consistent with this opinion;
2. The FDIC’s Request for Oral Argument (ECF No. 13) is DENIED AS MOOT; and
3. This appeal (ECF No. 6) is TERMINATED.
. The FDIC moved for oral argument. (ECF No. 13.) This Court reviewed the transcript of the parties' oral argument before the Bankruptcy Court and found that it answered all of the questions this Court would ask. The Court therefore denies the request for oral argument.
. Oddly, this appeal contains both a Record on Appeal (EOF No. 6) as well as an Appendix (EOF Nos. 10-2 through 10-5). The parties’ briefs generally cite to the Appendix, so the Court will as well, using the abbreviation "App.”
. The TAA is in the record at App. 41-46. Whenever possible, the Court will cite to the TAA by its internal section numbers, e.g., "TAA § H.4,” but a few important matters bear no section numbers. In that case, the Court will cite directly to the "App.” page number.
. The drafters of the TAA inserted this provision, of all places, in the final section of the document under the heading ''Miscellaneous,” (App, 44.)
. The record does not reveal what happened in 2009.
. Considering the many cases cited by the parties where refund allocation has been litigated in bankruptcy or receivership proceed-tags, it is rather astonishing that the tax bar has not yet agreed upon some sort of standard clause to address these precise circumstances.
. The appellant based its argument on the Sixth Circuit's AmFin decision, which had been announced after the appellant’s opening brief was filed. But the Eleventh Circuit’s NetBank decision, which predated the appellant’s opening brief, also characterized the Bob Richards rule as federal common law; and in any event, the federal common law issue should be obvious to any lawyer reasonably experienced in federal choice-of-law questions. Thus, there was a basis to reject the argument as untimely. But the Tenth Circuit never announced as much.
. All ECF page citations are to the page number in the ECF header, which does not always match the document's internal pagination, especially where the document has prefatory material such as a table of contents or table of authorities.
. This opinion was a report and recommendation, which was subsequently adopted by the Central District of California in an unpublished disposition, see 2012 WL 1951474 (C.D. Cal. May 30, 2012), which was in turn affirmed by the Ninth Circuit also in an unpublished disposition, see 554 Fed.Appx. 668 (9th Cir. 2014).
. To be clear, the Court is not convinced that Colorado law requires a party to fit any trust-like relationship into the labels of "express,” “constructive,” or "resulting.” The Colorado Supreme Court has held, for example, that a public utility company holds overcharges recovered from wholesalers "in trust” pending refund to the ratepayers, without fitting the relationship into a predefined category of trust. See Colo. Office of Consumer Counsel v. Pub. Serv. Co. of Colo., 877 P.2d 867, 873 (Colo. 1994). The FDIC actually cited this case below (see App. 255), but has not continued to press the argument here.