Opinion
No. 09 C 136.
2013-01-4
Frederick J. GREDE, not individually but as Liquidation Trustee of the Sentinel Liquidation Trust, Plaintiff, v. FCSTONE, LLC, Defendant.
Chris C. Gair, Angela M. Allen, Christine L. Childers, Jeffrey Scott Eberhard, Michael H. Margolis, Vincent E. Lazar, Jenner & Block LLP, Chicago, IL, for Plaintiff. Stephen Patrick Bedell, Geoffrey S. Goodman, Jill L. Nicholson, Joanne Lee, Robert S. Bressler, Thomas Kelly Anderson, Thomas Paul Krebs, William J. McKenna, Jr., Foley & Lardner, Chicago, IL, for Defendant.
Chris C. Gair, Angela M. Allen, Christine L. Childers, Jeffrey Scott Eberhard, Michael H. Margolis, Vincent E. Lazar, Jenner & Block LLP, Chicago, IL, for Plaintiff. Stephen Patrick Bedell, Geoffrey S. Goodman, Jill L. Nicholson, Joanne Lee, Robert S. Bressler, Thomas Kelly Anderson, Thomas Paul Krebs, William J. McKenna, Jr., Foley & Lardner, Chicago, IL, for Defendant.
MEMORANDUM OPINION AND ORDER
JAMES B. ZAGEL, District Judge.
I. INTRODUCTION
Sentinel Management Group, Inc. (“Sentinel”) filed under Chapter 11 of the Bankruptcy Code in August 2007. In September 2008, Plaintiff Liquidation Trustee filed adversary proceedings in the Bankruptcy Court for the Northern District of Illinois for avoidance and recovery of pre and post-petition transfers made by Sentinel to or for the benefit of certain customers. On October 29, 2008, I withdrew the reference to the Bankruptcy Court, finding that the adversary proceedings raised “significant open and unresolved issues” of non-bankruptcy law regarding the applicability of common law trust principles to statutory trusts, and the duty of futures commission merchants (“FCMs”) to cover customer segregation shortfalls under the Commodity Exchange Act (“CEA”) and its regulatory provisions. Grede v. Fortis Clearing Americas LLC, No. 09 C 138, 2009 WL 3518159, at *3–4 (N.D.Ill. Oct. 28, 2009).
The instant adversary proceeding was chosen as a “test case” (as least in part) to resolve common legal issues among the Trustee's actions. Here, the Trustee seeks to avoid or reduce the transfer of approximately $15.6 million to Defendant FCStone. He alleges five counts: 1) avoidance and recovery of post-petition transfers (11 U.S.C. §§ 549(a) and 550); 2) avoidance and recovery of preferential transfers (11 U.S.C. §§ 547(b) and 550); 3) declaratory judgment that cash and securities held by Sentinel in allegedly segregated bank accounts is property of the Debtor's estate; 4) unjust enrichment; and 5) disallowance or reduction of claims (11 U.S.C. § 502(d)).
A bench trial was held on October 1 through 17, 2012. Pursuant to Fed.R.Civ.P. 52(a), my findings of fact and conclusions of law are laid out below.
Both parties moved for summary judgment on Counts I–IV before trial. I considered the motions but ultimately decided to take them with the case. This opinion addresses all issues raised at summary judgment and trial.
II. FINDINGS OF FACT
To the extent that any findings of fact may be deemed conclusions of law, and vice versa, they should be considered as such; the labels used herein are not controlling. See9 Wright & Miller, Federal Practice & Procedure, § 2579 (3d ed. 2008).
Parties
1. The Sentinel Liquidation Trust (the “Trust”) is a liquidating trust created under the Fourth Amended Chapter 11 Plan of Liquidation (the “Plan”) for Sentinel. The effective date of the Plan was December 17, 2008. Plaintiff Frederick J. Grede was formerly the Chapter 11 trustee for Sentinel. On December 17, 2008, pursuant to the terms of the Plan, Grede was appointed Liquidation Trustee of the Trust (the “Trustee”).
2. Defendant FCStone is an Iowa limited liability company with its principal place of business in Chicago, Illinois. FCStone is a futures commission merchant (“FCM”). As an FCM, FCStone maintains accounts and clears trades for customers in the futures markets; FCStone acts as a financial intermediary between its customers and the futures markets.
Sentinel's Business
3. Sentinel was an Illinois corporation headquartered in Northbrook, Illinois. Sentinel managed investments for various clients, including FCMs, hedge funds, financial institutions, pension funds, and individuals.
4. Sentinel offered its customers several different portfolios as investment options. Sentinel represented to its customers that all of its portfolios met the dual objectives of low risk and high liquidity. Sentinel's marketing materials described the allowable investments in the three primary portfolios as follows:
• Treasury Only Portfolio—Direct obligations of the U.S. Treasury.
• 1.25 Portfolio—Obligations of the U.S. Treasury, short term commercial paper rated A1/P1, medium and long term debt rated AA or higher, bank time deposits and repurchase agreements collateralized by the above.
• Prime Portfolio—Short term commercial paper rated A1/P1, investment grade corporate bonds, bank time deposits, repurchase agreements collateralized by the above and other highly rated marketable securities.
5. Sentinel classified its customers into three distinct segments or “SEGs” based on their regulatory status and the source and nature of their investments. The SEGs were comprised as follows:
SEG 1: Comprised of FCMs' customer funds required to be invested in compliance with CFTC Rule 1.25 and held in compliance with CEA and CFTC segregation requirements;
SEG 2: Comprised of FCMs' foreign futures and foreign options customer funds required to be invested in compliance with CFTC Rule 1.25 and held in separate accounts in compliance with CFTC Rule 30.7;
SEG 3: Comprised of hedge funds, other public and private trading funds, individual investors and FCMs investing proprietary or “House” funds.
6. Within each SEG, Sentinel further divided its customers into 11 groups, each of which consisted of customers with the same risk and return goals. Each customer participating within a specific group held an indirect beneficial ownership interest based on its pro rata share of the value of the securities held in that group's portfolio. The breakdown of the 11 customer groups by SEG, and their investment guidelines, were as follows:
SEG 1: FCM customers trading on U.S. exchanges
Group 1: Rule 1.25—Overnight reverse-repo government securities only
Group 7: Rule 1.25—Government securities, corporate bonds, cash
Group 8: Rule 1.25—Direct obligations of the U.S. Treasury only
Group 9: Rule 1.25—Government securities (no agency), corporate bonds, cash
SEG 2: FCM customers trading foreign futures and options
Group 5: Rule 30.7—Cash only
Group 6: Rule 30.7—Government securities and cash
SEG 3: Hedge funds, trusts, individual investors, FCM proprietary or “House” funds
Group 2: Prime—Government, corporate, sovereign debt rated as “investment grade” by an NRSRO.
Group 3: TOP—Direct obligations of the U.S. Treasury only
Group 4: Prime
Group 10: Rule 1.25—Government securities, corporate bonds, cash
Group 11: Prime—Government, corporate, sovereign debt rated as “investment grade” by an NRSRO.
7. Defendant FCStone was a Sentinel customer. Defendant's funds were invested in the SEG 1, Group 7 customer portfolio.
This group consisted of a single customer, Lakeshore Asset Management Ltd. In July 2007, this Court, in an unrelated proceeding, ordered Sentinel to invest all Lakeshore funds under its management in Rule 1.25 compliant investments only.
8. Sentinel also managed a “House” or “Street” portfolio comprised of securities that were managed on a proprietary basis on behalf of Sentinel and certain employees, insiders and investors.
9. Prior to 2004, Sentinel entered into Investment Advisory Agreements, and post 2004 entered into Investment Management Agreements (collectively, “Customer Agreements”), with each of its investing customers. The Customer Agreements governed the terms of Sentinel's relationship with its customers. The Customer Agreements appointed Sentinel as a discretionary investment adviser with respect to assets deposited by customers. The Customer Agreements specified that the client's assets in a particular program would be invested along with the assets of other Sentinel clients in the same program and that the client would own an indirect interest in the segregated portfolio of the relevant program.
10. Sentinel's customers did not own any particular securities and were entitled only to redemptions of cash. All of Sentinel's transactions with customers were cash transactions.
The Regulatory Framework that Governed Sentinel's Business
11. Sentinel was registered with the Securities and Exchange Commission (“SEC”) as an investment adviser and with the Commodity Futures Trading Commission (“CFTC”) as a futures commission merchant. FCM registration was necessary for Sentinel to provide its investment advisory services to FCMs investing funds of their commodity customers. Sentinel did not itself execute or clear futures transactions, as registered FCMS typically do.
12. The Commodity Exchange Act (CEA) and CFTC Rules promulgated thereunder required Sentinel to segregate commodity customers' funds from those of other customer groups and from Sentinel's own assets. The CEA and its related CFTC rule applied to Sentinel's SEG 1 FCM customers with respect to the funds of the FCMs' commodity customers.
13. 17 C.F.R. 275.206 (the “SEC Custody Rule”), a regulatory provision promulgated under the Investment Advisers Act (IAA,) required Sentinel to segregate its customers' assets from those of other customer groups and from Sentinel's own assets. The SEC Custody Rule applied to all of Sentinel's customers.
Sentinel's Account Structure at the Bank of New York and JP Morgan
14. Sentinel maintained several accounts at the Bank of New York (“BONY”) to process daily transactions related to securities trading and customer cash deposit and withdrawal activity. BONY also functioned as the custodian of securities held on behalf of Sentinel's customers.
15. Sentinel maintained several accounts at JP Morgan, which functioned as the custodian for customer cash.
16. Specifically, Sentinel maintained three segregated cash accounts at BONY that were held for Sentinel's customers in SEGs 1, 2 and 3. The three BONY cash accounts were the transactional accounts through which all of Sentinel's customer deposits and withdrawals were received and paid. Customer deposits and withdrawals were wired in and out of these accounts on a daily basis.
17. Sentinel also maintained a House cash account at BONY.
18. Sentinel established three segregated securities accounts at BONY to hold government and governmental securities for customers in SEGS 1, 2 and 3.
19. Sentinel established three segregated securities accounts at BONY to hold Depository Trust Company registered corporate securities (“DTC securities”) for customers in SEGs 1, 2 and 3.
20. The BONY account structure also included the SEN and SLM accounts. The SEN account was a lienable account that served as the central settlement account at BONY for Sentinel's investment and trading activity. All purchases and sales of government securities were processed through the SEN, whether for SEGs 1, 2, 3 or the House account. In addition to securities settlements, Sentinel used the SEN account for cash management. Cash from all SEGS as well as the House was commingled in the SEN account. The SEN account was active only during the business day and did not hold securities or cash overnight.
21. BONY provided an overnight loan to Sentinel. The loan's original purpose was to provide Sentinel with liquidity for customer redemptions and failed trades. Later, Sentinel used the BONY loan to fund its own proprietary repurchase agreements as part of a leveraged trading strategy.
22. The SLM account was Sentinel's lienable overnight loan account at BONY. At the close of each trading day, Sentinel would reset its overnight loan in the BONY system. BONY required an offsetting amount of securities to be held as collateral for Sentinel's loan each night. After the amount of the overnight loan was determined, Sentinel would transfer securities via the SEN account to the SLM account at an amount equal to or greater than the amount of the overnight loan. The following morning, the securities in the SLM account were returned to the SEN account.
23. The BONY account structure also included a lienable, non-segregated clearing account used to settle all DTC securities transactions (the “DTC Securities/Clearing Account” or “FC1”), including those made for the House. The FC1 account was not used for cash transactions. Cash transactions relating to DTC corporate transactions were processed in the SEN account.
24. Sentinel established a single clearing account at BONY for securities registered with Euroclear. Sentinel also established segregated Euroclear accounts at BONY, but never activated them.
25. Sentinel established three non-interest bearing cash accounts at JP Morgan as well as three interest bearing cash accounts that were linked to the corresponding non-interest bearing cash accounts. Two of the non-interest bearing accounts and their interest bearing counterparts were available to be used for both SEG 1 and SEG 3 customer funds. The other non-interest bearing account and its interest bearing counterpart was used to hold SEG 2 funds.
26. The JPMorgan cash accounts were non-transactional. Their sole purpose was to hold cash in segregation. Sentinel allocated interest earned on deposits at JP Morgan to customer accounts on a daily basis. SEG 1 and SEG 3 cash was pooled in at least one of the JP Morgan cash accounts. No cash belonging to Sentinel was held in the JP Morgan cash accounts.
Sentinel's Investment Model
27. In order to invest with Sentinel, customers would wire cash to the applicable segregated cash account at BONY and, in exchange, receive a pro rata beneficial interest in securities held by Sentinel at BONY.
28. Sentinel managed customer group investments in securities on a daily basis by allocating suitable securities held by Sentinel to each group according to the group's investment guidelines and applicable regulatory restrictions.
29. The purpose of the allocation process was to invest customer funds deposited with Sentinel in a pool of securities. Sentinel allocated securities it held to each customer group with the total market value (less a small deduction called a “haircut”) of the securities in the pool equaling the total value of the customers' accounts in that pool.
30. The daily allocation process was based upon: 1) changes in individual customer account balances due to deposit and withdrawal activity and the resulting changes in each group's total balance; 2) changes in market value of securities due to changing market conditions; and 3) changes in securities holdings of Sentinel due to securities trading and settlement activity.
31. Generally, the same or similar securities were allocated to each pool daily. However, Sentinel was free to move securities between customer groups without customer permission so long as those securities met the investment standards of the customer group portfolios. Sentinel was also free to liquidate securities held for customers at any time without customer permission.
32. During the relevant time period, Sentinel held billions of dollars in securities that it did not allocate to customers.
33. Sentinel did not generally buy and sell securities in response to daily customer deposits and redemptions. Rather, it operated under a pooled investment model in which one customer's withdrawal or another's deposit would affect the total balance of the customer group and shift each customer's pro rata interest in the group portfolio.
34. Sentinel also allocated interest income to each customer on a daily basis. The interest income was an approximation based on the interest earned by the entire pool of securities that Sentinel managed, not a calculation based on customers' indirect ownership interest in their group portfolio. This included interest earned on billions of dollars of securities that did not appear on any customer statements and the interest earned on securities listed on the account statements of other customer groups.
35. Sentinel issued daily statements (“Customer Statements”) to its customers that summarized the daily account activity and detailed the net equity and net interest earned by the customer. The Customer Statements included a description of the securities reported to be held within the customer's group portfolio on the date of the statement, the number of units held for the customer, the cost per unit and the current market value of the securities.
36. The securities reflected on a given Customer Statement generally were not purchased with the cash the customer deposited, but instead came from Sentinel's large pool of pre-existing unallocated securities that was financed by the BONY loan and cash received from repo counterparties.
A repo is a form of short-term borrowing for dealers in government securities in which a party purchases a security and then immediately loans it to a dealer in exchange for cash equal to the value of the security, less a “haircut,” with an agreement to repurchase the security at a given date for the amount loaned plus interest. The “haircut” is the difference between a security's market value and the amount of the repo loan. Lending parties in repo transactions require haircuts as a cushion against possible decline in the market value of the loaned security. A “reverse repo” is a transaction in which a party borrows a security from a dealer in exchange for cash collateral. Sentinel also engaged in reverse repo transactions as part of its leveraged trading strategy.
37. Customer redemptions were generally funded by other customer deposits or with proceeds of the BONY loan.
38. Sentinel represented to its customers and to regulators that all of its customer funds were properly held in segregation.
39. In fact, Sentinel treated its own and its customers' assets as a single, undifferentiated pool of cash and securities.
40. All sources of cash, including cash deposited by customers, proceed from the BONY loan, cash received from repo counterparties, proceeds of securities transactions, and interest income received on securities were commingled in the unsegregated SEN Account on a daily basis.
Sentinel's Accounting Systems
41. Sentinel used two main accounting systems to track its customers' indirect beneficial ownership interests in its pool of cash and securities: FoxPro and Excel spreadsheets.
42. The FoxPro system was the primary ledger system used by Sentinel in recording and tracking daily accounting and transaction activity. FoxPro was comprised of two primary ledgers: 1) the Customer Ledger, which tracked customer transactions and balances, and formed the book-entry accounting system for Sentinel's customer accounts; and 2) the Securities Inventory, which recorded all securities held or controlled by Sentinel (whether for customers or the House).
43. The FoxPro system generated daily reports that provided detailed information on customer activity and accounts, including account deposits and withdrawals, beneficial interests in securities, interest received, and management fees and other expenses. Sentinel's daily Customer Statements were also produced from the Customer Ledger in FoxPro.
44. Sentinel also maintained a number of detailed Excel spreadsheets that complemented the FoxPro system. The Excel spreadsheets were, among other things, used to summarize the allocation of interest earned for the day by group; compute the current market value for all securities held in Sentinel's inventory; summarize characteristics of securities allocated to each customer group portfolio; reconcile cash and securities activity at BONY and JP Morgan to the FoxPro Securities Ledger; reconcile the FoxPro Securities Ledger and the FoxPro Customer Ledger; and summarize and aggregate daily customer activity by SEG.
45. Using the FoxPro System and the Excel spreadsheets, it is possible to identify: 1) the custodial location of every Sentinel security held at BONY for all relevant time periods; and 2) the indirect beneficial ownership interest in these securities that Sentinel assigned its customers.
The BONY Loan and Sentinel's Leveraged Trading Strategy
46. Beginning in 2001, and increasingly in 2004 and onwards, Sentinel entered into a number of repo trades with counterparties such as FIMAT USA and Cantor Fitzgerald & Co (“Repo Counterparties”). Sentinel used the overnight BONY loan to cover the haircuts associated with these trades.
47. In order to secure the overnight loan, BONY required Sentinel to place securities into its lienable accounts to serve as collateral.
48. Sentinel routinely pledged hundreds of millions in securities that were reflected in the FoxPro ledger as being allocated to SEG 1 and SEG 3 customers, and which were also reported on the daily Customer Statements to SEG 1 and SEG 3 customers as being held in segregation for their benefit, as collateral for the BONY loan.
49. Sentinel's use of securities allocated to customers as collateral for the BONY loan left its segregated accounts chronically underfunded. In other words, the amount of securities held in segregation on behalf of Sentinel's customers was consistently far below what it owed its customers.
50. As Sentinel expanded its leveraged trading strategy, the BONY loan, and the number of securities allocated to customers needed to secure the loan, grew. Sentinel's guidance line for the BONY loan, a preapproved amount for nightly loans, increased from $55 million in May 2004 to $300 million in September 2006. The average loan balance from June 1, 2007 to August 13, 2007, was $369 million.
51. By early 2007, Sentinel held more than $2 billion in securities through repo arrangements.
52. Between December 2004 and June 2007, Sentinel's segregation shortfalls—the difference between what Sentinel owed its customers and the amount of assets actually held in segregation for those customers—grew from roughly $150 million to over $800 million.
53. Prior to June 2007, Sentinel mostly used government securities to secure the BONY loan. However, DTC securities were occasionally used as collateral during this period.
Sentinel's Collapse
54. During the summer of 2007, credit markets contracted significantly due in part to the collapse of the subprime mortgage industry. The unavailability of credit led to a liquidity crisis throughout U.S. financial markets.
55. In response to this credit/liquidity crisis, the Repo Counterparties began to close out positions in Sentinel's repo portfolio, returning securities that Sentinel had loaned out and demanding cash payments in return. The widespread return of repos required Sentinel to repay the Repo Counterparties hundreds of millions of dollars. Sentinel did not have enough of its own funds to meets its repayment obligations to the Repo Counterparties, so it drew even more heavily on the BONY Loan. In late June 2007, the BONY loan topped out at $573.8 million.
56. On June 26, 2007, BONY accepted physical securities (CDOs) as collateral in addition to government securities in order to cover the increased loan. On June 29, 2007, BONY notified Sentinel that it no longer would accept physical securities as collateral.
57. In order to provide adequate collateral that BONY deemed acceptable for the increasing loan, on June 29, 2007, Sentinel moved $166 million of corporate securities from the SEG 1 segregated custodial account to the FC1 Account.
58. In July 2007, Sentinel was able to sell securities to pay down the BONY Loan. Many of the securities Sentinel sold to pay down the loan had been allocated to SEG 1 and SEG 3 customers.
59. Proceeds from the sale of SEG 1 and SEG 3 securities were used to pay down the loan largely because Sentinel was unable to sell the securities that were returned by Repo Counterparties. The securities that had been out on repo tended to be high risk, illiquid CDOs that had been allocated to the House and had not appeared on Customer Statements.
60. As Sentinel's pool of assets decreased, it had to allocate the CDOs that had been out on repo to SEG 3 customers since the higher quality government and corporate securities had been already sold or were allocated to customers with Rule 1.25 compliant portfolios.
61. On July 17, 2007, Sentinel moved $84 million of additional SEG 1 corporate securities to the FC1 account from the SEG 1 segregated custodial account.
62. On July 30, 2007, the BONY Loan was $362.2 million, collateralized by $384.2 million in securities. The collateral chosen by BONY was comprised of government securities with a collateral value of $354.0 million and two DTC securities with a collateral value of $30.2 million. The total par value of DTC securities held in the FC1 account and subject to lien on July 30 was $366.7 million.
63. On July 30 and 31, 2007, Sentinel moved securities with a par value of $263.9 million from the FC1 account to the SEG 1 DTC securities account. Most of the securities returned to the SEG 1 DTC securities account on July 30, 2007, were the same ones that had been moved from the SEG 1 DTC securities account to the FC1 account on June 28 and July 17, 2007. None of these transfers was related to customer activity.
64. To maintain sufficient collateral following the transfer of SEG 1 DTC securities back into segregation, Sentinel transferred securities with a par value of $289.9 million from the SEG 3 DTC securities account to the FC1 account. As a result of this transfer, the SEG 3 segregated security accounts were virtually emptied. The transfer of SEG 3 DTC securities into the FC1 account was not related to customer activity.
65. The swap of SEG 1 securities for SEG 3 securities to meet Sentinel's collateral obligations to BONY resulted in a massive shift of loss exposure in the weeks preceding Sentinel's bankruptcy. Specifically, between July 27 and 31, 2007, Sentinel's SEG 1 customer segregation shortfalls decreased by $543.8 million, while its SEG 3 customer segregation shortfalls increased by $289.7 million. This shift in loss exposure was not based on customer activity, differing legal obligations to customers, or any other legitimate economic or legal grounds.
66. Sentinel's liquidity and solvency problems continued into August as Repo Counterparties continued to unwind their positions. On August 13, 2007, Sentinel sent its customers a letter informing them that it had asked the CFTC for permission to halt redemptions to customers (“No Redemption Letter”). On that same date, the par value of securities and cash controlled by Sentinel was 80.4% of total Customer Equity.
67. Despite the No Redemption Letter, Sentinel paid out full and partial redemptions to its SEG 1 customers.
68. On August 15, 2007, Sentinel distributed $111,229,456.41 to the accounts of SEG 1 customers participating in Group 1 and 8, which constituted the full values of those accounts.
69. On August 16, 2007, BONY sent a letter to Sentinel requesting that it immediately repay its loan in full, and notifying Sentinel of its intention to commence liquidating the collateral that was pledged to secure the loan.
70. On August 17, 2007, BONY sent another letter to Sentinel notifying it that due to its failure to repay the loan, on or after August 22, 2007, BONY would liquidate the collateral pledged to secure the loan.
71. On August 17, 2007, Sentinel filed for bankruptcy.
The Citadel Sale
72. On August 16, 2007, Sentinel sold to Citadel a portfolio of 98 securities that Sentinel had allocated to SEG 1 customers for a purchase price equal to the purported market value of the securities ($384 million), less a “market uncertainty concession” of $47.1 million (the “Citadel Sale”). 92 of the securities settled on August 16, 2007, while six failed to settle. The proceeds of the 92 securities sold were deposited in the SEG 1 segregated cash account at BONY.
73. On August 17, 2007, Sentinel distributed $22.5 million of cash held at JP Morgan to the customer segregated accounts of SEG 1 customers participating in Group 7 and Group 9. One security that failed to settle on August 16, 2007 was settled, and the related $4.9 million in proceeds were deposited in the SEG 1 segregated cash account at BONY.
74. FCStone received $1,097,925 of the August 17, 2007 distribution.
75. The same day, Sentinel filed a voluntary petition for relief under Chapter 11 of the Bankruptcy Code.
76. On August 20, 2007, Sentinel filed an emergency motion with the United States Bankruptcy Court for the Northern District of Illinois in which it sought entry of an order approving the turnover and distribution by BONY of the proceeds from the Citadel sale to the customer segregated accounts of Sentinel's remaining SEG 1 FCM customers. The Bankruptcy Court issued an order stating that BONY “may” distribute the Citadel sale proceeds to Group 7 and Group 9 customers without violating three extant TROs. The order also stated that it was “without prejudice to all rights, defenses, claims and/or causes of action, if any, of the Debtor ... against the Distributee ... with respect to any claim for priority under Section 761–767, or other applicable law.”
77. On August 21, 2007, $297 million in proceeds from the Citadel sale were distributed to the customer segregated accounts of Sentinel's remaining SEG 1 customers in Groups 7 and 9. FCStone received $14,479,039 of the August 21, 2007 distribution. Following the distribution, $20.5 million remained in the SEG 1 account at BONY, which consisted of a $15.6 million holdback and $4.9 million in proceeds received on August 17, 2007, from the security sold to Citadel that failed to settle on the first attempt.
78. FCStone received a total $15,576,964 in proceeds from the Citadel Sale.
79. SEG 3 customers were unable to redeem any of their funds from Sentinel prior to bankruptcy.
80. The August 15 to 21, 2007, SEG 1 customer redemptions, coupled with the August 17, 2007 lien asserted by BONY over the $286 million in the SEG 3 securities pledged as collateral, left SEG 3 customers with a tiny pool of assets (relative to what they were owed) from which their funds could be redeemed post-filing. This resulted in a substantially lower recovery rate for SEG 3 customers than SEG 1 customers.
81. If the August 15–21, 2007 distributions of $430.8 million to SEG 1 customers had been made on a pro rata basis to all customers, the distributions would have represented 32% of total Sentinel customer obligations. Had Defendant received a 32% distribution, the total amount of this distribution would have been $6,977,653.
III. DISCUSSION AND CONCLUSIONS OF LAW
A. Count I and III—Avoidance of Post–Petition Transfer and Declaratory Judgment
The Trustee's first claim seeks to avoid and recover as a postpetition transaction the August 21, 2007 distribution of $14,479,039 from the SEG 1 BONY cash account to Defendant. To avoid this transfer, the Trustee must prove (1) that the debtor made a transfer of property of the estate; (2) that it occurred after the commencement of the bankruptcy case; and (3) that it was not authorized under the Bankruptcy Code or by the bankruptcy court. 11 U.S.C. § 549(a). To recover the transfer once avoided, the Trustee must further prove that the Defendant is (1) the initial transferee of such transfer, or (2) the entity for whose benefit such transfer was made. 11 U.S.C. § 550(a)(1).
Defendant argues that the Trustee cannot avoid and recover the post-petition transfer because (1) the funds were not property of the estate; (2) the Bankruptcy Court authorized the transfer; and (3) Defendant was neither the initial transferee nor the entity for whose benefit the transfer was made. I address each issue in turn.
1. Whether the Proceeds of the Citadel Sale are Property of the Estate
11 U.S.C. § 541(a)(1) provides that “property of the estate” includes “all legal or equitable interests of the debtor in property as of the commencement of the case.” Section 541(d) provides:
“Property in which the debtor holds, as of the commencement of the case, only legal title and not an equitable interest ... becomes property of the estate under subsection (a) of this section 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.”
The Bankruptcy Code itself does not set forth rules for determining whether or to what extent a debtor has an ownership interest in property. This determination is made by looking to “applicable non-bankruptcy law,” which typically means state property law. Butner v. U.S., 440 U.S. 48, 55, 99 S.Ct. 914, 59 L.Ed.2d 136 (1979). However, cases that involve overriding federal interests may require application of federal law. Id. Because federal interests are central to the regulation of the securities and commodity futures markets, federal law should be used to the greatest extent possible in determining ownership interests in this case.
Defendant argues that the CEA and its regulatory provisions constitute the applicable non-bankruptcy law that governs ownership interests in this case. Specifically, Defendant points to Sections 6d(a)(2) and 6d(b) of the CEA and regulatory provisions 17 C.F.R. §§ 1.20–1.29, which prescribe rules for the treatment of commodity customer funds by registered FCMs. Section 6d(a)(2) requires that FCMs “treat and deal with all money, securities, and property received ... to margin, guarantee, or secure the trades or contracts of any customer of such person, or accruing to such customer as the result of such trades or contracts as belonging to such customer.” Section 6d(a)(2) and 17 CFR § 1.20 require that customer funds be kept segregated from the funds of the FCM. Section 6d(b) requires that depositories of commodity customer funds not “hold, dispose of, or use any such [assets] as belonging to the depositing futures commission merchant or any person other than the customers of such futures commission merchant.”
Courts have widely recognized that the CEA and its regulatory provisions create a statutory trust over FCM customer funds. See In re Sawyer, 112 B.R. 386, 390–91 (D.Colo.1990); In re Scheuer, 125 B.R. 584, 590–92 (Bankr.C.D.Cal.1991). Because commodity customer assets in SEG 1 accounts were subject to this trust, Defendant argues, Sentinel held only bare legal title to the Citadel securities and not an equitable interest. Therefore, the proceeds of the securities cannot be considered “property of the estate” under 11 U.S.C. § 541(a)(1).
i) Begier v. IRS and Defendant's Argument that Funds Held in Trust Cannot be Property of the Estate
In support of its argument, Defendant points to Begier v. Internal Revenue Service, 496 U.S. 53, 110 S.Ct. 2258, 110 L.Ed.2d 46 (1990). In Begier, a Chapter 7 trustee brought suit seeking to avoid as preferences certain prepetition tax payments to the IRS. The debtor, American International Airways, Inc. (AIA), a commercial airline company, was subject to several federal income taxes levied against employers and airliners, the total amount of which was “held to be a special fund in trust for the United States,” under 26 U.S.C. § 7501. Taxes subject to § 7501 were referred to as “trust-fund taxes.” Id. at 56, 110 S.Ct. 2258. Prior to filing for bankruptcy, AIA had fallen behind on its trust-fund tax payments to the IRS. As a result, the IRS ordered AIA to deposit all subsequently collected trust-fund taxes into a separate, segregated bank account. AIA established the account but did not deposit adequate funds to cover its trust-fund tax obligations. Id. Nevertheless, AIA remained current on its tax obligations by covering the shortfall in its segregated account with payments of roughly $950,000 from its general operating funds. Id. at 56, 110 S.Ct. 2258. AIA and the IRS agreed that the payments from both the segregated bank account and the general operating funds would be allocated to specific trust-fund tax obligations. Id.
At the outset of its opinion, the Court stated that “[b]ecause the debtor does not own an equitable interest in property he holds in trust for another, that interest is not ‘property of the estate.’ ” Id. at 59, 110 S.Ct. 2258. The question for the Court, then, was whether the money AIA had transferred from its general accounts to the IRS—money which was not traceable to a segregated account established for the purposes of holding money in trust—was property that AIA had, nevertheless, held in trust for the IRS. Id. at 59, 110 S.Ct. 2258.
The Court held that the money AIA transferred from its general accounts was trust property based on a number of specific findings. First, the Court found that it was not necessary to segregate funds to establish a § 7501 trust. Based on the timing of the collection or withholding of funds referenced in the statute, the Court concluded that the § 7501 trust was created “at the moment the relevant payments (from customers to AIA for excise taxes and from AIA to its employees for FICA and income taxes) were made.” Id. at 61–62, 110 S.Ct. 2258. If segregation was needed to create a § 7501 trust, the Court noted, the IRS would be protected “only insofar as dictated by the debtor's whim [in deciding whether or not to segregate funds]”—a result Congress clearly did not intend. Id. at 61, 110 S.Ct. 2258.
Second, the Court found that § 7501 trusts are “radically different” from common law trusts in terms of specific property requirements. Id. 62–63, 110 S.Ct. 2258. Whereas common law trusts require the settlor to designate particular property as the trust res, “§ 7501 creates a trust in an abstract ‘amount’—a dollar figure not tied to any particular assets—rather than in the actual dollars withheld.” Id. 62, 110 S.Ct. 2258. This was a critical finding because it meant that common law tracing rules—which, in the context of bankruptcy, require a trust beneficiary to identify particular trust property to exempt it from the estate, Cunningham v. Brown, 265 U.S. 1, 44 S.Ct. 424, 68 L.Ed. 873 (1924)—were inapplicable.
“Tracing” is a right afforded to trust beneficiaries to follow and identify converted or commingled trust property in order to recover such property. See generallyRestatement Second, Trusts, § 202 (Following trust property into its product); Restatement, Restitution, §§ 202 to 215. It is based on a fundamental principle of English common law that a change in the form of that which is owned does not change who owns it. Bogert's Trust and Trustees, § 921. Generally, a trust is terminated once the trust property ceases to exist. Restatement Third, Trusts, § 2. Thus, when trust property has been converted or commingled, it is necessary for a beneficiary seeking to recover his property to demonstrate that the trust still exist by, identifying the trust res.
Third, examining the legislative history of the Bankruptcy Reform Act of 1978, the Court determined that, while common law tracing requirements did not apply to § 7501 trusts, “Congress expected that the IRS would have to show some connection between the § 7501 trust and the assets sought to be applied to a debtor's trust-fund tax obligations.” Begier, 496 U.S. at 65–66, 110 S.Ct. 2258 (emphasis in original). In determining precisely what connection was necessary, the Court found that Congress specifically intended courts to apply reasonable presumptions—or tracing fictions (discussed in greater detail below)—in identifying § 7501 trust property. Begier, 496 U.S. at 66–67, 110 S.Ct. 2258. Based on the House Report, the Court determined it was reasonable to assume that trust-fund taxes had been properly held for payment if the debtor is able to make the payments. Id. at 66–67, 110 S.Ct. 2258. In other words, the IRS did not need to trace specific tax property in order to exempt it from the estate; it merely had to show the debtor held an amount sufficient to satisfy its § 7501 obligations.
Since Begier involved only pre-petition transfers, the AIA's voluntary payment of its trust-fund tax obligations met this requirement.
Using Begier as its backdrop, Defendant makes the following arguments as to why the Citadel proceeds cannot be considered “property of the estate.” First, the CEA's language and legislative history demonstrate that segregation violations and commingling of funds do not destroy a CEA trust. Thus, regardless of any segregation violations that occurred, Sentinel never held more than bare legal title to the Citadel securities. Second, as a federal statutory trust, CEA trusts are not subject to common law tracing principles, so Defendant need not identify specific property to exempt the transferred funds. Third, the CEA's legislative history makes clear that Congress intended that CEA trust beneficiaries (FCM customers) would not be subject to any tracing requirements. Fourth, if Defendant is required to trace, it should at least be afforded the Begier “nexus” presumption. Fifth, if necessary, Defendant can trace under common law tracing principles. Thus, Defendant argues, anyway you slice it the Citadel proceeds cannot be considered property of the estate.
ii) Whether SEG 3 Customers Have an Equally Valid Trust Claim
The Trustee does not dispute that SEG 1 customer funds were protected by a statutory trust under the CEA. The problem with Defendant's theory, the Trustee argues, is that it ignores the fact that Sentinel's advisory clients in SEG 3 were also protected by a statutory trust under the IAA. Thus, ownership interests cannot be determined by looking to the CEA alone.
The custody rule promulgated by the SEC pursuant to the IAA's antifraud provision, Section 206(4), provides:
it is a fraudulent, deceptive, or manipulative act, practice or course of business ... [for an investment adviser registered or required to be registered under section 203 of the IAA] to have custody of client funds or securities unless: ... [a] qualified custodian maintains those funds and securities: ... [i]n a separate account for each client under that client's name; or ... [i]n accounts that contain only [the investment adviser's] clients' funds and securities, under [the investment adviser's] name as agent or trustee for the clients.
17 C.F.R. § 275.206 (“Rule 275.206”). Courts, including this one, have concluded that this language creates a specific statutory trust that protects customer funds from the investment advisor and its creditors, just as the CEA protects customer funds from the FCM and its creditors. See, e.g., Grede v. Fortis Clearing Americas LLC, No. 09 C 138, 2009 WL 3518159, at *3 (N.D.Ill. Oct. 28, 2009); Griffiths v. Peterson (In re Peterson), 96 B.R. 314, 323 (Bankr.D.Colo.1988) ( [t]he import of [the custody rule,] as well as [a recordkeeping rule,] is that client funds never lose that character merely because an investment adviser ... takes possession of them”); Investment Advisors Act Release No. 123, 27 Fed. Reg. 2149, 2149 (Mar. 6, 1962) (“[IAA custody rule is designed to require] an investment advisor who has custody of funds or securities of any client to maintain them in such a way that they will be insulated from and not be jeopardized by financial reverses, including insolvency, of the investment advisor.”).
Defendant agrees that Rule 275.206 imposes a federal statutory trust over Sentinel's advisory client funds, but argues that the protections afforded by the CEA are stronger than the IAA. Defendant makes three arguments in support of this assertion. First, Defendant argues that the CEA trust is stronger because it stems from a statute, whereas the IAA trust stems from a regulation. Second, Defendant argues that because, in addition to segregation, the CEA requires that all customer property be treated “as belonging to” customers regardless of location, the CEA creates the same type of “floating” trust that existed in Begier. By contrast, the IAA custody rule requires only segregation, which, according to Defendant, demonstrates that Congress did not intend to protect advisory client funds to the same extent as commodity customer funds. Finally, Defendant argues that risks unique to the commodity futures market demand that FCM customer funds be afforded heightened protections over investment adviser funds.
The CFTC, as Amicus Curiae, initially raised the first argument (and hinted at the second) in its Supplemental Memorandum (CFTC Supp. Mem at 9–10, 11 n.8). Defendant, apparently realizing it could no longer ignore the IAA trust as it did in its summary judgment filings, adopted (rather reticently) the arguments at trial.
I am not persuaded by these arguments. There is absolutely no basis in law for elevating one federal statutory trust over another absent the tracing of specific property. Defendant's suggestion that the CEA trust trumps the IAA trust because it stems directly from a statutory provision is simply incorrect. See Chrysler Corp. v. Brown, 441 U.S. 281, 295, 99 S.Ct. 1705, 60 L.Ed.2d 208 (1979). The IAA custody rule is a “properly promulgated, substantive agency regulation” and thus has the same “force and effect” as if it had been passed directly by Congress. Id. at 295, 99 S.Ct. 1705.
I am also unable to discern from the relevant legislative history and text any Congressional intent to provide CEA-protected customer property heightened protection over IAA customer property. As the SEC stated in its supplemental Amicus Curiae memorandum, the IAA custody rule and the commodities law “share a common purpose and a similar methodology.” (SEC Supp. Mem. at 2). Both sets of custody rules were designed specifically to protect client property against 1) misuse by the custodian; and 2) use as payment to the custodian's general creditors in the event of insolvency. Compare IAA Release No. 2176, 68 Fed. Reg. 56690–01, 56692 (Oct. 1, 2003) (“[IAA custody rule] requires advisers that have custody of client securities or funds to implement a set of controls designed to protect those clients assets from being lost, misused, misappropriated or subject to the advisers' financial reverses.”); withS.Rep. No. 90–947 (Jan. 18, 1968), 1968 U.S.C.C.A.N. 1673 (Section 6(b) ... is [designed] to prohibit expressly customers' funds from being used to offset the liabilities of the futures commission merchant or otherwise being misappropriated.”). The demonstrably common purpose behind the CEA and IAA custody rules renders Defendant's claim to uncommon treatment difficult to maintain.
Both the SEC and the CFTC have the gratitude of the Court for their candid and insightful Amicus Curiae briefings.
It is true that, unlike the CEA, the IAA custody rule does not explicitly state that client assets are always to be treated as belonging to the client. But that is hardly probative of Congress's intent to elevate protections for customer funds regulated by the CEA over customer funds regulated by the IAA. Notably, the CEA and IAA rely on the exact same methodology—segregation—for implementing protections to client property. If Congress intended this methodology to function differently between CEA-protected customers and IAA-protected customers (i.e. segregation violations destroy the latter trust but not the former, which is where Defendant's argument ineluctably leads), one would expect to see a clearer indication from Congress regarding the specific interplay between the IAA and the CEA than Section 6d's “belonging to” provision.
Defendant next argues that Congress's intent to elevate the protections afforded by the CEA over the IAA can be found in the legislative history of the commodity broker subchapter of the 1978 amendments to the Bankruptcy Code. I disagree. The 1978 amendments included provisions that clarified protections for both securities and commodities customers; nowhere is a clear intent to elevate the interests of one over the other evidenced. Cf.H.R. Rep. No. 95–595, at 6 (1978), 1978 U.S.C.C.A.N. 5963, 5967 (“Chapter 7 contains two subchapters to handle the unique problems of stockbrokers and commodity brokers.”). The Senate report states that commodity customer claims are “granted the highest priority against the bankrupt's estate,” but this does not appear to add any protections beyond those afforded by the CEA. S. Rep. No. 95–989, at 8 (1978), 1978 U.S.C.C.A.N. 5787, 5794 (“This policy maintains consistency with the Commodity Exchange Act, which establishes customer protection as a primary objective.”). Specifically, this “highest priority” status appears to simply elevate commodity customer claims over those of general creditors, which is not helpful here. S. Rep. 95–989, at 106, 1978 U.S.C.C.A.N. 5787 at 5892, (“[as a result of the 1978 amendments a commodity] customer need not trace any funds in order to avoid treatment as a general creditor.”).
Sentinel's bankruptcy is not a commodity-broker liquidation, and thus the commodity-broker liquidation rules do not apply. The rules are relevant to this case only insofar as they evince Congressional purpose to elevate commodity customer claims over customers protected under the IAA.
The fact is that the CEA and its regulatory provisions and the IAA custody rule create similar segregation obligations, but neither provides any guidance as to what happens where, as here, the property of multiple statutory trusts is commingled and recoverable assets are insufficient to satisfy the claims of all beneficiaries. Again, if Congress intended one federal trust to trump all others—a concept that has no basis in past acts of Congress or the common law of trusts—one would expect to see some very clear statutory language. Congress either did not contemplate the scenario of competing federal trust beneficiaries with equal claims to an inadequate pool of recoverable assets in bankruptcy , or it intended courts to apply common law trust principles if and when such a scenario arose. Either way, there are no grounds in the relevant statutory and regulatory texts or the legislative history to elevate the CEA-protected claimants over those protected under the IAA.
See Corcoran Dep. (150.21–151.4).
This would be unsurprising given Sentinel's unique regulatory status as both an FCM and an investment advisor.
Finally, Defendant raises a number of policy arguments as to why I should treat SEG 1 commodity customer claims as superior to the claims of Sentinel's SEG 3 advisory clients. Many of these policy arguments were presented through the expert testimony of Andrea Corcoran, who, among other qualifications, served for 27 years as a CFTC regulator. Ms. Corcoran's report can be summarized as a list of reasons why the CFTC believes commodity customers ought to have “super priority” in bankruptcy. Corcoran Rep. at 105. Her arguments are based largely on distinguishing traits of the futures market that make it, in her view, subject to greater systemic risk than securities markets in the event of loss of customer funds due to a custodian's insolvency. I regard Ms. Corcoran's testimony as cogent, if perhaps a bit biased by her specific agency service. The problem is that her arguments are better suited for a Congressional committee considering amendments to the federal bankruptcy laws than this Court. My job is to say what the law is, not what it should be, and the policy arguments are not so one-sided as to compel the result Defendant urges. Until Congress demonstrates a clear intention to give commodity customers so-called “super priority” in bankruptcy, I have no basis for elevating the interests of CEA-protected customers over IAA-protected customers.
I note that the Trustee could very well have brought in its own veteran SEC regulator to make arguments as to why such a so-called “super priority” would be damaging to securities markets. That it did not, I think, is more out of recognition that such policy arguments are not relevant to my Section 541 determination than an indication that equally compelling arguments cannot be made from the other side. Whatever arguments may or may not exist they are for Congress to weigh, which perhaps this litigation will prompt it to do.
In this case it might be argued that whatever collision there is between the CFTC rules and the SEC rules should be resolved in favor of applying the SEC rules because Sentinel functioned as an investment advisor and was an FCM in name only. This would be an equally misguided resolution, and the Trustee does not offer it. Sentinel submitted itself to CFTC control. As a licensee it came under the jurisdiction of the CFTC and bound itself to obey the rules and regulations of the Commission even if it never worked as a futures commission merchant. Practically speaking, allowing the SEC rules to trump would not change the outcome of this case—SEG 1 and SEG 3 customers would still be similarly situated, since SEG 1 customers were also protected under the IAA. I raise the point simply to illustrate the competing policy arguments raised as a result of Sentinel's dual registration as an FCM and an investment advisor.
FIRST CONCLUSION OF LAW: The IAA's custody rule creates statutory trust protections as robust as the CEA's. Sentinel's SEG 3 customers therefore have an equally forceful claim to trust protection as SEG 1 customers.
iii) Cunningham and the Law of Similarly Situated Trust Claimants in Bankruptcy
Absent clear statutory guidance on how to deal with competing CEA and IAA trust claimants, I turn to the common law. See Begier, 496 U.S. at 62, 110 S.Ct. 2258. In Cunningham v. Brown, 265 U.S. 1, 44 S.Ct. 424, 68 L.Ed. 873 (1924), the Supreme Court first addressed the scenario of multiple trust claimants competing over an insufficient pool of commingled assets in bankruptcy. Cunningham involved the recovery rights of Charles Ponzi's duped investors following the collapse of his now famous investment scam. Ponzi induced thousands of investors to lend him money in exchange for promissory notes in which he guaranteed payment within 90 days of $150 for every $100 loaned. Ponzi told his investors he was able to earn such high returns through the targeted sale of international postal coupons that capitalized on excessive exchange rate differences following World War I. Id. at 7–9, 44 S.Ct. 424. Ponzi, of course, did not use the funds to purchase postal coupons; in fact, he did not invest the funds at all. He simply used new investors' funds to redeem the promissory notes of prior investors. Ponzi's initial ability to meet his extraordinary promised returns rapidly attracted new investors. But as the business became more “successful,” his pit of insolvency grew deeper. Eventually, Ponzi drew the attention of law enforcement and word got out about the true nature of his business. This led to a run on the bank by holders of unmatured promissory notes. Soon, Ponzi's accounts were emptied and he filed for bankruptcy. Some investors were able to recover their money before Ponzi's accounts were drained, most were not. In the end, millions were lost.
Cunningham concerned six suits brought by the trustees of Ponzi's bankruptcy estate to recover certain pre-petition payments made during the run on the bank as unlawful preferences. The Supreme Court rejected the lower courts' position that the early redemptions constituted rescissions for fraud that left the defendants in a different legal position than those investors who had not redeemed their notes in time and whose funds remained under Ponzi's control at the time he filed for bankruptcy. Once it became clear that Ponzi was insolvent, the Court stated, there were only two ways for the defendants to recover their original loan without running afoul of statutory rules against preference. The first was to trace their specific property, i.e. the actual currency given by the lender to Ponzi in exchange for a promissory note, and thereby assert a constructive trust. The second option was to establish a lien over any fund to which the investor could trace his specific property. Id. at 9–12, 44 S.Ct. 424.
The Court thought it permissible for defendants to trace their property and impose a constructive trust over it because, in doing so, the defendants “would have been endeavoring to get their own money, and not the money in the estate of the bankrupt.” Id. However, the Court found that the defendants could not trace their property because the specific funds they had deposited with Ponzi were withdrawn prior to defendants' application for and payment on their unmatured notes. The defendants, the Court found, had been paid with funds that Ponzi had transferred from outside bank accounts once the run on the bank had begun. Id. Because the redemptions had not been made with the defendants' initially deposited funds, they could not assert a constructive trust over them. Id. The Court stated that “[i]n such a case [where the original trust res has been destroyed], the defrauded lender becomes merely a creditor to the extent of his loss ...” Id.
The Court next considered the appropriateness of applying two common law tracing fictions. A tracing fiction is a presumption about the identity of trust property that courts may apply, in certain contexts, when it is not possible to trace the actual trust res. The first tracing fiction the Court considered was the presumption that Ponzi withdrew against customer deposits in the order in which the deposits were made (first withdrawals charged against first deposits, etc.). Under that presumption, customers would recover in the inverse order in which they made their deposits, with the last depositor given first claim to recoverable assets and so forth. The Court found that the inverse payment rule was inapplicable to the case because Ponzi had already withdrawn defendants' specific property prior to their giving any indication of purpose to rescind. Id. at 12–14, 44 S.Ct. 424.
The so-called “ ‘rule of Clayton's Case,” extracted from Clayton's Case, [1816] Ch. 1 Merivale, 572.
The second tracing fiction the Court considered was what is today called “the lowest intermediate balance test,” or the presumption that trust property is the last to leave a pool of commingled assets. The Court found that this rule was also inapplicable because the remaining pool of funds was composed entirely of customer property: “The rule is useful to work out equity between a wrongdoer and a victim; but when the fund with which the wrongdoer is dealing is wholly made up of the fruits of fraud perpetrated against a myriad of victims, the case is different.” Id. at 12–14, 44 S.Ct. 424.
Because tracing the specific trust res was not possible and the application of tracing fictions was deemed inappropriate, the Court decided to treat the defendants and those who had not redeemed their notes in time as a single class with equal rights to the remaining funds: “It is a case the circumstances of which call strongly for the principle that equality is equity, and this is the spirit of the bankrupt law.” Id. This resulted in a clawback of defendants' redemptions, which were then distributed pro rata amongst all Ponzi's victims.
Cunningham itself does not stand for the principle that trust rules cannot be used to promote individual recovery when recoverable assets are insufficient to satisfy all the claims of similarly situated investors, as the Trustee asserts. To the contrary, the Court explicitly stated that it was the defendants' “inability to identify their payments” that left them in the position of a general creditor. Id. In other words, if the Cunningham defendants could have traced their specific property, the Court would have imposed a constructive trust over the property and exempted it from the estate. However, what Cunningham has come to stand for is that tracing fictions should not be applied to elevate any one claimant above others similarly situated. Id. at 12–14, 44 S.Ct. 424.See also Hill v. Kinzler, 275 F.3d 924, 928 (10th Cir.2001) (“a tracing fiction should not be employed to elevate [an investor's] claim over other [investors] if those [investors] are similarly situated.”); Hatoff v. Lemons & Assoc., Inc. (In re Lemons & Assoc., Inc.), 67 B.R. 198, 214 (Bankr.D.Nev.1986) (“[T]he Cunningham line of cases rejects the fictional tracing rule which allows a claimant to trace the trust res through a commingled account.”).
Cunningham remains the essential legal framework that governs the distribution of commingled funds between competing trust claimants in bankruptcy. See Lemons, 67 B.R. 198, 213;Distral Energy Corp. v. Michigan Boiler and Engineering Co. (In re Michigan Boiler and Engineering Co.), 171 B.R. 565, 571 (Bankr.E.D.Mich.1993); SEC v. Wealth Management, LLC, 628 F.3d 323, 333–34 (7th Cir.2010) (approving Cunningham-based pro rata distribution plan in securities-fraud receivership because, “[t]he goal in both securities-fraud receiverships and liquidation bankruptcy is identical—the fair distribution of the liquidated assets.”). Although the case dealt specifically with constructive trusts, courts have applied its equitable principles to determine distribution rights in cases involving commingled funds protected under express statutory trusts.
Defendant argues that Cunningham is inapplicable to this case because Sentinel was not running a Ponzi scheme. That is incorrect. While it is true that Sentinel was not technically a Ponzi scheme, Cunningham is about the rights of similarly situated claimants to insufficient commingled funds, not the nature of the fraud that led to the commingling. See Michigan Boiler, 171 B.R. at 576. Ponzi schemes may provide the most straightforward context for Cunningham' s application, but its equitable principles are not so limited.
For example, in Michigan Boiler, 171 B.R. 565, a debtor commingled its own funds with funds that were required under state law to be held in trust for the payment of several of the debtor's subcontractors. One subcontractor brought an adversary proceeding seeking the declaration of a trust fund in the amount it was owed and a determination the funds held in trust were not property of the estate. Id. The court denied the requested relief based on the equally valid claims of the multiple trust beneficiaries:
[I]n a multiple commingled trust fund situation ... any presumptions as to which monies were used first, and for what, are baseless and purely arbitrary and would lead to inequitable results which favor one similarly situated trust fund cestui over another for no cognizable reason ... All were equally innocent in the aftermath of the debtor's failure to deal appropriately with each of their trust funds, and all are (or were) in a position to make similar arguments to those of [the plaintiff]” Id. at 573.
Instead, the court relied on Cunningham' s “equality is equity” principle and allowed the trust funds to remain in the bankruptcy trustee's hands to be distributed pro rata. Id. at 576.
Defendant asserts, without explaining, that Michigan Boiler is inapposite because it involved a statutory trust created by state law, whereas the CEA trust is created by federal law. True enough, but for purposes of this case that is a distinction without a difference. The common law of trusts is no more binding on state legislatures than it is on Congress. The pertinent question for determining tracing requirements is whether a statutory trust of any kind has been created that alters common law tracing requirements. Cf. Begier, 496 U.S. at 62 n. 4, 110 S.Ct. 2258.
Several courts have extended Cunningham to hold that tracing is per se inequitable in certain contexts and should not be allowed even when claimants can identify their specific property. See Lemons, 67 B.R. at 213 (“Cunningham and its progeny stands for the proposition that ... a creditor cannot sufficiently identify or trace the trust res through a commingled fund where the fund is too small to satisfy the claims of similarly situated parties,” because “[t]o do so would allow that claimant to benefit at the expense of those who have equally strong equitable claims to the same fund.”); U.S. v. Durham, 86 F.3d 70, 73 (5th Cir.1996) (upholding district court's decision not to trace as a permissible exercise of discretion despite fact that most funds could be traced to particular claimants); SEC v. Forex Asset Mgmt., LLC, 242 F.3d 325, 331–32 (5th Cir.2001) (affirming pro rata distribution even where objecting investors' funds were never commingled, noting that whether funds are commingled or traceable is “a distinction without a difference”); SEC v. Credit Bancorp, Ltd., 290 F.3d 80, 89 (2d Cir.2002) (tracing inappropriate in Ponzi scheme because “whether at any given moment a particular customer's assets are traceable is a result of the merely fortuitous fact that the defrauders spent the money of the other victims first.” (internal quotations omitted).); SEC v. Elliott, 953 F.2d 1560, 1569–70 (11th Cir.1992) (approving pro rata distribution where tracing would lead to inequitable result.).
Defendant correctly points out that most cases in which courts have declined to trace trust property in order to avoid disproportionate losses to similarly situated claimants have occurred in the context of equitable receivership, not bankruptcy. Those courts were therefore not bound by Section 541's “property of the estate” definition. But, as the Seventh Circuit recently noted, “[t]he goal in both securities-fraud receiverships and liquidation bankruptcy is identical—the fair distribution of the liquidated assets.” Wealth Management, 628 F.3d at 334. Where tracing would produce an outcome clearly at odds with this goal, a strong case can be made that courts should not allow it, Section 541 notwithstanding. See U.S. v. Ron Pair Enterprises, Inc., 489 U.S. 235, 242, 109 S.Ct. 1026, 103 L.Ed.2d 290 (1989) (“[I]n the rare cases in which the literal application of a statute will produce a result demonstrably at odds with the intentions of its drafters ... the intention of the drafters, rather than the strict language controls.”); but see Bonded Financial Services, Inc. v. European American Bank, 838 F.2d 890, 894 (7th Cir.1988) (expressing general doubt about “the propriety of judges' declining to enforce statutes that produce inequitable results,” and stating “[b]ankruptcy statutes are not special cases.”). This case does not squarely present the issue because, as explained below, Defendant cannot trace. Because I believe it is a question better left for Congress, I do not decide the issue.
Applying Cunningham and its progeny to the instant case, I make the following conclusions. First, I reject Defendant's argument that it should not be required to trace at all. The argument is based on an attempted analogy between CEA trusts and trusts created under the Perishable Agricultural Commodities Act (PACA). The analogy fails. PACA trusts are explicitly designed to operate as nonsegregated “floating” trusts. 7 C.F.R. § 46.46(b) (“Trust assets are to be preserved as a nonsegregated ‘floating’ trust. Commingling of assets is contemplated.”). In other words, the PACA trust does not require a specific trust res—it covers an abstract dollar figure based on the value of the underlying agricultural products and their derivatives. See, e.g., C.H. Robinson Co. v. Alanco Corp., 239 F.3d 483, 486–87 (2d Cir.2001). Courts do not require PACA trust claimants to trace precisely because there is no specific trust property to trace. The PACA trust is similar to the § 7501 trust in Begier, which, as the Supreme Court explained, “creates a trust in an abstract ‘amount’—a dollar figure not tied to any particular asset.” Begier, 496 U.S. at 62, 110 S.Ct. 2258.
Similar, but not the same—unlike PACA beneficiaries, “Congress expected that the IRS would have to show some connection between the § 7501 trust and the assets sought to be applied to a debtor's trust-fund tax obligations.” Begier, 496 U.S. at 66–67, 110 S.Ct. 2258.
The CEA trust, by contrast, requires a specific trust res, specifically “all money, securities, and property” received from FCM customers. There is no language in the CEA or its regulatory provisions to indicate that the trust protects an abstract dollar figure like PACA and § 7501 trusts. This, of course, is unsurprising: if the CEA created a floating trust, as Defendant contends, its elaborate segregation requirements would be entirely superfluous.
Defendant also points to the Senate Report accompanying the 1978 passage of the commodity-broker liquidation rules, 11 U.S.C. § 761 et seq., as evidence that Congress intended to excuse CEA beneficiaries from any tracing requirement. The report, in relevant part, states, “a customer need not trace any funds in order to avoid treatment as a general creditor.” S. Rep. No. 95–989, 95th Cong., 2d Sess. 106 (1978), 1978 U.S.C.C.A.N. 5787 at 5892 (emphasis added). But in finding that the Defendant bears at least some burden to trace, I am not treating it as a general creditor; I am treating it as one of several similarly situated trust claimants. See In re Sentinel Management Group, Inc., 398 B.R. 281, 298 (Bankr.N.D.Ill.2008) (finding that the claims of SEG 1 and SEG 3 customers “are similar in their legal nature, character, and effect.”). If the facts of this case were different, and I were merely determining distribution rights between SEG 1 customers and unsecured creditors whose property was not held in trust, Defendant would be on solid ground. But that is not the case.
The closer question is whether I ought to apply a tracing fiction to preserve SEG 1 customer property. Defendant points to no case in which a trust claimant is excused from tracing where multiple trust beneficiaries can assert co-equal claims over an insufficient pool of commingled assets. Begier did not involve similarly situated claimants so, coupled with the fact that the CEA trust does not protect an abstract dollar figure like the § 7501 trust, there is no basis for applying its “nexus” fiction. And because the lowest intermediate balance test is inappropriate where “the fund with which the wrongdoer is dealing is wholly made up of the fruits of the frauds perpetrated against a myriad of victims,” I likewise decline to apply that fiction. Due to the co-equal trust claims of SEG 1 and SEG 3 customers, I find that Cunningham and its progeny rule out the application of tracing fictions to this case.
The CFTC points to Old Republic Nat'l Title Ins. Co. v. Tyler (In re Dameron), 155 F.3d 718 (4th Cir.1998), as a case in which tracing fictions were employed to allocate an insufficient pool of commingled funds among multiple trust claimants. Old Republic is distinguishable because the claimants in that case did not have contemporaneous claims. The very small deficit in the commingled account in Dameron was created when only a single claimant had funds in the account. Thus the withdrawals that created the deficit necessarily depleted the funds of that single claimant. This case, by contrast, involves a commingled account to and from which dozens of customers made contemporaneous deposits and withdrawals over a period of several years. On top of that, and unlike Dameron, the cash deposits in this case were converted into indirect ownership interests in a commingled securities pool, which adds another layer of fiction in purporting to link customer redemptions to original deposits.
To be clear: if I was merely dealing with competing claims of SEG 1 customers and unsecured creditors not protected by statutory trust, I believe the law would require me to apply every reasonable tracing fiction available to preserve the CEA trust. But because the claimants are similarly situated, equity prevents the application of any fiction. Thus, in order to exempt the proceeds of the Citadel sale from the property of the estate, Defendant must demonstrate that the Citadel securities can be traced back to the actual deposited funds of Group 7 customers.
SECOND CONCLUSION OF LAW: Defendant is subject to common law tracing requirements due to the co-equal claims of the competing trust claimants.
iv) Whether Defendant Has Met its Burden of Tracing the Citadel proceeds to Group 7 Customer Deposits
Defendant claims that, through the report of its expert Frances McCloskey, it has met the common law requirement of tracing specific trust property. Defendant is wrong. What Ms. McCloskey has done is identify the custodial location of all securities that Sentinel recorded as being allocated to customers on its internal ledgers and on customer statements during relevant time periods. In her own words, she “trace[d] ... customers' indirect beneficial ownership in securities.” McCloskey Dep at 163.19–22. This statement is nonsensical in the context of common law tracing rules, and it is indicative of why tracing is not possible in this case. Ms. McCloskey's report demonstrates only that: 1) the securities that Sentinel listed on its ledgers and customer statements actually existed (i.e. there were no fictitious securities that appeared on Sentinel's ledgers or customer statements); and 2) the allocation of securities reflected in Sentinel's internal ledgers consistently matched—with a few limited exceptions—the information reflected on customer statements.
I do not say this to disparage the work of Ms. McCloskey, who is an accountant not an attorney. As such, she should not be expected to understand arcane common law tracing rules. Her work, like the work of all the experts in this case, was of value for my understanding of the facts.
The Trustee does not dispute any of this. Instead, he argues it is irrelevant because neither the ledgers nor the customer statements reflect the massive segregation violations that occurred at Sentinel, or the consequent jeopardy all customer funds were placed under while held in lienable accounts. True enough. But for tracing purposes the critical shortcoming of Ms. McCloskey's report is that it fails to adequately account for the fact that none of Sentinel's customers held specific ownership interests in securities. Rather, they owned pro rata portions of investment portfolios, which Sentinel was free to fill with any of the securities in its pool of assets so long as those securities met the portfolio's investment criteria. Further, these securities were generally purchased with commingled funds from the BONY SEN account or the JP Morgan cash accounts. The upshot is that the securities held in a given customer group portfolio at any time were not necessarily—indeed, were most improbably—the converted form of the original trust property (i.e. cash deposits) of the customers within that group.
This is why Defendant's attempt to demonstrate that the securities sold to Citadel were generally kept in segregation has very little meaning—the custodial location of a security at any given time was not related to the source of funds used to purchase that security.
The fungible nature of cash alone makes it impossible to trace specific securities back to original customer deposits in this case. And it gets worse for Defendant. As explained in Ms. McCloskey's report, Sentinel's “buy and hold” strategy meant that “Sentinel did not generally buy and sell securities in response to daily customer deposits and withdrawals. The nature of Sentinel's pooled investment concept meant that one customer's withdrawal and another's deposit affected the total balance in the group along with every customer's pro-rata (proportional) interest in the group.” McCloskey Rep. at 16. So, commingling aside, Sentinel's investment model makes tracing essentially impossible because, upon deposit, customer funds were immediately converted into an abstract ownership interest. In other words, Sentinel's pooled investment model renders tracing impracticable because there is no specific form of converted trust property to trace.
In summary, the removal of SEG 1 cash from segregation did not itself destroy the CEA trust. See Dameron, 155 F.3d at 723–24 (“courts have consistently rejected the notion that commingling of trust property, without more, is sufficient to defeat tracing.”). It would be paradoxical, to say the least, if Sentinel could abrogate a federal statutory trust and obtain an ownership interest over customer funds simply by removing the funds from segregation. But the commingling of customer deposits made tracing specific customer property impossible. Anytime fungible property is commingled it is, by definition, necessary to apply a tracing fiction to recover it. As explained above, if SEG 1 cash had been commingled only with House cash, Defendant would simply have to point to the particular fund into which the CEA-protected cash had gone in order to recover an equivalent amount. See, e.g., Marcus v. Otis, 169 F.2d 148, 149–50 (2d Cir.1948); Corpus Juris Secondum, Trusts, § 738: Money or other property of fungible nature. But because the SEG 1 and SEG 3 claimants are similarly situated in every meaningful respect, tracing fictions are inappropriate in this case.
Defendant argued at trial that SEG 3 customers should bear heavier losses than SEG 1 customers because of the riskier nature of SEG 3 investment portfolios. This is a misleading argument for two reasons. First, the vast majority of SEG 3 funds were held in the Group 10 portfolio, which, like all SEG 1 portfolios, was Rule 1.25 compliant. Second, the SEG 3 prime portfolio customers were stuck holding highly illiquid CDOs as a result of Sentinel's reckless leveraged trading strategy which it hid from its customers—hardly the type of risk for which Prime Portfolio customers contracted.
So the bottom line is that Defendant cannot meet its burden of tracing the Citadel proceeds to its initial deposits of customer funds and therefore cannot exempt the proceeds from the estate. See, e.g., In re United Cigar Stores Co., 70 F.2d 313, 316 (2d Cir.1934) (“There can be no recovery ... where all that can be shown is enrichment of the trustee. [The trust property] must be clearly traced and identified in specific property.”). Defendant, in its inability to identify Group 7 customer deposits, is a similarly situated creditor, and nothing more. Cunningham, 265 U.S. at 13, 44 S.Ct. 424.
THIRD CONCLUSION OF LAW: The Citadel proceeds are property of the estate under 11 U.S.C. § 541.
JUDGMENT for the Trustee and against Defendant on Count III.
1. Whether the Post–Petition Transfer was Authorized under the Bankruptcy Code or by the Bankruptcy Court
I next turn to the question of whether the August 21, 2007, transfer of funds from BONY to Defendant was authorized by the Bankruptcy Code or the Bankruptcy Court. 11 U.S.C. § 549(a)(2)(B). The Bankruptcy Code does not allow for unequal treatment of similarly-situated creditors. See11 U.S.C. § 726; 11 U.S.C. § 1123(a)(4). Based on my finding that SEG 1 and SEG 3 customers had co-equal trust claims over the assets held by Sentinel on the day it filed for bankruptcy, I conclude that the payout of the Citadel proceeds to Group 7 customers was not authorized by the Bankruptcy Code.
The question, then, is whether the Bankruptcy Court authorized the distribution. On August 20, 2007, Sentinel filed an emergency motion with the United States Bankruptcy Court for the Northern District of Illinois in which it sought entry of an order approving the turnover and distribution by BONY of the proceeds from the Citadel sale to the customer segregated accounts of Sentinel's remaining SEG 1 FCM customers. The Bankruptcy Court issued an order stating that BONY “may” distribute the Citadel sale proceeds, less a $15.6 million holdback, to Sentinel's clients in accordance with an approved pro rata plan. In re Sentinel Management Group, Inc., Case No. 07 B 14987 (Bankr. N.D. Ill., Dkt. (hereinafter “Bankr. Dkt.”) No. 978–1). The order also stated that it was “without prejudice to all rights, defenses, claims and/or causes of action, if any, of the Debtor ... against the Distributee ... with respect to any claim for priority under Section 761–767, or other applicable law.” Id. The order said nothing about whether the proceeds were property of the estate, and the Bankruptcy Judge explicitly stated in open court that he was not deciding that issue because he did not have enough information. (Bankr. Dkt. No. 978–2 at 39–47).
On August 8, 2008, the Trustee filed a Motion to Clarify or in the Alternative to Vacate or Modify the Court's August 20, 2007 Order. (Bankr. Dkt. No. 978). After briefing and oral argument, the bankruptcy judge granted the Trustee's motion to clarify. The judge explained in open court that the August 20, 2007 Order did not “authorize,” within the meaning of 11 U.S.C. § 549(a)(2)(B), the August 20, 2007 transfer, because such authorization is premised on a determination of whether the transferred property belonged to the estate, which the court had made clear it was not deciding back on August 20, 2007.
Defendant argues that the plain language of the August 20, 2007 Order indicates that the Bankruptcy Court did in fact authorize the transfer. This is a fair argument, but I am not in a position to second guess the bankruptcy judge's interpretation of his own order. See Ill. Inv. Trust No. 92–7163 v. Allied Waste Indus., Inc., (In re Resource Tech. Co.), 624 F.3d 376, 386 (7th Cir.2010) (“[Reviewing courts] owe substantial deference to the bankruptcy court's interpretation of its own orders and will not overturn that interpretation unless [the reviewing court is] convinced that it amounts to an abuse of discretion.”). Given the limited information available to the Bankruptcy Court on August 20, 2007, the potentially damaging impact a denial of the distribution could have had on futures markets, and the explicit language, both in the August 20, 2007 Order itself and the bankruptcy judge's in court statements that the transfer remained subject to challenge under applicable law, I regard the Bankruptcy Court's actions in this case as eminently reasonable. More to the point, the Court's interpretation of its own order does not even approximate an abuse of discretion.
FOURTH CONCLUSION OF LAW: The August 21, 2007 transfer of the Citadel proceeds was not authorized under the Bankruptcy Code or by the Bankruptcy Court.
2. Whether Defendant was the initial transferee or the entity for whose benefit such transfer was made
The final showing the Trustee must make to avoid the post-petition transfer is that Defendant was “the initial transferee of such transfer or the entity for whose benefit such transfer was made.” 11 U.S.C. § 550(a)(1). The Bankruptcy Code does not define the term “initial transferee.” Absent statutory guidance, the Seventh Circuit has developed a “dominion and control” test to determine initial transferee status. Bonded Fin. Servs., Inc. v. European Am. Bank, 838 F.2d 890, 893 (7th Cir.1988). Under this test, a party is considered an initial transferee only if it has “dominion over the money or other asset, the right to put the money to one's own purpose.” Id. If a party is the initial recipient of funds, but does not exercise dominion and control over them, the party is not an initial transferee. See, e.g., In re Joy Recovery Tech. Corp., 286 B.R. 54, 80 (Bankr.N.D.Ill.2002).
The Trustee argues that Defendant qualifies as an initial transferee because, although the post-petition transfer was made to a denominated segregated account for Defendant's customers, that account held some $75 million in funds over and above the amounts required to be segregated for the benefit of Defendant's customers. Because CFTC regulations allow FCMs to “draw upon such [excess] segregated funds to its own order, to the extent of its actual interest therein,” 17 C.F.R. § 1.23, the Trustee asserts that Defendant's initial custody of the Citadel proceeds satisfies the “dominion and control” test.
Defendant raises two counterarguments. First, Defendant argues that it functioned as a mere conduit between its customers and the futures markets, and cites to a number of cases in which stock brokers acting in like capacity were found not to be initial transferees. See, e.g., In re Toy King Dist., Inc., 256 B.R. 1, 145 (Bankr.M.D.Fla.2000) (listing cases in which courts have found securities and investment brokers to be mere conduits and not initial transferees); In re Dominion Corp., 199 B.R. 410, 415 (9th Cir. BAP 1996) (stock broker was a mere “conduit, not a transferee”); In re Blinder, Robinson & Co., Inc., 162 B.R. 555, 562 (D.Colo.1994) (stock broker exercised insufficient dominion over shareholders' funds to qualify as initial transferee).
The difficulty with this argument is that none of the stock broker cases involve the transfer of funds into customer accounts that contained excess segregated funds both before and after the challenged transfer. That distinction is critical to this case because, since 1997, the CFTC has treated segregated customer property as entirely fungible. See Corcoran Rep. at ¶¶ 88, 132. This means that an FCM meets its CEA segregation obligations by maintaining a minimum segregated account balance equal to or greater than the amount of its total customer obligations, not by keeping specific property in segregation. See Securities Representing Investment of Customer Funds Held in Segregated Accounts by Futures Commission Merchants, 62 F.R. 42398–01 (August 7, 1997); Financial and Segregation Interpretation 7–1, Comm. Fut. L. Rep. (July 23, 2008); Corcoran Rep. at ¶ 132 (“[Since September 1997], from the CFTC's perspective, what was critical to segregation calculations was ... whether the full amount of required customer segregated assets were maintained in cash and/or securities at any given point in time.”) (emphasis added).
I interpret the 1997 amendments to Rule 1.23 and 1.25 to mean that, because Defendant had sufficient excess funds in segregation to meets all of its customer obligations, it was not required to segregate the incoming proceeds of the Citadel sale, but instead was immediately free to use the proceeds to its own purposes. See62 F.R. 42398, 42398 (eliminating the requirement that proceeds from the sale of segregated securities be redeposited into a segregated account); Financial and Segregation Interpretation 7–1 (“The amended rules permit an FCM to ... sell permissible securities that are in a segregated account and directly deposit the proceeds from such sale into a nonsegregated account ... Any such transfer of securities to, or deposit of proceeds into, nonsegregated account can only be made if the remaining funds in segregated accounts are sufficient to cover the FCM's obligations to its commodity customers.”). And although Defendant chose to have the Citadel proceeds deposited into a segregated account, after the funds hit Defendant was free to move those funds (or, more accurately, the same amount of funds) out of segregation to use for its own purposes. CFTC Rule 1.23. Based on the above, I find that Defendant exercised “dominion and control” over the Citadel proceeds at all times that the proceeds were in its customer segregated account. To find otherwise would require me to interpret the CEA's segregation requirement in a manner that is at odds with the agency charged with enforcing it.
Second, Defendant argues that regardless of its rights over the funds, it did not actually exercise dominion and control over the Citadel proceeds because it kept them in a segregated customer account at all times. This argument fails. The court in Bonded Financial stated that the dominion and control test turns on whether a party has “ the right to put the money to [its] own purposes;” Bonded Financial, 838 F.2d at 893 (emphasis added). There is no language in the opinion to suggest that a party must actually exercise that right before it can be considered an initial transferee. Courts have interpreted Bonded Financial accordingly. See, e.g., CLC Creditor's Grantor Trust v. Howard Sav. Bank (In re Commercial Loan Corp.), 396 B.R. 730, 743 n. 8 (Bankr.N.D.Ill.2008) (“Under Bonded, however, the relevant question is whether the recipient had ‘the right to put the money to use for its own purposes' ... not whether the recipient actually exercised that right”) (emphasis in original) (quoting Bonded Financial, 838 F.2d at 893; citing Universal Serv. Admin. Co. v. Post–Confirmation Comm. Of Unsecured Creditors (In re Incomnet, Inc.), 463 F.3d 1064, 1070 (9th Cir.2006)); see also Geltzer v. D'Antona (In re The Cassandra Group), 312 B.R. 491, 496–97 (Bankr.S.D.N.Y.2004) (dominion and control analysis turns on whether party ever “had a legal right to put the payments to [its] own personal use ...”). It may very well be that “prudence and financially responsible business practice” militated against Defendant putting the Citadel proceeds to its own purposes, but best business practices do not alter the relevant legal landscape.
I also conclude that Defendant is the “entity for whose benefit [the] transfer was made.” 11 U.S.C. § 550(a)(1). This conclusion is based mostly on my finding that Defendant functioned as a de facto guarantor for all of its customer funds invested with Sentinel. See Bonded Financial, 838 F.2d at 894. Although Defendant points out that it contractually disclaimed liability for Sentinel-related losses, there is strong evidence in the record that 1) the disclaimer was not enforceable ( See, e.g., Corcoran Dep. At pp. 98–100 (explaining that an FCM cannot, by contract, shift investment losses to its customers)); CFTC Supp. Brief at p.6 n.7 (“[H]aving elected to invest customer funds through Sentinel and benefit from the expected earnings, [Defendant was] responsible for any losses resulting from [its] election”); and 2) even if it was enforceable, Defendant would have gone out of business had it tried to shift Sentinel-related losses to its customers. Thus, for all intents and purposes, the August 21, 2007 post-petition transfer had the effect of relieving Defendant from an actual, quantifiable legal liability. See In re McCook Metals, LLC, 319 B.R. 570, 590 (Bankr.N.D.Ill.2005).
There was extensive testimony to this effect at trial from Corcoran, William Dunaway, and Professor Jerry Markham.
Defendant argues, incorrectly, that the Trustee must establish that Sentinel actually intended to benefit FCStone by making the transfer of the assets. In McCook, the Bankruptcy Court stated, “[f]ollowing ... Bonded Financial ..., it appears that transfer beneficiary status depends on three aspects of the “benefit”: (1) it must actually have been received by the beneficiary; (2) it must be quantifiable; and (3) it must be accessible to the beneficiary.” 319 B.R. at 590. The court went on to consider whether demonstrating “the transferor's intent to convey a benefit is sufficient in itself” to constitute an actual benefit received (the court held it was not sufficient). Id. at 590–91. The McCook court did not hold that the transferor's intent to benefit the transferee was a separate element that must be proven to establish beneficiary status, nor has any other court in this circuit as far as I can tell.
FIFTH CONCLUSION OF LAW: Defendant was the initial transferee and the party for whose benefit the August 21, 2007 post-petition transfer was made under 11 U.S.C. § 550(a)(1).
JUDGMENT for the Trustee and against Defendant on Counts I and V.
Long after the approval of the Fourth Amended Plan, Defendant argues that the SEG 1 reserve is underfunded by approximately $32 million, which, according to Defendant, more than offsets any disproportionate distribution it received from the Citadel sale. To the extent that this is an actual claim by Defendant, it is disallowed by § 502(d). To the extent that it is simply a new argument that Defendant wants me to weigh in determining an equitable outcome, it comes too late and is barred by collateral estoppel. See Grede v. Bank of New York Mellon Corp., 598 F.3d 899, 902 (7th Cir.2010).
B. Count II—Avoidance of Preferential Transfer
The Trustee next seeks to avoid and recover the August 17, 2007, pre-petition transfer of $1,097,925 to FCStone as an unlawful preference pursuant to 11 U.S.C. § 547(b). To avoid the transfer the Trustee must demonstrate the August 17, 2007 transfer was 1) to or for the benefit of Defendant; 2) for or on account of an antecedent debt owed by the debtor before such transfer was made; 3) made while the debtor was insolvent; 4) made on or within 90 days before the date of the filing of the petition; and 5) enabled Defendant to receive more than it would have in Chapter 7 liquidation. 11 U.S.C. § 547(b). Based on my findings above, Defendant has made the necessary initial showing to avoid the August 17 transfer.
The real dispute between the parties on Count II is whether the transfer is shielded from avoidance by Section 546(e)'s safe harbor provision, which provides:
Notwithstanding sections 544, 545, 547, 548(a)(1)(B), and 548(b) of this title, the trustee may not avoid a transfer that is a margin payment, as defined in section 101, 741, or 761 of this title or settlement payment, as defined in section 101 or 741 of this title, made by or to (or for the benefit of) a commodity broker, forward contract merchant, stockbroker, financial institution, financial participant or securities clearing agency, or that is a transfer made by or to (or for the benefit of) a commodity broker, forward contract merchant, stockbroker, financial institution, financial participant, or securities clearing agency, in connection with a securities contract, as defined in section 741(7), commodity contract, as defined in section 761(4), or forward contract, that is made before the commencement of the case, except under section 548(a)(1)(A) of this title.
11 U.S.C. § 546(e).
Defendant asserts that the August 17, 2007 transfer was both a settlement payment made to a commodity broker as well as a transfer made in connection with a securities contract, to wit, the Investment Advisory Agreement that governed the relationship between Sentinel and Defendant. The trustee argues that the August 17, 2007 transfer was not made “in connection with a securities contract” because the Investment Advisory Agreement does not, in and of itself, “contract for the purchase, sale, or loan of a security,” 11 U.S.C. § 741(7). The trustee also contends that the August 17, 2007 transfer was not a “settlement payment” because it was not “made to complete [a] securities transaction.” The relevant securities transaction, the Trustee argues, was between Sentinel and Citadel; the transfer merely constituted redemption of Group 7 and 9 customers' indirect beneficial interest in their respective portfolios (which happened to include proceeds from the Citadel sale).
I decline to address these specific arguments because, regardless of whether the distribution of the Citadel proceeds fits under a literal interpretation of § 546(e), I find it inconceivable that Congress intended the safe harbor provisions to apply to the circumstances of this case. There are two main bases for my finding. First, applying the safe harbor to shield Sentinel's distributions to its SEG 1 customers would create the very type of systemic market risks that Congress sought to prevent with its passage. Second, failing to apply the safe harbor in this case will not result in the unwinding of completed securities and commodities transactions that Congress sought to protect. Thus, applying the safe harbor here would produce a result “demonstrably at odds with the intentions of its drafters.” United States v. Ron Pair Enter., 489 U.S. at 242, 109 S.Ct. 1026.
Congress enacted § 546(e)'s safe harbor as a means of “minimizing the displacement caused in the commodities and securities markets in the event of a major bankruptcy affecting those industries.” H.R. Rep. 97–420, at 1 (1982), 1982 U.S.C.C.A.N. 583, 583. The safe harbor functions as a firewall that insulates legitimate securities and commodities transactions from avoidance because of the potential destabilizing effects that unwinding such trades could have on the broader market. Section 546(e) reflects a clear policy choice by Congress to carve out an exception to the general principles of equity that underlie American bankruptcy law in order to protect the nation's financial systems. See generally In re Resorts Intern., Inc., 181 F.3d 505, 515 (3d Cir.1999) (“[S]ection 546 [stands] at the intersection of two important national legislative policies on a collision course—the policies of bankruptcy and securities law.”) (internal quotations omitted).
When Debtor sells a security to Buyer immediately prior to filing for bankruptcy, it is logical to shield that transaction from avoidance in order to prevent damaging ripple effects from spreading across the securities market. See In re Enron Creditors Recovery Corp., 651 F.3d 329, 334 (2d Cir.2011) ( “If a firm is required to repay amounts received in settled securities transactions, it could have insufficient capital or liquidity to meet its current securities trading obligations, placing other market participants and the securities markets themselves at risk.”).
But a different set of ripple effects arise where Debtor is a financial institution that sells securities on behalf of third party customers, and § 546(e) is invoked not to shield the actual exchange between Debtor and Buyer but to uphold the manner in which Debtor distributes exchange proceeds to its customers. If Debtor distributes proceeds in an uneven and arbitrary manner (i.e. favoring certain customers over others with an equally forceful legal claim to the funds), extending § 546(e) safe harbors to uphold the distribution would destabilize the financial system by making it utterly unpredictable how losses will be apportioned in the event that an FCM or investment advisor goes bankrupt.
Consider the following hypothetical. If the safe harbor provisions applied to this case, then an insolvent investment advisor, in the eleventh hour prior to filing for bankruptcy, could drain its accounts to pay out all customers with names beginning with letters in the first half of the alphabet, while shifting all losses to customers with names in the latter half, and § 546(e) would render courts powerless to do anything about it. Could a Congress concerned with systemic market risks have intended the safe harbor to shield this type of arbitrary and destructive conduct simply because the distributions were made “in connection with a securities contract,” or (less clear in this case) may be described as a “settlement payment”? I do not think so.
But the risks do not end with introducing uncertainty into the market and the investment determent this would cause. The problems laid out in the above scenario are compounded where, as here, all of the customers are themselves financial institutions. That a few of these customers could be forced to bear all losses, rather than their pro rata share, in the event of an FCM or investment advisor's bankruptcy raises the likelihood of institutional collapse and associated systemic fallout. Any systemic risks caused by future bankruptcies of investment advisers will likely be mitigated if losses are proportionately spread across the entire customer base, rather than arbitrarily foisted on a small group of financial institutions that may be unable to bear them.
Defendant warns of damaging market fallout that will occur if I allow the transfers to be avoided. I am unconvinced. First, the argument tends to lose its force where, as here, all claimants are financial institutions and can make the “ripple effect” argument with equal force. In fact, given that investment advisers have a much larger stake in the economy than FCMs, the ripple effects argument probably cuts against Defendant. Second, I do not believe that the losses inflicted by a claw back of funds will spread beyond Defendant and some similarly situated enterprises. I agree with the CFTC that Defendant will be legally unable to seek redress from their customers—“having elected to invest customer funds through Sentinel and benefit from their expected earnings, the FCMs are responsible for any losses resulting from their election.” CFTC Amicus Memorandum at 2 n.1 ( citing Craig v. Refco, Inc., 624 F.Supp. 944, 947 (N.D.Ill.1985)). Even if Defendant did have a viable cause of action against its customers, it is constrained from bringing suit by the practical consequences such action would have on Defendant's business.
Arguments about the uniquely fragile nature of the futures market also tend to break down in this case. As Professor Markham explained at trial, many of the SEG 3 customers faced margin calls similar to the SEG 1 customers. Hedge funds, for example, are also active in the futures market—their failure could have the exact same consequences as an FCM's failure.
Perhaps an even more important consideration in determining the applicability of § 546(e) to this case is whether failing to apply it will cause the type of unwinding of completed security transactions the safe harbor was designed to prevent. The answer is no. The relevant security transaction in this case was between Sentinel and Citadel; no securities were ever exchanged between Sentinel and FCStone. ( See Figure 1 below). The safe harbor is very clearly designed to protect Citadel from having to return the securities it received from Sentinel, as those securities may be necessary to meet its current trading obligations. See In re Enron Creditors Recovery Corp., 651 F.3d at 334. Allowing the Trustee to avoid the August 17, 2007 transfer would not affect Citadel in any way.
The legislative record is devoid of indication that Congress intended § 546(e) to govern how the debtor distributes proceeds from a completed securities transaction if that debtor happens to be trading on behalf of third parties. Defendant's argument that this secondary transaction between Sentinel and Defendant is subject to § 546(e) is further weakened when considered in the context of Sentinel's investment model. As described above, customer deposits and redemptions at Sentinel were not directly tied to the purchase or sale of securities—customer deposits were not necessary to settle security purchases, nor were customer redemptions necessary to settle security sales—and therefore did not affect the settlement chain that § 546(e) is designed to protect. See Wieboldt Stores Inc. v. Schottenstein, 131 B.R. 655, 664 (N.D.Ill.1991) (“Congress exempted settlement payments in the commodities (and later the securities) industry out of concern that the bankruptcy of one party in the clearance and settlement chain could spread to other parties in that chain.”). So, assuming Sentinel's customer redemptions can be squeezed within the broadest literal interpretation of § 546(e)'s terms, from a practical standpoint it is clear that they were separate transactions unrelated to the completed trade between Citadel and Sentinel that is protected by the safe harbor.
The Southern District of New York has come down differently on this issue in two recent cases involving the Ponzi scheme perpetrated by Bernard Madoff. See Picard v. Katz, 462 B.R. 447 (S.D.N.Y.2011); Picard v. Greiff et al., No. 11 Civ. 3775, 2012 WL 1505349 (S.D.N.Y. Apr. 30, 2012). Madoff Securities operated an investment advisory unit that purported to trade securities on behalf of customers. In fact, it made no such trades—the unit was operated as a pure Ponzi scheme. As with the instant case, the Trustee of the Madoff Securities estate sought to avoid certain pre-petition payouts to customers. The Picard Court found that § 546(e) “precludes the Trustee from bringing any action to recover from any of Madoff's customers any of the monies paid by Madoff Securities to those customers except in the case of actual fraud.” Katz, 462 B.R. at 452. The Picard decisions were based largely on an “extremely broad” interpretation of § 546(e)'s “settlement payment” provision given by the Second Circuit in Enron Creditors, 651 F.3d at 335. There, the Second Circuit considered whether an issuer's early redemption of commercial paper met § 741(e)'s definition of “settlement payment.” The Enron Court examined the type of financial instruments whose trade qualifies as a securities transaction subject to § 546(e)—it did not deal with the distribution of proceeds from that transaction to third party customers. Id. at 335. I agree with the Second Circuit that Congress intended § 546(e) to reach a broad scope of financial instruments exchanged in Transaction 1 (see Figure 1 below). I do not see, however, how this relatively uncontroversial holding allows for extending § 546(e)beyond the securities transaction to subsequent, indirectly related cash transactions to customers.
SIXTH CONCLUSION OF LAW: 11 U.S.C. § 546(e)'s safe harbor provisions do not apply to this case. Assuming that the Investment Advisor Agreement qualifies as a “securities contract,” and/or the August 17, 2007 transfer qualifies as a “settlement payment” under a literal interpretation of 11 U.S.C. § 741, applying § 546(e) to exempt the August 17, 2007 transfer from avoidance would produce a result “demonstrably at odds with the intentions of its drafters.” Ron Pair Enter., 489 U.S. at 242, 109 S.Ct. 1026.
JUDGMENT for the Trustee and against Defendant on Count II.
Figure 1
Image 1 (5.46" X 6.47") Available for Offline Print
C. COUNT IV—UNJUST ENRICHMENT
As an alternative to the avoidance claims, the Trustee seeks to recover funds transferred to Defendant under a theory of unjust enrichment. I find, as a matter of law, the unjust enrichment claim is preempted by the bankruptcy laws, regardless of on whose behalf the Trustee brings the claim. See B.E.L.T., Inc. v. Wachovia Corp., 403 F.3d 474, 477 (7th Cir.2005) ( “Calling the receipt of a preference ‘unjust enrichment’ does not change matters; a preference by any other name is still a preference and cannot be recovered outside of bankruptcy.”).
SEVENTH CONCLUSION OF LAW: The Trustee's unjust enrichment claim is preempted by the Bankruptcy Code.
JUDGMENT for the Defendant and against the Trustee on Count IV.
IV. CONCLUSION
This is an extraordinary case for a number of reasons. It involves unprecedented violations of federal segregation rules and enormous losses of customer funds. Never before has a firm with dual registration as an FCM and an investment advisor filed for bankruptcy, raising difficult questions over the competing rights of two groups of trust claimants assigned special protection in bankruptcy under federal law. In fact, this court is unable to find past cases of any kind involving such a clash between two federal trusts.
Yes, the facts are extraordinary, but at the end of the day it is a bankruptcy case. And if there is one prevailing principle that underpins American bankruptcy laws, it is that “equality is equity.” Cunningham, 265 U.S. at 13, 44 S.Ct. 424. This is the starting principle in cases where investors' assets are commingled and recoverable property in bankruptcy is insufficient to fully repay those investors. Until Congress determines otherwise, it remains the starting principle even when the commingled funds are protected under competing federal trusts.
The pro rata distribution model that stems from Cunningham is particularly appropriate for this case because a pro rata interest in Sentinel's commingled pool of securities and cash is precisely what Defendant, and all other Sentinel customers, owned. Divvying up pro rata shares on the group level (i.e. what happened to be left in each customer portfolio when Sentinel filed for bankruptcy), as Defendant urges, however, is illogical from a tracing standpoint, because the securities in a given folder are not the converted form of the group customers' original trust property. More importantly, dividing pro rata interests by group would be grossly inequitable, because the allocation of customer securities at the time of Sentinel's bankruptcy filing was the product of Sentinel's “baseless and purely arbitrary” decisions as to which customer securities to 1) liquefy during the summer of 2007 to pay down the BONY loan; and 2) which securities to stake as collateral for the BONY loan in the eleventh hour. Michigan Boiler, 171 B.R. at 573.
That pure happenstance governed which Sentinel customers received payouts and which customers bore disproportionate losses is most dramatically highlighted by the collateral swap that took place between July 30 and 31, 2007. If Sentinel had kept the securities allocated to SEG 1 customers in the FC1 account rather than replace them with securities allocated to SEG 3 customers in the SLM account, I imagine the very same parties with the very same claims would be before the court, the only difference being which side of the “v” the parties stood in the case caption. If that were so, I would apply the same principles to ensure that SEG 1 customers did not bear an unfair share of Sentinel's losses.
The most apt description I have heard of Sentinel's collapse and last minute distributions of customer funds came at trial when a witness likened it to a game of musical chairs—whichever customers' funds happened to be in segregation when the music stopped received redemptions; those whose funds were not in segregation received nothing. To allow Sentinel's management's baseless, eleventh hour choices over how to steer the company ship, sinking under the weight of their own fraud, to dictate the outcome of this case would fly in the face of justice and do nothing to advance any plausible Congressional purpose. With no legal requirement to do so, I refuse to give such arbitrary and inequitable conduct the imprimatur of this Court.
Based on my findings of fact and seven conclusions of law, I find in favor of the 890Trustee and against Defendant on Counts I, II, III, and V; and in favor of Defendant and against the Trustee on Count IV. It is HEREBY ORDERED that Defendant return $15,576,964.00, the full amount it received in proceeds from the Citadel sale, to the Sentinel Liquidation Trust to be distributed pro rata in accordance with the Fourth Amended Chapter 11 Plan of Liquidation.