United States Tax Court

Agency decision

Ask Donna

What actually matters in this document.

Text

T.C. Memo. 2025-34

GWA, LLC, GEORGE A. WEISS, TAX MATTERS PARTNER,

Petitioner

COMMISSIONER OF INTERNAL REVENUE,

Respondent

__________

Docket No. 6981-19.

Filed April 16, 2025.

__________

Eric J. Albers-Fiedler, William F. Nelson, Drew A. Cummings, Thomas

V. Linguanti, Sheri A. Dillon, James G. Steele III, Jennifer E. Breen, and

Maya A. Hairston, for petitioner.

Lisa M. Goldberg, Elizabeth P. Flores, Michael A. Sienkiewicz, Byron M.

Huang, and Oleida Sullivan, for respondent.

TABLE OF CONTENTS

MEMORANDUM FINDINGS OF FACT AND OPINION ..................... 4

FINDINGS OF FACT .............................................................................. 7

Introduction ................................................................................... 7

II.

GWA, LLC ..................................................................................... 8

III.

GWA’s Desire for Leverage ........................................................... 9

IV.

GWA’s Affiliates .......................................................................... 11

“Specially Tailored Financial Instruments” ............................... 11

VI.

Features of Standard Call Options ............................................. 13

VII.

The RBC Transactions ................................................................ 16

Served 04/16/25

[*2]

VIII. Deutsche Bank’s “Managed Account Product Structure” .......... 19

IX.

GWA’s Negotiations with Deutsche Bank .................................. 20

The Investment Advisory Agreement ......................................... 23

XI.

Term of Barrier Contract #1 ....................................................... 24

XII.

Payout on Barrier Contract #1 ................................................... 25

XIII. Trading and Management of the Securities Basket .................. 27

XIV. Related Agreements .................................................................... 28

XV.

Barrier Contract #2 ..................................................................... 29

XVI. Weiss Multi-Strategy Advisors ................................................... 30

XVII. Termination of Barrier Contract #2 and Execution of

Barrier Contracts #3 through #6 ................................................ 32

XVIII. Cross Trading and Position Journaling ..................................... 33

XIX. Replacement of Barrier Contracts #3 through #6 by

Barrier Contracts #7 through #10 .............................................. 34

XX.

Financial Turbulence and “New MAPS” .................................... 35

XXI. Termination of Barrier Contract #1 ........................................... 37

XXII. Termination of Barrier Contracts #7 through #10 .................... 38

XXIII. IRS Legal Advice Memorandum ................................................. 39

XXIV. Mark-to-Market Election ............................................................ 40

XXV. Supervisory Approval of Penalties ............................................. 42

XXVI. Issuance of the FPAAs ................................................................ 43

OPINION ................................................................................................ 44

Burden of Proof............................................................................ 44

II.

Expert Testimony ........................................................................ 44

[*3]

III.

Proper Characterization of the Barrier Contracts ..................... 45

IV.

Economic Realities of the Barrier Contracts ................... 46

Consideration ......................................................... 47

Pricing .................................................................... 49

Option Term ........................................................... 52

Reference Property ................................................ 52

Early Termination ................................................. 55

Treatment of Dividends ......................................... 59

Absence of Risk to Deutsche Bank ........................ 60

Lack of “Optionality” for GWA .............................. 67

Ownership of the Underlying Securities ......................... 70

Risk of Investment Loss ........................................ 72

Opportunity for Investment Gain ......................... 74

Control over Investment Assets ............................ 76

Other Benefits and Burdens of Ownership........... 79

Ability to Extract Cash .......................................... 82

Other Benefits ........................................................ 86

Conclusion .............................................................. 87

Petitioner’s Leverage Theory ........................................... 88

Mark-to-Market Election ............................................................ 92

Statutory and Regulatory Background ............................ 93

GWA’s Mark-to-Market Election ..................................... 95

OGI’s Alleged Mark-to-Market Election .......................... 98

[*4]

Change in Accounting Method .................................................. 114

VI.

Validity of the Election ................................................... 109

Governing Legal Principles ............................................ 115

Purpose and Operation of Section 481 ................ 115

Changes in Accounting Method........................... 116

Analysis ........................................................................... 117

Petitioner’s Arguments................................................... 121

Penalties .................................................................................... 127

Penalty Approval ............................................................ 127

Accuracy-Related Penalties............................................ 129

Negligence ............................................................ 129

Substantial Understatement of Income Tax....... 133

Applicability of Penalties to Section 481(a)

Adjustments .................................................................... 134

APPENDIX ........................................................................................... 137

MEMORANDUM FINDINGS OF FACT AND OPINION

LAUBER, Judge: This case involves GWA, LLC (GWA), a TEFRA

partnership, of which George A. Weiss, a hedge fund manager, is the tax

matters partner. 1 In the 2000s GWA executed with Deutsche Bank AG

(Deutsche Bank) ten transactions to which we will refer as the Barrier

Contracts. GWA was the nominal buyer and Deutsche Bank was the

1 Before its repeal, the Tax Equity and Fiscal Responsibility Act of 1982

(TEFRA), Pub. L. No. 97-248, §§ 401–407, 96 Stat. 324, 648–71, governed the tax treatment and audit procedures for many partnerships, including GWA.

[*5] nominal seller. GWA treated the Barrier Contracts as call option

contracts under sections 1234 and 1234A. 2

Each Barrier Contract referenced a basket of securities, and the

payout on each “option” depended on the value of those securities on the

expiration date. The securities were nominally owned by a Deutsche

Bank affiliate. But GWA directed trading in the securities basket on a

daily or hourly basis, employing the same complex strategies it used in

its other portfolios.

For each Barrier Contract, GWA racked up large trading gains in

the underlying securities basket. But for Federal tax purposes it took

the position that these profits were not taxable on an annual basis as

short-term gains. Rather, it contended that tax on its profits should be

deferred until it exercised or terminated the “option.” Because each “option” had a term of 12+ years, the tax deferral could continue for quite a

while. And the tax would then be imposed, not at ordinary income rates,

but at the lower rates applicable to long-term capital gains.

In 2010 the Internal Revenue Service (IRS or respondent) published a memorandum identifying transactions resembling the Barrier

Contracts as abusive. The Senate Permanent Subcommittee on Investigations (PSI) subsequently opened an investigation into these transactions. The PSI conducted interviews, held hearings, and collected more

than one million pages of documents from five custodians, including

Deutsche Bank and GWA.

On July 22, 2014, the PSI completed its investigation and published a 96-page report, concluding that Deutsche Bank had promoted

the Barrier Contracts to help hedge funds “avoid [F]ederal taxes and

leverage limits on buying securities with borrowed funds.” Staff of S.

Perm. Subcomm. on Investigations, 113th Cong., Abuse of Structured

Financial Products: Misusing Basket Options to Avoid Taxes and Leverage Limits 1 (Comm. Print 2014). The PSI estimated that Deutsche

Bank helped GWA and other funds avoid more than $3 billion in Federal

income tax. The PSI specifically identified GWA as one of “the two largest participants” in this endeavor.

2 Unless otherwise indicated, statutory references are to the Internal Revenue

Code, Title 26 U.S.C. (Code), in effect at all relevant times, regulation references are

to the Code of Federal Regulations, Title 26 (Treas. Reg.), in effect at all relevant times,

and Rule references are to the Tax Court Rules of Practice and Procedure. We round

monetary amounts to the nearest dollar.

[*6] The IRS selected GWA’s 2009 and 2010 returns for examination.

On December 3, 2018, it issued petitioner a Notice of Final Partnership

Administrative Adjustment (FPAA) for each year. The FPAAs determined (among other things) that, for Federal income tax purposes, the

Barrier Contracts were not “options” and that GWA was, in substance,

the owner of the basket securities. The FPAAs determined total ordinary income adjustments in excess of $500 million for 2009 and 2010,

plus accuracy-related penalties for each year.

On May 1, 2019, petitioner petitioned this Court for readjustment

of the partnership items. The case presents three principal questions,

which are interrelated in terms of their bottom-line tax effects:

● Whether the “option” form of the Barrier Contracts should be

disregarded, with GWA being treated, in substance, as owning the basket securities for Federal income tax purposes.

● Whether a “mark-to-market” election that GWA made on its

1998 tax return required that it mark to market the basket securities

(or the “option”) on an annual basis under section 475(f)(1), with the result that any gain or loss would be taxed annually as ordinary income

or loss under section 475(d)(3)(A) and (f)(1)(D).

● If respondent’s position on one or both of the foregoing questions

is sustained, whether the Commissioner’s action constitutes a change to

GWA’s method of accounting to clearly reflect income under section 446,

requiring one or more section 481 adjustments to prevent amounts from

being duplicated or omitted.

We answer these questions as follows:

● The Barrier Contracts were not “options,” and GWA in substance was the owner of the basket securities.

● GWA made a mark-to-market election on its 1998 tax return,

but this election was invalid because it purported to cover only a subset

of the securities trading activities in which GWA then engaged (or might

in future engage).

● The Commissioner’s adjustments to GWA’s income, premised

on the determination that it owned the basket securities, constituted a

change in method of accounting that necessitates an adjustment under

section 481 to prevent omission of income.

FINDINGS OF FACT

[*7]

The following facts are derived from the Pleadings, 12 Stipulations of Facts with attached Exhibits, and the testimony of fact and expert witnesses admitted into evidence at trial. GWA had its principal

place of business in Connecticut when its Petition was timely filed. Absent stipulation to the contrary, this case would be appealable to the

U.S. Court of Appeals for the Second Circuit. See § 7482(b)(1)(E), (2). 3

Introduction

George Weiss has been a financial services professional for more

than 50 years. He is a graduate of the Wharton School and holds numerous professional licenses. In 1978 he founded George Weiss Associates, Inc. (Weiss Associates), a Connecticut corporation, and has been

its sole shareholder ever since. During its earliest years Weiss Associates engaged in securities trading and brokerage with a primary focus

on domestic utility companies. Its main clients were Connecticut-based

financial institutions and insurance companies that sought reliable, if

conservative, returns.

In 1986 Weiss Associates began offering investors the opportunity

to participate in a hedge fund. A hedge fund is a pool of money invested

in stocks and other securities to which managers apply complex trading

and risk management techniques. Investing in hedge funds appeals to

investors seeking to outperform market averages, while protecting

against the risk of large losses during market downturns.

The hedge fund that Weiss Associates offered to investors employed a “relative value long/short strategy,” in which pairs of stocks,

typically from the same industry, were bought and sold in roughly equal

3 On April 29, 2024, GWA and certain of its affiliates filed a chapter 11 petition

for bankruptcy in the U.S. Bankruptcy Court for the Southern District of New York.

A bankruptcy petition operates as a stay of “the commencement or continuation of a

proceeding before the United States Tax Court concerning a tax liability of a debtor

that is a corporation.” 11 U.S.C. § 362(a)(8). That provision does not apply here because this case does not concern “a tax liability” of GWA. As a TEFRA partnership,

GWA has no entity-level tax liability. Rather, this case relates to the tax liabilities of

GWA’s partners, who will be affected by any adjustments to GWA’s partnership items.

See § 701; 1983 W. Rsrv. Oil & Gas Co. v. Commissioner, 95 T.C. 51, 56–60 (1990)

(holding that a partnership’s petition for bankruptcy did not stay a TEFRA partnership

proceeding because it was not a proceeding “concerning the debtor”), aff’d, 995 F.2d 235

(9th Cir. 1993) (unpublished table decision). Petitioner has confirmed that none of

GWA’s partners has filed for bankruptcy.

[*8] dollar amounts. A “long” position refers to a security that an investor buys and holds for a period of time, generally because the purchaser

believes that the security will increase in value. A “short” position refers

to a transaction where the investor borrows a security and then sells it,

generally in the belief that it will decrease in value.

Implementing this strategy, Weiss Associates acquired long positions in stocks it believed were undervalued and took short positions in

stocks it believed were overvalued. The strategy identified pairs of companies whose stock performance was expected to correlate generally (because in the same industry), but not perfectly, and sought to exploit temporary differences in the price movements of these stocks. Weiss Associates expected its long/short strategy to earn investors stable annual

returns in the range of 6–7%.

As of the mid-1990s Weiss Associates had $1.7 billion in assets

under management and was one of the largest hedge fund organizations

in the world. The firm gradually diversified its utilities-focused approach to encompass other types of securities that could be “paired” in a

manner consistent with its long/short strategy.

II.

GWA, LLC

Mr. Weiss and his colleagues eventually began investing their

own money—referred to as “inside money” or “proprietary capital”—using the same long/short strategy that Weiss Associates used to generate

returns for third-party investors. In 1996 Mr. Weiss caused GWA to be

formed as a vehicle for such investment. Although this “inside money”

venture was originally expected to be temporary, it proved so lucrative

that it became a major focus of Mr. Weiss and his colleagues during the

1990s and 2000s.

Mr. Weiss was GWA’s sole manager, and he held (directly or

through affiliates) majority ownership of GWA at all relevant times.

GWA’s operating agreement gave him plenary discretion to decide who

else could be a member. He used this authority to reward and retain

key employees of Weiss Associates by offering them a stake in GWA.

Prospective members of GWA were generally required to have a

specified level of industry experience and meet certain income thresholds. Members were generally forbidden to transfer their membership

interests without Mr. Weiss’s consent, and they were required to sell

their interests back to the firm if their employment with GWA or one of

its affiliates ended. GWA had fewer than 50 members at all times.

[*9] Frederick Doucette was a longstanding associate of Mr. Weiss.

He joined Weiss Associates in 1990, became a member of GWA in 2003,

and eventually served as its president and chief operating officer (COO).

In that capacity he was responsible for overseeing GWA’s legal, compliance, technology, accounting, and tax functions. He and Mr. Weiss provided extensive testimony during the trial of this case.

Mr. Weiss focused his attention on the firm’s investment portfolios, leaving stewardship of GWA’s day-to-day operations to Mr.

Doucette. Mr. Weiss chaired the firm’s executive committee, which was

responsible for making final decisions on new products and approving

portfolio strategies, and its allocation committee, which was responsible

for allocating investment dollars among the various portfolio managers.

GWA’s portfolio managers were New York based and reported to Mr.

Weiss directly.

III.

GWA’s Desire for Leverage

Like many hedge funds, GWA earned revenue by charging fees to

investors under the “two-and-twenty” model. Under this fee structure,

the hedge fund levies a flat management fee, calculated as 2% of the

assets under management, and a performance fee, calculated as 20% of

the annual gain enjoyed by the fund. Investment managers are incentivized to grow their assets under management because doing so offers

a greater opportunity for fees. But the long/short strategies employed

by GWA generated modest (albeit stable) returns relative to the amount

of capital invested. To achieve greater profits, GWA’s managers looked

for ways to enhance their exposure to financial markets by increasing

the quantum of capital invested.

A common technique that hedge funds employ to increase their

market exposure is to borrow money. The term “leverage” refers to this

tactic of using borrowed capital to increase market exposure and (it is

hoped) investment returns. Obtaining leverage was key to the profitability of Weiss Associates and GWA. Mr. Doucette had primary responsibility for investigating ways to secure greater leverage.

Prime brokerage is a major resource available to investment managers seeking leverage. In a prime brokerage account, the broker lends

the customer cash, which the customer can then use to purchase stocks

or other securities. Prime brokers derive revenue by charging interest

on loans to customers and by offering customers fee-based services, such

as trade executions and cash management. They may also “internalize”

[*10] the equities that are held long in customers’ accounts and derive

revenue by lending those equities to other investors intending to sell

them short. Prime brokerage is an established business for global financial firms, which compete for hedge fund customers by negotiating rates

on lending, fees for services, and the amounts of leverage they are willing to make available.

A risk faced by prime brokers who offer leverage is that the customer may be unable to repay the loan if market conditions deteriorate.

To protect themselves against this risk, prime brokers require customers to supply collateral in the form of cash or securities, commonly

known as “margin.” A loan that is made available in a prime brokerage

account is usually called a “margin loan.” Prime brokers monitor the

amount of margin in a customer’s account daily. If the margin’s value

falls below a set level, the broker will require the customer to contribute

additional cash or securities (or to sell stock to reduce the margin debt).

This is dubbed a “margin call,” a call no investor wants to receive.

Beginning in the 1930s, rules issued by the Federal Reserve

Board—commonly called Regulation T and Regulation U—placed limits

on the amount of leverage that prime brokers and other lenders could

offer customers. See 12 C.F.R. §§ 220.12, 221.7 (1998). Regulation X

extended the limits imposed by Regulations T and U to cover credit from

foreign lenders. See 12 C.F.R. § 224.3 (1998). Under these rules customers investing in U.S. equities had to supply margin equal to 50% or

more of the value of the securities in the account. In other words, the

customer was required to maintain a leverage-to-collateral ratio that did

not exceed two to one. These margin restrictions prevailed through at

least December 12, 2006.

Beginning in early 2007 the rules surrounding portfolio margin

requirements changed. Whereas Regulations T and U had imposed

fixed margin requirements on most prime brokerage accounts, the new

rules were more flexible and keyed allowable leverage to portfolio risk.

In February 2007 a pilot regulatory program permitted member firms

to receive up to 6.5 times leverage on equities in their prime brokerage

accounts. See Notice of Filing and Immediate Effectiveness of Proposed

Rule Change Relating to Making the Portfolio Margin Pilot Permanent,

Exchange Act Release No. 34-58251, 73 Fed. Reg. 45,506, 45,507 (July

30, 2008). This pilot became permanent in mid-2008. See ibid.

Investment managers who were designated “specialists” or “alternate specialists” on major U.S. stock exchanges were exempt from some

[*11] margin requirements. Weiss Associates took advantage of this exemption in 1996 when it became an alternate specialist on the Philadelphia Stock Exchange. Through a joint back-office agreement with Merrill Professional Clearing Corp. (Merrill Pro), Weiss Associates could enjoy leverage of 20 to 1 on the 150 stocks for which it ultimately served

as an alternate specialist. But this benefit was available only for trading

in those 150 stocks, and GWA viewed the interest rate charged by Merrill Pro (sometimes approaching 10%) as rather high.

IV.

GWA’s Affiliates

GWA conducted a portion of its trading activity through affiliates.

One such affiliate was George Weiss & Co., LLC (Weiss & Co.), a brokerdealer and securities trader treated as a partnership for Federal income

tax purposes. GWA had a controlling interest in Weiss & Co. during

1997 and 1998. Like Weiss Associates, Weiss & Co. had a joint backoffice agreement with Merrill Pro that enabled it to obtain extra leverage in its portfolios.

GWA conducted another portion of its securities trading business

through OGI Associates, LLC (OGI), a Connecticut company. OGI was

formed in 1994 for the purpose of trading in securities using GWA’s proprietary capital, which it did at all relevant times. As of May 28, 1998,

GWA was OGI’s sole member, and Mr. Weiss was its sole manager.

OGI was a single-member limited liability company (LLC) wholly

owned by GWA, and OGI did not elect to be classified as a corporation.

For Federal tax purposes, therefore, OGI was “[d]isregarded as an entity

separate from its owner.” See Treas. Reg. § 301.7701-3(b)(1)(ii). As a

rule, “if [an] entity is disregarded, its activities are treated in the same

manner as a sole proprietorship, branch, or division of the owner.” Id.

§ 301.7701-2(a).

In 1998 OGI entered into an investment banking services agreement with Weiss Associates. Under this agreement Weiss Associates

committed to “perform investment operations and investment banking

functions for [OGI], using whatever leverage is available under regulatory constraints.” OGI maintained a prime brokerage account at

Deutsche Bank during all relevant years.

“Specially Tailored Financial Instruments”

GWA’s Operating Agreement, dated March 11, 1998, expressed a

commitment to employing a leveraged investment strategy focusing on

[*12] stocks of utilities, financial institutions, and companies in other

industries. This strategy included investment in “specially tailored financial instrument[s]” (STFIs). The Operating Agreement defined

STFIs as “investment[s] . . . pursuant to which, through the use of an

option, swap, or other derivative structure . . . [GWA] obtains or increases a desired amount of leverage and/or deferral of income.” (Emphasis added.)

As indicated in the Operating Agreement, GWA pursued STFIs

for two major reasons. First, they offered access to leverage at levels

greater than those available through traditional prime brokerage. Second, they portended the hope of deferring income by converting shortterm trading profits into long-term capital gains taxable at much lower

rates many years down the road.

From 1998 through the tax years in issue, GWA invested hundreds of millions of dollars in STFIs that it characterized as “call options.” GWA initially invested in this type of product with Royal Bank

of Canada (RBC) and later with Deutsche Bank. These products lacked

many (or most) features of standard call options. And the products GWA

purchased from Deutsche Bank included stop-loss features or “barriers”

that significantly reduced, if they did not entirely eliminate, the risks

that buyers and sellers of call options typically face.

The putative “call options” had long terms, with expiration dates

ranging from 5 to 12 years into the future. The asset underlying each

“option” was a basket of securities that included hundreds or thousands

of different stocks (“basket securities”), which could vary from day to day

(or hour to hour). The basket securities were nominally owned by the

bank. But GWA was entitled to trade the basket securities as it wished,

subject to very minor constraints. And it traded them with great gusto,

employing the same long/short strategies that its affiliates deployed in

their other portfolios.

If GWA’s investment strategy was successful, its annual trading

profits would increase the value of the securities in the underlying basket. But if GWA did not exercise its “option” on the basket securities

until maturity, the accumulated trading profits could escape taxation

for many years, and would ultimately be taxed, not at ordinary income

rates, but at more favorable long-term capital gain rates. And because

the basket securities were not held in a prime brokerage account titled

to GWA, regulatory limits on leverage would not apply.

[*13] VI.

Features of Standard Call Options

A call option is a contract that provides the buyer (optionee) the

right, but not the obligation, to purchase an asset from the seller (optionor) at a specified price, known as the “strike price.” The underlying

asset is typically stock, but it could be a bond, a commodity, a derivative,

or anything else of value. For simplicity, we will describe the features

of call options assuming that the underlying asset is corporate stock.

A call option enables the optionee to secure exposure to a stock’s

upside potential without requiring him to pay the full price of the stock.

The price paid by the optionee is called the “premium.” The premium

compensates the optionor for giving the optionee the opportunity to purchase the stock—i.e., to “call” it away from the optionor—if it closes

above the strike price before the option expires.

During the option period, the optionee has no ownership of (or

control over) the stock. The owner of the stock—typically the optionor,

except in the case of a “naked” call option—remains entitled to receive

all dividends (or other distributions) paid on the stock during the option

period. The owner of the stock likewise retains all other rights incident

to ownership of the stock, e.g., the right to vote the shares at annual

meetings and the right to sue the company in a shareholder derivative

action. For Federal income tax purposes, gain or loss is generally realized only when the call option is sold, is exercised, or expires worthless.

A standard call option provides the optionee with asymmetric exposure. The optionee participates in gains if the stock climbs above the

strike price. But if the stock declines in value, the optionee has no loss

exposure beyond the price he paid for the option. For this reason, the

optionee is said to enjoy “downside protection,” as compared with an investor who owns the stock outright. Conversely, the optionor bears “upside risk,” i.e., the risk that the stock will increase in value and be taken

away from him for less than it is then worth. In that event, the optionor

will keep the option premium, but he will experience an economic loss

versus the position he would have occupied if he had never written the

option and simply held the stock.

A call option is said to be “at the money” when the stock is trading

at the strike price, “in the money” when the stock is trading above the

strike price, and “out of the money” when the stock is trading below the

strike price. If the option is “out of the money” on the expiration date, it

[*14] expires worthless. If the option is “in the money” on the exercise

date, it may be “cash settled” or “physically settled.”

If the optionee were to choose physical settlement upon exercise

of a call option, he would pay the strike price and receive the optioned

shares, typically by debit/credit to his brokerage account. But publicly

traded options in financial markets are almost always “cash settled.” In

that event, the optionee receives cash equal to the value of the option on

the exercise date. That value normally equals the amount by which the

price of the stock on the exercise date exceeds the strike price, multiplied

by the number of optioned shares.

In U.S. financial markets, the optionee typically can exercise a

call option at any time up to the option’s expiration date. These contracts are called “American-style” options. A “European-style” option is

one that the optionee can exercise only on the expiration date specified

in the contract.

“Optionality” measures the degree of certainty that an optionee

will or will not exercise the option. An option has low optionality where

the likelihood of the optionee’s exercising it approaches 0% or 100%. A

call option so far in the money that it is virtually certain to be exercised—a so-called “deep-in-the-money” option—has optionality that approaches zero. The same is true for an option so far out of the money

that no rational investor would be likely to exercise it.

The price of a call option represents the premium the optionor

demands for granting the option. This premium reflects the risk to the

optionor that the stock will close above the strike price at expiration. In

a standard call option the premium is paid to the optionor at the outset

of the contract and is never refunded or returned to the optionee.

The premium that an optionor demands for granting an option is

the sum of its “intrinsic value” and its “time value.” An option’s “intrinsic value” is its current value assuming it were to expire immediately.

An option that is out of the money has zero intrinsic value—no rational

optionee would exercise such an option because he would lose money by

doing so.

An option’s “time value”—sometimes called its “extrinsic value”—

is essentially a measure of its optionality. The time value is greatest

when there is significant uncertainty as to whether the option will expire in or out of the money. An option typically loses time value as it

approaches its expiration date because the choice about whether to

[*15] exercise becomes increasingly clear. A deep-in-the-money call option will have time value close to zero because it is virtually assured that

the optionee will exercise it. In this circumstance, the option’s value

consists entirely of its intrinsic value.

The following example illustrates how a standard call option is

priced. Assume the optionor writes a call option on 100 shares of Corp. A

stock, currently trading at $103 per share. Assume that the strike price

is $100 and the option expires in 120 days.

This option has an intrinsic value of $300—its value if it were to

expire immediately ($3 × 100 shares). The optionor will of course demand additional premium to account for time value—the possibility that

the stock will close substantially above $100 per share during the ensuing 120 days, causing the optionee to exercise the option. Calculation of

the time value will depend on numerous factors, including the number

of days left before expiration, the price volatility of Corp. A stock, prevailing interest rates, etc. If the time value determined by buyers and

sellers in the marketplace is $350, the option’s total value (and thus its

premium) will be $650.

Greek-letter variables, called “the Greeks,” are commonly used to

express different components of risk in the options market. Risk informs

an option’s time value and thus the premium that an optionor should

demand. Rho (ρ) measures an option’s sensitivity to interest rates.

Theta (θ) measures sensitivity to changes in the time remaining until

expiration. “Vega” (not actually a Greek letter) measures an option’s

price sensitivity to expected volatility in the price of the underlying asset.

Delta (δ) measures an option’s sensitivity to changes in the price

of the underlying asset. Gamma (γ) is the mathematical derivative of

delta. Gamma measures an option’s sensitivity to changes in the rate of

change in the price of the underlying asset.

Delta is measured on a scale of 0 to 1. As the delta of an option

approaches 1, its value begins to change dollar for dollar with changes

in the value of the underlying asset. The pricing relationship between

a “delta-1” option and its underlying asset is thus said to be “linear.” An

option so far in the money that its time value is zero will have a delta

of 1 because it is virtually certain to be exercised.

Since the 1980s, sophisticated actors have fashioned derivative

products that modify some features of standard options. These products

[*16] are traded over the counter (OTC), rather than on centralized exchanges, and they are commonly called OTC derivatives. OTC derivatives include “exotic” options that alter certain features of standard options to modify the payoffs and risks to the parties.

One type of nonstandard option is the “knock-out barrier option.”

An option with a knockout feature will pay out to the optionee only if the

value of the underlying asset has not hit a specified price barrier during

the option’s life. If a “down-and-out” barrier option hits a barrier and

“knocks out,” it will never come back to life, even if the underlying asset

ultimately closes above the strike price. Some options with a knockout

feature provide that the optionor will make a cash payment to the optionee, generally called a “rebate,” if the option hits a barrier and

“knocks out” before the normal expiration date.

VII.

The RBC Transactions

Beginning in the 1990s or earlier, RBC offered customers a financial product that it styled as a “cash-settled, out-performance, index call

option.” In exchange for the payment of a “premium,” the buyer would

be entitled to receive, at the contract’s expiration, a cash payment that

would depend on the performance of an underlying basket of securities.

The cash that RBC would pay the customer at expiration was determined by the amount by which the “Reference Index” exceeded the

“strike price.” The “Reference Index” was defined as the difference in

the percentage changes of two price indices, multiplied by 100. “Index #1” was a basket of securities titled to an RBC affiliate but managed

and traded by an investment advisor selected by the customer. “Index

#2” was the S&P 500 Index.

In late 1997 GWA entered into negotiations with RBC about investing in this derivative product. Mr. Doucette was chiefly responsible

for negotiating on GWA’s behalf. RBC proposed that the amount of capital it would make available for investment in Index #1 would equal ten

times what GWA would be expected to pay in “premium.” In effect, RBC

offered GWA leverage of 10 to 1.

Mr. Doucette viewed the RBC product as a way of expanding

GWA’s investment in the same long/short strategies it was already pursuing, but with certain advantages. Whereas GWA faced margin restrictions on its prime brokerage investments, it would be eligible for

“ten times leverage” through the RBC product. The RBC product would

enable GWA to “lock in” the contractual terms for a defined period,

[*17] whereas prime brokers could change on short notice the terms on

which they extended financing. And although RBC offered only half the

leverage GWA enjoyed as an alternate specialist on the Philadelphia

Stock Exchange—10 to 1 versus 20 to 1—the RBC product offered access

to a wider array of securities.

On April 15, 1998, GWA and RBC entered into the first of six

transactions involving this derivative product. For every $10X of “option

premium” paid by GWA, RBC deposited $100X into a Merrill Lynch

prime brokerage account titled to an RBC affiliate. For the first transaction the parties agreed on a “premium” of $10.5 million, and RBC accordingly deposited $105 million into the prime brokerage account. The

value of the securities ultimately held in this account became “Index #1.”

The “strike price” for this putative call option was the negative of

the premium. Since the “premium” was $10.5 million, the strike price

was −$10.5 million. This structure ensured that 100% of the “premium”

would be returned to GWA upon termination or exercise of the “option.” 4

Because the “option premium” was returned to GWA, RBC’s compensation as the putative “optionor” consisted essentially of interest.

GWA was obligated to pay RBC a fee that was equivalent to interest on

the $105 million RBC had placed into the basket. And RBC could earn

interest on the $10.5 million “premium,” which resembled an ordinary

bank deposit (or contribution to an investment account). In economic

terms, the $10.5 million “bank deposit” served as collateral for a $105

million loan.

This first “option” had a five-year term, with a stated expiration

date of April 15, 2003. The product was described as a European-style

option, so that GWA supposedly could not exercise it before the expiration date. But an earlier termination could be triggered by an “extraordinary event.” One such trigger was a “cash event.” A cash event would

occur if the securities in the prime brokerage account were liquidated,

4 As a simplified example, assume that the value of Index #1, managed by

GWA’s advisor, had risen by 35%, while the S&P 500 Index had risen by 20%, as of the

expiration date. The “Reference Index” would thus equal $15 million, and this amount

would exceed the “strike price” by $25.5 million ($15 million minus negative $10.5 million). GWA would thus get its entire premium back and keep the $15 million surplus.

Conversely, assume that the value of Index #1 had risen by only 15%, while the S&P

500 Index had risen by 20%. The “Reference Index” in this instance would be negative

$5 million, resulting in a $5.5 million excess over the “strike price” (negative $5 million

minus negative $10.5 million). GWA would again get its entire premium back, less the

$5 million deficit.

[*18] leaving the account consisting solely of U.S. dollar cash equivalents.

GWA in effect selected itself to trade the securities in the prime

brokerage account. Mr. Doucette and two other members of GWA

formed Quaker Partners, LLC, to serve as investment advisor for Index #1. Quaker Partners had no employees, so it delegated to Weiss

Associates all rights to manage the account. Weiss Associates managed

the account using the same long/short strategies that GWA’s affiliates

deployed in their other portfolios. RBC was aware that the account was

being “managed by GWA.”

Quaker Partners could terminate the investment management

agreement after one year. Termination of the agreement would trigger

liquidation of the assets in the prime brokerage account, causing a “cash

event.” In effect, GWA thus could unilaterally terminate the five-year

“option” after one year.

Between May 1998 and March 2001 GWA and RBC executed five

more “call options.” The “premiums” paid by GWA ranged from $5.5

million to $20 million, and the cash deposited by RBC in the underlying

prime brokerage account concomitantly ranged from $55 million to $200

million. In each case the “strike price” for the “option” was the negative

of the “premium,” guaranteeing that the “premium” would be refunded

to GWA at expiration.

Two of the five contracts had the same stated expiration date as

the first contract (April 15, 2003). The other three had stated expiration

dates between April 2005 and March 2006. Although RBC was the nominal owner of the securities in the reference baskets underlying the “options,” GWA received settlement payments for class action lawsuits filed

against companies that issued those securities.

GWA discussed its investment in STFIs, including the RBC product, in a May 2001 private placement memorandum (PPM). The PPM

noted that GWA was “seeking to achieve a profit from [STFIs] by employing the same strategies as are currently employed in [its] Leveraged

Investment Strategy,” i.e., the long/short strategy. It noted that these

investments “may be made in a manner designed to lessen and/or defer

the taxation of income on such investments, or to otherwise tax advantage such investments.” But it warned investors that “[t]here is no

assurance that such position will be sustained” by the IRS and that investors could be liable for interest and penalties if these tax results could

[*19] not be achieved. GWA estimated that, as of 2000, its trading gains

in the portfolios underlying the RBC “options” had reached into the tens

of millions of dollars.

VIII. Deutsche Bank’s “Managed Account Product Structure”

In the late 1990s Deutsche Bank was building a prime brokerage

business designed to attract hedge fund customers. It accordingly developed financial products offering greater leverage than was available

through margin-restricted accounts. For example, customers could obtain up to 200 times leverage from Deutsche Bank through interest rate

swaps and repurchase agreements. Deutsche Bank’s Global Prime Finance division (GPF) managed the prime brokerage business and offered

standard prime brokerage services to customers.

Beginning in 1998 Deutsche Bank developed a Managed Account

Product Structure, or “MAPS,” which was marketed by GPF. It characterized MAPS as involving “barrier call options,” to which we refer as

Barrier Contracts. GPF marketed this product to hedge fund customers

as an “amortizing call option [that] provides delta-1 exposure to [an] underlying reference portfolio” of securities. A customer’s “delta-1 exposure” to the reference basket ensured that changes in the value of the

basket securities would be reflected dollar for dollar in the value of the

“option.”

The customer would be required to pay a “premium” when purchasing the “barrier call option.” The “premium” would purportedly be

priced according to the investment strategy used to manage the basket

securities, including the volatility and liquidity of the assets. In practice, the “premium” was almost always 10% of the “notional amount,”

i.e., the amount of cash Deutsch Bank made available to the customer

for investment in the securities basket.

The customer could choose the initial composition of the securities

portfolio, and Deutsche Bank’s London branch would purchase those securities and place them (at least notionally) into the basket. Although

the securities were titled to Deutsche Bank, the customer could appoint

the investment advisor, who would direct all trading activity in the account. At the termination of the contract, the customer would be entitled to receive a cash settlement amount corresponding to the gains that

had accumulated in the basket.

The Barrier Contract had a knockout feature such that, if the

value of the basket fell below a specified barrier, the contract would be

[*20] terminated and the securities would be liquidated. But Deutsche

Bank built in a fail-safe mechanism that kicked in before that point was

reached. If the value of the portfolio fell to a level that approached the

knockout barrier, Deutsche Bank could demand payment of an “additional premium.” The demand for “additional premium” resembled an

anticipatory margin call in a traditional prime brokerage account.

If the customer declined to supply additional premium, Deutsche

Bank could immediately terminate the contract and the securities would

be liquidated to cash. These features were designed to ensure that the

Barrier Contract would be terminated, and the underlying assets converted to cash, before investment losses in the reference basket exceeded

the “premium” paid by the customer. Deutsche Bank was thus insulated

from downside risk on the investment portfolio.

As with the RBC product, the “premium” paid for the “barrier call

option” did not constitute compensation to Deutsche Bank because the

premium was ultimately refunded to the customer. See infra pp. 25–26,

47–49. Deutsche Bank derived revenue from the Barrier Contracts in

three ways. First, it levied a trade fee, called a “ticket charge,” for each

trade executed in the reference basket. Second, it levied a financing fee

keyed to the amount of capital actively invested in the basket, minus the

stated premium. This fee was the economic equivalent of the interest

that a prime broker charges customers for a margin loan. 5 Third,

Deutsche Bank was free to earn interest on the “option premium” before

that sum was returned to the customer. As with the RBC product, the

“premium” thus resembled a bank deposit that served as collateral for a

margin loan.

IX.

GWA’s Negotiations with Deutsche Bank

Mr. Doucette commenced negotiations with Deutsche Bank about

MAPS in September 2002, roughly six months before GWA’s first three

RBC contracts were set to expire. He understood that MAPS, like the

RBC product, offered “all the benefits of prime brokerage with the benefits of tax deferral [and] long-term treatment.” But whereas RBC

charged its financing fee on 100% of the investment capital that it supplied through the prime brokerage account, Deutsche Bank charged its

financing fee only on the portion of the capital that was actively invested

in the reference basket. This fee structure better suited GWA, which

5 The financing fee, which GWA called the “leverage fee,” was not charged to

the customer on a periodic basis. Rather, it reduced the cash settlement amount to be

paid out to the customer at the end of the contract.

[*21] generally sought to keep 20% of the account’s value uninvested,

i.e., in the form of cash equivalents.

Deutsche Bank and GWA agreed that Quaker Partners would be

appointed as investment advisor for the reference basket of securities.

GWA provided a sample securities basket, ostensibly so that Deutsche

Bank could evaluate the basket risk and offer appropriate pricing.

Deutsche Bank promptly offered to fund the reference basket at ten

times the amount of GWA’s stated premium. This was the same 10 to 1

leverage that GWA enjoyed through its RBC investment. But in practical effect GWA would have access to double this leverage: For each basket security held in a long position, GWA could hedge the position by

selling the same security short. GWA thus regarded the Barrier Contracts as affording it leverage of up to 20 to 1.

On March 6, 2003, Deutsche Bank presented GWA with a “pricing

proposal” setting forth its fees for “Equity Prime Services,” which included prime brokerage, swaps, and MAPS. The pricing proposal indicated that Deutsche Bank would charge financing fees for the capital it

supplied in the Barrier Contracts at one of three interest rates, depending on the value of the basket securities. These financing fees were identical to the fees Deutsche Bank charged customers for leverage in its

prime brokerage accounts. The “ticket charge” for each trade executed

in the securities basket, $3, was also identical to the commission

Deutsche Bank charged for trades in a prime brokerage account.

Mr. Doucette approved the pricing proposal, and in early 2003

GWA’s executive committee authorized the firm’s participation in the

Barrier Contracts. Three of the RBC contracts expired on April 15, 2003.

GWA concurrently decided to terminate the other three RBC contracts,

whose stated expiration dates fell during 2005 and 2006.

On April 15, 2003, Deutsche Bank acquired via “cross trade” the

portfolio securities held in the securities baskets underlying the RBC

contracts set to expire on that date. 6 The securities thus transferred to

Deutsche Bank were used to populate the reference basket for the first

Barrier Contract. GWA then instigated a “cash event” in the other three

RBC contracts, causing their expiration dates to be accelerated to April

23, 2003. On its Form 1065, U.S. Return of Partnership Income, for

6 A cross trade is the practice of matching buy and sell orders for the same

instrument without engaging in an open-market transaction. Its use permits an investor to avoid certain costs that open-market transactions entail.

[*22] 2003, GWA reported $59,439,344 in long-term capital gain stemming from termination of the six RBC contracts.

On April 1, 2003, GWA and OGI each signed a Margin Lending,

Securities Lending, Custody Account, and Sweep Account Agreement

with Deutsche Bank. On the same day each signed a Prime Broker Margin Account Agreement with Deutsche Bank Securities Inc., the brokerdealer that cleared and settled transactions on Deutsche Bank’s behalf

for both MAPS and prime brokerage clients.

GWA entered into its first Barrier Contract with Deutsche Bank

on April 15, 2003 (Barrier Contract #1). It was described as a Europeanstyle “barrier call option” with a “notional amount” of $500 million and

a stated premium of $50 million. 7 The amount to be paid to GWA at the

end of the contract (cash settlement amount) was to be calculated by

reference to the performance of a basket of U.S. equities. At commencement the securities basket held 919 positions, divided into long and

short stock positions, as well as some positions in bonds and derivatives.

All of these positions were transferred via cross trade from the RBC contracts that had expired on April 15, 2003.

Deutsche Bank held title to the securities in the reference basket.

But GWA or one of its affiliates could (and did on occasion) instruct

Deutsche Bank as to how voting rights associated with the shares should

be exercised. (Deutsche Bank was not required to follow this advice.)

Deutsche Bank was obligated to prepare periodic reports showing the

performance of the securities in the reference basket and indicating

what the cash settlement amount would be if the Barrier Contract were

terminated on that date. These reports resembled the monthly statements that Deutsche Bank delivered to its prime brokerage customers.

As was true for the RBC “options,” the customer would receive settlement payments from class action and shareholder-derivative lawsuits

filed on behalf of companies whose stock was held in the reference baskets. See supra p. 18. In fact, on its 2009 return GWA reported $127,925

in class action settlement proceeds—all received in connection with basket securities—as “other long term capital gains.”

7 In at least some of its promotional materials, Deutsche Bank advertised the

barrier call options as “American-style,” suggesting that a MAPS customer could exercise the option at any time before its expiration date.

[*23] X.

The Investment Advisory Agreement

Although Deutsche Bank was entitled to choose the investment

advisor for the reference basket, it agreed that Quaker Partners, a GWA

affiliate, would be selected for this role. On April 15, 2003, GWA and

Deutsche Bank executed an Investment Advisory Agreement (IAA #1)

providing that Quaker Partners would receive, for providing advisory

services, a quarterly fee equal to 0.25% of the stated premium. Because

the stated premium for Barrier Contract #1 was $50 million, the “advisory fee” was $125,000 per quarter. 8

GWA and Deutsche Bank negotiated the terms of IAA #1, which

supplied guidelines and restrictions governing trading in the reference

basket. The guidelines stated that trading was to follow a “long/short

statistical arbitrage” strategy and specified general parameters regarding the maximum size of long/short positions and the acceptable classes

of investments. The guidelines concerning the size of positions ensured

that there was sufficient liquidity to facilitate unwinding the positions

if necessary.

Quaker Partners was not required to seek permission from

Deutsche Bank before executing any trade, and nothing prevented it

from liquidating the basket entirely to cash. However, Quaker Partners

was precluded from trading any securities appearing on a “Trade Restricted List,” which Deutsche Bank updated daily. The purpose of this

restriction was to ensure that Deutsche Bank did not violate any conflict-of-interest rules. IAA #1 also prohibited Quaker Partners from

trading securities designated “hot issues.” 9

IAA #1 obligated Quaker Partners, as investment advisor, to take

remedial action if a restricted security was inadvertently included in the

reference basket. In that event, Quaker Partners could dispose of the

problematic security through an ordinary market transaction or transfer it to another account at Deutsche Bank that was customer owned.

8 Barrier Contract #1 stated that GWA “shall not contact directly the investment advisor regarding the terms or subject matter of th[e] transaction.” But this

prohibition was meaningless because Quaker Partners had no employees and had delegated all of its investment management responsibilities to Weiss Associates. See supra p.18.

9 Under SEC rules, a “hot issue” is a stock issued in an initial public offering

(IPO) whose market price rises 5% or more above the IPO price within the first five

minutes of trading. “Hot issues” are regarded as risky investments.

[*24] OGI’s prime brokerage account at Deutsche Bank was designated

the other account.

Under IAA #1, Deutsche Bank could terminate the advisory

agreement for any reason, or for no reason, upon written notice to

Quaker Partners. If Deutsche Bank terminated the agreement within

12 months of its effective date and the performance of the MAPS account

was positive, then Deutsche Bank was required to pay Quaker Partners

$200,000.

Shortly after executing IAA #1, Quaker Partners subcontracted

to Weiss Associates its role as investment advisor. The parties thereby

agreed that Weiss Associates would receive 95% of Quaker Partners’ fee

for providing investment advisory services in connection with Barrier

Contract #1.

XI.

Term of Barrier Contract #1

Barrier Contract #1 had a term of 12+ years, running from April

15, 2003, to April 30, 2015. Deutsche Bank could accelerate the expiration date to 3, 6, or 9 years preceding the stated expiration date, provided that it gave GWA 30 days’ notice of its decision to do so.

GWA had no explicit right to terminate the contract early. But it

had the de facto ability to do so by causing Quaker Partners, the investment advisor, to manufacture a “cash event.” Quaker Partners could

generate a “cash event” by selling all securities in the reference basket,

reducing it to cash. Or Quaker Partners could terminate IAA #1 (after

giving Deutsche Bank sufficient notice), which would require the basket

securities to be liquidated and converted to cash “in a prompt and orderly manner.”

Upon occurrence of a “cash event,” Deutsche Bank had the right

to immediately accelerate the option termination date. Deutsche Bank

would have a strong economic incentive to exercise this right because

the cash in the reference basket would begin accruing interest at the

Federal Funds Rate plus 5%. Marcus Peckman, GWA’s chief financial

officer (CFO), acknowledged that this rate would be “punitive” for a financial institution like Deutsche Bank. Moreover, because none of

Deutsche Bank’s capital would be actively invested in the reference basket following a cash event, Deutsche Bank would be entitled to receive

no further financing fees. For both reasons, Mr. Peckman viewed GWA’s

ability to generate a cash event as a de facto “out provision” that it could

employ to terminate a Barrier Contract at any time of its choosing.

[*25] Barrier Contract #1 could also terminate early if the basket value

reached a knockout barrier, defined as an “early expiration event.” Such

an event would occur if the Net Asset Value (NAV) Index Level, set at

100 at the outset of the contract, declined to 94 (the “Expiration Price”).

A decline of that magnitude would translate to a 6% reduction in the

value of the reference basket.

If the NAV Index Level declined to 97, Deutsche Bank was required to provide GWA an “early expiration notice.” GWA would then

have four hours to notify Deutsche Bank, via a “buyer election notice,”

that it intended to continue with Barrier Contract #1. If so, GWA was

required to pay Deutsche Bank an “additional premium amount” of $15

million, i.e., 3% of the “notional amount.” The additional premium was

due by 4 p.m. on the next business day following delivery of the “early

expiration notice.” If GWA declined to pay additional premium, the “option” would terminate and the basket securities would be liquidated.

Barrier Contract #1 would terminate automatically in any event if the

NAV Index Level reached 94.

XII.

Payout on Barrier Contract #1

Barrier Contract #1 stated that GWA was to pay a “premium” of

$50 million for the “option.” The stated premium consisted of two parts:

a “fixed premium” of $44 million, and an “amortizable premium” of

$6 million. The fixed premium was payable to Deutsche Bank on the

third business day following commencement of the contract. The amortizable premium accrued as a daily amount and was spread over the life

of the contract.

Upon expiration of the “option” Deutsche Bank was required to

pay GWA a cash settlement amount. This was expressed by a complex

formula. In essence, GWA was entitled to receive upon expiration an

amount equal to the cumulative performance of the basket securities

(“Basket Base Performance”) plus a “Premium Settlement Amount.”

The “Basket Base Performance” was the amount by which “Basket Gains and Income” exceeded “Basket Losses and Expenses.” “Basket

Gains and Income” included realized and unrealized gains in the underlying securities basket, plus “dividends in respect of the Basket Long

Positions.” “Basket Losses” included realized and unrealized losses in

the underlying securities basket. “Basket Expenses” included “dividends in respect of the Basket Short Positions,” investment advisory fees

paid to Quaker Partners, “ticket charges” paid to Deutsche Bank, and

[*26] the financing or “leverage” fees paid to Deutsche Bank for use of

its capital.

The “Premium Settlement Amount” for Barrier Contract #1

equaled the stated premium ($50 million) minus the “total amortized

premium.” The latter amount was calculated as the sum of the “amortized daily premium” charged for each calendar day of the contract. Barrier Contract #1 ran from April 15, 2003, through April 30, 2015, i.e., for

4,398 days. Since the “amortizable premium” was $6 million, the “amortized daily premium” was $1,364 ($6 million ÷ 4,398). The “total amortized premium” would thus be exactly $6 million if the contract expired

as scheduled, but it would be less than $6 million if the contract terminated early.

The calculation described in the preceding paragraph suggests

that GWA would be refunded only a portion of the $50 million stated

premium because of the downward effect of the “total amortized premium.” But the “total amortized premium” would also be refunded, albeit in a different manner, i.e., via calculation of the Basket Base Performance.

In calculating the cash settlement amount, the leverage fee paid

to Deutsche Bank reduced the Basket Base Performance. But the leverage fee itself was reduced by $1,364 for every day that the amount of

capital invested in basket securities exceeded the stated premium ($50

million). The amount of Deutsche Bank capital that could be invested

in basket securities could be as high as $500 million, and the amount so

invested invariably exceeded $50 million by a very healthy margin.10

For every day that a Barrier Contract was in place, therefore, the leverage fee was reduced by $1,364.

In short, the Basket Base Performance would be adjusted upward

via reduction of the leverage fee at a rate of $1,364 per day. This upward

adjustment would precisely offset the downward effect of the “total

amortized premium,” which was also calculated at a rate of $1,364 per

day. The record disclosed no reason for reducing the leverage fee by

$1,364 per day, other than to create this offset. These neutralizing adjustments ensured that GWA would receive, upon exercise of the

10 As noted supra pp. 20–21, GWA typically sought to keep 20% of the reference

basket in cash, so it appears that up to $400 million would usually be actively invested.

[*27] “option,” 100% of the accumulated net gains in the reference basket plus 100% of the $50 million stated premium. 11

If Barrier Contract #1 were terminated by an “early expiration

event,” GWA would be refunded at least some portion of its “premium.”

If the NAV Index Level hit 97, representing a 3% decline in the value of

the reference basket, and if GWA declined to pay additional premium,

Deutsche Bank would begin “orderly liquidation” of the basket securities. Unless Deutsche Bank was unable to liquidate the securities before

the portfolio had declined by another 7%—an extremely unlikely scenario, given that most positions in the reference basket were hedged

long/short positions—GWA would be refunded up to 70% of its “premium,” or $35 million. The premium refund would vanish only if the

NAV Index Level fell to 90 by the time the portfolio had been fully liquidated. In that event, the Basket Base Performance would be negative

$50 million, exactly offsetting the $50 million “premium.”

XIII. Trading and Management of the Securities Basket

Weiss Associates, pursuant to delegation from Quaker Partners,

directed trading in the reference basket. It pursued trading strategies

that precisely mirrored the long/short investment strategies that GWA

and its affiliates deployed in their other portfolios.

Each night Weiss Associates would send Deutsche Bank trade

files through an electronic file transfer protocol. These trade files would

be entered directly into Deutsche Bank’s order management system for

booking and execution on the following business day. On an average

trading day, Weiss Associates initiated trades of 268 unique securities

in the reference basket. During 2005 it initiated 89,075 trades involving

more than two billion units of stock.

Weiss Associates occasionally requested the trade of a security

that appeared on the “Trade Restricted List.” When this occurred,

Deutsche Bank’s order management system automatically redirected

that trade to OGI’s prime brokerage account. As of April 2006 OGI’s

trading activity primarily involved securities that could not be traded in

the Barrier Contract reference baskets.

11 In the event GWA had paid an “additional premium,” see supra pp. 25–26,

that “additional premium” would also be refunded 100% through the “Premium Settlement Amount.”

[*28] Weiss Associates was responsible for identifying violations of the

investment guidelines and so informing Deutsche Bank. On several occasions, however, Deutsche Bank was the first to discover the violation

and urged Weiss Associates to remediate it. The urgency with which

Weiss Associates did so varied.

XIV. Related Agreements

An International Swaps and Derivatives Association (ISDA)

agreement is typically used by a derivatives dealer and its counterparty

before executing a derivatives trade. GWA and RBC had signed an

ISDA agreement in 1998 before executing the RBC “options.” Deutsche

Bank required that all derivatives customers sign ISDA agreements,

and that a parent and its subsidiary sign separate ISDA agreements

even if both were existing clients. OGI and Deutsche Bank executed an

ISDA agreement in 1998 and amended it in 2003. But the record contains no evidence that GWA ever executed an ISDA agreement with

Deutsche Bank, notwithstanding their shared view that MAPS was a

type of derivative product.

On April 18, 2003, GWA and OGI entered into a Master Netting

Agreement (MNA) with Deutsche Bank. An MNA allows a customer to

use positive equity in one account as collateral to support borrowing in

another account. Such agreements typically cover accounts that have

the same beneficial owner. They mitigate risk for the investment firm

by bringing multiple entities under a single agreement, so that the firm

has recourse against one entity for the liabilities of the other. On June

16, 2003, the MNA was amended so that it also applied to Barrier Contract #2.

As initially drafted, the MNA governed three agreements between GWA and Deutsche Bank (including Barrier Contract #1) and five

agreements between OGI and Deutsche Bank (including their prime

brokerage contract). The MNA provided that all of these agreements

constituted “a single business and contractual relationship among the

parties.” This agreement permitted (for example) the netting of

amounts that OGI owed Deutsche Bank (such as interest that had accrued on the margin loan in OGI’s prime brokerage account) against

amounts that Deutsche Bank owed GWA (such as a Barrier Contract’s

cash settlement amount).

The MNA also permitted cross-collateralization between the

MAPS account and the accounts that other GWA affiliates held at

[*29] Deutsche Bank. GWA could thus pledge the equity value in a Barrier Contract reference basket as collateral for the margin loan that

Deutsche Bank extended to OGI through the latter’s prime brokerage

account. OGI often drew on this line of credit, then lent the proceeds

back to GWA. In this and other ways GWA had de facto access to the

cash value of the barrier “option” at any time of its choosing.

As Mr. Doucette acknowledged, “[h]istorically we have been able

to fund the operating expenses of our business by borrowing against the

excess equity value of the [barrier] option.” The operating expenses thus

funded included payroll, rent, and employee bonuses. GWA used borrowed funds—all collateralized by the equity value in the Barrier Contracts—to acquire positions in OGI’s prime brokerage account that could

not be maintained in a Barrier Contract reference basket without violating investment guidelines. GWA also used OGI-borrowed funds to

acquire positions at other financial institutions, which had the effect of

reducing GWA’s counterparty exposure to Deutsche Bank.

GWA provided Deutsche Bank with a guaranty, dated April 15,

2003, by which GWA guaranteed repayment of all of OGI’s liabilities

and obligations to Deutsche Bank. GWA thus assumed secondary liability for any deficits in the line of credit that Deutsche Bank extended

to OGI in the latter’s prime brokerage account.

XV.

Barrier Contract #2

In May 2003 GWA received the proceeds from its termination of

the final three RBC “options.” See supra p. 21. Mr. Doucette approached

Deutsche Bank about investing these assets through MAPS. Deutsche

Bank presented Mr. Doucette with several possible scenarios for doing

this.

One scenario involved terminating Barrier Contract #1 and striking a new “option” using the combined proceeds from that contract and

the final three RBC “options.” But GWA was advised that termination

of Barrier Contract #1 in May 2003—one month after the “option” was

entered into—would trigger recognition of capital gain taxable at the

short-term rate (35%) instead of the long-term rate (15%) applicable to

assets held longer than one year. That outcome was not appealing to

GWA.

Instead, GWA agreed to purchase a second “option” whose performance would be tied to trading activity in the same reference basket

that underlay Barrier Contract #1. On May 22, 2003, Deutsche Bank

[*30] and GWA entered into Barrier Contract #2 on substantially the

same terms as Barrier Contract #1. The “notional amount” was again

$500 million, but the “premium” was revised to $52.8 million. This revised premium roughly equaled the cash that became available to GWA

following termination of the final three RBC “options.” The parties concurrently amended IAA #1 to provide that Quaker Partners would receive a quarterly investment advisory fee of $257,000. That fee equaled

0.25% of the aggregate “premium” for Barrier Contracts #1 and #2, or

$102.8 million.

XVI. Weiss Multi-Strategy Advisors

By the mid-2000s GWA’s investment of “inside money” through

MAPS had proven lucrative. In 2005 GWA launched Weiss Multi-Strategy Partners, LLC (WMSP), as a hedge fund dedicated to investing outside money. This hedge fund employed the same long/short strategies

used in the reference baskets underlying the Barrier Contracts.

GWA decided that there should be a single entity to serve as investment advisor for its “inside money” and “outside money” portfolios.

On May 9, 2005, Weiss Multi-Strategy Advisors, LLC (WMSA), was

formed for this purpose. WMSA provided advisory services for GWA’s

MAPS accounts, OGI’s prime brokerage account at Deutsche Bank, and

the “outside money” accounts held through WMSP. WMSA pooled the

capital from these sources, deploying its investment strategies across

what was essentially a single aggregated fund.

From time to time WMSA issued “due diligence questionnaires”

to provide current and prospective investors with information about its

products. In one of these documents WMSA stated that “[GWA’s] principals have generally not invested any capital in [WMSP]. For tax purposes, [GWA’s] principals . . . invest their capital in a separate legal

structure [i.e., the Barrier Contracts] which is managed pari passu to

[WMSP].”

GWA held a 99.9% ownership interest in WMSA. The remaining

0.1% was owned by Mr. Weiss directly. From 2006 through 2010 Mr.

Weiss served as chairman and chief executive officer (CEO) of WMSA,

and Mr. Doucette served as its president, COO, and head of risk management. Weiss Associates gradually transferred its operations, including its investment advisory activities, to WMSA. By late 2006 Weiss

Associates had become a shell.

[*31] On January 1, 2006, Quaker Partners redelegated to WMSA the

investment advisory services that Weiss Associates had previously performed for the MAPS reference baskets. The agreement contained

roughly the same terms as the prior agreement between Quaker Partners and Weiss Associates. WMSA was the investment advisor for all

GWA-affiliated accounts, including the Barrier Contracts, from 2006

through 2010.

Upon receipt of its quarterly advisory fee, Quaker Partners would

transfer 95% of that sum to WMSA. The remaining 5% was distributed

to Quaker Partners’ members—Mr. Doucette and two other employees

of WMSA. But when these individuals received a distribution from

Quaker Partners, their WMSA salaries were reduced by the amount of

the distribution. In effect, therefore, GWA and Mr. Weiss—who together owned 100% of WMSA—received (directly or indirectly) all of the

investment advisory fees that Deutsche Bank paid in connection with

the Barrier Contracts.

Although 100% of the advisory fees eventually flowed up to GWA

and Mr. Weiss, GWA returned those sums to Deutsche Bank at the expiration of a Barrier Contract. The advisory fees were included in “Basket Losses and Expenses,” which were subtracted from “Basket Gains

and Income” to determine the payout on the “option.” See supra pp. 25–

26. Because this reduction to the cash settlement amount offset GWA’s

advisory fees virtually dollar for dollar, those advisory fees had no economic significance.

Mr. Weiss managed the WMSA investment teams, which typically consisted of a portfolio manager, a trader, and quantitative analysts. Each team was responsible for managing one of the investment

strategies that WMSA deployed. The “allocation committee,” chaired by

Mr. Weiss, decided what proportion of the total funds under management would be allocated to each “strategy.”

From 2006 through 2010, each team deployed its particular strategy across all GWA-affiliated accounts, including the Barrier Contracts,

“outside money,” and OGI prime brokerage. WMSA’s traders did not

know the account or fund to which their trades would be settled. Rather,

once a trade had been executed, a computer-based accounting system

allocated the trade pari passu (i.e., proportionally) across all of the

funds.

[*32] XVII. Termination of Barrier Contract #2 and Execution of Barrier Contracts #3 Through #6

GWA and Deutsche Bank agreed that Barrier Contract #2 would

be terminated in December 2005. On December 21, 2005, Deutsche

Bank issued GWA a letter asserting that a “cash event” had occurred

and that Deutsche Bank was accelerating the expiration date of the “option” to that day. GWA received $130,569,181 in proceeds from the termination of Barrier Contract #2 and reported $76,907,731 in long-term

capital gain on its Form 1065 for 2005.

In fact, the securities in the reference basket underlying Barrier

Contract #2 had not been liquidated as of December 21, 2005. And no

“cash event,” as defined in the contract, had occurred as of that date.

Having noticed this problem, GWA in February 2006 requested from

Deutsche Bank a report showing that a “cash event” had occurred on the

desired date. GWA noted that, “in order for us to terminate the option,

the account has to be all cash.” GWA accordingly requested “[f]or tax

purposes . . . a report for Option 2 [that] shows only a cash balance” and

“all positions . . . [having been] liquidated prior to the exercise of the

option” on December 21, 2005.

On December 21, 2005, the same day Deutsche Bank terminated

Barrier Contract #2, GWA and Deutsche Bank entered into four new

“options” (Barrier Contracts #3 through #6). Barrier Contracts #3 and

#5 had “notional amounts” of $184 million and “premiums” of $18.4 million; Barrier Contracts #4 and #6 had “notional amounts” of $276 million

and “premiums” of $27.6 million. Each “option” had a 12-year term, with

a stated expiration date of December 21, 2017.

That same day Deutsche Bank and Quaker Partners entered into

a new investment advisory agreement (IAA #2) for these four contracts.

It resembled IAA #1, except that it did not limit trading to U.S. equities.

Rather, the reference baskets were permitted to include foreign equities,

bonds, derivatives, futures contracts, and other securities. The quarterly advisory fee was $230,000, i.e., the same 0.25% rate but applied

against the $92 million aggregate stated premium for Barrier Contracts

#3 through #6 ([$18.4 million × 2] + [$27.6 million × 2] = $92 million).

GWA updated its May 2001 PPM in an addendum dated April 20,

2006. The addendum noted that GWA had total capital of $149,558,658,

and “[s]ubstantially all of [these] assets” were “devoted to STFIs, in particular the barrier options.” It further stated:

[*33] [GWA] expects that it will not report gain or loss from its

investment in the barrier options until such options are exercised or terminated and that gain or loss will be treated

as gain or loss from the sale or exchange of a capital asset.

Nevertheless, the Company is unaware of any case law,

regulations or rulings of the [IRS] dealing with financial

instruments similar to the barrier options purchased by

the Company. There is a risk that the [IRS] or the courts

could conclude that some other less favorable tax treatment is appropriate for [GWA’s] barrier options.

GWA incorporated this same statement into three more PPM addenda that it issued between June 2007 and July 2008.

XVIII. Cross Trading and Position Journaling

GWA and Deutsche Bank regularly used “cross trading” to move

securities between the Barrier Contract reference baskets and OGI’s

prime brokerage account. Because cross trades do not take place on the

open market, discrepancies between the “bid” and “ask” prices are eliminated, and ticket charges and commissions do not apply. See supra pp.

21–22 & note 6.

Cross trading was beneficial to GWA because it facilitated the

speedy extraction of gains from its Barrier Contract investments. Without the use of cross trading, securities in the reference basket would

need to be liquidated, and those transactions settled, before GWA could

access the cash. By cross trading basket securities to OGI, GWA could

realize a return on its investment without relinquishing control of the

underlying securities and without causing market disruptions through

open-market transactions.

GWA and Deutsche Bank also used a technique called “position

journaling,” or “position rolling,” beginning in 2006 or earlier. Position

journaling refers to the movement of a securities position via book entry

between two separate accounts that have the same legal owner. Like

cross trading, position journaling avoids the need to execute an openmarket transaction. Deutsche Bank used position journaling to transfer

securities from a MAPS basket to OGI’s prime brokerage account, even

though the accounts had different legal owners.

[*34] XIX. Replacement of Barrier Contracts #3 Through #6 by Barrier Contracts #7 Through #10

In December 2006 GWA wished to extract cash from Barrier Contracts #3 through #6 without causing the securities in the associated

reference baskets to be liquidated. GWA hoped to accomplish these objectives by use of “position journaling.” If the securities positions associated with those four contracts could be “journaled” into separate accounts tied to four new contracts, no investment positions would need to

be changed.

GWA had no unilateral right to “terminate” Barrier Contracts #3

through #6. Nevertheless, on December 11, 2006, GWA notified

Deutsche Bank of its intention to “exercise its rights with Deutsche

Bank to terminate Options 3, 4, 5 & 6.” On the following day, GWA

entered into four new “options” with Deutsche Bank (Barrier Contracts

#7 through #10).

The terms of the four new contracts were substantially identical

to the terms of the contracts they replaced, including the aggregate “premium” ($92 million for all four “options”). The portfolio positions in the

securities baskets associated with Barrier Contracts #3 through #6 were

replicated in new accounts associated with Barrier Contracts #7 through

#10. Deutsche Bank “journaled” the positions in the old accounts to the

new accounts on December 12, 2006. Three days later GWA directed

Deutsche Bank to wire $92 million from OGI’s prime brokerage account

“[t]o reflect payment of option premiums.” Deutsche Bank agreed to do

this even though the debit balance in OGI’s account then exceeded $200

million. Quaker Partners and Deutsche Bank executed a new investment advisory agreement (IAA #3) to cover trading in the four new contracts.

On December 22, 2006, Deutsche Bank issued a letter to GWA

asserting that a “cash event” had occurred with respect to Barrier Contracts #3 through #6 and that it was terminating them immediately.

GWA treated the four “options” as terminating on December 22, 2006—

exactly one year and one day after the “options” had been entered into.

On December 28, 2006, Deutsche Bank deposited $124,191,610

into OGI’s prime brokerage account. Of this deposit, $92 million was

designated as replacing the $92 million that OGI had transferred two

weeks earlier “[t]o reflect payment of option premiums.” On its Form

1065 for 2006, GWA reported gross proceeds of $124,191,610 from

[*35] disposition of the “options” and an aggregate cost basis of

$92,036,098. It thus reported $32,155,512 as long-term capital gain

from the termination of Barrier Contracts #3 through #6.

XX.

Financial Turbulence and “New MAPS”

In August 2007, in an event known as the “Quant Quake,” several

hedge funds engaged in a massive selloff that shook financial markets.

During the financial crisis of 2008–2009, stock market prices declined

by more than 50%. These events caused banks and investment firms to

engage in deleveraging and other risk-averse behaviors.

During this period Deutsche Bank took steps to mitigate its risk

exposure. In December 2008 Deutsche Bank reduced the “gross leverage” that was available for investment in the MAPS reference basket—

i.e., the total “long-side” plus “short-side” leverage—from 20 to 12 times

the stated premium. It made this change unilaterally, even though the

Barrier Contracts’ terms were supposedly “locked in” for the duration of

the agreement. 12

Deutsche Bank also became very concerned about the debt that

GWA was running up in OGI’s prime brokerage account. In late November 2008 Deutsche Bank officers noted that OGI’s margin debt exceeded

$400 million and that the “MAPS/OGI cross-collateralization arrangement is very low on equity.” Deutsche Bank informed GWA that “the

cross-collateralization has to end.” Believing that GWA would nevertheless “try to hang on to the options,” Deutsche Bank concluded that “we

can/should force early exercise of the oldest option [Barrier Contract #1]

in Apr 09.” (Deutsche Bank in fact terminated Barrier Contract #1 on

April 30, 2009, facilitating the transfer of $380 million into OGI’s account. See infra p. 37.)

During 2007 and 2008 Deutsche Bank’s chief risk officer and general counsel became concerned that its arrangements with GWA exposed the bank to excessive financial and legal risks. Deutsche Bank

12 It appears that GWA generally did not need more than 12 times gross leverage in the reference basket. In mid-March 2009 a GPF employee stated in an email to

GPF’s head of risk that GWA’s investment strategy “ha[d] a normal range of 3×–5.5×”

leverage. Later that month the same employee sent an email to Mr. Doucette noting

that, since 2005–2006, the account had not required more than 5 times leverage per

side (10 times on a gross basis), “even with a buffer.” Mr. Doucette likewise testified

that the accounts managed by WMSA generally had “four to five times [leverage] per

side.” Dr. Montgomery determined that the leverage ratio in the reference basket as

of May 2003 was approximately 10.7 times on a gross basis.

[*36] accordingly approached GWA about entering into a new version of

MAPS, which would retain a similar structure but exhibit features more

akin to those of standard call options. Under “New MAPS” Deutsche

Bank proposed that:

● The Barrier Contract would have a term of 13 to 18 months, as

opposed to 12 years under the existing contracts. “When pushed,” Mr.

Doucette noted, Deutsche Bank “said they might be able to do [a] 24

months term.”

● The “knockout barrier” would occur at an NAV Index Level of

97.7, as opposed to 94 under the existing contracts.

● It would no longer be possible for GWA to avert a knockout by

paying an “additional premium.”

● If the option did knock out, GWA would no longer be entitled to

a refund of the “amortizable premium.” The “amortizable premium,”

moreover, could be as high as 20% of the stated premium (as opposed to

12% under the existing Barrier Contracts).

● The leverage fee would be calculated on the full “notional value”

of the Barrier Contract, rather than being imposed only on the amount

of capital actively invested in the reference basket.

● GWA would no longer be permitted to engage in cross trading

between the MAPS reference basket and OGI’s prime brokerage account.

● GWA would no longer be permitted to cause OGI to borrow

against the “excess equity” in the MAPS reference basket. In other

words, GWA could no longer pledge the equity value in a Barrier Contract reference basket as collateral for the margin loan that Deutsche

Bank extended to OGI through the latter’s prime brokerage account.

Deutsche Bank made clear that this change “is not negotiable.”

● GWA would no longer be able to manufacture early termination

of a Barrier Contract (e.g., by generating a “cash event” or ending an

investment advisory agreement). Rather, as with a true European style

option, GWA would be able to exercise the option only on the stated expiration date.

Deutsche Bank later proposed a further modification to address

what it called “optionality value.” Under this proposal, Deutsche Bank

[*37] would retain a portion of the stated premium—perhaps as much

as 20%—if a Barrier Contract “terminated in a situation in which the

purchaser of a ‘true’ option would not expect to receive back its premium.” Deutsche Bank’s counsel believed that this modification would

require the customer to bear a degree of risk that better aligned with

the risk incident to “‘true’ option[s].” In a February 25, 2009, email to

Deutsche Bank, Mr. Doucette called several of the proposed changes “potential deal breakers.”

Negotiations about the terms of New MAPS continued through

the end of 2010. GWA proposed that New MAPS include a “tax out”

provision, whereby GWA could terminate a barrier contract if there was

a “change in the tax law” that “adversely impacts the . . . tax treatment

of [MAPS] to [GWA].” Deutsche Bank did not oppose that idea, but it

insisted on a further agreement that, if such a change occurred, GWA

would not report a New MAPS barrier contract as “an option, forward

contract, or other open transaction.” Deutsche Bank also insisted that

“change in the tax law” be defined to exclude GWA’s “realization that

[it] has misconstrued current law.” The record of this case contains no

evidence that a final agreement regarding “New MAPS” was ever

reached.

XXI. Termination of Barrier Contract #1

In April 2009 Deutsche Bank accelerated termination of Barrier

Contract #1 to April 30, 2009, one of the “early termination dates” permitted in that contract. Deutsche Bank insisted that the cash settlement for Barrier Contract #1 be used to reduce the massive deficit in

OGI’s prime brokerage account (caused in part by new margin requirements Deutsche Bank had imposed in December 2008). But Deutsche

Bank agreed that the payment would first be made to GWA so as “to

show the proper transaction trail.”

Barrier Contract #1 was terminated effective April 30, 2009, with

a cash settlement amount of $387,324,387. On May 5, 2009, that sum

was wired to GWA’s prime brokerage account at Deutsche Bank, and

$380 million was then journaled to OGI’s prime brokerage account at

Deutsche Bank. On its Form 1065 for 2009, filed August 30, 2010, GWA

reported gross proceeds of $387,324,387 from disposition of Barrier Contract #1 and an adjusted basis of $53,182,269. It thus reported

$334,142,118 as long-term capital gain from the termination of that “option.”

[*38] XXII. Termination of Barrier Contracts #7 Through #10

In August 2009 GWA reiterated its interest in terminating the

four remaining Barrier Contracts, noting that it “suspect[ed] a change

[in] tax laws and want[ed] to crystallize [its] gains.” GWA feared that

MAPS may “no longer [be] a viable investing instrument due to changes

in Washington” that would eliminate the “long term tax advantages” associated with the Barrier Contracts. GWA also noted the parties’ continuing impasse over the terms of “New MAPS” and GWA’s desire to

reduce its counterparty exposure to Deutsche Bank.

In October 2009 Mr. Kleinman emailed Mr. Doucette and Robert

Gendreau (GWA’s tax director) expressing his concern about proposals,

then pending in Congress, regarding “codification of the ‘economic substance doctrine.’” Mr. Kleinman stated his view that codification “could

have serious implications with respect to the [Deutsche Bank] option

transaction.” He noted that, “[w]hile this proposal will not completely

eliminate the benefit of the option structure, nevertheless, this will be a

powerful tool for the IRS.” In reply Mr. Gendreau “agreed that the codification of the ‘economic substance doctrine’ would be a powerful tool

for the IRS.” Mr. Doucette forwarded these messages to Deutsche Bank

with an inquiry about “the risk of passage and its affects [sic] on the

MAPS product.”

GWA wished to unwind the last four Barrier Contracts by use of

cross trading or position journaling, whereby the securities positions

would be transferred to OGI’s prime brokerage account (or another account under GWA’s control). But Deutsche Bank would not agree to use

these techniques to transfer the positions unless the positions were

transferred to a “New MAPS” account. Unwilling to accept that condition, GWA acquiesced in liquidation of the securities in the reference

baskets. But in the hope of ensuring an “orderly liquidation” and minimizing any possible market disruption, GWA requested that the securities baskets underlying Barrier Contracts #7, #8, and #10 be liquidated

first.

In letters to GWA dated May 14, 2010, Deutsche Bank stated that

“cash events” had occurred in Barrier Contracts #7, #8, and #10 and that

it was accelerating the “option termination dates” accordingly. (In fact,

no “cash event” had yet occurred because the reference baskets were still

fully populated with securities.) On May 17, 2010, WMSA began liquidating the positions in those reference baskets using open-market transactions. Most of the securities (valued at $790 million) were liquidated

[*39] that same day, and all positions (other than de minimis fractional

shares) were liquidated by May 19, i.e., within three days.

Upon liquidating positions in the three reference baskets, WMSA

replicated the exact same positions—generally within 15 minutes—in

OGI’s prime brokerage account at Deutsche Bank. WMSA refrained

from replicating positions only when it regarded the original position as

“fully matured,” i.e., where that position had reached a value that

aligned with GWA’s price target.

Barrier Contract #7 had a cash settlement amount of

$57,469,367, and Barrier Contracts #8 and #10 each had a cash settlement amount of $86,204,046. On May 19, 2010, the cumulative cash

settlement amounts ($229,877,460) were wired to GWA’s prime brokerage account at Deutsche Bank. Later that day, GWA instructed

Deutsche Bank to wire this same amount to OGI’s prime brokerage account at Deutsche Bank.

In a letter to GWA dated May 21, 2010, Deutsche Bank stated

that a “cash event” had occurred in Barrier Contract #9 and that it was

accelerating the “option” termination date accordingly. Barrier Contract #9 had a cash settlement amount of $56,210,572. That same day

Deutsche Bank wired $43 million to GWA’s prime brokerage account,

and then to OGI’s prime brokerage account at Deutsche Bank. Another

$13 million followed the same path on May 24–26, and a final $133,948

on June 1. The remainder of the $56,210,572 cash settlement amount,

$76,625, was paid to Quaker Partners as its final advisory fee.

XXIII. IRS Legal Advice Memorandum

On November 12, 2010, the IRS released Generic Legal Advice

Memorandum No. AM2010-005 on the subject of “Hedge Fund Basket

Option Contracts.” 13 It posited a scenario in which a hedge fund entered

into a contract with a foreign bank. The contract was styled a “call option,” with a payout linked to the value of an underlying reference basket of securities. The contract addressed in the IRS memorandum was

substantially similar to the Barrier Contracts. The memorandum concluded that the contract in question was not an option and that the

hedge fund in substance owned the basket securities.

13 Generic legal advice memoranda are nonprecedential legal opinions written

by the National Office of the IRS Office of Chief Counsel. They are intended to assist

IRS personnel in administering the tax laws.

[*40] On the following business day Deutsche Bank emailed Mr.

Doucette a copy of the IRS memorandum. On January 14, 2011, Mr.

Doucette met with a Deutsche Bank official and was informed that “New

MAPS” was in grave danger. Although the GPF team believed in the

product, they were “under a lot of pressure from the tax people” at

Deutsche Bank to abandon it.

GWA filed its Form 1065 for 2010 on September 1, 2011. Messrs.

Weiss and Gendreau were aware of the IRS Memorandum, and the conclusions it reached, before that return was filed. GWA nevertheless took

the same position on that return, with respect to the termination of Barrier Contracts #7 through #10, that it had taken on prior returns with

respect to the termination of the other six contracts. On its Form 1065

for 2010, GWA reported $192,679,910 as long-term capital gain stemming from the termination of Barrier Contracts #7 through #10 (aggregate amount realized of $286,011,407 less aggregate adjusted basis of

$93,331,497).

XXIV. Mark-to-Market Election

On its Federal income tax return for 1997, Weiss & Co. made a

“mark-to-market” election under section 475(f). It thus elected to recognize gain or loss on any security held at the close of the taxable year as

if that security had been sold for its fair market value on the last business day of that year.

GWA made the same mark-to-market election on its Form 1065

for 1998, which bears the signature of its return preparer dated May 28,

1999. First, GWA included with that return a Form 3115, Application

for Change in Accounting Method, to request a change from the cash to

the accrual method of accounting. On the Form 3115 GWA stated that

its primary business activity was as an investment company and that it

“also engage[d] in a trader activity through a wholly owned limited liability company,” viz., OGI.

Line 15 of the Form 3115 asked whether the taxpayer had “more

than one trade or business” and (if so) directed the taxpayer to attach a

description of “each trade or business.” In the attached statement GWA

identified its two businesses as “investment activity” and “trader activity.” In the case of its “trader activity,” it stated that it was “[a]dopting

the accrual method of accounting in its initial year of operation.”

Second, GWA included with its 1998 return a statement captioned “Election Under [Section] 475(f) for OGI, LLC (a Wholly Owned

[*41] Limited Liability Company of GWA, LLC).” As noted earlier, OGI

was a “disregarded entity” of GWA. The Election bore the header “GWA,

LLC” followed by GWA’s mailing address and EIN. GWA stated that

OGI was “engaged in a trade or business as a trader in securities and

elects to have [section] 475(f)(1) apply to such trade or business.”

Mr. Kleinman, who replaced Mr. Peckman as CFO of GWA,

pointed to the existence of a mark-to-market election during discussions

surrounding the execution of Barrier Contract #2. During a May 22,

2003, meeting between GWA and Deutsche Bank, Mr. Doucette recorded

in his notes that “we are at 35% vs 15%”—referring to the tax rates on

short-term versus long-term capital gains—and “currently have mark to

market election.”

During the examination of GWA’s returns in this case, GWA sent

the IRS examination team a letter captioned “Change in Method of Accounting Analysis.” This letter, dated September 20, 2013, stated that,

“[i]n 1998, GWAL [viz., GWA, LLC] made an election under section 475

to report its trading gains and losses on the mark-to-market method.”

The letter reported that “[o]ne of GWAL’s principal activities, which it

conducts through OGI, is trading securities for its own account using

various proprietary long-short trading strategies.” It then said that,

“[f]or 1998, and all subsequent years, GWAL (through OGI) directly

traded equity and debt securities using long-short trading strategies.”

In December 2013 GWA provided responses to an IRS Information Document Request (IDR). The IDR responses acknowledged

that the “Barrier Options” executed with Deutsche Bank “are securities

and are subject to the mark-to-market election that GWAL made, unless

the Barrier Options can satisfy the exception set forth in section

475(f)(1)(B).” Section 475(f)(1)(B) provides that a mark-to-market election by a securities trader shall not apply to any security that is “clearly

identified in such person’s records” as “having no connection to the activities of such person as a trader.”

In its IDR response GWA stated that it had “made the mark-tomarket election on behalf of its wholly owned, disregarded subsidiary,

OGI.” It initially believed that “it could make a ‘separate’ mark-to-market election for its trading business conducted through OGI, as distinguished from GWAL as an entity, and therefore was not required to satisfy the exception listed in section 475(f)(1)(B).” However, it later concluded that its initial view was incorrect. It accordingly acknowledged

in its IDR response that, “unless the Barrier Options met the exception

[*42] under section 475(f)(1)(B), they were subject to the mark-tomarket election made by GWAL on behalf of OGI.”

XXV. Supervisory Approval of Penalties

Susan Chambers (RA Chambers) was the revenue agent who

served as senior team coordinator for the IRS examination of GWA’s

2009 and 2010 returns. Her immediate supervisor was Keneth Hetzel.

Mr. Hetzel was a supervisor in the IRS Global High Wealth Department

during 2014 and 2015.

Philip Yarberough was an attorney in the IRS Office of Chief

Counsel during 2014 and 2015. He was assigned to offer advice to RA

Chambers in connection with the GWA examination. Mr. Yarberough’s

immediate supervisor was Associate Area Counsel John Guarnieri.

On December 17, 2014, Mr. Yarberough drafted a memorandum

advising RA Chambers about the applicability of penalties in connection

with GWA’s reporting of the Barrier Contracts. He recommended that

penalties be determined for underpayments due (in the alternative) to

negligence and substantial understatements of income tax. See

§ 6662(a) and (b)(1) and (2). Mr. Guarnieri approved this recommendation, indicating his approval by initialing the memorandum on December 17, 2014.

On December 19, 2014, Mr. Yarberough sent his memorandum,

thus approved, via email to RA Chambers. That same day she emailed

Mr. Hetzel, her immediate supervisor, requesting approval to assert the

section 6662 penalties. She attached to her email Mr. Yarberough’s

memorandum recommending that these penalties be asserted. Mr. Hetzel approved assertion of both penalties by return email on December

19, 2014.

On March 3, 2015, RA Chambers sent Mr. Hetzel draft Forms

886–A, Explanation of Items, that included penalties for underpayments

due (in the alternative) to negligence and substantial understatements

of income tax. Mr. Hetzel approved her penalty recommendations that

same day by placing his initials on the “Penalty Lead Sheet.” He again

approved her penalty recommendations two days later in an email stating that her request to impose the penalties was “approved.”

On June 22, 2015, the IRS issued GWA so-called 60-day letters

for 2009 and 2010. These letters indicated (among other things) that

the IRS intended to assert penalties for each year (in the alternative) for

[*43] negligence and substantial understatement of income tax. These

letters constituted the first formal communication to GWA that the IRS

intended to assert these penalties.

XXVI. Issuance of the FPAAs

On December 3, 2018, the IRS timely mailed FPAAs to petitioner

for tax years 2009 and 2010. The FPAAs made three principal determinations that are the focus of the parties’ dispute. First, the IRS determined that the Barrier Contracts “are not options for [F]ederal [income]

tax purposes, and that the partnership [GWA] is the owner of the security positions in the Reference Baskets for [F]ederal [income] tax purposes.”

Second, the IRS determined that the mark-to-market election

that GWA made on its 1998 return “applies to both GWA LLC and OGI

(as GWA LLC’s disregarded entity).” The IRS concluded that GWA had

failed to establish, “to the satisfaction of the Secretary,” that either the

“barrier options” or the securities in the reference baskets had “no connection to the activities of [GWA] as a trader” or that those securities

were “clearly identified in [GWA’s] records” as having no such connection. See § 475(f)(1)(B). Because the “exception” set forth in section

475(f)(1)(B) therefore did not apply, GWA was required to mark the reference basket securities (or the “options”) to market on an annual basis,

rather than deferring realization of its profits to the year in which the

“options” were terminated or exercised.

Third, the IRS determined that “requiring [GWA] to account for

gains and losses from the security positions in the Reference Baskets

under the . . . mark-to-market method of accounting [or] to recognize

gains and losses [on the underlying securities] under I.R.C. § 1001 . . .

constitutes a change to [GWA’s] method of accounting to clearly reflect

income under I.R.C. § 446.” The IRS further concluded that “an adjustment under I.R.C. § 481 is necessary solely by reason of the change in

order to prevent amounts from being duplicated or omitted.” See

§ 481(a)(2). The FPAA for 2009 determined a section 481 adjustment of

$337,170,142 on this ground. The FPAAs asserted a variety of alternative positions, depending on how the three questions listed above are

decided. Finally, the FPAAs asserted for each year a 20% accuracy-related penalty for an underpayment due to negligence or (in the alternative) a substantial understatement of income tax. See § 6662. These

were the same penalties that the examination team had communicated

to GWA in the 60-day letters. See supra pp. 42–43.

OPINION

[*44]

Burden of Proof

The IRS’s determinations in a notice of deficiency or an FPAA are

generally presumed correct, though the taxpayer can rebut this presumption. See Rule 142(a); Welch v. Helvering, 290 U.S. 111, 115 (1933);

Republic Plaza Props. P’ship v. Commissioner, 107 T.C. 94, 104 (1996);

Genecure, LLC v. Commissioner, T.C. Memo. 2022-52, 123 T.C.M. (CCH)

1271, 1276. Section 7491 provides that the burden of proof on a factual

issue may shift to the Commissioner if the taxpayer satisfies specified

conditions. Among these conditions are that the taxpayer must have

“introduce[d] credible evidence with respect to [that] factual issue,”

§ 7491(a)(1), and have “complied with the requirements under this title

to substantiate any item,” § 7491(a)(2)(A). Petitioner does not contend

that the burden of proof should shift to respondent on any question of

fact.

II.

Expert Testimony

To support their positions regarding the proper characterization

of the Barrier Contracts, the parties retained experts who testified at

trial. We assess an expert’s opinion in the light of his or her qualifications and the evidence in the record. See Parker v. Commissioner, 86

T.C. 547, 561 (1986). When experts offer competing opinions, we weigh

them by examining the factors the experts considered in reaching their

conclusions. See Casey v. Commissioner, 38 T.C. 357, 381 (1962).

We are not bound by an expert opinion that we find contrary to

our judgment. Parker, 86 T.C. at 561. We may accept an expert’s opinion in toto or accept aspects of his or her testimony that we find reliable.

See Helvering v. Nat’l Grocery Co., 304 U.S. 282, 295 (1938); Boltar,

L.L.C. v. Commissioner, 136 T.C. 326, 333–40 (2011) (rejecting expert

opinion that disregards relevant facts). And we may resolve the disputed factual questions on the basis of our own examination of the record evidence. See Silverman v. Commissioner, 538 F.2d 927, 933 (2d Cir.

1976), aff’g T.C. Memo. 1974-285.

We have listed the experts who testified in this case, along with

brief summaries of their credentials, in the Appendix to this Opinion.

In the pages that follow, we discuss their testimony to the extent it is

relevant to our analysis.

[*45] III.

Proper Characterization of the Barrier Contracts

The first question we must decide is whether the “option” form of

the Barrier Contracts should be disregarded for Federal income tax purposes, and whether GWA should be treated, in substance, as owning the

securities in the underlying reference baskets. If GWA is determined to

have been the owner of the basket securities for Federal income tax purposes, it would be required to recognize, on an annual basis, the profits

it realized from trading those securities each year. See § 1001. By contrast, the holder of a standard option contract will recognize gain or loss

only for the taxable year when the option is exercised, is terminated, or

expires worthless. See Fed. Home Loan Mortg. Corporation v. Commissioner (Freddie Mac), 125 T.C. 248, 267 (2005); Westall v. Commissioner,

T.C. Memo. 1988-421, 56 T.C.M. (CCH) 66, 73. 14

It has long been established that substance, not form, determines

the proper characterization of a transaction (or group of transactions)

for Federal income tax purposes. Frank Lyon Co. v. United States, 435

U.S. 561, 573 (1978); Commissioner v. Court Holding Co., 324 U.S. 331,

334 (1945); Benenson v. Commissioner, 910 F.3d 690, 699 (2d Cir. 2018),

rev’g and remanding T.C. Memo. 2015-119; Altria Grp., Inc. v. United

States, 658 F.3d 276, 284 (2d Cir. 2011). “[I]n tax law, . . . substance

rather than form determines tax consequences.” Raymond v. United

States, 355 F.3d 107, 108 (2d Cir. 2004) (quoting Cottage Sav. Assn. v.

Commissioner, 499 U.S. 554, 570 (1991) (Blackmun, J., dissenting)). In

applying the substance-over-form doctrine, courts look to the “the objective economic realities of a transaction rather than to the particular

form the parties employed.” Altria Grp., 658 F.3d at 284 (quoting Frank

Lyon, 435 U.S. at 573).

Labels do not determine tax consequences when they are inconsistent with economic realities. Bank of N.Y. Mellon Corp. v. Commissioner, 140 T.C. 15, 40 (2013), supplemented by T.C. Memo. 2013-225,

aff’d, 801 F.3d 104 (2d Cir. 2015). But “[i]f substance follows form then

this Court will respect the form chosen by the taxpayer.” Turner Broad.

Sys., Inc. & Subs. v. Commissioner, 111 T.C. 315, 326–27 (1998). Deciding whether the form of a transaction should be disregarded in favor of

its substance requires a factual determination. Harris v. Commissioner,

14 Section 1234A provides that gain or loss attributable to the termination of

an option with respect to property which is a capital asset in the hands of the taxpayer

is treated as gain or loss from the sale of that capital asset. Because every Barrier

Contract was terminated before its expiration date, section 1234A would presumably

govern if we were to decide that the Barrier Contracts were true “options.”

[*46] 61 T.C. 770, 783 (1974); Endeavor Partners Fund, LLC v. Commissioner, T.C. Memo. 2018-96, 115 T.C.M. (CCH) 1540, 1551, aff’d, 943

F.3d 464 (D.C. Cir. 2019). 15

Economic Realities of the Barrier Contracts

As the Supreme Court emphasized in Frank Lyon, 435 U.S. at

584, the answer to the substance-over-form inquiry “in any particular

case will necessarily depend upon its facts.” The Second Circuit has described the analysis mandated by Frank Lyon as a “wide-ranging and

fact-intensive” inquiry. Altria Grp., 658 F.3d at 286, 288 (ruling that

the district court properly “instructed the jury to consider ‘all the relevant facts and circumstances’”).

“A contract is an option contract when it provides (A) the option

to buy or sell, (B) certain property, (C) at a stipulated price, (D) on or

before a specific future date or within a specified time period, (E) for

consideration.” Freddie Mac, 125 T.C. at 261 (citing W. Union Tel. Co.

v. Brown, 253 U.S. 101, 110 (1920)); see Halle v. Commissioner, 83 F.3d

649, 654 (4th Cir. 1996), rev’g and remanding Kingstowne LP v. Commissioner, T.C. Memo. 1994-630; Estate of Franklin v. Commissioner, 64

T.C. 752, 762–63 (1975), aff’d, 544 F.2d 1045 (9th Cir. 1976). Characterization of an agreement as an option contract depends not only on the

“contractual language” but also on “the economic substance of the agreement.” Freddie Mac, 125 T.C. at 261; see Old Harbor Native Corp. v.

Commissioner, 104 T.C. 191, 201 (1995) (citing Frank Lyon, 435 U.S. at

573).

We have undertaken the fact-intensive inquiry required to ascertain “the substance and economic realities of the [Barrier Contract]

transaction[s].” See Frank Lyon, 435 U.S. at 582. We conclude that the

Barrier Contracts were not options in substance because they lacked the

essential economic and legal characteristics of genuine options. When

15 The substance-over-form doctrine is related to, but distinct from, the “economic substance” doctrine. See Benenson v. Commissioner, 910 F.3d at 699 n.8; Altria

Grp., 658 F.3d at 291 (recognizing doctrines as distinct); Neonatology Assocs., P.A. v.

Commissioner, 299 F.3d 221, 230 n.12 (3d Cir. 2002) (same), aff’g 115 T.C. 43 (2000).

But see Summa Holdings, Inc. v. Commissioner, 848 F.3d 779, 785 (6th Cir. 2017) (appearing to conflate the doctrines), rev’g T.C. Memo. 2015-119. Under the economic

substance doctrine, the court will consider whether the taxpayer (1) had an objectively

reasonable expectation of profit from the transaction, apart from tax benefits, and

(2) had a subjective nontax business purpose in entering the transaction. Bank of N.Y.

Mellon Corp. v. Commissioner, 801 F.3d at 115. Given our disposition, we need not

address the economic substance doctrine.

[*47] the self-serving labels are stripped away, the true substance of the

arrangements is clear. GWA held and traded the basket securities

through a prime brokerage account, and Deutsche Bank financed GWA’s

investment in those securities by extending a margin loan at 10-to-1 leverage, with the putative “premium” serving as collateral for that loan.

Consideration

For a contract to be an option contract, it must provide for “consideration.” Freddie Mac, 125 T.C. at 261. In a standard equity option

contract, the consideration paid to the optionor is the option premium.

The premium compensates the optionor for accepting the investment

risk that the stock, at a future date, will be called away from him for less

than the stock is then worth. Stated differently, the premium compensates the optionor for bearing upside investment risk. The option premium is paid to the optionor at the outset of the contract, and the premium is never refunded or returned to the optionee.

In stark contrast to option contract norms, the Barrier Contracts

provided that Deutsche Bank would refund the premium to GWA upon

its exercise of the “option.” Barrier Contract #1 was typical. It specified

a nominal premium of $50 million, divided into a $44 million “fixed premium” and a $6 million “amortizable premium.” The $44 million fixed

premium (as well as any “additional premium” GWA might have paid)

was refunded to GWA as part of the “Premium Settlement Amount.” See

supra pp. 25–27 & note 11. The “amortizable premium,” which accrued

at a rate of $1,364 per day, was refunded to GWA via upward adjustment

to the Basket Base Performance, at the identical rate of $1,364 per day

for every day the contract was outstanding. See supra p. 26.

These neutralizing adjustments ensured that GWA would be refunded, upon exercise of the “option,” 100% of the $50 million stated premium. The other nine Barrier Contracts were structured the same way.

All in all, GWA received premium refunds totaling $286.8 million upon

termination of the Barrier Contracts, the exact amount of the “premiums” it paid.

Besides departing from recognized option norms, Deutsche

Bank’s agreement to refund 100% of the premium to GWA sheds light

on how the bank actually viewed the arrangement. The owner of stock

is entitled to enjoy the stock’s full upside potential. By refunding the

premium to GWA upon exercise of the “option,” Deutsche Bank waived

all consideration for surrendering to GWA 100% of the upside potential

[*48] of the basket securities. This suggests that Deutsche Bank regarded the upside potential of those securities as belonging, not to it,

but to GWA. But if GWA owned 100% of the upside potential, that is a

strong indication that GWA owned the securities in substance.

The “premium refund” feature of the Barrier Contracts, while inconsistent with their characterization as “call options,” is perfectly consistent with what we believe to be their substance—namely, prime brokerage accounts funded by margin loans from Deutsche Bank. As Peter

Tufano, respondent’s expert in financial economics and engineering, cogently explained, the economic role of the “premium” was essentially

identical to that of margin (collateral) in a prime brokerage account.

Like margin in a prime brokerage account, the premium supplied a

cushion that protected Deutsche Bank from downside risk if GWA’s

trading generated losses. But if GWA’s trading generated profits, GWA

would get 100% of its collateral back through refund of its “premium.”

In substance, the “premium” thus functioned as the collateral Deutsche

Bank required as a condition of extending a margin loan to GWA at 10to-1 leverage.

Petitioner contends that the premium refund provision is not fatal to option characterization. Timothy Weithers, petitioner’s expert in

financial economics, asserts that exotic versions of knockout barrier options occasionally display a similar feature, providing for “an independent cash payment (from the barrier option seller to the barrier option

buyer) should the barrier be breached.” Dr. Weithers indicates that this

type of cash payment “is generally known as a ‘rebate.’”

We are not persuaded by this line of argument. First, the exotic

products to which Dr. Weithers refers appear to be rare in the option

universe. Second, petitioner has not demonstrated that such contracts,

any more than the Barrier Contracts, would be characterized as true

“options” for Federal income tax purposes. Third, these exotic products

appear to provide for a partial cash rebate, rather than the 100% premium refund that occurs upon exercise of a Barrier Contract. As respondent’s expert Tanya Beder explained: “Occasionally, the owner of

the down-and-out [knockout barrier] option may receive a portion of the

premium back if the option is cancelled.”

Finally, and perhaps most importantly, the cash rebate cited by

Dr. Weithers works very differently from the Barrier Contract refund.

The cash rebate occurs in a loss scenario, i.e., where the option hits a

barrier and terminates before its expected expiration date. It is not

[*49] wholly illogical to provide for a cash rebate in this scenario. According to a source cited by Dr. Weithers, the rationale for such a rebate

is as follows: “When a knock-out option knocks out, all hopes of participating in the upside of the vanilla option payoff are dashed. To soften

the blow, contracts are sometimes modified to include a feature whereby

a fixed payment is made if the option knocks out.” See Zareer

Dadachanji, FX Barrier Options 25 (2015).

The Barrier Contracts themselves provided for a partial premium

refund in a loss scenario. See supra p. 27. If a contract hit a barrier and

“knocked out,” GWA would always be refunded a portion of its premium,

so long as the NAV Index Level did not decline to 90. We need not decide

whether this type of rebate—a partial rebate in a loss scenario—is fatal

to option characterization.

The problem with the Barrier Contracts is that they provided for

a full premium refund in a gain scenario, i.e., where the option finishes

“in the money” and is exercised. Where the optionee has made a profit,

there is no logic behind a cash rebate “to soften the blow.” By refunding

the premium in this scenario, Deutsche Bank waived all consideration

for surrendering to GWA 100% of the upside potential of the basket securities. Petitioner has offered no explanation as to why a rational optionor would do that. And petitioner’s experts cited no example of an

option—however exotic—that that would offer a 100% premium refund

in a gain scenario. Every source cited by Dr. Weithers indicates that

cash rebates are paid only when the option hits a barrier and “knocks

out.”

Pricing

The pricing of the Barrier Contracts exhibited none of the risk

characteristics that inform the pricing of true options. As respondent’s

experts cogently explained, the pricing of call options—at least in theory—is a complicated affair. Factors that determine the magnitude of

the premium include time until expiration (theta), sensitivity to the volatility of the underlying asset (vega), prevailing interest rates (rho), sensitivity to changes in the price of the underlying asset (delta), and sensitivity to changes in the rate of change in the price of the underlying

asset (gamma). Sophisticated option traders call these risk factors “the

Greeks.”

“The Greeks” were utterly irrelevant to the pricing of the Barrier

Contracts. Each “premium” was calculated exactly the same way—as a

[*50] flat 10% of the “notional amount.” Barrier Contract #2 was a

slight exception to the rule, with the “premium” calculated as 10.56% of

the “notional amount.” 16

Petitioner did not attempt to show how “the Greeks” would (or

could) produce premiums of this sort. Interest rates and asset volatility,

which inevitably vary over a multiyear period, are highly influential in

how an option is priced. The Barrier Contracts had extremely long

terms—12+ years—and the risks attributable to interest rate and asset

volatility would thus be at their apogee. Barrier Contracts #1, #3

through #6, and #7 through #10, respectively, were executed on three

different trade dates between April 2003 and December 2006. But each

was assigned exactly the same premium—10% of the contract’s “notional amount.” It seems obvious that standard option pricing methods

would not yield identical premium calculations at such divergent points

in time. 17

The multiplicand in the Barrier Contract pricing formula—the

number that was multiplied by 10% to generate the “premium”—also

shows that these arrangements were not true “options. The price of a

true option will be heavily influenced by the characteristics of the underlying asset. For example, assume an investor writes a call option on

IBM stock, currently trading at 245, with an option strike price of 250.

The premium demanded by the optionor will be influenced to a limited

degree by factors exogenous to IBM stock, e.g., the option term and general market conditions. But it will be heavily influenced by the salient

characteristics of that underlying asset, e.g., the price volatility of IBM

stock, the company’s expected earnings, its price/earnings ratio, its current dividend, etc.

Because the securities in the Barrier Contract reference baskets

changed daily or hourly, Deutsche Bank could not know what the

16 Barrier Contract #2, like Barrier Contract #1, had a “notional amount” of

$500 million, but its “premium” was $52.8 million rather than $50 million. That may

have been because $52.8 million was the amount of cash that happened to be available

for carryover to Deutsche Bank following termination of the final three RBC “options.”

See supra p. 30. Respondent’s expert Ms. Beder explained that, “[a]fter adjustment to

align Barrier Options #1 and #2 for the difference in start dates, Barrier Option #2 also

had a Notional Amount ten times its Total Premium.”

17 As Prof. Tufano showed, a well-known measure of market volatility decreased from 3.44% in April 2003 to 1.86% in December 2006. And the interest rate

on one-year Treasury securities (often called the “risk-free rate”) increased from 1.32%

to 4.91% between those dates. Yet the pricing on the Barrier Contracts remained exactly the same.

[*51] “underlying asset” actually was. (This is a distinct problem we

discuss infra pp. 52–55.) But even if Deutsche Bank had known what

the underlying asset was, the “premium” it charged was not determined

with respect to that asset. The “premium” was calculated as 10% of the

“notional amount,” i.e., the maximum amount of capital Deutsche Bank

was prepared to make available to GWA for acquisition of basket securities. Petitioner has cited no example, and we know of none, in which

the premium for a genuine call option was dictated, not by the characteristics of the underlying asset, but by the amount of financing a bank

was willing to make available to facilitate purchase of that asset.

A third component of the Barrier Contract pricing—Deutsche

Bank’s ability to demand “additional premium” if the NAV Index Level

fell to 97—was likewise inconsistent with true option pricing. For a genuine call option, the price is determined and paid at the outset of the

contract. By paying that price, the optionee acquires the right to exercise the option until its expiration date. Requiring the optionee to pay

additional premium to preserve that right would constitute a retroactive

increase to the agreed-upon price, depriving the optionee of the benefit

of his bargain. And it would violate a basic principle underlying all call

options: that the optionee bears no downside risk beyond the premium

he pays. If the optionee is required to pay additional premium to retain

his bargained-for rights, he is forced to bear additional downside risk. 18

These pricing features, while making no sense for a genuine call

option, make perfect sense if the Barrier Contracts are recharacterized

to match their substance. Pricing GWA’s required payment by reference

to the amount of capital Deutsche Bank made available—rather than by

reference to the characteristics of the underlying assets—was completely logical, because Deutsche Bank was providing financing. Computing the “premium” as 10% of the “notional amount” was completely

logical, because Deutsche Bank agreed to provide financing with 10-to1 leverage. And requiring GWA to pay “additional premium” if the securities declined in value was completely logical, because that requirement was equivalent to a margin call in a prime brokerage account.

18 If the NAV Index Level fell to 97 and GWA paid a $15 million additional

premium, the “knock-out level” would be reset downward from 94 to 91. This was also

inconsistent with standard option norms. As Ms. Beder explained, “[i]n a typical [barrier] option, the knock-out level does not change during the option’s life.”

[*52]

Option Term

Each Barrier Contract had a term of 12+ years. A term of this

length is not absolutely inconsistent with “option” characterization. But

setting the expiration date 12 years away is—at the very least—highly

unusual for an equity call option.

Publicly traded call options commonly have terms of 3, 6, or 9

months. So-called long-term options may have terms of 12 to 18 months.

It is thus no accident that Deutsche Bank, when proposing terms for

“New MAPS,” specified that future barrier contracts would have terms

between 13 and 18 months. “When pushed,” Mr. Doucette noted,

Deutsche Bank “said they might be able to do [a] 24 months term.”

These changes were part of Deutsche Bank’s effort to make the arrangements look more like “‘true’ option[s].” See supra pp. 35–37.

The reason equity call options with 12-year terms are difficult to

find is not hard to guess. Key factors in pricing call options include the

price of the underlying security, the volatility of the underlying security,

general stock market conditions, and prevailing interest rates. Needless

to say, these factors vary considerably over time. For example, in the

12-year period beginning January 1, 2010, the S&P 500 Index reached

a low of 1,034 and a high of 4,766, and the Dow Jones Industrial Average

ranged between 9,774 and 36,338. Interest rates were likewise variable,

with the Federal funds rate touching a low of 0.05% and a high of 2.42%.

The direction of stock prices and interest rates is hard to predict

over the short term. Twelve years is an eternity in the stock market.

Equity call options with 12-year terms are unicorns because no investor

could rationally price them using established option pricing methods.

On the other hand, if the Barrier Contracts are recharacterized

to match their substance, the 12-year term is not surprising or odd. In

substance, those contracts constituted an agreement by Deutsche Bank

to lend money at ten times leverage for securities investment in a prime

brokerage account controlled by GWA. Loan agreements with 12-year

terms are hardly uncommon. Banks routinely offer home mortgages

with 15- and 30-year terms, and corporations routinely issue bonds with

distant maturity dates.

Reference Property

The reference property specified for the Barrier Contracts was

fundamentally inconsistent with option norms. As respondent’s expert

[*53] Ms. Beder explained, “[a] barrier option provides a payout dependent on the value of a specific underlying [asset].” Typically, the underlying asset takes the form of a “well-defined equity, fixed income, commodity, currency, credit, or other instrument.” A contract is an option

contract when it provides the option to buy or sell “certain property . . .

at a stipulated price.” Freddie Mac, 125 T.C. at 261 (emphasis added).

The underlying asset need not be a single, discrete, or fixed investment item. But it must be sufficiently well defined to enable the

optionor, using standard option pricing methods, to set a price that reasonably reflects the option’s risk. A purported option whose underlying

property is ill defined or constantly changing cannot be a true option if

it is impossible to assign that option a rational market price.

The reference property for each Barrier Contract was a huge basket of equities, plus some bonds and derivatives. The IAAs gave GWA—

acting through its affiliates, Quaker Partners, Weiss Associates, and

WMSA—wide discretion to trade those securities as it saw fit, with little

or no oversight by Deutsche Bank. GWA traded the securities with such

gusto that the contents of the reference baskets changed daily, hourly,

or minute by minute. On an average trading day during 2003–2010,

GWA initiated trades of 268 unique securities. During 2005 it initiated

89,075 trades involving more than two billion units of stock. Because

the ultimate identity of the “underlying asset” was unknowable at the

outset of each Barrier Contract, determining a rational premium for an

option would be challenging, to say the least.

Petitioner seeks to analogize the Barrier Contracts to options

written on an index of securities, such as the S&P 500 Index or the Dow

Jones Industrial Average. As petitioner notes, the stocks included in

those indices occasionally change. Yet options on those indices are “common in the derivatives market” and “well accepted.”

The comparison is unconvincing. As Prof. Tufano explained, familiar market indices are occasionally “rebalanced” by removing the

stock of one company and replacing it with another. Such rebalancing

occurs very infrequently, and any proposed rebalancing is announced

publicly in advance. The rebalancing is conducted mechanically or is

based on a specified methodology established by an independent third

party (e.g., Standard & Poor’s). This episodic form of stock substitution

is at the opposite end of the spectrum from the incessant and unpredictable trading in which GWA engaged.

[*54] The rebalancing of equity indices, moreover, is typically done because the index sponsor believes rebalancing necessary to keep the index representative of what it is supposed to represent. The S&P 500

Index, for example, is a market-capitalization-weighted index of 500 major corporations in the United States. Every sophisticated investor

knows exactly what the S&P 500 Index stands for. If Standard & Poor’s

concludes that Company A should be removed from the Index and be

replaced by Company B, that does not make the Index less “well defined.” Quite the contrary: The substitution is intended to ensure that

the stocks in the Index continue to mirror its well-defined objective.

Dr. Weithers opined that a true option need not be tied to the performance of a single asset or even a defined pool of assets. Rather, he

suggested that an option could be tied (at least in theory) to a “welldefined activity,” such as a specific trading strategy. But petitioner

came up with virtually no real-world examples of call options structured

in that way.

Assuming arguendo that a genuine call option could be written on

a “trading strategy” as opposed to an “underlying asset,” the trading

strategy would have to be—at the very least—specific and well defined.

But not only were the securities in the reference baskets wholly unpredictable, the strategies that GWA pursued in trading them were numerous and varied. According to PPMs issued between 2006 and 2007, GWA

was pursuing 31 different trading strategies as of December 2007, a twothirds increase over the 19 different trading strategies that it was pursuing in 2006. GWA’s “allocation committee,” chaired by Mr. Weiss, allocated funds among the various trading strategies as it saw fit. See

supra p. 31. None of petitioner’s experts could explain how a rational

market participant would go about pricing a 12-year call option, on 31

different trading strategies, which were being implemented on a subjective proprietary basis that was invisible to the market.

In a typical option contract, the underlying asset is a security or

group of securities outside the control of the optionor and the optionee,

e.g., shares of IBM stock, Treasury bonds, or the S&P 500 Index. Under

the Barrier Contracts, the underlying assets were subject to the complete control of GWA, which (through its affiliates) selected and traded

the securities in the reference baskets. See infra pp. 76–79. As Prof.

Glasserman, respondent’s expert in derivatives, financial engineering,

and risk analysis, noted, “it would be unusual to have an option contract

where the underlying asset is under the option buyer’s control,” because

[*55] the buyer could potentially manipulate the reference property to

the seller’s disadvantage.

On the other hand, if the Barrier Contracts are recharacterized

to match their substance, the ill-defined and indeterminate nature of

the reference basket, and GWA’s control over the reference assets, are

not the least problematic. In substance, the contracts constituted an

agreement by Deutsche Bank to lend money to GWA to acquire securities positions in a prime brokerage account. It was immaterial to

Deutsche Bank what those positions were, so long as GWA adhered to

the investment guidelines and the reference baskets contained no securities on the “restricted list.” Deutsche Bank’s only concern was the

risk—an infinitesimal risk, as we explain infra pp. 62–67—that the

value of the reference basket would decline so precipitously as to wipe

out the margin that GWA supplied.

Early Termination

An option contract affords the right to buy or sell specific property

“on or before a specific future date or within a specified time period.”

Freddie Mac, 125 T.C. at 261. Unlike standard call options, the Barrier

Contracts permitted Deutsche Bank, the putative optionor, to terminate

the “options” at virtually any time. And whereas European-style options

permit exercise only on the stated expiration date, GWA essentially terminated nine Barrier Contracts early. Significantly in our view, GWA

was allowed to do so without being required to pay anything to Deutsche

Bank for being granted this early-exercise privilege.

Deutsche Bank could instigate early termination of a Barrier

Contract in two ways. First, it could accelerate termination to various

dates preceding the stated expiration date, provided it gave GWA 30

days’ notice of its decision. See supra p. 24. Deutsche Bank availed itself

of this right when it accelerated the termination of Barrier Contract #1

to April 30, 2009. See supra p. 37. Second, Deutsche Bank could terminate the “option” by causing a “cash event,” e.g., by canceling an IAA.

Deutsche Bank could cancel an IAA “for any reason or for no reason,”

and subject only to written notice and payment of a termination fee of

at most $200,000. See supra p. 24. In effect, Deutsche Bank thus could

terminate a Barrier Contract at essentially any time.

Deutsche Bank’s unilateral ability to terminate the contract was

inconsistent with option norms. The price of a call option is heavily influenced by the length of the option period—the longer the option period,

[*56] the higher the premium. By paying that price, the optionee acquires the right to exercise the option until it expires. By accelerating

expiration to an earlier date—e.g., a date on which the option is “out of

the money”—the optionor would deprive the optionee of his bargainedfor rights. See Halle v. Commissioner, 83 F.3d at 654; Freddie Mac, 125

T.C. at 259 (noting that an essential feature of an option is an agreement

by the optionor “to leave the offer open for a specified or reasonable period of time” (quoting Old Harbor Native Corp., 104 T.C. at 201)); Saviano v. Commissioner, 80 T.C. 955, 970 & n.20 (1983) (citing Restatement

(Second) of Contracts § 25 and other authorities), aff’d, 765 F.2d 643 (7th

Cir. 1985).

GWA also had the de facto ability to terminate a Barrier Contract

early, enabling it to receive a payout before the stated expiration date.

This would not be problematic for an American-style option, which permits the optionee to exercise at any time during the option term. But

GWA and Deutsche Bank ostensibly entered into European-style options. A European-style option may be exercised only on the stated expiration date. Because European-style options impose greater risk on

the optionee, they are typically priced differently—i.e., less expensively—than American-style options with similar features.

Although the Barrier Contracts did not afford GWA an explicit

right to terminate, it could manufacture early termination at essentially

any time. First, it could cause Quaker Partners to liquidate the basket

securities to U.S. dollar cash equivalents, creating a “cash event.” Second, it could direct Quaker Partners to cancel the current IAA upon 30

days’ notice, triggering the requirement that the basket be liquidated

“in a prompt and orderly manner.” That would likewise cause a “cash

event.” See supra p. 24.

Upon occurrence of a cash event, Deutsche Bank had the immediate right to accelerate the option termination date. It would have a

strong economic incentive to exercise this right because the cash in the

reference basket would begin accruing interest at the Federal funds rate

plus 5%. Mr. Peckman, GWA’s CFO, acknowledged that this rate would

be “punitive” for a financial institution like Deutsche Bank. Moreover,

because none of Deutsche Bank’s capital would be actively invested in

the reference basket following a cash event, Deutsche Bank would be

entitled to receive no further financing fees.

For both reasons, Mr. Peckman viewed GWA’s ability to generate

a cash event as a de facto “out provision” that it could employ to

[*57] terminate a Barrier Contract at a time of its choosing. And GWA

evidently believed that the Barrier Contracts afforded it a right to terminate. On December 11, 2006, it informed Deutsche Bank of its intention to “exercise its rights . . . to terminate Options 3, 4, 5 & 6 in the

MAPS account.” 19

Petitioner contends that GWA had no right to terminate a Barrier

Contract early because Deutsche Bank had ultimate “discretion as to

whether to terminate.” As a supposed example of the exercise of such

discretion, petitioner asserts that Deutsche Bank declined GWA’s request to terminate a Barrier Contract in October 2008 because the bank

was allegedly reluctant to pay out cash during a time of financial stress.

The evidence leads us to a different explanation. Internal

Deutsche Bank emails indicate that GWA, as part of its request to terminate, asked that the underlying portfolio positions be “journaled” to

other accounts under GWA’s control. Deutsche Bank declined to permit

this: It acknowledged that GWA could terminate the Barrier Contract

but insisted that it would “hav[e] to put the account to cash,” i.e., liquidate the underlying positions. In short, Deutsche Bank was not demurring to termination, as petitioner contends, but merely refusing to accede to GWA’s extracontractual request that the underlying positions be

rolled into other accounts under GWA’s control.

In practice, GWA and Deutsche Bank negotiated the early termination of every Barrier Contract, with GWA initiating the negotiations

whenever it wished to extract cash from MAPS. On several occasions,

Deutsche Bank declared that a “cash event” had occurred, even though

the reference basket was still populated with securities. Noting one instance of this problem, GWA in February 2006 requested a report from

Deutsche Bank showing that a “cash event” had occurred the previous

December. See supra p. 32. GWA noted that, “in order for us to terminate the option, the account has to be all cash.” It accordingly requested

“[f]or tax purposes . . . a report for Option 2 [that] shows only a cash

19 Petitioner appears to contend that GWA could not effect early termination

in the manner described above because it was prohibited from “contact[ing] directly

the investment advisor [i.e., Quaker Partners] regarding the terms or subject matter

of th[e] [MAPS] transaction.” See supra note 8. But this prohibition was meaningless

because Quaker Partners had no employees and delegated all of its investment management responsibilities to Weiss Associates and later to WMSA, both of which were

owned and operated by GWA and/or Mr. Weiss.

[*58] balance” and “all positions . . . [having been] liquidated prior to the

exercise of the option” on December 21, 2005.

In short, while the Barrier Contracts were European-style options

in form, the substance differed significantly from the form. Although

GWA supposedly could exercise each “option” only on the stated expiration date, it could terminate the option (and demand payment of the

proceeds) at any time of its choosing. All ten Barrier Contracts were in

fact terminated long before the option expiration dates, and the terminations were sometimes accomplished in a manner that did not comply

with contractual requirements.

Petitioner contends that the deficiencies described above are not

fatal to “option” characterization, asserting that standard call options

and European-style options may permit early termination in some circumstances. But petitioner’s experts cite no examples of genuine call

options that can be terminated by the optionor at virtually any time.

And while it appears that European-style options occasionally permit

early exercise by the optionee, early exercise invariably comes with a

financial cost that was not imposed on GWA when it terminated the

Barrier Contracts.

European-style options impose greater risk on the optionee. The

value of the underlying asset, for example, may rise substantially above

the strike price 60 days into the option period, but it may close below the

strike price on the expiration date, causing the option to expire worthless. Because of this greater risk to the optionee, the optionor will accept

a lower premium for writing a European-style option than for writing a

comparable American-style option.

Having agreed to accept a lower premium in consideration of the

optionee’s being restricted to exercise on a single date, the optionor will

naturally demand compensation for releasing the optionee from that restriction. This compensation might take the form of a financial penalty

or a “haircut” on the proceeds that would be payable if the option were

exercised at maturity in the normal way. 20

Nine of the Barrier Contracts were terminated early at GWA’s

request. But on no occasion did Deutsche Bank insist that GWA pay a

20 Respondent’s expert Ms. Beder acknowledged that an option seller in some

instances “may be willing to negotiate an early termination with the buyer.” But she

credibly testified that “this is subject to price, including add-on costs for hedges, risk

management, lost opportunity and operational costs among others.”

[*59] penalty or fee of any kind for the privilege of accelerating the exercise date. Rather, upon termination of each contract, GWA received

exactly the same proceeds it would have received if it had exercised the

“option” on the expiration date. Because GWA was allowed to exercise

the “options” early, and because it was required to pay nothing for securing the ability to do so, the substance of the Barrier Contracts did not

match the form of genuine European-style options.

Treatment of Dividends

When an investor writes a call option on stock he owns, he remains the owner of the stock unless and until the option is exercised.

The stock owner is entitled to receive all dividends declared with respect

to the stock during the life of the option. As the nominal optionee on a

call option, GWA had no right to any dividends paid on shares held in

the reference baskets.

But that is not how the Barrier Contracts worked. In calculating

the option payout to GWA, the Basket Base Performance was increased

by the aggregate amount of “dividends in respect of the Basket Long

Positions.” See supra pp. 25–26. This means that GWA, rather than

Deutsche Bank, received the economic value of all dividends paid on

stock held in the reference baskets. This is an indication that GWA, not

Deutsche Bank, in substance owned those shares.

A similar anomaly existed (in reverse) with respect to short positions in the reference baskets. When an investor borrows shares to sell

them short, the investor becomes liable for dividends declared on the

stock while the securities loan is outstanding. As the nominal holder of

the short positions in the reference baskets, Deutsche Bank in theory

was the “borrower” of those shares and it should have been liable for the

dividends. But in calculating the option payout to GWA, “dividends in

respect of Basket Short Positions” were included among “Basket Losses

and Expenses.” Those dividends thus decreased Basket Base Performance and hence reduced the payout GWA received. See supra p. 26.

The fact that GWA bore economic liability for dividends on the shares

sold short is a strong indication that GWA was in substance the borrower, and hence the short-seller, of those shares. 21

21 As respondent’s expert Prof. Tufano explained, one factor that affects the

pricing of a call option is the “dividends to be paid out by the reference asset prior to

exercise.” In theory, the optionor conceivably could agree to assign to the optionee the

[*60]

Absence of Risk to Deutsche Bank

An investor who writes a call option on stock bears two kinds of

investment risk. He bears upside risk on the option, and he bears downside risk on the underlying stock position. The Barrier Contracts were

not true options because Deutsche Bank bore neither type of risk.

Upside Risk

In a genuine call option, the premium compensates the optionor

for accepting upside investment risk—the risk that the stock, at a future

date, will be called away from him for less than it is then worth. Suppose

an investor writes a call on 100 shares of Company A stock, currently

trading at $100. Assume that the strike price is $100 and that the premium is $1,000, or $10 per share. The optionor bears upside risk because he has surrendered to the optionee, for the life of the option, the

stock’s upside potential beyond $110 per share, including the possibility

that it could rise to $120 or $150 per share. The $1,000 premium compensates him for accepting that risk.

The economics of the Barrier Contracts show that Deutsche Bank

bore no upside risk. If it had borne upside risk, it would have demanded

compensation for doing so. By agreeing to refund 100% of the premium

to GWA upon exercise of the “option,” Deutsche Bank in effect waived

any such compensation. See supra pp. 47–49. No rational investor

would do that. By its behavior, Deutsche Bank thus acknowledged that

it bore no upside risk.

Petitioner asserts that Deutsche Bank bore upside risk because it

could have chosen not to purchase the basket securities. Instead,

Deutsche Bank allegedly could have made notional trades in a notional

securities basket, with the cash settlement amount being based on the

cumulative performance of the theoretical securities positions. In effect,

petitioner argues that Deutsche Bank could have converted a Barrier

Contract into what is commonly called a “naked” call option.

A “naked” call option occurs when an investor sells a call on stock

he does not own. “Naked” call options are extremely risky. Because the

dividends paid on the underlying stock during the option period, and the premium

price could be adjusted accordingly. But petitioner supplied no evidence that this occurred here. Rather, the Barrier Contracts were priced at a flat 10% of the “notional

amount,” i.e., the maximum amount of capital Deutsche Bank agreed to make available for investment in the reference baskets. See supra pp. 19, 50–51.

[*61] optionor has no stock to surrender when the optionee exercises the

option, the optionor must pay cash out of pocket for every dollar by which

the stock’s closing price at expiration exceeds the strike price.

We reject this argument out of hand. First, the Barrier Contracts

required that the underlying securities be purchased, providing that

“[t]he Basket shall be comprised of Shares which shall be traded by the

[Investment] Advisor.” (Emphasis added.) When asked whether the

Barrier Contracts permitted a naked call strategy, petitioner’s expert

Fabio Savoldelli and respondent’s expert Prof. Tufano both opined that

the Barrier Contracts did not.

Second, there is no evidence that Deutsche Bank ever considered

pursuing a naked call strategy, which would have subjected it to unlimited upside risk. The securities basket was under GWA’s control, and

its composition changed daily according to GWA’s investment strategies.

See supra pp. 23, 27. Whenever GWA wished to extract gains from the

basket, it could quickly manufacture the early termination of a Barrier

Contract. See supra pp. 55–59. Under these circumstances, it is utterly

implausible that a publicly traded bank would write naked call options

on a $500 million investment portfolio.

Third, in the unlikely event that Deutsche Bank would choose to

pursue a “naked” option strategy, the risk it would assume thereby

would be of its own making. It would then face the possibility that it

would need to come out of pocket for gains realized in the reference basket during a Barrier Contract’s lifetime. But the decision to pursue a

“naked” call strategy would be a decision Deutsche Bank would make

wholly apart from its execution of the Barrier Contract. The “naked”

call risk, in other words, would be extrinsic to the Barrier Contract. It

would have nothing to do with the risk (if any) inherent in the “call option” itself.

Alternatively, petitioner contends that Deutsche Bank faced upside risk because it supposedly could “internalize” the basket’s long positions and lend those securities to other customers who wished to sell

the securities short. Petitioner hypothesizes a scenario in which

Deutsche Bank had outstanding loans of basket securities on the “option” expiration date. If Deutsche Bank were unable to replace the lent

securities with other securities in its inventory, and instead had to go

into the market to repurchase the securities, it could theoretically be at

risk from upward market movements in the interim.

[*62] We find this argument wholly unconvincing, for at least four reasons:

● As explained infra pp. 81–82, there is no evidence that Deutsche

Bank in fact lent to short sellers any shares held in any of the Barrier

Contract reference baskets. The contract indicates that Deutsche Bank

could not lend basket securities to short sellers unless GWA explicitly

consented, and GWA was free to withhold its consent.

● If Deutsche Bank were to lend shares held in basket long positions, the risk that it would be unable to replace those shares in timely

fashion would seem extremely small. Most basket securities were highly

liquid, and the investment guidelines limited the size of individual stock

positions. Prudent risk-management practices would dictate that

Deutsche Bank find replacement shares well before the Barrier Contract

expiration date. Petitioner’s experts made no effort to quantify this alleged risk or ascertain whether it was meaningful.

● To the extent Deutsche Bank incurred any risk from securities

lending, that risk was of its own making. The Barrier Contracts did not

require Deutsche Bank to lend basket securities. If it did so, that would

be a wholly unrelated business decision. Any risk it incurred thereby

had nothing to do with the risk (if any) inherent in the “call option.”

● The evidence established that Deutsche Bank’s London office

routinely derived income by lending securities held in its customers’

prime brokerage accounts. Most prime brokers engage in this practice.

See supra p. 10. Like any prime broker, Deutsche Bank thus bore a

theoretical risk that, on any given day, the customer would decide to

liquidate its long position, requiring Deutsche Bank to replace the securities before the closing date or come out of pocket for their cash value.

If Deutsche Bank did lend any basket securities, the risk it incurred

thereby was exactly the same risk that all prime brokers face when they

lend securities in their customers’ accounts. Needless to say, bearing

this risk supplies no evidence that the prime broker “owns” the securities in the customer’s account.

Downside Risk

The owner of stock bears downside risk—the risk that the shares

will decline in value. By writing a call option on his stock, the investor

secures a degree of protection from downside risk, to the extent of the

premium he receives. Returning to our example above, if Company A

stock closed at 90 on the option expiration date, the option would expire

[*63] worthless. The optionor would realize a $1,000 gain on the option,

which would precisely offset his $1,000 investment loss on the stock.

The optionor would be protected from net downside risk as long as the

stock did not close below 90, but he would remain exposed to the risk of

the stock’s declining below that price point.

The structure of the Barrier Contracts shows that Deutsche Bank

bore no cognizable downside risk with respect to the securities positions

in the reference baskets. That is because, in a loss scenario, the “option”

would terminate automatically, with the basket securities being converted into cash before the “premium” had been exhausted. The cash

plus the remaining “premium” would ensure that Deutsche Bank was

repaid in full for the capital it advanced to GWA.

The economics may be illustrated most easily if we simplify the

numbers somewhat. Assume that Deutsche Bank supplied capital of

$100X in exchange for a “premium” of $10X. If the NAV Index Level

declined to 97, Deutsche Bank would demand additional premium of

$3X. If GWA paid the additional premium, Deutsche Bank would retain

the $10X cushion with which it started (aggregate premium of $13X minus investment loss of $3X). The $10X cushion would continue to protect Deutsche Bank from downside risk.

If the NAV Index Level declined to 97 and GWA refused to pay

additional premium, the contract would terminate and liquidation of the

basket securities would begin. Assuming that liquidation of the basket

securities was completed by the time the NAV Index Level reached 94,

Deutsche Bank would get at least $94X in cash and would keep $6X of

premium, refunding $4X to GWA. Deutsche Bank would thus be repaid

$100X, the full amount of the capital it supplied for investment in the

reference basket.

Large securities portfolios, of course, cannot be liquidated instantaneously, and it was possible that the NAV Index Level might decline

below 94 before the reference basket was fully converted to cash. Suppose that it took several additional days in a brutal market to close out

all the positions, by which time the NAV Index Level had declined to 92.

Deutsche Bank would then get at least $92X in cash and would keep

$8X of premium, refunding $2X to GWA. Deutsche Bank would again

be repaid $100X, the full amount of the capital it supplied for investment

in the reference basket.

[*64] In each of these scenarios, Deutsche Bank would be insulated

from any downside risk on its $100X capital investment. In asserting

that Deutsche Bank nevertheless bore downside risk, petitioner urges

the possibility that, under extremely distressed market conditions, the

NAV Index Level might decline to (say) 88 before the securities in the

reference baskets could be reduced to cash. If that scenario were to occur, the premium would be fully exhausted, and Deutsche Bank would

face a loss of $2X (premium of $10X minus investment loss of $12X).

To assess the probability that this “nightmare scenario” might

happen in the real world, both parties offered testimony from expert witnesses. We found the testimony of respondent’s expert, Prof. Glasserman, most persuasive. Since 2000 he has held an endowed chair at Columbia Business School. He is the author of a treatise titled Monte Carlo

Methods in Financial Engineering, a widely used reference for valuing

derivative securities. He has written more than 100 articles in refereed

journals focusing on statistical and probabilistic methodologies for financial applications.

To calculate the possibility that Deutsche Bank would ever suffer

a loss on a Barrier Contract, Prof. Glasserman performed a “bootstrap

simulation methodology.” “Bootstrapping” is a widely used technique

for conducting statistical tests and analyzing the distributional properties of data. A “bootstrap simulation methodology” generates a large

number of potential paths for a securities portfolio by sampling returns

from the portfolio’s historical distribution of daily returns.

To implement this methodology Prof. Glasserman used GWA’s

trading data to compute daily returns on the NAV Index Level. He focused his analysis on Barrier Contract #1, which he determined to be

the riskiest of the 10 contracts. Because it was the riskiest, using it for

his analysis benefited petitioner.

Before being terminated, Barrier Contract #1 spanned a 6-year

period. That period included bouts of extremely volatile market conditions, including the 2007 “Quant Quake” and the 2008–2009 financial

crisis. Prof. Glasserman projected sample paths for the NAV Index

Level throughout the full 12-year contract term by using a technique

called “sampling with replacement.”

Prof. Glasserman performed bootstrap simulations under fourday and seven-day liquidation scenarios. He chose a four-day period because the Barrier Contracts specified a four-day averaging period for

[*65] securities settlements in the reference baskets. He chose an alternative seven-day period because, when GWA decided to terminate Barrier Contracts #7, #8, and #10, it told Deutsche Bank that liquidation of

the portfolios would likely take five or six days, so as to minimize market

disruptions. GWA’s prediction proved pessimistic: WMSA began liquidating those positions on May 17, 2010, and most of the securities (valued at $790 million) were successfully liquidated that same day. All

positions other than fractional shares were liquidated by May 19 (i.e.,

within three days). See supra pp. 38–39.

In his first set of simulations, Prof. Glasserman assumed that liquidation of the portfolio would begin when the NAV Index Level hit 97,

triggering an “early expiration notice” to GWA. See supra p. 25. If GWA

declined to pay additional premium, liquidation of the portfolio would

begin immediately. Thus, liquidation of the securities beginning at NAV

Index Level 97 was a very likely scenario.

Prof. Glasserman generated one million sample paths starting at

97, then counted how many paths ever reached 90. Assuming a fourday liquidation period, he found that only 35 of one million paths declined below 94, and that none declined below 90. Assuming a sevenday liquidation period, he found that the lowest NAV Index Level

reached by any path was 92.34, and that no path declined below 90.

Prof. Glasserman observed that the most extreme negative oneday return for Barrier Contract #1 over its six-year life was −1.74%. In

the highly unlikely event that GWA were to experience that maximum

negative return four days in a row, the total loss would be less than 7%.

Assuming the worst of all possible outcomes, therefore, the NAV Index

Level would not decline from 97 to 90 during a four-day period even if

no securities were liquidated.

As a “sanity check” on these findings, Prof. Glasserman used a

“Black-Scholes” model to estimate the likelihood that the NAV Index

Level would move from 97 to 90 in a 7-day period. He found the likelihood of this occurring to be essentially zero. As a further sanity check

he assumed a 30-day liquidation period, an extremely unlikely scenario.

He found that the NAV Index Level declined from 97 to 90 on between

[*66] 0.0001% and 0.0069% of the sample paths (i.e., between 1 and 69

times out of one million paths). 22

On the basis of these statistical results, Prof. Glasserman concluded that Deutsche Bank’s risk of loss on the Barrier Contracts was

“de minimis.” This conclusion is not surprising given the composition of

the reference basket. The investment guidelines reduced risk by requiring diversification of positions across numerous issuers, industries, and

economic sectors. The guidelines also limited the size of individual stock

positions. As Prof. Glasserman explained, this “helped ensure sufficient

liquidity to facilitate unwinding the portfolio, if necessary.”

Significantly, the positions in the reference baskets were mostly

hedged long/short positions. “For portfolios following a Long/Short

strategy,” Prof. Glasserman observed, “market-wide movements should

result in the long and short positions moving in opposite directions,

thereby reducing the risk and volatility associated with general market

moves.” Prof. Tufano agreed that these investment restrictions, in conjunction with the expiration barriers, “ensured that the likelihood of

[Deutsche Bank’s] incurring a loss was expected to be remote.”

Given the low-risk nature of the portfolio, Prof. Glasserman found

it very unlikely that the NAV Index Level would ever decline even as

low as 97, the point at which liquidation would begin if GWA did not pay

additional premium. Using “single-day” and “block” bootstrapping approaches, he found that the NAV Index Level declined to 97 on only 6.5%

to 8.9% of the million sample paths. “These results show that not only

was it implausible that Deutsche Bank had any risk of loss associated

with the Barrier Contracts, but also that the likelihood of even reaching

[NAV Index Level 97] was small.”

For these reasons, we conclude that Deutsche Bank bore no upside risk with respect to the Barrier Contracts and bore no cognizable

22 Prof. Glasserman performed another set of simulations in which he made

the unlikely assumption that liquidation of the portfolio would not begin until the NAV

Index Level had fallen all the way to 94, the point at which the “option” automatically

terminated. For this purpose he used two different types of simulation methodologies,

“single-day” and “block” bootstrapping. Block bootstrapping caters to the possibility

that short-term market trends may persist, and it thus samples “blocks” of consecutive

returns rath

This text is long and has been trimmed here. Open the source document for the complete record.

This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.