PRINCIPAL LIFE INSURANCE COMPANY AND SUBSIDIARIES v. USA
REPORTED OPINION denying 110 Motion for Partial Summary Judgment; granting, in part, 129 Cross-Motion for Partial Summary Judgment. Signed by Judge Francis M. Allegra. (si) Copy to parties.
In The United States Court of Federal Claims
Nos. 07-06T, 07-706T, 08-135T, & 08-605T
(Filed: May 9, 2014)
PRINCIPAL LIFE INSURANCE
COMPANY AND SUBSIDIARIES, et al.,
THE UNITED STATES,
Tax refund suit; Cross-motions for partial
summary judgment; Loss deduction –
section 165 of the Code; “Actual economic
loss”; Basis allocation – Treas. Reg. § 1.616(a); “Carved-out income interests”;
Allocation of basis in “Perpetuals”
transaction violated Treasury Regulation;
“Investment trusts”; “Sears Regulations” –
Treas. Reg. § 301.7701-4; Multi-class
“investment trust” not “trust” for tax
purposes; Loss disallowed; Custodial
arrangements and investment trusts not
“trusts” for tax purposes; Income from
Custodial Share Receipts includible in
Jay H. Zimbler, Sidley Austin, LLP, Chicago, IL, for plaintiffs.
Bart Duncan Jeffress, United States Department of Justice, Washington, D.C., with
whom was Acting Assistant Attorney General David A. Hubbert, for defendant.
“[T]he tax could not be escaped by anticipatory arrangements and contracts
however skillfully devised . . . by which the fruits are attributed to a
different tree from that on which they grew.” 1
Before the court, on cross-motions for partial summary judgment, is the next leg of this
complex tax refund suit.2 At issue is the tax treatment of two distinct, but structurally-similar,
Lucas v. Earl, 281 U.S. 111, 115 (1930) (Holmes, J.).
series of investments made by Principal Life Insurance Company and Subsidiaries (PLIC). Eight
of these investments related to so-called “custodial share receipts” or “CSRs,” while three others
involved so-called “perpetual securities” or “Perpetuals.” The CSRs and Perpetuals were similar
in that both sets of transactions involved carving out an interest in future income payments from
a security. In the case of the CSRs, those interests were carved out by a third party, which sold
the residual equity interest to PLIC. In the case of the Perpetuals, PLIC itself bought the
securities, carved out and retained the income interest, and then sold the residual principal
interests to a third party.
Springing from these transactions, so PLIC claims, are twin tax benefits: the exclusion of
approximately $21 million in income on the CSRs, and an approximately $291 million loss
deduction generated by the sale of the Perpetuals principal interests. PLIC asserts that its
treatment of these items was impelled by the relevant provisions of the Internal Revenue Code of
1986, 3 and associated “common law” principles. Not so, defendant remonstrates. It asserts that
PLIC was required to report income from the CSRs, which it claims involved partnerships under
the controlling Treasury regulations. It further argues that PLIC’s claimed loss deduction stems
from a gross misapplication of the loss provisions of section 165 of the Code and the rules for
calculating the adjusted basis of the certificates sold. For the reasons that follow, the court
concludes that defendant is right.
A recitation of the underlying facts sets the context for this decision.
PLIC, an Iowa corporation with principal offices in Des Moines, is engaged, and at all
times relevant to this action, was engaged, in the business of writing various forms of individual
and group life and health insurance and annuities. During the years in question (1996-2001), it
filed consolidated returns as the parent corporation of a consolidated group of corporations.
During these years, and at all times relevant to this action, PLIC was a calendar-year, accrualbasis taxpayer subject to tax under the provisions of Subchapter L of the Code.
The CSRs. Between September 1996 and October 2001, PLIC purchased residual
interests in money market mutual fund shares from six separate sellers in eight separate
transactions. PLIC engaged a variety of entities to act as custodians or trustees associated with
these transactions, including Wilmington Trust Company, the United States Trust Company of
New York, the First Union Trust Company, and Chase Manhattan Bank.
See Principal Life Ins. Co. & Subs. v. United States, 95 Fed. Cl. 786 (2010); Principal
Life Ins. Co. v. United States, 70 Fed. Cl. 144 (2006).
All references herein are to the Internal Revenue Code of 1986 (26 U.S.C.), as
amended and in force during the years in question.
In these transactions, PLIC acquired a residual interest in each investment that entitled it
to all dividends, appreciation, and voting rights in the specified shares, except for dividends paid
out on the shares during a prescribed time period.
These transactions took one of three forms:
Six of these investments used a custodial arrangement, in which an
unrelated financial institution (the Depositor) transferred money market
fund shares (or the money to buy such shares) to a custodian. In return,
the custodian issued to the Depositor both CSRs and Custodial Dividend
Receipts (CDRs). The Depositor then sold the CSRs to PLIC.
One of the investments employed a trust in place of the custodial
arrangement. In this instance, the Depositor deposited the money market
shares directly into a trust, which issued Dividend and Corpus Certificates.
The Depositor then sold the Corpus Certificates to PLIC.
The eighth and final investment employed two trusts. The Depositor
deposited money market fund shares (or the money to buy such shares)
into a trust in exchange for Principal and Dividend Certificates. The
Principal Certificates were transferred to a second trust in exchange for a
“Principal Unit” and a “Termination Unit.” PLIC then purchased the
In each of these transactions, the Depositor, i.e., the financial institution, retained a carved-out
income interest in the underlying money market shares. Via that interest, the Depositors were
entitled to all dividends paid in connection with the money market shares for a prescribed period
of between 20 and 23 years (the Restricted Period). At the end of the Restricted Period, the
Depositors had no future right to the shares. The interests purchased by PLIC represented a
residual interest in the money market shares and entitled PLIC, after the expiration of the
Restricted Period, to either the money market shares or any amount paid with respect to those
To guard against the possibility that the money market shares held in the custodial
arrangement would be prematurely redeemed or lose their money market status, PLIC entered
into a “Termination Agreement” with each Depositor. Under this agreement, upon the
occurrence of an adverse event, PLIC was required to purchase the CDRs (or the equivalent)
from the Depositor at a price designed to prevent the Depositor from losing the value of the
dividends for the Restricted Period. Absent bad faith, nothing in any other agreement created
obligations running from the Depositor to PLIC. Nonetheless, PLIC had the right to terminate
certain of the custodial arrangements at any time, and take possession of the corresponding
money market shares, as long as it provided the Depositors with a valid, perfected, first-priority
security interest in the shares.
The annual internal economic yields for the eight CSR investments, before taxes, ranged
from 7.225 to 9.903 percent. PLIC claimed that in addition to the expected yields, the CSRs
presented attractive benefits for its long-term portfolio. For example, in contrast to holding
money market shares directly, the CSRs did not present reinvestment risk, in that the yields built
up internally at a fixed rate without creating cash flows that might have had to be reinvested at
lower market interest rates. As a result, PLIC asserts that they could be confident that its initial
investments in the CSRs would yield fixed amounts at pre-specified times in the future, at which
time that income would be needed to satisfy specific long-term liabilities such as payouts under
life insurance policies. On its returns for its tax years 1999 through 2001, PLIC reported no
income from the CSRs.
The Perpetuals. PLIC entered into three Perpetuals investments in 2000 and 2001.
These transactions were known as “Asgard,” “Seve,” and “Evergreen,” named after the primary
trusts involved. As will be seen, Morgan Stanley & Co. (Morgan Stanley) was involved in each
of the Perpetuals transactions in a variety of ways. The Perpetuals were similar to the CSRs, in
that PLIC retained the carved-out interests in the underlying perpetual securities while selling the
residual equity interests. The CSR and Perpetuals transactions differed, however, in that in the
former the financial institutions stripped out the carved-out income interests from the residual
interests, while in the latter, PLIC did so. Each of the Perpetuals transactions involved a doubletrust structure similar to that employed in the eighth CSR transaction described above.
The Perpetuals transactions were similar in structure. At the inception of each
transaction, PLIC engaged an investment banker – Morgan Stanley – to buy a portfolio of eight
to ten perpetual floating-rate securities from third parties in the secondary market assertedly at
arm’s length prices. Morgan Stanley then sold the securities to PLIC, earning a spread on the
transaction. PLIC held the securities in its portfolio for a relatively brief period of time (one to
two months). At the end of this holding period, on the “Transaction Date,” PLIC deposited the
securities into a “Primary Trust.” Chase Manhattan Bank was the trustee of each of these
Primary Trusts. The Primary Trust issued to PLIC a series of “Interest Certificates” and
“Principal Certificates.” Each of the Interest Certificates entitled the holder to the interest paid
on the underlying perpetual security from the inception of the Primary Trust to a specified
“Redemption Date,” 16 to 18 years after the Transaction Date, unless a defined “Reference
Event” occurred. The Principal Certificates entitled the holder to the underlying perpetual
security on the Redemption Date and any other distributions or payments received by the
Primary Trust, other than the interest payable to the Interest Certificate holder. On the
Transaction Date, PLIC sold the Principal Certificates to Morgan Stanley; it retained the Interest
To protect itself against the risk of a Reference Event occurring before the Redemption
Date, PLIC entered into Termination Agreements with Morgan Stanley Credit Products, Ltd.
(MSCPL), a Cayman entity. The Termination Agreements were documented through a master
agreement between PLIC and MSCPL; MSCPL’s ultimate parent, Morgan Stanley Dean Witter,
guaranteed its subsidiary’s obligations under the master agreement. Under the Termination
Agreements, PLIC was obligated to pay MSCPL a one-time premium. 4 In exchange, MSCPL
became obliged to pay PLIC a calculated amount upon the occurrence of a Reference Event.
There were three Reference Events under the agreements: (i) a default with respect to the
perpetual securities; (ii) a bankruptcy, insolvency, liquidation or winding up of the perpetual
security issuer; or (iii) a mandatory redemption of the perpetual securities at the election of the
issuer. 5 PLIC admits that it likely would not have entered into the Perpetual transactions without
the Termination Agreements (or at least would have done so at a different price).
On each of the Transaction Dates, Morgan Stanley deposited the Principal Certificates
into an “Issuer Trust” and, in exchange, received two units – a Principal Unit and a Termination
Unit. The Termination Units entitled the holder to a declining percentage of any payments
received by the Issuer Trust during a time period specified in the Trust Agreements. The
Principal Units entitled the holder to all payments received by the Issuer Trust in connection with
the Principal Certificates, other than the payments required to be made to the Termination Units.
Morgan Stanley sold the Principal Units to unrelated financial institutions. Morgan Stanley
transferred the Termination Units to its affiliate, Morgan Stanley International Ltd., a United
The Perpetuals produced a positive yield while having a “negative duration.” The latter
refers to the price sensitivity an asset has to changes in market interest rates. An asset has a
“negative duration,” for this purpose, when its rate of return floats with the market. The Interest
Certificates were projected to yield a before-tax economic return from 6.059 to 7.517 percent per
annum. In each of these transactions, PLIC allocated all its tax basis in each underlying
perpetual security to the corresponding Principal Certificate – even though the Interest
Certificate reflected 80 percent of the cost of the overall security. As a result, when it sold the
Principal Certificates, PLIC claimed a capital loss equal to the difference between its basis in the
Principal Certificates and the price at which PLIC sold them. Under this approach, PLIC
Defendant notes that, at closing, PLIC paid the premium owed under the three
Termination Agreements ($45,000 for Asgard/Valhalla, $80,000 for Seve/Olivia, and $50,000
for Evergreen/Holly), but that Morgan Stanley then paid these amounts back to PLIC as part of
the purchase price of the Principal Certificates.
The Primary Trust agreements defined a “Reference Event” more specifically, as
(a) a Default with respect to such Notes;
(b) a bankruptcy, insolvency, liquidation or winding up (or other similar event)
with respect to the Note Issuer (excluding any merger or non-bankruptcy related
corporate reorganization of such issuer); or
(c) a mandatory redemption of all outstanding Notes prior to the applicable
Redemption Date, for tax reasons or otherwise at the election of the Note Issuer
pursuant to the terms of the underlying trust deed or any other instrument
governing the Notes (a “Call Event”).
recognized ordinary income from the Interest Certificates, without any reduction for basis
recovery. Owing to this practice, PLIC claimed loss deductions totaling approximately $291
million on its tax returns for 2000 and 2001.
On December 29, 2004, the Internal Revenue Service (IRS) mailed a statutory notice of
deficiency to PLIC for its tax years 1996 through 2001. 6 The notice disallowed the
approximately $291 million in capital losses claimed by PLIC with respect to the three
Perpetuals. It also determined that PLIC was taxable on approximately $21 million on CSRrelated income for its taxable years 1999-2001. Overall, the notice determined that PLIC owed
additional tax and penalties for its tax years in the amounts of $362,030,347 (tax) and
$27,377,423 (penalties). On May 27, 2005, the IRS assessed the deficiencies determined in the
notice of deficiency. Thereafter, taxpayer paid the assessed amounts and filed refund claims
seeking the refund of income taxes and penalties attributable to the adjustments made by the IRS.
After receiving a notice of partial disallowance of its claims, PLIC filed a tax refund suit
in this court on January 4, 2007. Subsequently, the court, at the parties’ request, broke this case
into a series of tranches, allowing decisions on certain issues to proceed while discovery is
occurring on other issues. On July 23, 2012, PLIC filed a motion for partial summary judgment
as to one of those tranches, that involving the Perpetuals and CSR transactions. 7 On November
While the notice of deficiency in the record relates to PLIC’s taxable years 1996
through 2000, it appears from that notice that PLIC’s taxable year 2001 was also at issue and the
parties have proceeded on that assumption.
Prior to filing its summary judgment motion, PLIC, on December 14, 2011, filed a
motion to compel the production of certain documents by defendant relating to the Perpetuals. It
claimed that the IRS had improperly withheld various internal documents that addressed the
same subjects as documents that had been disclosed, namely, internal IRS documents discussing
the reasonableness of PLIC’s position with respect to the Perpetuals transactions. On December
22, 2011, the court denied PLIC’s motion, indicating that “[a]n opinion explaining the basis for
this ruling will follow.”
Federal Rule of Evidence 502(a) provides that when a disclosure is made in a federal
proceeding with a waiver of the attorney-client privilege, “the waiver extends to an undisclosed
communication or information . . . only if” the waiver is intentional, the disclosed and
undisclosed communications or information concern the same subject matter, and “they ought in
fairness to be considered together.” Fed. R. Evid. 502(a). This waiver “is limited to situations in
which a party intentionally puts protected information into the litigation in a selective,
misleading and unfair manner.” Fed. R. Evid. 502(a), adv. comm. notes 2007. A party that
makes such a presentation “opens itself to a more complete and accurate presentation.” Id.
While “[t]here is no bright line test for determining” when this waiver applies, Eden Isla
Marina, Inc. v. United States, 89 Fed. Cl. 480, 503 (2009), it is safe to assume that a party may
not compel the production of materials that are irrelevant simply because the other side has
8, 2012, defendant filed a cross-motion as to these issues, expanding the scope of the motion to
include penalties. Oral argument on the cross-motions was held on November 12, 2013.
We begin with common ground. Summary judgment is appropriate when there is no
genuine dispute as to any material fact and the moving party is entitled to judgment as a matter
of law. See RCFC 56; Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 247-48 (1986). Disputes
over facts that are not outcome-determinative will not preclude the entry of summary judgment.
Id. at 248. However, summary judgment will not be granted if “the dispute about a material fact
is ‘genuine,’ that is, if the evidence is such that a reasonable [trier of fact] could return a verdict
for the nonmoving party.” Id.; see also Matsushita Electric Indus. Co., Ltd. v. Zenith Radio
Corp., 475 U.S. 574, 587 (1986); Becho, Inc. v. United States, 47 Fed. Cl. 595, 599 (2000).
When making a summary judgment determination, the court is not to weigh the evidence,
but to “determine whether there is a genuine issue for trial.” Anderson, 477 U.S. at 249; see also
Agosto v. INS, 436 U.S. 748, 756 (1978) (“a [trial] court generally cannot grant summary
judgment based on its assessment of the credibility of the evidence presented”); Am. Ins. Co. v.
United States, 62 Fed. Cl. 151, 154 (2004). The court must determine whether the evidence
presents a disagreement sufficient to require fact finding, or, conversely, is so one-sided that one
party must prevail as a matter of law. Anderson, 477 U.S. at 251-52; see also Ricci v. DeStefano,
557 U.S. 557, 586 (2009) (“‘Where the record taken as a whole could not lead a rational trier of
fact to find for the nonmoving party, there is no genuine issue for trial.’” (quoting Matsushita,
475 U.S. at 587)). Where there is a genuine dispute, all facts must be construed, and all
inferences drawn from the evidence must be viewed, in the light most favorable to the party
opposing the motion. Matsushita, 475 U.S. at 587-88 (citing United States v. Diebold, Inc., 369
U.S. 654, 655 (1962)); see also Stovall v. United States, 94 Fed. Cl. 336, 344 (2010); L.P.
Consulting Grp., Inc. v. United States, 66 Fed. Cl. 238, 240 (2005). Where, as here, a court
already disclosed related materials that are also irrelevant. The internal IRS documents PLIC
sought – and the internal IRS documents already produced that were the basis for PLIC’s motion
– are both irrelevant as far as the court is concerned and, hence, are not subject to discovery. See
RCFC 26(b)(1); see also Pinkard v. Baldwin Richardson Foods, Inc., 2013 WL 1308713, at *8
(W.D.N.Y. Mar. 28, 2013) (fairness did not require production of witnesses’ report where
defendant did not “proffer any reason to believe that any other portion of the report, let alone
the entirety of the report, is relevant”). Tax refund cases are de novo proceedings. Lewis v.
Reynolds, 284 U.S. 281, 283 (1932). “As such, this court’s determination of plaintiff’s tax
liability must be based on facts and merits presented to the court and does not require (or even
ordinarily permit) this court to review findings or a record previously developed at the
administrative level.” Vons Cos., Inc. v. United States, 51 Fed. Cl. 1, 6 (2001). Why a given
IRS employee felt that penalties should be imposed on PLIC has nothing to do with this
considers cross-motions for (partial) summary judgment, it must view each motion, separately,
through this prism. 8
PLIC claims a loss deduction, pursuant to section 165(a) of the Code, allegedly generated
upon its sale of the Principal Certificates associated with the three Perpetuals transactions. It
also seeks to exclude from taxable income the income generated on the eight CSR transactions.
Defendant objects on both counts, contending that PLIC was entitled neither to the loss
deduction claimed, nor the income exclusion it seeks. Although the CSR transactions somewhat
predate the Perpetuals, the court, for reasons that will become obvious, will deal with the proper
tax treatment of the latter first.
PLIC argues that the losses it claims on the Perpetual transactions were deductible on its
2000 and 2001 returns under section 165 of the Code. That section allows for a deduction for
“any loss sustained during the taxable year and not compensated for by insurance or otherwise.”
26 U.S.C. § 165(a). 9 To qualify for this deduction, a taxpayer must prove the existence and
amount of the claimed loss. Boehm v. Comm’r of Internal Revenue, 326 U.S. 287, 294 (1945);
Stivers v. Comm’r of Internal Revenue, 360 F.2d 35, 40 (6th Cir. 1966); Goeller v. United States,
109 Fed. Cl. 534, 539 (2013). While defendant offers a host of reasons why the losses claimed
here are not deductible, its primary assertions are that: (i) the losses claimed were not “bona
fide” within the meaning of section 165 and the related Treasury Regulations, as they did not
relate to an “actual economic loss;” (ii) PLIC failed to calculate its loss properly because, in
subdividing the perpetual securities, it wrongly allocated all of its costs to the tax basis of the
Principal Certificates; and (iii) the transactions involving the Perpetuals produced no losses
because, under the so-called “Sears Regulations,” they led to the creation not of trusts, but of
partnerships. The court will consider these claims, and PLIC’s responses thereto, seriatim.
See Chevron U.S.A. Inc. v. Mobil Producing Tex. & N.M., 281 F.3d 1249, 1252-53
(Fed. Cir. 2002); see also Estate of Hevia v. Portrio Corp., 602 F.3d 34, 40 (1st Cir. 2010);
Travelers Prop. Cas. Co. of Am. v. Hillerich & Bradsby Co., Inc., 598 F.3d 257, 264 (6th Cir.
2010); Northrop Grumman Computing Sys., Inc. v. United States, 93 Fed. Cl. 144, 148 (2010).
The deduction for losses has one of the oldest lineages of any provision in the Code.
The first section authorizing such a deduction was included in the Revenue Act of 1867, which
allowed as a deduction “losses actually sustained during the year arising from fires, shipwreck,
or incurred in trade.” Revenue Act of 1867, ch. 169, § 13, 14 Stat. 471, 477 (1867). That Act
famously was declared unconstitutional in Pollock v. Farmers’ Loan & Trust Co., 157 U.S. 429
(1895), leading to the passage of the Sixteenth Amendment. Following that amendment,
language similar to the current section 165(c) was added to the statute in the Revenue Act of
1916, ch. 463, § 5(a)(4), 39 Stat. 756 (1916). See John Seidman, Seidman’s Legislative History
of the Federal Income Tax Laws (1938-1861) 962-63 (1938); see also Ambrose v. United States,
106 Fed. Cl. 152, 156 n.5 (2012).
Defendant first contends that by allowing a deduction only for “any loss sustained,”
section 165(a) requires that there be an “actual economic loss” before a deduction is permitted.
Treasury Regulation § 1.165-1(b) states that “[o]nly a bona fide loss is allowable,” adding that
“[s]ubstance and not mere form shall govern in determining a deductible loss.” See also Cottage
Sav. Ass’n v. Comm’r of Internal Revenue, 499 U.S. 554, 567-68 (1991). But, this regulation
does not, in so many words, say that only “actual economic losses” are deductible. Nor, for that
matter, does it define how one determines whether a loss is “bona fide.” See Cottage Sav., 499
U.S. at 568; Higgins v. Smith, 308 U.S. 473, 475-76 (1940); see also Marvin A. Chirelstein &
Lawrence A. Zelenak, “Tax Shelters and the Search for a Silver Bullet,” 105 Columb. L. Rev.
1939, 1952 n.45 (2005). In asserting that there is, nonetheless, an overarching “actual economic
loss” requirement for deductibility under section 165, defendant offers two cases – United States
v. Flannery, 268 U.S. 98 (1925) and Centex Corp. v. United States, 395 F.3d 1283 (Fed. Cir.
2005). But, the court is unconvinced these cases support defendant’s claims.
In Flannery, the taxpayer purchased stock prior to March 1, 1913, and sold it in 1919 for
more than its cost, but for less than its value on March 1, 1913. Section 202(a)(1) of the
Revenue Act of 1918, 40 Stat. 1060, stated that in computing loss on the sale of property
acquired prior to March 1, 1913, the fair market value as of that date should be used to calculate
gain or loss. Although the taxpayer actually realized a profit upon the sale in 1919, he claimed a
deductible loss because the sale price was less than his basis, i.e., the March 1, 1913, value of the
stock. The Supreme Court held, however, that the special basis provisions for property acquired
prior to March 1, 1913, could not be utilized to provide a deduction on account of loss. This was
because “the Act of 1918 imposed a tax and allowed a deduction to the extent only that an actual
gain was derived or an actual loss sustained from the investment,” with the reference to market
value on March 1, 1913, constituting “a limitation upon the amount of the actual gain or loss that
would have otherwise have been taxable or deductible.” 268 U.S. at 103; see also McCaughn v.
Ludington, 268 U.S. 106, 107 (1925) (Flannery held “that the Act allowed a deduction to the
extent only that an actual loss was sustained from the investment, as measured by the difference
between the purchase and sale prices of the property”).
Flannery is among an assortment of Supreme Court decisions in the 1920s and early
1930s construing section 202(a)(1) (or its predecessor provision, the Revenue Act of 1916, §
5(a), 39 Stat. 756). These cases presented varying fact patterns concerning how section
202(a)(1) should be applied where the cost of the asset sold varied from its fair market value on
March 1, 1913. Some of them focused on whether a transaction yielded a deductible loss, while
others focused on whether a transaction resulted in a taxable gain. 10 Curiously, Flannery has not
See Burnet v. Houston, 283 U.S. 223, 227 (1931) (extent of deductible loss, if any, on
disposition of stock acquired in 1906 and sold in 1920, was to be measured against lower of fair
market value of stock on March 1, 1913 and price of stock when acquired); Lucas v. Alexander,
279 U.S. 573 (1929) (taxpayer liable for gain equal to value of insurance policy upon disposition
minus value of policy in 1913); United States v. Ludey, 274 U.S. 295 (1927) (taxpayer liable for
gain on the sale of oil wells and equipment, resulting from application of depreciation rules,
been cited by the Supreme Court since 1931 – which is particularly puzzling if, as defendant
contends, that case inaugurated a broad and enduring rule governing the nondeductibility of
losses. 11 So why have precedents like Flannery fallen into disuse, even in the face of many
cases involving tax-advantaged transactions driven by losses? As is often the case in tax law, the
answer is multi-faceted.
To start, a careful reading of these cases suggests that they did not enunciate some broad
rule of nondeductibility, but instead focused, more narrowly, on the transition provisions passed
by Congress to phase in the modern income tax, provisions like section 202(a)(1) of the Revenue
Act of 1918. The Court interpreted these transition provisions consistent with their limited
purpose – to avoid the Constitutional problems that would have arisen had Congress taxed gains
accruing prior to the passage of the Sixteenth Amendment – and refused to allow taxpayers to
reap additional tax benefits through transactions that bridged the transitional period. Other cases
decided around this time recognized that the Supreme Court’s focus was limited and refused to
treat its opinions as somehow supplanting the Code’s provisions for calculating gain and loss.
See Basch v. Comm’r of Internal Revenue, 30 B.T.A. 305, 306-07 (1934) (finding that Flannery,
McCaughn and Burnet, were “not in point here” and applying instead the provisions of the
Revenue Act of 1926); see also Rands, Inc. v. Comm’r of Internal Revenue, 34 B.T.A. 1094,
1104 (1936). So held this court’s predecessor, in Davison v. United States, 6 F. Supp. 236, 239
(Ct. Cl. 1934), where Judge Green, commenting on Ludey, wrote that “[t]he decision . . . has
sometimes been treated as if it had ignored the provisions of the statute which require the basis to
be the March 31, 1913, value, but the language used in the opinion in this case as well as that in
[Flannery] and in the companion case, [McCaughn], should, as we think, be considered only as
applying each instance to the case then before the court.” See also Pfleghar Hardware Specialty
Co. v. Blair, 30 F.2d 614, 618 (2d Cir. 1929).
notwithstanding that sale price was lower than purchase price); McCaughn v. Ludington, 268
U.S. 106, 107 (1925) (taxpayer, upon disposing of stock in 1919, could deduct the value of the
stock less its purchase price, rather than the value of the stock less its price on March 1, 1913);
Walsh v. Brewster, 255 U.S. 536, 537-38 (1921) (first transaction: upon disposition in 1916,
taxpayer did not have gain when the sale price was the same as the 1909 purchase price, even
though the price on March 1, 1913 was lower than the sale and purchase prices; second
transaction: upon disposition in 1916, taxpayer had gain in the amount of the difference between
sale price and purchase price, notwithstanding that March 1, 1913 price was lower than either
sale or purchase price); see also Goodrich v. Edwards, 255 U.S. 527 (1921); Pioneer Cooperage
Co. v. Comm’r of Internal Revenue, 53 F.2d 43, 44 (8th Cir. 1931), cert. denied, 284 U.S. 686
Some of the other cases cited above dealing with the 1916 and 1918 Revenue Acts
have been cited by the Supreme Court more recently, but not for any proposition remotely like
the one defendant posits. See Steel Co. v. Citizens for a Better Env’t, 523 U.S. 83, 134 (1998)
(citing Lucas, 279 U.S. at 577, for the doctrine of constitutional avoidance).
- 10 -
Now Flannery might be viewed as presaging the anti-abuse doctrines enunciated by the
Supreme Court in the late 1930s, in now-famous cases like Gregory v. Helvering, 293 U.S. 465,
469-70 (1935); Minn. Tea Co. v. Helvering, 302 U.S. 609, 613 (1938); and Griffiths v. Helvering,
308 U.S. 355, 357 (1939). 12 But, the very existence of these familiar precedents (none of which
cites Flannery, by the way) casts doubt on the notion that, a decade or so earlier, the Court
established a far broader anti-abuse doctrine, i.e., that only “actual economic losses” are
deductible. Why craft these other anti-abuse doctrines – substance over form; sham transaction;
step transaction – if this field were already superintended by Flannery, particularly since a few of
the Helvering cases also involved loss deductions? See, e.g., Griffiths, 308 U.S. at 357; see also
Helvering v. Southwest Consol. Corp., 315 U.S. 194 (1942). At the least, one must surmise that
the Supreme Court felt that the refinements that came with these later anti-abuse doctrines were
important. True, like defendant’s rule, these anti-abuse doctrines focus on “‘economic realities,
not legal abstractions,’” PPL Corp. v. Comm’r of Internal Revenue, 133 S. Ct. 1897, 1905 (2013)
(quoting Comm’r of Internal Revenue v. Southwest Exploration Co., 350 U.S. 308, 315 (1956)),
and admittedly do so with a degree of imprecision. 13 But, they are neither as sweeping, nor as
fuzzily-defined, as the rule defendant would have this court “distill” from Flannery. 14
Indeed, in its zeal to prevent the perceived abuse of the loss deduction, defendant glosses
over the difficulties associated with determining whether a given transaction results in an “actual
economic loss.” Congress, after all, has supplied a myriad of rules on this subject, which belie
the notion that the deductibility of a loss as to a given asset turns simply on subtracting the price
received from the price paid. This is particularly true where the asset involved is wasting, e.g.,
amortizable or depreciable, and, especially so, where, as here, the asset originally acquired is
later subdivided (and part sold).
For these reasons, neither Flannery nor its progeny should be regarded as laying down
any overarching “actual economic loss” rule that governs, as a matter of law, the deductibility of
losses under section 165(a). Centex – the other case upon which defendant relies – certainly is
not to the contrary. As an aside, Centex was not a tax refund suit, but a Winstar-type, breach of
Guy Tresillian Helvering was appointed Commissioner of Internal Revenue by
President Franklin D. Roosevelt in 1933, and served in that capacity until he was appointed to
the federal bench in 1943.
See, e.g., Frank Lyon Co. v. United States, 435 U.S. 561, 584 (1978) (economic
substance doctrine); Stewart v. Comm’r of Internal Revenue, 714 F.2d 977, 988 (9th Cir. 1983)
(substance-over-form doctrine); King Enters., Inc. v. United States, 418 F.2d 511, 516 (Ct. Cl.
1969) (step transaction doctrine).
In H.J. Heinz Co. & Subs. v. United States, 76 Fed. Cl. 570, 580 (2007), this court
observed that “while these various doctrines overlap, they also have different criteria that bring
into relief the nuances of various transactions, as well as the importance of particular features
therein.” See also Lewis R. Steinberg, “Form, Substance and Directionality in Subchapter C,”
52 Tax Law. 457, 458 n.8, 499 (1999); Marvin A Chirelstein, “Learned Hand’s Contribution to
the Law of Tax Avoidance,” 77 Yale L.J. 440, 472 (1968).
- 11 -
contract case, 15 in which the Federal Circuit reviewed this court’s damage calculations. There,
the appellate court agreed that “the ordinary rule [is] that a loss deduction may not be taken in
the absence of actual economic loss.” 395 F.3d at 1293. But, it hastened to add that there are
exceptions to this “ordinary rule.” Id. Indeed, it held that it was “far from clear that those
general principles” applied to the loss deductions in question, which the court held were
controlled by “special tax rules” that Congress had enacted during the 1980s. Id. at 1294.
Ultimately, the Federal Circuit “reject[ed] the government’s submission that this case can be
resolved simply by applying the general rule that reimbursed losses do not give rise to the right
to take a deduction.” Id. at 1294-95. Hence, in Centex, the Federal Circuit eschewed the very
rule that defendant would have this court apply here – to wit, that a taxpayer must always end up
poorer as a precondition to taking a loss deduction. 16
To be fair, Centex is not alone in essaying that tax losses, to some degree, should
correspond to “actual” or “genuine” economic losses. See ACM Partnership v. Comm’r of
Internal Revenue, 157 F.3d 231, 252 (3d Cir. 1998), cert. denied, 526 U.S. 1017 (1999) (“Tax
losses such as these . . . which do not correspond to any actual economic losses, do not
constitute the type of ‘bona fide’ losses that are deductible under the Internal Revenue Code and
regulations.”); Scully v. United States, 840 F.2d 478, 486 (7th Cir. 1988) (to be deductible, a loss
must be a “genuine economic loss”); Shoenberg v. Comm’r of Internal Revenue, 77 F.2d 446,
448 (8th Cir. 1935), cert. denied, 296 U.S. 586 (1935) (to be deductible, a loss must be “actual
and real”). But, the question is – to what degree? Tellingly, all these cases share an important
feature with Centex – the “economic loss” phraseology ultimately played no role in the
decisions. Instead, all these cases ultimately determined whether a loss was deductible by
employing the loss calculation rules of the Code (and the regulations thereunder), supplemented,
See United States v. Winstar, 518 U.S. 839 (1996).
Several examples serve to illustrate why defendant’s “actual economic loss” concept is
overly simplistic. For example, section 1014(a) of the Code provides generally for a stepped-up
basis for property transferred from a decedent. 26 U.S.C. § 1014(a). Such a basis can far exceed
the original cost of acquiring the property. Yet, there is no doubt that if upon a sale or
disposition, the amount realized is less than the stepped-up basis, the seller may be entitled to a
loss deduction even though in actual economic terms the property was sold at an extraordinary
gain. How can this be? It is a simple matter of applying the Code by its terms. See Janis v.
Comm’r of Internal Revenue, 469 F.3d 256 (2d Cir. 2006) (discussing how these rules work).
And similar results occur in various other settings where the adjusted basis of an asset is not its
historical cost, such as like-kind exchanges under 26 U.S.C. § 1031, or gifts under 26 U.S.C. §
1015(a). One must wonder whether defendant has considered the ramifications if its “actual
economic” rule were applied to gains taxable under section 61 of the Code. After all, the
Flannery line of cases emphasized that the transition rules were a two-way street, applicable to
losses as well as gains. What if, under such an approach, a taxpayer argued that no income is
produced on the sale of a capital asset if the appreciation of that asset has not kept pace with
inflation? In the latter instance, there is no “actual economic” gain, yet the IRS undoubtedly
would take the position that if the amount realized exceeds the asset’s tax basis, gain, for
purposes of section 61 of the Code, is produced.
- 12 -
as necessary, by the traditional anti-abuse doctrines described above. See ACM Partnership, 157
F.3d at 260; Scully, 840 F.2d at 484; Shoenberg, 77 F.2d at 448-49. This approach makes good
sense as it recognizes that the Code and the application of the anti-abuse doctrines are not
“‘independent’ of one another.” United States v. Woods, 134 S. Ct. 557, 567 (2013). And that
approach should be ours, too.
So what are the statutory and regulatory rules that apply here? The Code generally takes
account of increases and decreases in the value of property only when gains or losses are realized
and recognized. See John Mertens, 4 Mertens Law of Fed. Income Tax’n § 22:7 (2014)
(hereinafter “Mertens”). The realization requirement is found in section 1001(a) of the Code,
which provides that –
The gain from the sale or other disposition of property shall be the excess of the
amount realized therefrom over the adjusted basis provided in section 1011 for
determining gain, and the loss shall be the excess of the adjusted basis provided in
such section for determining loss over the amount realized.
26 U.S.C. § 1001(a); see also Treas. Reg. § 1.1001-1(a) (“Except as otherwise provided in
subtitle A of the Code, the gain or loss realized from the conversion of property into cash, or
from the exchange of property for other property differing materially either in kind or in extent,
is treated as income or as loss sustained”); see also Cottage Sav. Ass’n, 499 U.S. at 559. The
recognition requirement derives from Section 1001(c) of the Code, which states that “[e]xcept as
otherwise provided in this subtitle, the entire amount of the gain or loss, determined under this
section, on the sale or exchange of property shall be recognized.” 26 U.S.C. § 1001(c); see also
Treas. Reg. § 1.1002-1(a) (“The general rule with respect to gain or loss realized upon the sale or
exchange of property as determined under section 1001 is that the entire amount of such gain or
loss is recognized except in cases where specific provisions of subtitle A of the code provide
To apply these provisions to a sale or exchange, one must determine a property’s
“adjusted basis.” Section 1012 defines “basis” of property as “the cost of such property, except
as otherwise provided in this subchapter and subchapters C . . ., K . . ., and P . . .” 26 U.S.C. §
1012. Section 1011(a) defines “adjusted basis” as “the basis (determined under section 1012
. . .), adjusted as provided in section 1016.” Id. at § 1011(a). Section 1016 then lists a lengthy
Like section 165, section 1001 of the Code has a long lineage. See Revenue Act of
1924, ch. 234, § 203(a), 43 Stat. 256.; H.R. Rep. No. 179, at 13 (1924); S. Rep. No. 398, at 14
(1924). The statutory distinction between realization and recognition was, perhaps, easier to see
in the 1954 version of the Code, in which the recognition rule of present-day section 1001(c)
stood apart in its own section of the Code, Section 1002. The provisions of old section 1002
became section 1001(c) in 1976. Tax Reform Act of 1976, Pub. L. No. 94-455, Tit. XIX, §
1901(a)(121), 90 Stat. 1784. This amendment was not intended to have any substantive effect.
See S. Rep. No. 938, at 512 (1976).
- 13 -
set of potential adjustments, none of which, however, is pertinent here. Id. at § 1016. The
Treasury Regulations under section 165 make explicit reference to these provisions, stating that
“[t]he amount of loss allowable as a deduction under section 165(a) shall not exceed the amount
prescribed by § 1.1011-1 as the adjusted basis for determining the loss from the sale or other
disposition of the property involved.” Treas. Reg. § 1.165-1(c). Importantly, a long-standing
regulation, Treas. Reg. § 1.61-6(a), further provides, in unequivocal language, that
[w]hen a part of a larger property is sold, the cost or other basis of the entire
property shall be equitably apportioned among the several parts, and the gain
realized or loss sustained on the part of the entire property sold is the difference
between the selling price and the cost or other basis allocated to such part.
Treas. Reg. § 1.61-6(a). This regulation indicates that “[t]he sale of each part is treated as a
separate transaction and gain or loss shall be computed separately on each part.” Id. As this
court stated in Fisher v. United States, 82 Fed. Cl. 780, 784 (2008), aff’d, 333 Fed. Appx. 572
(Fed. Cir. 2009), the apportionment required by the regulation “is done by dividing the cost basis
of the larger property among its components in proportion to their fair market values at the time
they were acquired.” 18
Courts have been relatively steadfast in applying the apportionment rules in Treas. Reg. §
1.61-6(a) to all forms of property. With the exception of an outlier or two (discussed below),
courts, including the Tax Court, have rejected the notion that the regulation governs only the
subdivision of real property, concluding instead that real property is the main, but not the sole,
focus of the regulation. See Fisher, 82 Fed. Cl. at 785 (discussing the history of the
regulation). 19 Rather, as its plain language suggests, 20 the regulation applies to any “collection
See Beaver Dam Coal Co. v. United States, 370 F.2d 414, 416–17 (6th Cir. 1966);
Fasken v. Comm’r of Internal Revenue, 71 T.C. 650, 656-57 (1979); Fairfield Plaza, Inc. v.
Comm’r of Internal Revenue, 39 T.C. 706, 712 (1963); Ayling v. Comm’r of Internal Revenue, 32
T.C. 704, 711 (1959); Cleveland–Sandusky Brewing Corp. v. Comm’r of Internal Revenue, 30
T.C. 539, 545 (1958); see also Gladden v. Comm’r of Internal Revenue, 262 F.3d 851, 853 (9th
In Fisher, this court noted, inter alia, that the progenitor of Treas. Reg. § 1.61-6
(Treas. Reg. 45, art. 43 (1921)) dealt only with the subdivision of real estate, but that the
language of the regulation was changed in 1957 to encompass all forms of property. Fisher, 82
Fed. Cl. at 785, 789 (citing Computation of Taxable Income, 22 Fed. Reg. 9419, 9422 (Nov. 26,
1957)). The Supreme Court discussed the pre-1957 version of the regulation in Heiner v.
Mellon, 304 U.S. 271, 280-81 (1938). Notably, while the Court there indicated that Article 43 of
Treas. Reg. 45 was limited to real estate, it pointed out that “[a] like rule has been applied where
the taxpayer had purchased personal property in block and was engaged in selling it in parcels.”
304 U.S. at 275-76.
- 14 -
or bundle of rights with respect to . . . property” and “is not limited to the subdivision of real
property.” Fasken v. Comm’r of Internal Revenue, 71 T.C. 650, 656 (1979); see also Norwest
Corp. & Subs. v. Comm’r of Internal Revenue, 111 T.C. 105, 139 (1998) (Fasken holds that
Treas. Reg. § 1.61-6(a) “is not limited to the severance of realty into two or more parcels, but
applies with respect to parts of the bundle of rights comprising property”). 21 Consistent with this
view, the apportionment rules have been directed to a wide range of tangible and intangible
property: shares of stock received in the demutualization of insurance companies, 22 shares of
stock purchased in bulk but sold in lesser parcels, 23 water rights, 24 oil leases, 25 ground rents, 26
materials used in shipbuilding, 27 portions of a pipeline, 28 antique furniture, 29 revenue rights
obtained in the purchase of a basketball team, 30 gold coins, 31 cement mixing equipment, 32 and
Black’s Law Dictionary (9th ed. 2009) (property: “the right to possess, use, and enjoy
a determinate thing (either a tract of land or a chattel) . . . ; [a]ny external thing over which the
rights of possession, use, and enjoyment are exercised.”); John Salmond, Jurisprudence 423-24
(Glanville L. Williams ed., 10th ed. 1947) (“property” includes “proprietary rights” in things like
“land, chattels, shares, and the debts due him”).
See also Anthony P. Polito, “Borrowing, Return of Capital Conventions, and the
Structure of the Income Tax: An Essay in Statutory Interpretation,” 17 Va. Tax Rev. 467, 534
(1998) (“If a taxpayer disposes of a portion of an asset, the general rule calls for apportionment
of basis between the portion of the asset sold and the portion retained.”).
Dorrance v. United States, 2013 WL 1704907, at *5 (D. Ariz. Apr. 19, 2013); Reuben
v. United States, 2013 WL 656864, at *4 (C.D. Cal. Jan. 15, 2013); Fisher, 82 Fed. Cl. at 784.
Bancitaly Corp. v. Comm’r of Internal Revenue, 34 B.T.A. 494, 503-05 (1936).
Gladden v. Comm’r of Internal Revenue, 262 F.3d 851, 853 (9th Cir. 2001).
Columbia Oil & Gas Co. v. Comm’r of Internal Revenue, 118 F.2d 459, 461 (5th Cir.
Welsh Homes, Inc. v. Comm’r of Internal Revenue, 279 F.2d 391, 395 (4th Cir. 1960).
Am. Industrial Corp. v. Comm’r of Internal Revenue, 20 B.T.A. 188, 197-98 (1930).
Santa Maria Gas Co. v. Comm’r of Internal Revenue, 10 B.T.A. 1412, 1414-15
Andrew Crispo Gallery, Inc. v. Comm’r of Internal Revenue, 63 T.C.M. (CCH) 2152
(1992), aff’d in part, and vacated in part, on other grounds, 16 F.3d 1336 (2d Cir. 1994).
First Northwest Industries of Am., Inc. v. Comm’r of Internal Revenue, 649 F.2d 707,
710 (9 Cir. 1981).
Daniels v. Comm’r of Internal Revenue, 2014 WL 700347, at *3-4 (U.S. Tax Court
Feb. 24, 2014).
Kunz v. Comm’r of Internal Revenue, 21 T.C.M. (CCH) 1454 (1962), aff’d, 333 F.2d
556 (6 Cir. 1964).
- 15 -
accounts receivable. 33 In some of these cases, the property was divided into interests described
spatially, while in others the division was temporal. 34 Careful adherence to this fractional,
divisible view of property is important, as it avoids having the sale of a part of a larger property
become an opportunity for income deferral. See Foster v. Comm’r of Internal Revenue, 80 T.C.
34, 216-17 (1983), aff’d in part, vacated in part on other grounds, 756 F.2d 1430 (9th Cir. 1985),
cert. denied, 474 U.S. 1055 (1986); Fasken, 71 T.C. at 655-57; see also Stephen B. Cohen,
“Apportioning Basis: Partial Sales, Bargain Sales and the Realization Principle,” 34 San Diego
L. Rev. 1693, 1703 (1997).
But, despite these many cases, PLIC insists that the apportionment rules do not apply to
“a carved-out income interest like the one under consideration.” It asserts, in effect, that the
regulation has a tacit exception, that is, it does not address situations in which an income interest
is carved out from a financial instrument. In that situation, PLIC claims, the proper tax treatment
is governed not by the Treasury Regulations, but by “80 years of common law, which Congress
and the Treasury have knowingly left in place.” PLIC cites, as evidence of this, a line of
authority that it claims demonstrates not only the existence of carve-out interests, but also the
fact that the normal basis allocation rules do not apply to them. It contends that this lineament
well-illustrates that the basis allocation performed by PLIC here – in which all of its cost in
acquiring the Perpetuals was allocated to the residual equity interest – was quite appropriate.
But, as will be seen, PLIC’s invocation of these cases – and the supposed “common law” rules
they embody – turns out to be something of a clupeidae roseus (or perhaps a school of them). A
brief overview of how income is taxed under the Code helps explain why.
Section 61(a) of the Code provides that “gross income means all income from whatever
source derived.” A fundamental principle underlying this provision is that income must be taxed
to the one who earns it – “that income is taxed to the party who earns it and that liability may not
be avoided through an anticipatory assignment of that income.” United States v. Basye, 410 U.S.
441, 447, 449-50 (1973); see also Lucas v. Earl, 281 U.S. 111, 114-15 (1930). Under section 61,
a taxpayer realizes income if he controls the disposition of that which it could have received,
even if it diverts the income to another. See Helvering v. Horst, 311 U.S. 112, 116-17 (1940);
Wheeler v. United States, 768 F.2d 1333, 1335-36 (Fed. Cir. 1985), cert. denied, 474 U.S. 1081
(1986). In such circumstances, the receipt of income by the third party stems from the taxpayer’s
economic gain – and that gain, therefore, is included in the gross income of the taxpayer, not that
of its assignee. See Comm’r of Internal Revenue v. Sunnen, 333 U.S. 591, 605-06 (1948); Horst,
R.M. Smith v. Comm’r of Internal Revenue, 69 T.C. 317, 335 (1977), aff’d, 591 F.2d
248 (3d Cir. 1979), cert. denied, 448 U.S. 828 (1979).
Compare Welsh Homes, 279 F.2d at 395 (temporal – ground rents) with Santa Maria
Gas, 10 B.T.A. at 1414-15 (spatial – pipeline); see, however, Jeffrey L. Kwall, “The Income Tax
Consequences of Sales of Present Interests and Future Interests: Distinguishing Time from
Space,” 49 Ohio St. L.J. 1 (1988) (arguing that spatial and temporal divisions should be treated
- 16 -
311 U.S. at 116-17; see also Yankee Atomic Elec. Co. v. United States, 782 F.2d 1013, 1016-17
(Fed. Cir. 1986). The exceptions to this rule are purposely narrow. For example, the rule does
not apply where the taxpayer which may be thought to have earned the income is precluded from
receiving it by operation of law. See Comm’r of Internal Revenue v. First Sec. Bank, 405 U.S.
394, 406-07 (1972); Yankee Atomic Elec., 782 F.2d at 1016.
So how do these rules apply where B owns an income-producing asset and conveys a
present interest to A for cash considerations, retaining either the remainder of the asset itself or a
reversion? Is this a sale by B, requiring A to report the income until B’s reversion takes effect,
or is it to be treated merely as a loan from A to B to be repaid by B from the income produced by
the asset and reported by B? How about the reverse – A owns an income-producing asset and
conveys the remainder therein to B, while reserving a present income interest in itself? To what
extent, if any, is B subject to federal income tax on amounts received by A with respect to the
property? See Kenneth F. Joyce and Louis A. Del Cotto, “The AB (ABC) and BA Transactions:
An Economic and Tax Analysis of Reserved and Carved Out Income Interests,” 31 Tax L. Rev.
121 (1975-1976) (extensively discussing these transactions); see also Jeffrey L. Kwall, “The
Income Tax Consequences of Sales of Present Interests and Future Interests: Distinguishing
Time from Space,” 49 Ohio St. L. J. 1 (1987). According to PLIC, these questions are readily
answered by its “80 years of common law.”
Of course, none of the dozen or so cases PLIC cites truly involve “common law,” at least
in the way that a tax lawyer would use the phrase. Instead, by and large, these cases involve the
application by courts of particular rules in the Code to the transactions presented. Though they
make occasional reference to the anti-abuse rules, like the substance-over-form doctrine, these
cases bottom on the assignment of income doctrine discussed above, which, in turn, has its roots
firmly fixed in section 61 itself. Be that as it may, PLIC asserts that a long and uninterrupted
line of cases holds that a taxpayer carving out an income interest and selling the residual is
required to allocate its entire basis in the original investment to the residual interest sold. But,
though its discussion of these cases is lengthy, PLIC is hard-pressed to quote anything from them
that actually says what it says. Indeed, as the following discussion reveals, one searches in vain
for anything in these cases that even approximates PLIC’s basis allocation rule.
PLIC begins with four cases in which the owner of stock transferred it to a third party,
but retained the right to receive dividends. Peck v. Comm’r of Internal Revenue, 77 F.2d 857 (2d
Cir. 1935), 77 F.2d 857 (2d Cir. 1935); Bettendorf v. Comm’r of Internal Revenue, 40 F.2d 49
F.2d 173, 174-75 (8th Cir. 1931); Morton v. Comm’r of Internal Revenue, 23 B.T.A. 930, 935
(1931), aff’d, 61 F.2d 1036 (2d Cir. 1932); Heminway v. Comm’r of Internal Revenue, 44 T.C.
96, 97-98 (1965). The courts held that the original owner of the stock was taxable on the
retained dividends. Those holdings, however, rested not upon any notion that the assignment-ofincome principle could be so easily avoided, but upon a finding that the recipient of the
transferred stock received the dividends in a trustee or fiduciary capacity, and not as a
beneficiary. See Peck, 77 F.2d at 858; Bettendorf, 49 F.2d at 175; Heminway, 44 T.C. at 100-01;
Morton, 23 B.T.A. at 935. The courts held that there was no anticipatory assignment of future
income in such an instance, as it was the “tree and not the fruit which [was] being sold.”
- 17 -
Heminway, 44 T.C. at 102. None of these cases suggests that the same result would obtain were
there no such trust relationship.
A fifth case cited by PLIC involving stock is Estate of Stranahan v. Comm’r of Internal
Revenue, 472 F.2d 867 (6th Cir. 1973). There, a father had a large interest deduction available to
use and wanted to accelerate his income to avoid losing that deduction. He assigned his son
$122,820 in anticipated stock dividends in exchange for the immediate payment of $115,000.
The father reported the $115,000 as ordinary income; the son reported the dividend income
received that same year (reduced by an amount he believed corresponded to his basis). Id. at
868. The Tax Court concluded that this transaction, “though conducted in the form of an
assignment of a property right, was in reality a loan to [the father] masquerading as a sale and so
disguised lacked any business purpose.” 30 T.C.M. (CCH) 1078 (1971). It held that the father
received income when the dividend was declared paid. Id. The Sixth Circuit reversed, holding
that the assignment-of-income principles in cases like Lucas and Horst did not apply because
“this was a transaction for good and sufficient consideration, and not merely gratuitous.” 472
F.2d at 870. As further grounds for this conclusion, the court emphasized that the transfer was
complete and that the seller (the father) bore the risk of not receiving the future income. Id. at
871. Accordingly, the court held that the dividends were taxable to the son when received and
that the son’s cash payment was income to the taxpayer in the year received. Id.; see also
Mertens, supra, at §§ 5:54, 38B:40.
Any notion that Stranahan represents some exception to the assignment-of-income
principles, applicable to carved-out income interests, rather than an isolated anomaly, requires
the court to ignore a number of similar cases that have come out differently. These cases have
refused to recognize the existence of an assignment where there was a risk as to whether the
future income assigned by the owner of an asset would be realized. Such was the holding in
Martin v. Comm’r of Internal Revenue, 56 T.C. 1255, 1259 (1971), aff’d, 469 F.2d 1406 (5th Cir.
1972), which involved a purported sale of future rents by the owner of an apartment building,
and in Hydrometals, Inc. v. Comm’r of Internal Revenue, 31 T.C.M. (CCH) 1260 (1972), aff’d,
485 F.2d 1236 (5th Cir. 1973), cert. denied, 416 U.S. 938 (1974), which involved a purported
sale of future manufacturing revenues by a manufacturing company. An assignment was
likewise disregarded in Mapco, Inc. v. United States, 556 F.2d 1107 (Ct. Cl. 1977). 35 There, the
taxpayer, in return for $4 million in cash, assigned to Rock Creek a 75 percent interest in its
future, unearned pipeline revenues, until such time as Rock Creek had received back $4 million
(plus interest). 36 The taxpayer attempted to accelerate its income in this fashion to make use of
an expiring net operating loss. Reviewing prior law, the Court of Claims distinguished the case
before it from Stranahan, and analogized it instead to Martin and Hydrometals. Illuminating the
“principal difference” between these cases, the court explained:
In Mapco, the Court of Claims, as it sometimes then did, adopted the decision of the
trial judge with modifications. 556 F.2d at 1108.
Rock Creek borrowed the $4 million from Chemical Bank; as security for this loan, it
assigned its interest in Mapco’s future revenues. Mapco used the $4 million to purchase
certificates of deposit issued by Chemical Bank. Mapco, 556 F.2d at 1109.
- 18 -
[I]n Stranahan, the seller of the future income was not under any obligation
whatever to produce such income for the benefit of the purchaser, who was
compelled to look solely to a third person . . . for future income that he had
purchased, whereas in the Martin and Hydrometals cases the purported sellers of
future income were themselves obligated to produce future income for the benefit
of the purported buyers. Thus, in substance, the transactions in the Martin and
Hydrometals cases were more like loans that were to be repaid, with interest, by
the borrowers out of their own productive efforts, than like true sales agreements.
Mapco, 556 F.2d at 1116-17. Finding that, as in Martin and Hydrometals, Mapco was obligated
to produce future revenues from the pipeline, the Court of Claims held that the assignment of
pipeline revenues was, in fact, “a loan-type investment secured by the right to future revenues.”
Id. at 1110; see also Schering-Plough Corp. v. United States, 651 F. Supp. 2d 219, 257-59
(D.N.J. 2009), aff’d, 652 F.3d 475 (3d Cir. 2011); Johnston v. Comm’r of Internal Revenue, 35
T.C.M. (CCH) 642 (1976). 37
The remainder of the cases PLIC cites in support of its “common law” vision are a
hodge-podge, particularly when viewed in the context of related decisions. For example, PLIC
cites Commissioner v. P.G. Lake, Inc., 356 U.S. 260 (1958) and Hort v. Comm’r of Internal
Revenue, 313 U.S. 28 (1941), suggesting that the Supreme Court, in excluding property
representing income items from the general definition of “capital asset,” more broadly signaled
that income carve-outs are not subject to the normal basis rules. But, the Court rejected a similar
spin on these precedents in Arkansas Best Corp. v. Comm’r of Internal Revenue, 485 U.S. 212,
217 (1988), which involved whether a loss arising from the sale of bank stock was a capital or
ordinary loss. The Court noted there that the results in P.G. Lake (which involved proceeds from
the sale of oil payments rights) and Hort (which involved payments to a lessor for the
cancellation of the unexpired portion of a lease), reflected the Code’s narrow definition of
“capital asset” and the unique features of the property at issue, rather than any broader
sentiments regarding the character of income-carve outs, observing that “these items are property
In so concluding, the court was persuaded by the fact that Mapco had invested the $4
million in proceeds in Chemical Bank certificates of deposit, which the court took as indication
that Mapco had “indirectly guaranteed repayment.” Mapco, 556 F.2d at 1111. The court noted
that the maturity dates of the certificates of deposit were specifically selected to coincide with
the anticipated dates for the repayment to Chemical Bank of the $4 million borrowed from that
bank by Rock Creek.
Taxpayers, in seeking to avoid having their sale of an income carve-out be viewed as a
loan, have sometimes analogized their situations to the receipt of production payments in
conjunction with oil and gas leases. That was the situation in another case cited by PLIC, Bryant
v. Comm’r of Internal Revenue, 399 F.2d 800 (5th Cir. 1968). There, the seller of a farm sought
to reserve a portion of the farm’s future income. Refusing to treat this transaction as a sale of a
carved-out income interest, the Fifth Circuit instead viewed it as involving a loan in which the
“production payments” constituted a part of the purchase price. Id. at 805. The court held that
the seller of the property was taxable on the farm income. Id.
- 19 -
in the broad sense of the word.” Arkansas Best, 485 U.S. at 217 n.5. 38 P.G. Lake and Hort were
focused primarily on preventing taxpayers from circumventing ordinary income tax rates by
“selling” rights to future ordinary income payments in exchange for a lump sum. And, as with
almost all of PLIC’s “common law” recitation, these cases shed no light on the basis allocation
issue presented here. See Womack v. Comm’r of Internal Revenue, 510 F.3d 1295, 1299-1300
(11th Cir. 2007); Prebola v. Comm’r of Internal Revenue, 482 F.3d 610, 611-12 (2d Cir. 2007).
In fact, only one case cited by PLIC, Apex Corp. v. Comm’r of Internal Revenue, 42 T.C.
1122 (1964), addresses the allocation of basis rule in Treas. Reg. § 1.61-6. In Apex, the taxpayer
purchased equipment, leased it to outside parties and then sold all of its rental and lease rights to
Murdock Acceptance Corporation. A new corporation was formed to which the taxpayer sold its
reversionary interest in the equipment subject to the leases. The taxpayer reported the sales of
the leases to Murdock as ordinary income and claimed ordinary losses on the sale to the new
corporation of the reversion rights, to the extent the price received for the reversionary interest
was less than the basis of the equipment. Among the issues presented to the court was whether
basis should be allocated between the equipment and the rental or lease rights the taxpayer had
sold, or instead should be allocated entirely to the equipment (allowing the taxpayer to deduct a
loss on the sale of the equipment). Id. at 1123. In holding that the taxpayer was entitled to use
his entire basis and deduct the loss from the sale, the Tax Court rejected the Commissioner’s
reliance on Treas. Reg. § 1.61-6, finding that the regulation related only to “the acquisition of a
tract of land followed by the sale of a portion of it.” 42 T.C. at 1126-27.
But, Apex’s holding on this point is simply wrong and cannot be squared with the plain
wording of the regulation, at least after it was modified in 1957 to encompass all forms of
property. See Fisher, 82 Fed. Cl. at 785, 789. This was later recognized by the Tax Court itself
in cases like Norwest and Fasken. In those reviewed opinions, 39 the Tax Court flatly rejected the
claim that the regulation applies only to real estate and held instead that it applies to all forms of
property. Norwest, 111 T.C. at 139; Fasken, 71 T.C. at 656-57; see also Gladden v. Comm’r of
Internal Revenue, 262 F.3d 851, 853 (9th Cir. 2001) (“This regulation tells us that when property
is acquired in a lump-sum purchase but then divided and sold off in parts, the cost basis of the
property should generally be allocated over the several parts.”). Apex thus provides little reason
See also United States v. Midland-Ross Corp., 381 U.S. 54, 56 (1965) (“not everything
which can be called property in the ordinary sense . . . qualifies as a capital asset”); Comm’r v.
Gillette Motor Transport, 364 U.S. 130, 133-35 (1960); see generally, Stanley S. Surrey,
“Definitional Problems in Capital Gains Taxation,” 69 Harv. L. Rev. 985, 988 n.7 (1956).
Decisions of the Tax Court are issued by a single judge, but the chief judge may direct
that a decision be reviewed by the full Tax Court. See 26 U.S.C. § 7460(b); see also Arbitrage
Trading, LLC v. United States, 108 Fed. Cl. 588, 603 n.23 (2013); see generally Ballard v.
Comm’r of Internal Revenue, 544 U.S. 40, 72 (2005). Under the conference procedures adopted
by the Tax Court, the affirmative vote of a majority of the judges may overrule a prior decision.
See Rent-A-Center, Inc. v. Comm’r of Internal Revenue, 142 T.C. 1, 20 (2014).
- 20 -
for this court to disregard the basis allocation rules in dealing with the transactions presented
So the upshot is this: nothing about this tour d’horizon convinces the court that it can –
or should – depart from the plain language of Treas. Reg. § 1.61-6 in analyzing the basis
allocation issue here. PLIC essentially ignores – and by ignoring fails to account for – the plain
language of the regulations. The court sees no reason why the allocation rules ought not to apply
where a taxpayer takes an income-producing security and subdivides it into two parts, selling one
and retaining the other. 41 In that instance, the original cost basis in the security must, in the
words of the regulation, be “equitably apportioned among the several parts” in accordance with
the fair market values of the respective parts. 26 C.F.R. § 1.61-6(a). This is so, even if, as a
result of the division, the taxpayer ends up with a residual interest whose primary value relates to
PLIC argues that the basis allocation rules were also disregarded in Moberg v. Comm’r
of Internal Revenue, 365 F.2d 337 (5th Cir. 1966). Moberg was one of a series of five conflicting
appellate cases dealing with transfers of Dairy Queen franchise rights in exchange for various
payments, including gallonage rights. The Fifth Circuit held that the gallonage payments did not
represent proceeds from the sale of sub-franchises. 365 F.2d at 339. On brief, PLIC claims that
the Fifth Circuit “required the taxpayer to allocate all of its tax basis to the transferred subfranchise rights in order to determine the gain or loss on the sale of those rights.” But, there is no
discussion of this basis allocation issue anywhere in the opinion. Rather, a passing reference in
the opinion merely indicates that, in response to an earlier remand order by the Fifth Circuit, see
305 F.2d 800 (5th Cir. 1962), the Tax Court permitted the Mobergs to allocate a part of the cost
of the master franchise to the subfranchises. 365 F.2d at 338. There is no indication whatsoever
that the court permitted the Mobergs to allocate all of the cost of the master franchise to the
subfranchises – or if it did, that the Fifth Circuit focused on this issue in the slightest.
PLIC retreats behind a fig leaf, to wit, that Congress has not passed a specific
provision that controls the basis allocation here, seemingly suggesting that this inaction blesses
its view of the “common law” in this area. To paraphrase Hand, this is “mere cant.” Comm’r of
Internal Revenue v. Newman, 159 F.2d 848, 850-51 (2d Cir. 1947) (dissenting). To give PLIC’s
argument on this count any credence, one must don blinders to 2004 legislative history
concerning modifications to the rules governing stripping transactions, in which Congress, in
describing present law, stated that transactions of the sort at issue here may be challenged under
existing “Code-based and common law-based authorities,” including “section 165 . . . and the
regulations thereunder,” “authorities that recharacterized certain assignments or accelerations of
future payments as financings,” and the anti-abuse doctrines (e.g., substance-over-form). H.R.
Rep. 108-548, at 277 n. 286 (2004); see also H.R. Conf. Rep. 108-755, at 617 n. 518 (2004).
PLIC cites a Senate Report accompanying the original passage of the interest-stripping rules in
1982, in which it is noted that under present law, it is “arguable” that all of a taxpayer’s basis in a
debt instrument “is allocated to corpus,” S. Rep. 97-494, at 215 (1982). But, of course, that was
a description of the law that Congress was modifying – an action it took because it felt that prior
law allowed “artificial loss[es].” Id. at 216; see also H.R. Conf. Rep. 97-760, at 554 (1982). In
general, the court finds the ambiguous tidbits of legislative history offered by plaintiff to be
unhelpful in deciding the technical issues presented here.
- 21 -
the appreciation of principal, and an income interest whose value lies in a stream of payments
that constitute ordinary income. Nor does it make any difference that the former of these is a
capital asset and the latter is not. 42 Nothing – nothing at all – suggests that in such a situation all
the taxpayer’s basis goes to principal and none to the carved-out income interest – by anyone’s
reckoning (even PLIC’s), there is no room for such an argument in the plain terms of the
Treasury Regulations. Indeed, even if equitable apportionment were not explicitly required by
that regulation, the court would imply such a requirement, if only to conform the interpretation
of section 61 with logic and common sense. See, e.g., Columbia Oil & Gas Co. v. Comm’r of
Internal Revenue, 118 F.2d 459 (5th Cir. 1941). Only in a parallel universe, where the “too good
to be true” rule of taxation reigns not, should the result be different.
So where does this leave us? In a refund suit, it is axiomatic that a taxpayer must prove,
by a preponderance of the evidence, that the assessment or determination is incorrect and that it
is entitled to a specific refund. See Helvering v. Taylor, 293 U.S. 507, 515 (1935)
(“[u]nquestionably the burden of proof is on the taxpayer”); Lewis v. Reynolds, 284 U.S. 281,
283 (1932), modified, 284 U.S. 599 (1932); see also Deseret Management Corp. v. United
States, 112 Fed. Cl. 438, 447 (2013). This rule incontestably extends to deductions. See
Interstate Transit Lines v. Comm’r of Internal Revenue, 319 U.S. 590, 593 (1943) (“[A]n income
tax deduction is a matter of legislative grace and . . . the burden of clearly showing the right to
the claimed deduction is on the taxpayer.”). PLIC, of course, is not required to prove its entire
case at this juncture. But nor can it stand pat in the face of clear indication that its position is
erroneous as a matter of law. To defeat defendant’s motion, PLIC must show that, as to the basis
allocation issue, there are genuine issues of material fact and that defendant is not entitled to
judgment as a matter of law. See, e.g., United States v. Humer, 1995 WL 653161, *3 (S.D. Fla.
Aug. 30, 1995); Burns v. United States, 242 F. Supp. 947, 949 (D.N.H. 1965). It has not. Since
the loss deduction PLIC seeks is based upon a basis allocation that is erroneous, as a matter of
law, and since it has offered no alternative to this allocation, the court finds that defendant has
established that it is entitled to judgment as a matter of law on the loss issue. Dorrance, 2013
WL 1704907, at *5 (“Taxpayers must prove their bases in property by a preponderance of the
evidence and substantiate the amount of the refund they seek”); Burns, 242 F. Supp. at 949
(“Since the amount plaintiff claims is based on a computation which is inappropriate under the
regulations, he cannot prevail.”).
In most cases, our discussion of PLIC’s loss deduction would be at an end. But this is not
most cases. Defendant contends that the deduction is faulty for another reason – one that proves
On brief, PLIC claims that “if a taxpayer [is] not permitted to offset the basis of
property out of which an income interest [is] carved against the proceeds from the sale of that
income interest, the basis must be allocable to the residual interest.” But, this syllogism is based
on a faulty premise, as there is nothing in the Code or the regulations that prohibits a taxpayer
from allocating basis to a carved-out income interest and using that basis for whatever purposes
the Code permits. Whether the asset is sold or retained, in the short or long run – so that its basis
can be fully recovered by a particular taxpayer – is quite irrelevant.
- 22 -
important, particularly as it also impacts the second issue in this case, involving the CSR
Defendant asserts that the investment trusts employed in the Perpetuals transactions do
not qualify as trusts for income tax purposes. Under this theory, PLIC’s sale of the Principal
Certificates to Morgan Stanley would be viewed as the sale of a membership interest in a singlemember disregarded entity, transforming that entity into a partnership. The transaction would
then be treated as a sale of a portion of PLIC’s ownership interest in each of the Perpetuals,
followed by the contribution by both PLIC and Morgan Stanley of their respective interests in
the securities to a partnership in exchange for partnership interests. See Treas. Reg. § 301.77012(a) (defining a partnership as a business entity that is not a corporation and that has at least two
members). Because its pro rata basis in the portion of each security sold in this fashion would
be the same as the amount it received from the sale, PLIC would be viewed as having sustained
no loss. All of this depends, of course, on whether the Primary Trusts qualify as trusts – while
defendant contends that under the controlling regulation, they do not, PLIC, not surprisingly,
The controlling regulation in this regard is Treas. Reg. § 301.7701-4, commonly known
as the “Sears Regulations,” which are part of the entity classification rules used to distinguish
among trusts, partnership and associations taxable as corporations for federal income tax
purposes. See 26 U.S.C. § 7701; Treas. Regs. §§ 301-7701-1 through 301.7701-4. 43 An
“ordinary trust” is defined broadly under subsection (a) of this provision as “an arrangement
created either by a will or by an inter vivos declaration whereby trustees take title to property for
the purpose of protecting or conserving it for the beneficiaries . . .” Treas. Reg. § 301.7701-4(a).
Subsection (b) of this provision recognizes that there are other arrangements, known as business
or commercial trusts, that are not classified as trusts for federal tax purposes because they are not
simple arrangements to protect or conserve property, but rather are created by the beneficiaries to
carry on a profit-making business that normally would be carried on through business
organizations, like a corporation or a partnership. Id. at 4(b). Subsection (c) of the regulation
then defines what are known as “investment trusts,” that sometimes qualify as trusts and other
times are classified as business entities. Id. at 4(c); see also Boris I. Bittker & Lawrence Lokken,
Fed. Tax’n Income, Est. & Gifts ¶ 58.2 (2014) (hereinafter “Bittker & Lokken”).
Under the regulations, an investment trust with two or more classes of ownership
interests is generally classified as a business entity. Treas. Reg. § 301.7701-4(c)(1); Bittker &
Lokken, ¶ 58.2. By way of distinguishing these situations, subsection (c)(1) of Treas. Reg. §
301.7701-4 states that –
An “investment” trust will not be classified as a trust if there is a power under the
trust agreement to vary the investment of the certificate holders. . . . An
investment trust with a single class of ownership interests, representing undivided
beneficial interests in the assets of the trust, will be classified as a trust if there is
“The Internal Revenue Code prescribes the classification of various organizations for
federal tax purposes.” Treas. Reg. § 301.7701-1(a)(1).
- 23 -
no power under the trust agreement to vary the investment of the certificate
holders. An investment trust with multiple cases of ownership interests ordinarily
will be classified as a business entity under § 301.7701-2; however, an investment
trust with multiple classes of ownership interests, in which there is no power
under the trust agreement to vary the investment of the certificate holders, will be
classified as a trust if the trust is formed to facilitate direct investment in the
assets of the trust and the existence of multiple classes of ownership interests is
incidental to that purpose.
Treas. Reg. § 301.7701-4(c)(1). So, a trust with single class of ownership interests may be
classified as a trust if, under the trust agreement, no one may vary the investment of the
certificate holders (the “no vary” rule). 44 An investment trust with multiple classes of ownership
may be classified as a trust if it meets the “no vary” rule and, in addition, is formed to facilitate
direct investment in the assets of the trust, with the existence of multiple classes of ownership
being only “incidental” to that purpose (the “incidental” rule). Id.; see also Wendy S. Goffe,
“Oddball Trusts and the Lawyers Who Love Them or Trusts for Politicians and Other Animals,”
46 Real Prop. Tr. & Est. L. J. 543, 563 (2012); Alan S. Acker, 852-3d Tax Mgmt. Portfolio
(BNA), Income Taxation of Trusts and Estates, A-20 (2007); James M. Peaslee, “Investment
Trusts in the Age of Financial Derivatives,” 49 Tax L. Rev. 419, 431-32 (1995) (hereinafter
“Peaslee”); see generally, North Am. Bond Trust, 122 F.2d at 546.
To illustrate the application of the “incidental” rule, the regulations offer four examples.
In the first two of these, a corporation transfers a portfolio of residential mortgages to a bank,
which delivers back to the corporation certificates evidencing rights to payments from the pooled
mortgages. In both examples, two classes of certificates are created.
In Example 1, the holders of class A certificates are entitled to all
payments of mortgaged principal until their certificates are retired; holders
of class B certificates receive payments of principal only after all class A
certificates have been retired. The example notes that the different rights
of the class A and B certificates serve to shift to the holders of the class A
certificates the risk that mortgages in the pool will be prepaid, giving the
holders of the class B certificates “call protection.” 45 Based on this, the
example concludes that “[t]he trust thus serves to create investment
interests with respect to the mortgages held by the trust that differ
significantly from direct investment in the mortgages,” leading to the
Even prior to the issuance of the regulations, the existence of a power to vary the
beneficiaries’ investment was sufficient to “turn the venture into a ‘business.’” Comm’r of
Internal Revenue v. North Am. Bond Trust, 122 F.2d 545, 546 (2d Cir. 1941), cert. denied, 314
U.S. 701 (1942).
“In other words, those holders are protected against early retirement of their
investment by directing mortgage payments to the earlier maturing class.” Peaslee, supra, at
- 24 -
further conclusion that the existence of multiple classes of trust ownership
is not incidental to any purpose of the trust to facilitate direct investment.
This means, according to the example, that “the trust is classified as a
business entity under § 301.7701-2.”
In Example 2, the holders of class C certificates are entitled to receive 90
percent of the payments of principal and interest on the mortgages; class D
certificate holders are entitled to receive the other ten percent. The two
classes of certificates are otherwise identical, except that, in the event of a
default on the underlying mortgages, the payment rights of the class D
certificate holders are subordinated to the rights of class C certificate
holders. Under these facts, the example finds that though the trust has
multiple classes of ownership interests and gives greater security to
holders of the class C certificate holders, the interests of the certificate
holders “are substantially equivalent to undivided interests in the pool of
mortgages, coupled with a limited recourse guarantee running from [the
corporation] to the holders of class C certificates.” In these circumstances,
the example holds, “the existence of multiple classes of ownership
interests is incidental to the trust’s purpose of facilitating direct investment
in the assets of the trust,” leading the investment trust to be classified as a
In two additional examples, the drafters of the regulations again contrast two forms of
In Example 3, a promoter forms a trust into which shareholders can
deposit their stock. When they do, the participant receives two certificates
– one represents the right to dividends and the value of the underlying
stock up to a specified amount; the other represents the right to
appreciation in the stock’s value above the specified amount. Investors
can fulfill varying investment objectives by transferring one certificate and
retaining the other. The example, however, views this flexibility as
“creat[ing] investment interests with respect to the stock held by the trust
that differ significantly from direct investment in such stock.” In this
scenario, the trust is not viewed as being “formed to facilitate direct
investment in the assets of the trust,” and is thus “classified as a business
entity under § 301.7701-2.”
In Example 4, a corporation purchases a portfolio of bonds and transfers
them to a bank under a trust agreement. The trustee delivers to the
corporation certificates evidencing interest in the bonds, with each
certificate representing the right to receive a particular payment with
respect to a specific bond. The example states that under section 1286 of
the Code, stripped coupons and stripped bonds are treated as separate
- 25 -
bonds for federal income tax purposes. 46 As such, although the trust has
multiple classes of ownership, those classes “simply provide each
certificate holder with a direct interest in what is treated under section
1286 as a separate bond.” “Given the similarity of the interests acquired
by the certificate holders to the interests that could be acquired by direct
investment,” the example holds, “the multiple classes of trust interests
merely facilitate direct investment in the assets held by the trust.” This
causes the trust to be “classified as a trust.”
These examples yield several observations about the “incidental” rule. First, it appears
that whether the creation of multiple classes of trust interests is “incidental” to the trust’s purpose
of facilitating direct investment in the trust’s assets depends on the extent to which the attributes
of the trust interests differ from direct ownership of the trust assets. See Peaslee, supra, at 431;
Richard S. Millerick, “Federal Income Tax Aspects of Stripped Mortgage-Backed Securities,” 12
Va. Tax Rev. 219, 226-27 (1992) (hereinafter “Millerick”). This view is confirmed in the
preamble to the Sears Regulations, 47 which states that “[w]hether the existence of multiple
classes of ownership interests is incidental to the use of an investment trust as a vehicle to
facilitate direct investment, or instead reflects a purpose to provide investors with diverse
interests in the trust assets, generally depends on the extent to which the investment attributes of
interests in the trust diverge from direct ownership of the trust assets.” 51 Fed. Reg. 9950, 9951
(March 24, 1986). In this regard, the preamble further explains –
Multiple class trusts depart from the traditional form of fixed investment trust in
that the interest of the beneficiaries are not undivided, but diverse. The existence
of varied beneficial interests may indicate that the trust is not employed simply to
hold investment assets, but serves a significant additional purpose of providing
investors with economic and legal interests that could not be acquired through
direct investment in the trust assets. Such use of an investment trust introduces
the potential for complex allocations of trust income among investors, with
correspondingly difficult issues of how such income is to be allocated for tax
purposes. These issues are properly foreign to the taxation of trust income, where
See also Mertens, supra, at § 22:3.
Where the text of a regulation is less than clear, the court may look to the preamble of
the regulation to determine the administrative construction thereof. See Kingdomware Techs,
Inc. v. United States, 107 Fed. Cl. 226, 243 (2012); see also Fidelity Fed. Sav. and Loan Ass’n v.
dela Cuesta, 458 U.S. 141, 158 n. 13 (1982); Wyo. Outdoor Council v. U.S. Forest Serv., 165
F.3d 43, 53 (D.C. Cir. 1999) (“Although the preamble does not ‘control’ the meaning of the
regulation, it may serve as a source of evidence concerning contemporaneous agency intent.”);
see generally Lars Noah, “Divining Regulatory Intent: The Place for a ‘Legislative History’ of
Agency Rules,” 51 Hastings L.J. 255, 307 (2000). Although the preamble is not entitled to
Chevron deference, it is still entitled to deference insofar as it has “the power to persuade.”
Skidmore v. Swift & Co., 323 U.S. 134, 140 (1944); see also Kingdomware Techs., 107 Fed. Cl.
- 26 -
rules have not developed to accommodate the varied forms of commercial
investment, and no comprehensive economic substance requirement governs the
allocation of income for tax purposes.
Id.; see also Bittker & Lokken, supra, at ¶ 58.2. The extent to which the attributes of the trust
interests diverge from direct ownership of trust assets may, in turn, be revealed by determining
whether the interests of the investors in a multiple class trust could be reproduced outside the
trust without resort to the multiple classes of ownership. Peaslee, supra, at 431; Millerick, supra,
at 226. Additional guidance on this point can again be gleaned from the preamble, which states
that the extent of the “divergency may, in turn, be reflected by the extent to which the interests of
the investors in a multiple class trust could be reproduced without resort to multiple classes of
ownership.” 51 Fed. Reg. at 9951.
These principles can be seen at work in the examples in the regulation. Example 4
expressly sanctions the use of multiclass trust interests to facilitate the issuance of basic interestonly and principal-only securities. Treas. Reg. § 301.7701-4(c)(2) (Ex. 4). In the example, a
group of bonds is transferred to a trust in exchange for multiple certificates, each of which
represent the right to receive a particular payment of either principal or interest with respect to a
specific bond. Id. In concluding that the trust is classified as a trust, the example emphasized
that “the multiple classes simply provide each certificate holder with a direct interest in what is
treated under section 1286 as a separate bond.” Id. This suggests that the existence of the
coupon stripping rules was key to the analysis because, under those rules, the interests of the
investors in the trust could be reproduced without resort to multiple classes of ownership. This
leads to the conclusion that the use of multiple classes was “incidental” to the trust’s purpose of
facilitating direct investment in the trust assets. This analysis implies that the same conclusion
would not obtain if, outside of the trust regime, the coupon stripping rules did not exist.
Millerick, supra, at 227 (interest-only and principal-only “investment trust interests are permitted
solely because of the existence of the coupon stripping rules.”); Mertens, supra, at § 36:129. The
regulatory history surrounding the promulgation of the Sears Regulations confirms this. 48
As proposed in 1984, the Sears Regulations flatly prohibited multiple classes. See
Prop. Treas. Reg. § 301.7701, 49 Fed. Reg. 18741-02, 18743 (May 2, 1984). The sole exception
was for trusts formed to permit investors to own “specifically identifiable” stripped bonds and
coupons, governed by section 1232B of the Code, the predecessor of current section 1286. On
this point, a notice accompanying the proposed regulation stated, in reference to what would
become Example 4:
In a third type of arrangement, a custodian holds one or more issues of bonds and
issues certificates representing the right to specific interest or principal payments
on the bonds. Although this arrangement is similar to [one in which investors
divide the income and appreciation elements inherent in a share of common
stock], it is distinguishable in that the division of the property is only between the
bonds and specific interest coupons thereon. In section 1232B of the Code,
Congress has provided a method for taxing transactions involving such “stripped
bonds” and ‘stripped coupons.’” Thus, it would be inconsistent with section
1232B to treat typical “coupon stripping” arrangements in which bonds are held
- 27 -
The absence of a provision like section 1286 governing the splitting of common stock
into interests likely accounts for why, in Example 3, in which shares of stock are converted into
two classes of certificates (thereby allowing a separation of dividend income from capital
appreciation), the regulations conclude that the investment trust violates the “incidental” rule.
See Millerick, supra, at 229-30 (“In the case of corporate stock, however, there is no analogue to
section 1286 which clearly characterizes the transaction as a separation of ownership interests
[and] dictates how the investors are to be taxed . . . .”); Peaslee, supra, at 462 (“The fact that the
favorable outcome in example 4 is linked to the application of the bond stripping rules means
that similar arrangements for dividing up payments on a single bond that do not fall within those
rules may fail to qualify as trusts.”). 49 A similar pattern, illustrating the impact of not having a
provision like section 1286, is exhibited in the examples found in the Treasury Regulations
dealing with real estate mortgage investment conduits (REMICs) – which, like true investment
trusts are used to securitize real property mortgages. See Temp. Treas. Reg. § 1.67-3T(a) Ex.
by a custodian and interests in specifically identifiable stripped coupons or bonds
are sold as either associations or partnerships.
Id. at 779; see also Millerick, supra, at 227.
Consistent with this view, the investment trust described in Example 2 in the Sears
Regulations is considered a trust because the feature under which one class’s rights to principal
and interest are subordinated to the entitlement of the other not only can be reproduced outside
the trust, but, indeed, often are. See Treas. Reg. § 301.7701-4(c)(2) Ex. 2; Bittker & Lokken,
supra, at ¶ 58.2.
Example 2 under these regulations is similar to Example 2 above, except that the
REMIC includes a second regular interest represented by class E certificates. The example goes
on from there, as follows:
The principal purpose of M in structuring the REMIC to include class E
certificates is to avoid allocating allocable investment expenses to class C
certificate holders. The class E certificate holders are entitled to receive the
payments otherwise due the class D certificate holders until they have been paid a
stated amount of principal plus interest. The fair market value of the class E
certificate is ten percent of the fair market value of the class D certificate and,
therefore, less than one percent of the fair market value of the REMIC. The
REMIC would not be classified as an investment trust under § 301.7701-4(c)(1)
because the existence of the class E certificates is not incidental to the trust’s
purpose of facilitating direct investment in the assets of the trust.
Temp. Treas. Reg. § 1.67-3T(a) Ex. 2. By comparison, REMICs with certificates that provide
for floating or inverse floating rates of interest from a fixed rate REMIC certificate presumably
would meet the “incidental” rule as REMICs are permitted to issue such interests to investors
directly. See Treas. Reg. §1.860G-1(a)(2), 1(a)(3); see generally, Willard B. Taylor,
- 28 -
Perhaps, these examples in the Sears Regulations could be clearer in mapping out the
contours of the “incidental” rule. But, they are clear enough. Moreover, in the court’s view,
“any ambiguity [in the regulations] is dispelled by the preamble accompanying and explaining
the regulation.” Fidelity Fed. Sav. and Loan Ass’n, 458 U.S. at 158 & n.13. Moreover, it should
not be overlooked that an agency’s interpretation of its own regulations is “controlling unless
plainly erroneous or inconsistent with the regulations being interpreted.” Long Island Care at
Home, Ltd. v. Coke, 551 U.S. 158, 171 (2007) (internal quotation marks omitted); see also Auer
v. Robbins, 519 U.S. 452, 461 (1997); Thomas Jefferson Univ. v. Shalala, 512 U.S. 504, 512
(1994) (holding that an agency’s interpretation of its regulations is entitled to “substantial
deference” unless “an alternative reading is compelled by the regulation's plain language”);
Mason v. Shinseki, 743 F.3d 1370, 1374-75 (Fed. Cir. 2014). The court believes that regulations
and the preamble amply demonstrate that the creation of multiple classes of trust interests is not
“incidental” to the purpose of the arrangement if those interests could not be produced outside of
the trust environment.
Here, the multiple classes of ownership employed in the Perpetuals diverge from the
interests that could be obtained by direct investment in the underlying securities. The trust
arrangements serve to create investment interests with respect to the underlying securities that
differ significantly from what could be obtained from the direct investment in those securities.
For example, direct ownership of the perpetual securities entitled the holders to quarterly or
semi-annual interest payments from the date of investment into perpetuity, while the holder of an
Interest Certificate received interest only for the first 16 or 17 years (absent a Reference
Event). 51 The splitting of interests encountered appears to be similar to that described in
Example 3 of section 301.7701-4(c)(2). That the existence of multiple classes here is not
“incidental” is all the more obvious giving the looming presence of the Termination Agreements.
At the least, those agreements protect PLIC from the occurrence of a call or event of default with
respect to the perpetual securities – an event that would impact the flow of interest on PLIC’s
retained Interest Certificates. 52 In the court’s view, this type of protection is similar to the “call
protection” that was deemed disqualifying in Example 1 of the Sears Regulations. See Treas.
Reg. § 301.7701-4(c)(2) Ex. 1; see also Mertens, supra, at 38A:20 n.18; Bittker & Lokken,
‘“Blockers,’ ‘Stoppers,’ and the Entity Classification Rules,” 64 Tax Law. 1, 27-30 (2010)
(describing REMICs and comparing them to “investment trusts”).
PLIC also projected that, aside from any claimed tax savings, yields on the Interest
Certificates would be lower than those of the securities and, conversely, yields of the Principal
Certificates would be higher than those of the securities.
Defendant sees the Termination Agreements as accomplishing much more. It asserts
that, in the case of a Reference Event, the Termination Agreements potentially cause principal to
flow back to PLIC in the form of payments corresponding to the Termination Units created by
Morgan Stanley and assigned to MSIL. Because this interpretation of the agreements, however,
presents questions of fact, the court does not rely upon it as a basis for decision here. By
comparison, that the Termination Agreements provided some form of call protection seems
- 29 -
supra, at ¶ 58.2. In short, the “trust” arrangement in question was not formed to facilitate direct
investment in the assets and the existence of multiple classes of ownership interests is not merely
incidental to that purpose. Nothing in the preamble, the body of the regulations, or the examples
Under the regulations, the custodial arrangements employed in these investments are
properly classified as business entities under Treas. Reg. § 301.7701-2. Those arrangements are
not corporations, as defined under Treas. Reg. § 301.7701-2(b). Under the regulations, that
means that they are, as defendant contends, partnerships (or at least became so when Morgan
Stanley acquired its interests). Treas. Reg. § 301.7701-2(c). The partnerships were jointly
owned by PLIC and Morgan Stanley. Under section 721(a) of the Code, PLIC recognized no
gain or loss as a result of its contribution of property to the partnerships in exchange for interests
therein. See Rev. Rul. 99-5, 1999-1 Cum. Bull. 434 (Situation 1); see also Arthur B. Willis, John
S. Pennell, Philip F. Postlewaite, Partnership Tax. ¶12.01 (2014) (hereinafter “Willis”). PLIC’s
deemed sale to Morgan Stanley resulted in no gain or loss because, under section 1001 of the
Code, its basis allocated to the portion of the certificates sold equaled the amount realized. See
also Mertens, supra, at § 35A:47; Willis, supra, at ¶ 12.01.
Based on the foregoing, the court concludes that PLIC is not entitled to the loss deduction
it claims under section 165 of the Code. The court reaches this conclusion, as a matter of law,
without addressing various alternative claims defendant has made under the anti-abuse rules. 53
The court turns next to the tax treatment of PLIC’s CSR transactions, which, as will be seen,
incorporates some of the analysis above.
Recall that from 1996 to 2001, PLIC purchased interests in six custodial arrangements
and in two trusts that hold shares (the Shares) in money market funds (the Issuers). As described
at the outset, these transactions took several forms.
In the six custodial arrangements, a Depositor transferred to a Custodian either the Shares
or money to buy the Shares. In return, the Custodian issued to the Depositor both custodial
dividend receipts (CDRs) and custodial share receipts (CSRs). The CDR holders have the right
to dividends paid on the Shares until the maturity date (the Restricted Period). The CSR holders
represent other rights to the Shares exclusive of the right to receive dividends during the
Restricted Period. In connection with each of the custodial CSRs, PLIC entered into a Custody
Agreement with a Custodian, which agreement, inter alia, granted PLIC the right to sell its CSRs
to other investors. PLIC also entered into a Termination Agreement with the Depositor that
obliges it to buy the CDRs from the Depositor (or its successor) if the custodial arrangement
Defendant raises a host of other grounds upon which it believes the court should
disallow PLIC’s deduction, including arguments based on the so-called step-transaction doctrine.
Based on the conclusion reached above, the court need not reach these additional arguments.
- 30 -
terminates prematurely, either because the Issuer is about to liquidate or has failed to maintain its
status as a money market fund under the Investment Company Act of 1940, 15 U.S.C. § 80a-1 et
In the first of the two trust transactions, a Depositor placed the Shares into a trust (the
“REDI Trust”) which issued Dividend Certificates and Corpus Certificates. PLIC purchased the
Corpus Certificates, which entitles it to receive the Shares in January 2020 and all dividends
payable thereafter. In a premature termination, the trust agreements require the trustee to divide
trust assets between the holders of the Corpus and Dividend Certificates, under a schedule
designed to give the holders of the Dividend Certificates the approximate value of what they
would have received absent premature termination. In the latter of the two trust transactions, the
Depositor transferred either money or Shares into a trust (the “Primary Trust”), which issued a
Principal Certificate and a Dividend Certificate. The Depositor transferred the Principal
Certificate to the “Issuer Trustee,” which issued Principal Units and Termination Units. PLIC
purchased the Principal Unit, which generally entitled it to receive payments on the Principal
Certificates during the Trust’s existence and to receive the shares thereafter. As with the REDI
Trust, provisions were made in the trust agreement for premature termination.
On its federal income tax returns, PLIC treated the CSRs as long-term bonds. It reported
no taxable income on the accretions in the value of the CSRs that occurred during the Restricted
Period. In this regard, PLIC argued that because the money market shares are equity, it was not
required neither to accrue original issue discount during the Restricted Period under section 1272
of the Code, nor to treat the shares as stripped bonds under section 1286 of the Code. 54
In the court’s view, the CSR arrangements all give rise, in one fashion or another, to the
creation of “investment trusts,” which, in turn, must be analyzed under the Sears Regulations
discussed above. Because these trusts all have multiple classes of shares, they may be treated as
trusts only if they meet the “no vary” rule and the “incidental” rule.” As discussed above,
whether the creation of multiple classes of trust interests is “incidental” to a trust’s purpose of
facilitating direct investment in its assets depends upon the extent to which the trust attributes
differ from direct ownership of the trust assets. Correspondingly, the extent to which the
attributes of the trust interests diverge from direct ownership of trust assets depends upon
whether “the interests of the investors in a multiple class trust could be reproduced without resort
to the multiple classes of ownership.” T.D. 8080, 1986-1 C.B. 371. As demonstrated above,
these rules spring from the examples in Treas. Reg. § 301.7701-4(c)(2), as amplified by the
preamble to the 1986 regulations.
In the CSRs, the CDR holder has the right to dividends paid on the money market shares
for approximately 20 years, but no rights thereafter. The CSR holder’s rights to dividends pick
up at this point. This was true regardless of the CSR formats that PLIC employed. In the court’s
Original issue discount (OID) is “the excess (if any) of [a debt instrument’s] stated
redemption price at maturity, over . . . the issue price.” 26 U.S.C. § 1273(a)(1). Under the Code,
the gain from OID is included in gross income as interest over the obligation’s duration, rather
than entirely at the time the debt instrument is issued or redeemed. Id. at 1272(a)(1).
- 31 -
view, by virtue of this allocation of dividend rights, the certificates have attributes that diverge
from the direct ownership of the money market shares. Indeed, it appears, as defendant argues,
that the investors wanted the CSRs and the CDRs precisely because their attributes diverged
from the direct ownership of the trust assets. Moreover, PLIC can draw no solace from Example
4 of the regulations because, by its own admission, there is no analogue to section 1286 that
governs the money market shares in question. And while the interests of the CSR and CDR
holders could have been reproduced without a trust, that is not the standard. Indeed, via modern
derivatives, virtually any type of investment can be created. Rather, the question is whether the
investor interests could have been reproduced without resort to multiple classes of ownership.
And, despite PLIC’s protestations to the contrary, the simple fact is that they could not. To put it
in the words of the regulations, the CSR arrangements thus were not “formed to facilitate direct
investment in the assets of the trust.” Treas. Reg. § 301.7701-4(c). Accordingly, the
“investment trusts” employed in the CSR transactions are properly classified not as trusts, but as
business entities under Treas. Reg. § 301.7701-2.
The custodial arrangements and investment trusts employed in the CSR transactions are
not corporations, as defined under Treas. Reg. § 301.7701-2(b). Under the applicable
regulations, that means that they are, as defendant contends, partnerships. Treas. Reg. §
301.7701-2(c). 55 If that is so, for the years in question, PLIC must be allocated a portion of the
partnership income that corresponds to its interests in the partnership and thereby must include in
income a distributive share of each year’s money-market dividends. See 26 U.S.C. § 702(a);
Treas. Reg. § 1.702-1(d); see also id. at § 705 (describing how the adjusted basis of partner’s
interest is calculated); Prestop Holdings, Inc., LLC v. United States, 96 Fed. Cl. 244, 246 (2010).
That means that PLIC’s return positions, which reflect no receipt of income, are erroneous. 56
For the years in question, the IRS assessed against PLIC a substantial understatement
penalty under section 6662(d) of the Code, as well as a negligence penalty under sections
6662(a) and (b)(1) of the Code. In the court’s view, consideration of these penalties, including
PLIC acquired its first two CSRs in 1996. Those transactions are not covered by
current version of Treas. Reg. § 301.7701-2(c), but by an earlier regulation – Treas. Reg. §
301.7701-2(a) (1996). Under those regulations, the Wilmington Trust arrangements are not
corporations because they lack the corporate characteristics of limited liability and continuity of
life. Id. at § 301.7702-2(a)(1). They are thus also classified as partnerships. Id. at § 301.77012(b)(3).
Rather than urging that PLIC’s allocation be rejected outright, defendant indicates that
the allocation of taxable income allocable to PLIC is less than the amount the IRS determined in
its notice of deficiency. The court will permit the parties a reasonable period to perform and
agree upon any necessary recomputation of overpayment of tax liability on this issue.
- 32 -
PLIC’s defenses thereto, raise a number of material questions of fact that preclude this court
from ruling now. Those questions will be resolved at trial at an appropriate time. 57
Based on the foregoing, the court rules in defendant’s favor on the liability issues
associated with the Perpetuals and CSR transactions. It declines to address defendant’s penalty
arguments at this time. The court hereby DENIES plaintiffs’ motion for partial summary
judgment and GRANTS, in part, defendant’s cross-motion for partial summary judgment.
IT IS SO ORDERED.
s/Francis M. Allegra
Francis M. Allegra
For example, defendant argues that the Perpetual transactions were “tax shelters”
because a “significant purpose” of the transactions was to avoid or evade income tax, as per
section 6662(d)(2)(C)(iii), because “creating capital losses to reduce tax was at a minimum a
significant (i.e., important or meaningful) purpose of PLIC's deals.” Further, defendant argues
that PLIC cannot invoke any of the statutory defenses to the penalties because, in defendant’s
view, PLIC did not act in good faith with respect to the Perpetual transactions and did not have a
reasonable cause to take the deductions.
- 33 -
Disclaimer: Justia Dockets & Filings provides public litigation records from the federal appellate and district courts. These filings and docket sheets should not be considered findings of fact or liability, nor do they necessarily reflect the view of Justia.
Why Is My Information Online?