PRACTISING LAW INSTITUTE TAX STRATEGIES FOR CORPORATE ACQUISITIONS, DISPOSITIONS, SPIN-OFFS, JOINT VENTURES, FINANCINGS, REORGANIZATIONS AND RESTRUCTURINGS 2014

The Consolidated Unified Loss Rules

June 2014

By

Mark J. Silverman Steptoe & Johnson LLP Washington, D.C.

Copyright © 2014 Mark J. Silverman, All Rights Reserved.

- i -

TABLE OF CONTENTS Internal Revenue Service Circular 230 Disclosure: As provided for in IRS regulations, advice (if any) relating to federal taxes that is contained in this document (including attachments) is not intended or written to be used, and cannot be used, for the purpose of (1) avoiding penalties under the Internal Revenue Code or (2) promoting, marketing or recommending to another party any plan or arrangement addressed herein. Page

I. BACKGROUND AND HISTORY OF LOSS DISALLOWANCE AND LOSS DUPLICATION RULES ...... 1 A. Investment Adjustment Rules ...... 1 B. Targeted Problems ...... 2 C. Loss Disallowance Regulations for Dispositions Between 1/31/91 and 3/7/02 – History of Treasury Regulation Section 1.1502-20 ...... 9 D. Rite Aid Case ...... 13 E. Loss Disallowance Regulations for Dispositions Between 3/7/02 and 9/17/08 – Treasury Regulation Section 1.337(d)-2...... 17 F. Loss Duplication Regulations for Dispositions Between 3/7/02 and 9/17/08 – Treasury Regulation Section 1.1502-35 ...... 18 G. Current Unified Loss Rules for Dispositions After 9/17/08 – Treasury Regulation Section 1.1502-36 ...... 20 H. Avoiding the Unified Loss Rules or Prior Loss Disallowance and Duplication Rules...... 21 II. CURRENT UNIFIED LOSS RULES FOR LOSS ON SUBSIDIARY STOCK ...... 22 A. Background ...... 22 B. General Rule: Reg. § 1.1502-36 ...... 25 C. Basis Redetermination to Reduce Disparity ...... 29 D. Stock Basis Reduction to Prevent Non-Economic Loss ...... 33 E. Attribute Reduction to Prevent Duplication of Loss...... 37 F. Coordination with Loss Deferral and Other Loss Disallowance Rules ...... 45 G. Anti-Abuse Rule - Reg. § 1.1502-36(g)...... 46 III. OLD LOSS DISALLOWANCE RULES – ADDRESSING CONCERNS RELATING TO THE REPEAL OF GENERAL UTILITIES ...... 47 A. General Rule ...... 47 B. Deconsolidations ...... 48 B. Allowable Loss ...... 52 C. Netting Rule ...... 64

- ii - Page

D. Coordination with Loss Deferral and Other Loss Disallowance Rules ...... 68 E. Successor Rule ...... 71 F. Anti-Avoidance Rules ...... 72 G. No Tiering Up of Certain Adjustments ...... 75 H. Prior Reg. § 1.1502-20(i) – Transition Rules ...... 77 IV. OLD LOSS DISALLOWANCE RULES – ADDRESSING CONCERNS RELATING TO LOSS DUPLICATION...... 90 A. Background ...... 90 B. Basis Redetermination Rule ...... 92 C. Loss Suspension Rule ...... 97 D. Worthlessness and Dispositions Not Followed by Separate Return Years...... 100 E. Anti-Avoidance Rules ...... 104 V. EXAMPLES APPLYING THE UNIFIED LOSS RULES AND LOSS DISALLOWANCE AND DUPLICATION RULES...... 112 A. Basis Redetermination Examples ...... 112 B. Basis Reduction Examples ...... 124 C. Duplicated Loss Examples ...... 144 D. Intercompany Transfers of Subsidiary Stock Examples ...... 154

- iii -

TABLE OF EXAMPLES Page

Example 1 – Basic Investment Adjustment Rules ...... 2 Example 2 – Basic Son of Mirrors Transaction ...... 3 Example 3 – Son of Mirrors Transaction – “Bust Up” ...... 4 Example 4 – Wasting Assets ...... 5 Example 5 – Loss Duplication: Unrelated Taxpayers...... 6 Example 6 – Loss Duplication: Stuffing/Outside Loss Recognized Before Inside Loss ...... 7 Example 7 – Loss Duplication: Stuffing/Inside Loss Recognized Before Outside Loss...... 8 Example 8 – Overview of Unified Loss Rule ...... 27 Example 9 – Netting Rule Application (Allocation of Gain Amount to Determine Net Loss) ..36 Example 10 – Worthlessness Where S Continues as a Member ...... 45 Example 11 – Definition of Deconsolidation...... 49 Example 12 – Using the Deconsolidation Rule to Avoid Gain Recognition ...... 51 Example 13 – Built-In Loss Offsets Built-In Gain ...... 58 Example 14 – Netting Gains and Losses...... 66 Example 15 – Netting Under the Deconsolidation Rule ...... 67 Example 16 – Coordination With Loss Deferral Rules ...... 70 Example 17 – Successor Rule ...... 71 Example 18 – Shifting of Value ...... 72 Example 19 – Basic Stuffing Case ...... 74 Example 20 – Deconsolidation of Parent in Same Transaction as Subsidiary ...... 76 Example 21 – Reattribution Rule ...... 81 Example 22 – Worthless Stock Deduction...... 101 Example 23 – Transfer of Property to Avoid Basis Redetermination Rule ...... 104 Example 24 – Loss Reimportation ...... 107 Example 25 – Transfers to Avoid Gain Recognition ...... 110 Example 26 – Basis Redetermination To Prevent Non-Economic Loss ...... 112 Example 27 – Basis Redetermination To Prevent Duplicated Loss ...... 114 Example 28 – Increase In Basis of Transferred Loss Share ...... 115 Example 29 – No Investment Adjustments; Basis Redetermination Under Prior Rules But Not Unified Loss Rules ...... 117 Example 30 – Partial Duplicated Loss Allowed Under Prior Rules But Not Unified Loss Rules ...... 120 Example 31 – No Deconsolidation; Economic Loss Disallowed Under Prior Rules But Not Under Unified Loss Rules ...... 121 Example 32 – Deconsolidation; Economic Loss Disallowed Under Prior Rules But Not Under Unified Loss Rules ...... 122 Example 33 – Basis Redetermination to Eliminate an ELA ...... 123 Example 34 – Son-of-Mirrors Transaction ...... 125 Example 35 – Wasting Asset ...... 126 Example 36 – Post-Acquisition Appreciation Eliminates Stock Loss ...... 126 Example 37 – Distributions ...... 127 Example 38 – Loss Attributable to Post-Acquisition Loss ...... 128 Example 39 – Built-In Gain Asset Depreciates in Value ...... 129 Example 40 – Built-In Gain Asset Appreciates in Value ...... 130

- iv -

Example 41 – Built-In Gain in After-Acquired Asset ...... 132 Example 42 – Basis Reduction (Lower-Tier Subsidiary/No Transfer of Lower-Tier Stock) .....134 Example 43 – Carryover Basis in Stock ...... 136 Example 44 – Basis Reduction (Computing the Disconformity Amount – Unrecognized Loss Reflected in Stock Basis) ...... 138 Example 45 – Post-Acquisition Appreciation Removes Taint of Built-In Gain ...... 140 Example 46 – Post-Acquisition Appreciation Removes Taint of Built-In Gain – Lower Tier Subsidiary ...... 142 Example 47 – Computation of Attribute Reduction Amount/Transfer of All S Shares ...... 144 Example 48 – Transfer of Less than All S Shares ...... 145 Example 49 – Multiple Dispositions of S Stock ...... 146 Example 50 – Allocation of Attribute Reduction Amount Among Category D Assets ...... 148 Example 51 – Allocation of Attribute Reduction Amount Among Category A, Category B, and Category C ...... 149 Example 52 – Wholly Owned Lower-Tier Subsidiary ...... 151 Example 53 – Intercompany Sale with Duplicated Loss ...... 154 Example 54 – Intercompany Sale of Built-in Gain Stock ...... 156 Example 55 – Intercompany Sale Creates Built-in Gain Stock ...... 157 Example 56 – Subsidiary with Built-in Gain and Built-in Loss Assets ...... 158

I. BACKGROUND AND HISTORY OF LOSS DISALLOWANCE AND LOSS DUPLICATION RULES

A. Investment Adjustment Rules

1. The investment adjustment rules under Reg. § 1.1502-32 require that annual positive or negative adjustments be made to the basis of the stock of each subsidiary of a consolidated group to reflect gain or loss recognized by the subsidiary.

a. Specifically, basis is increased by S’s taxable income and tax- exempt income. Basis is decreased by S’s tax loss, nondeductible expenses, and distributions with respect to S’s stock. Reg. § 1.1502-32(b)(2).

b. If S has more than one class of stock outstanding, the adjustments must be allocated between the classes. An adjustment attributable to a distribution is allocated to the shares of S’s stock entitled to the distribution. If the remainder of the adjustments are positive, the adjustments are allocated first to preferred stock (and only to the extent of dividend arrearages and distributions to which the preferred stock becomes entitled), and second to S’s common stock. If the remainder of the adjustments not attributable to distributions is negative, they are allocated solely to the common stock. Reg. § 1.1502-32(c)(1).

c. The investment adjustment rules of Reg. § 1.1502-32 are based on certain assumptions regarding shareholders’ interests in the subsidiary. One assumption is that each share within a class is entitled to an equal portion of the subsidiary’s items of income and gain. Another assumption is that the subsidiary’s losses are borne by the holders of the common stock before the holders of the preferred stock.

d. The adjustments are designed to ensure that consolidated group members pay a single corporate tax on the group’s income and use losses only once.

- 2 -

2. Example 1 - Basic Investment Adjustment Rules

S Stock P X $120 $100 basis

S

a. Facts: P acquires all of the stock of S for $100. P and S elect to file a consolidated return. S earns $20 in Year 1. In Year 2, P sells its S stock to X for $120.

b. Under the investment adjustment rules, P’s basis in its S stock is increased by the $20 of taxable income in Year 1 to $120. Reg. § 1.1502-32(b)(2)(i).

c. Then, when P sells S in Year 2, it recognizes no gain or loss. Because of the $20 positive basis adjustment, the P group pays only one tax on its earned income.

B. Targeted Problems

1. Problems Relating to the Repeal of General Utilities

a. Son of Mirrors Transaction

(1) The Tax Reform Act of 1986, Pub. L. No. 99-514, repealed the General Utilities doctrine by requiring corporate-level gain recognition on a corporation’s sale or distribution of appreciated property, regardless of whether it occurs in a liquidating or nonliquidating context.1

(2) After the repeal of the General Utilities doctrine, the operation of the investment adjustment rules permitted consolidated groups to sell assets without paying a

1 In General Utilities and Operating Co. v. Helvering, 296 U.S. 200 (1936), the Supreme Court held that corporations could distribute appreciated property to their shareholders tax-free. The Tax Reform Act of 1986 repealed the General Utilities doctrine by amending section 311(b) of the Code. Section 311(b) imposes a corporate-level tax on the distribution of appreciated property to shareholders, as if the corporation sold such property for its fair market value.

- 3 -

corporate-level tax. The transaction became known as the “son of mirrors” transaction.

(3) Example 2 - Basic Son of Mirrors Transaction

(1) S Stock S Stock X P P Z $200 $200

$200 basis $400 basis (2) Land S S Y S $200 Land $200 Cash $200 Value $0 Basis

(a) Facts: P purchases the stock of S from X for $200. S’s only asset is land with a value of $200 and a basis of $0. P causes S to sell the land to Y for its fair market value of $200. P subsequently sells the S stock to Z for its fair market value of $200.

(b) S recognizes $200 gain on the sale of the land to Y. Under the investment adjustment rules, P increases its basis in the S stock by $200 to $400. Reg. § 1.1502-32(b)(2)(i).

(c) When P subsequently sells S, it recognizes a $200 loss, which offsets the gain recognized by S.

- 4 -

(4) Example 3 - Son of Mirrors Transaction - “Bust Up”

(1) S Stock S Stock X P P Y $200 $100 (2) $200 basis Wanted Asset $100 basis $100 Value $50 Basis S S S

Unwanted Asset Unwanted Asset $100 Value $100 Value $50 Basis $50 Basis

(a) Facts: P purchases the stock of S from X for $200. S has two assets, Unwanted Asset and Wanted Asset, each with a $100 fair market value and a $50 basis. P wants to keep Wanted Asset and dispose of Unwanted Asset. S distributes Wanted Asset to P, and P subsequently sells the S stock to Y for its fair market value of $100.

(b) When S distributes Wanted Asset to P, it results in $50 of section 311(b) gain to S, which is deferred. See Reg. § 1.1502-13(c)(2)(ii). P’s basis in its S stock is reduced by $100 to $100 as a result of the distribution. Reg. §§ 1.1502-13(f)(2)(ii), 1.1502- 32(b)(2)(iv).

(c) P’s subsequent sale of S stock triggers S’s $50 deferred gain. Reg. §§ l.1502-13(d)(1)(i), 1.1502- 13(f)(2)(iii). This gain increases P’s basis in its S stock by $50 to $150. Reg. § 1.1502-32(b)(2)(i). Thus, P recognizes a $50 loss on the sale of its T stock. The $50 loss offsets the $50 section 311(b) gain.

(5) In these examples, S’s built-in gain in its asset was already reflected in P’s initial cost basis in the S stock. Thus, the positive investment adjustment for the gain on the sale or distribution of S’s asset artificially increases P’s basis in its

- 5 -

S stock and permits P to recognize an offsetting loss, in effect eliminating S’s gain from corporate-level tax.

(6) Arguably, the lack of a tax on the disposition of S’s asset in these examples is appropriate. In Example 2, P invested $200 in S and receives $200 cash when S is sold. In Example 3, P invested $200 in S and receives a Wanted Asset worth $100 and $100 in cash when S is sold.

(7) However, the lack of a tax is inconsistent with the purpose of the repeal of the General Utilities doctrine: no corporate-level tax has been paid on the built-in gain in S’s asset, yet the asset has a stepped-up basis in the hands of the buyer.

b. Wasting Assets Problem

(1) The Internal Revenue Service (the “Service”) became concerned that the General Utilities repeal could also be avoided where assets are not actually disposed of but instead are used up in the process of earning income. Thus, where assets are expected to decline in value over time, all or a portion of the income earned from the asset is economically a return of capital.

(2) Example 4 - Wasting Assets

S Stock S Stock X P P Y $200 $200

$200 basis $300 basis

Income S S $20/year for 5 years S

Patent Patent $200 Value $100 Value $0 Basis $0 Basis

$100 Cash

(a) Facts: P purchases the stock of S from X for $200. S’s only asset is a patent with a value of $200 and a

- 6 -

basis of $0. S’s asset earns $20 and declines in value by $20 in each year over a five-year period. P subsequently sells the S stock to Y for its fair market value of $200.

(b) Under the investment adjustment rules, P increases its basis in the S stock by $20 each year, or $100 over the five-year period. Reg. § 1.1502- 32(b)(2)(i).

(c) When P subsequently sells S, it recognizes a $100 loss, which offsets S’s income. Note, however, that the time value of money reduces the effect of this offset.

2. Loss Duplication Problem

a. The Service is also concerned with the ability to duplicate losses of consolidated group members.

b. Example 5 - Loss Duplication: Unrelated Taxpayers

S Stock P P X $40

$100 S Stock $100 Cash Basis

S S

$60 NOL $60 NOL

(1) Facts: P forms S with a contribution of $100. S has an operating loss of $60, which the P group is unable to use on its consolidated return. P subsequently sells S to X for $40.

(2) P’s basis in its S stock remains at $100, because S’s loss has not been absorbed by the group. Reg. § 1.1502- 32(b)(3)(i)(A).

(3) When P sells the S stock to X for $40, P recognizes a $60 loss. S is apportioned its $60 net operating loss carryover

- 7 -

when it leaves the P group. Reg. §§ 1.1502-79(a), 1.1502- 21(b)(2).

(4) P’s loss on the sale of S is thus duplicated when S uses its loss after leaving the P group. However, S is restricted in its use of its apportioned losses by section 382, Reg. §§ 1.1502-21, 1.1502-22 (SRLY rules), etc.

(5) Loss duplication can also occur if S uses the $100 contributed by P to purchase an asset and the asset declines in value to $40.

(6) This sort of loss duplication is not unique to consolidated returns; it also exists when separate returns are filed. Arguably, Congress has already addressed the problem in sections 382, 384, and 269 (not to mention the Treasury’s own response in Reg. §§ 1.1502-15, 1.1502-21, and 1.1502-22).

c. Example 6 - Loss Duplication: Stuffing/Outside Loss Recognized Before Inside Loss

(2) S Preferred Stock P P X $20 (1) S Common Asset S Preferred $100 Cash Stock $20 Value Stock $50 Basis (3) Asset S S Y $20 Asset $20 Value $50 Basis

(1) Facts: In Year 1, P forms S with a contribution of $100 in exchange for all of the common stock of S. In Year 2, P contributes a built-in loss asset to S with a value of $20 and a basis of $50 in exchange for a separate block of S preferred stock. In Year 3, P sells the preferred stock to X for $20. In Year 4, S sells the asset to Y for $20.

- 8 -

(2) P recognizes a $30 loss on the sale of the S preferred stock. Because P continues to own all of the S common stock, S remains a member of the P consolidated group.

(3) When S sells its asset for $20, the P group may use the $30 loss on its return. P is required to reduce its basis in the S common stock by the amount of S’s loss absorbed by the P group, which would result in a gain if P subsequently disposes of the S common stock. Nonetheless, P could delay or avoid recognition of the gain through some planning.

(4) Loss Acceleration – In the context of this fact pattern, the Service also appears concerned with the ability of the group to accelerate economic losses by recognizing the outside loss before S recognizes its inside loss. See Notice 2002- 18, 2002-1 C.B. 644 (Mar. 7, 2002); Reg. § 1.1502-35.

d. Example 7 - Loss Duplication: Stuffing/Inside Loss Recognized Before Outside Loss

(3) S Preferred Stock P P X $20 (1) S Common Asset S Preferred $100 Cash Stock $20 Value Stock $50 Basis (2) Asset S S Y $20 Asset $20 Value $50 Basis

(1) Facts: Same facts as in Example 6, except that S sells its built-in loss asset to Y in Year 3, and P sells the S preferred stock to X in Year 4.

(2) When S sells its asset for $20, it recognizes a $30 loss, which offsets income on the P group’s return. Under the investment adjustment rules, P’s basis in each share of S

- 9 -

common stock is reduced by a pro rata share of the $30 loss. Reg. § 1.1502-32(c)(2)(i). P’s basis in the preferred shares is not, however, reduced. Reg. § 1.1502-32(c)(3).

(3) P then recognizes a $30 loss on the sale of the S preferred stock, which it uses to offset income on the P group’s return.

(4) Note that if P subsequently disposes of the S common stock, it would recognize $30 additional gain. Nonetheless, P could delay or avoid recognition of the gain through some planning.

C. Loss Disallowance Regulations for Dispositions Between 1/31/91 and 3/7/02 – History of Treasury Regulation Section 1.1502-20

1. Regulatory Authority – Congress granted the Service regulatory authority to protect the purposes behind the General Utilities repeal, including “regulations to ensure that such purposes may not be circumvented through the use of any provision of law or regulations (including the consolidated return regulations . . . ).” Section 337(d)(1).

2. Notice 87-14

a. The Service concluded that the result in the son of mirrors transaction undermined the repeal of the General Utilities doctrine and issued Notice 87-14, 1987-1 C.B. 445 (Jan. 6, 1987), in response.

b. In Notice 87-14, the Service announced that it intended to promulgate regulations that would deny positive basis adjustments for earnings and profits (under the former basis adjustment rules) attributable to the sale or distribution of built-in gain property, using a “tracing” method.

c. The Notice also indicated that regulations would be effective with respect to stock in a target that was acquired after January 6, 1987, the date of the Notice.

3. Original Set of Loss Disallowance Regulations

a. On March 9, 1990, temporary and proposed Reg. § 1.1502-20T was promulgated pursuant to Notice 87-14. Temp. Reg. § 1.1502- 20T disallowed any loss recognized on the disposition of a consolidated subsidiary by a consolidated group member.

- 10 -

(1) However, it contained a reattribution rule, which permitted the selling member to elect to reattribute net operating losses of the subsidiary to itself. See Prior Temp. Reg. § 1.1502-20T(f)(1).

(2) Thus, the regulations went far beyond Notice 87-14.

(a) Although Notice 87-14 was targeted to the son of mirrors problem, the original set of regulations reached all of the problems perceived by the Service: son of mirrors, wasting assets, and loss duplication.

(b) In Notice 87-14, the Service indicated that it would adopt a tracing rule to deny positive investment adjustments attributable to the recognition of built- in gain. Tracing would permit a seller of subsidiary stock to establish that it has a true economic loss. But in developing and revising the regulations, the Service rejected tracing as too burdensome on both taxpayers and the Service, because such a rule would require the appraisal of each asset of an acquired subsidiary to determine if built-in gain or built-in loss exists.

b. Temp. Reg. § 1.1502-20T applied to all consolidated subsidiary stock that was disposed of on or after March 9, 1990, regardless of when the stock was acquired.

(1) A transitional rule was promulgated in temporary and proposed Reg. § 1.337(d)-1T(a), which generally applied to subsidiaries acquired after January 6, 1987 and disposed of before March 9, 1990. Under this rule, losses on the sale of such stock were disallowed, except to the extent that the group established that the loss was not attributable to the recognition of built-in gain.

(2) Therefore, despite the reassurances of Notice 87-14, the loss disallowance rules (with one minor exception provided in Reg. § 1.337(d)-1T) did apply to subsidiaries acquired prior to January 7, 1987.

4. Amended Set of Regulations

a. After receiving numerous comments on the loss disallowance regulations, on November 19, 1990, Treasury and the Service promulgated revised regulations.

- 11 -

b. Temp. Reg. § 1.1502-20T was revoked,2 and Proposed Reg. § 1.1502-20 replaced Temp. Reg. § 1.1502-20T.

c. The proposed regulations generally contained the same rules as the original -20T regulations, except that subparagraph (c) added a limited loss allowance rule, which permitted losses to be recognized to the extent they exceeded (i) income or gain from extraordinary gain dispositions, (ii) the amount of positive investment adjustments, and (iii) the amount of any duplicated loss.

d. Prop. Reg. § 1.1502-20 generally applied to subsidiary stock disposed of after January 31, 1991.

e. The transitional rule in Temp. Reg. § 1.337(d)-1T was slightly amended and made final. Reg. § 1.337(d)-1. It is generally applicable to subsidiaries acquired after January 6, 1987 and disposed of before November 19, 1990.3

f. Temporary and proposed Reg. § 1.337(d)-2T generally carried forward the transitional rule from November 19, 1990 to January 31, 1991.4 Temp. Reg. § 1.337(d)-2T applied, however, to all subsidiaries, regardless of when they were acquired.

5. Final Regulations

a. On September 13, 1991, Treasury and the Service issued final regulations under Reg. §§ 1.1502-20 and 1.337(d)-2 and slightly

2 A group could, however, elect to apply the former Temp. Reg. § 1.1502-20T in lieu of the transitional rules of Reg. §§ 1.337(d)-1 and -2, or with respect to pre-effective date transactions, in order to take advantage of the reattribution rule. Reg. §§ 1.337(d)-1(e)(3), 1.337(d)-2(g)(3), 1.1502-20(h)(4).

3 However, this rule may apply to stock disposed of after November 18, 1990. If stock of a transitional subsidiary was deconsolidated before November 19, 1990, and the remaining subsidiary stock held by the group was not subject to Reg. §§ 1.337(d)-2 or 1.1502-20, then the subsidiary continued to be treated as a transitional subsidiary. Reg. § 1.337(d)-1(e)(1).

4 A selling group could, however, elect out of Reg. § 1.337(d)-2 and into the general loss disallowance regulations of Reg. § 1.1502-20. Reg. § 1.1502-20(h)(2).

- 12 -

modified Reg. § 1.337(d)-1 (collectively, the “final 1991 regulations”).5

b. Although the final 1991 regulations added some important provisions to Reg. § 1.1502-20, they retained the same approach as in the proposed regulations. The government acknowledged that the loss disallowance rule of Reg. § 1.1502-20 would disallow economic losses. Preamble to the final 1991 regulations, 56 Fed. Reg. 47,379, 47,380-82 (Sept. 13, 1991).

c. The final 1991 regulations did not change any of the effective dates. Consolidated groups could avoid the loss disallowance regulations and the “window period” transition rules in Temp. Reg. § 1.337(d)-2T by seeking permission to deconsolidate pursuant to Rev. Proc. 91-11, 1991-1 C.B. 470.

d. Rev. Proc. 91-11 originally set the final date for filing an application to discontinue filing consolidated returns as June 30, 1991.

e. However, in Rev. Proc. 91-39, 1991-2 C.B. 694, the Service modified Rev. Proc. 91-11 by providing that such applications must be filed no later than 90 days after the date proposed Reg. § 1.1502-20 was finalized (i.e., by December 12, 1991).

6. Summary – The following table summarizes the applicable loss disallowance provisions, which apply when a subsidiary is acquired and disposed of as follows:

Subsidiary acquired before Subsidiary acquired on or 1/7/87: after 1/7/87: Disposed of on or after 11/19/90 but § 1.337(d)-2 § 1.337(d)-2 before 2/1/91: Disposed of on or after 1/7/87 but LDRs do not apply § 1.337(d)-1 before 11/19/90:

7. Removal of -20 and Related Regulations – On January 23, 2007, the Service published proposed consolidated return loss disallowance rules that would both implement the repeal of the General Utilities doctrine and

5 See F.S.A. 199916007 (Jan. 2, 1999) (discussing the history of Reg. § 1.1502-20 and concluding that Reg. § 1.1502-20 was promulgated in accordance with the Administrative Procedure Act).

- 13 -

address the duplication of losses by members of a consolidated group. These “Unified Loss Rules” were finalized on September 17, 2007. The final regulations thus remove Reg. § 1.1502-20 and provide that Reg. § 1.337(d)-1, 1.337(d)-2, and 1.1502-35 do not apply to transactions subject to the Unified Loss Rules. T.D. 9424, 73 Fed. Reg. 53,933, 53,944 (Sept. 17, 2008); Reg. §§ 1.337(d)-1(a)(1), 1.337(d)-2(a)(1), 1.1502- 35(a)(2)(iii).

D. Rite Aid Case

1. Background

a. The approach of Reg. § 1.1502-20 had been widely criticized in that it disallows economic losses that would otherwise be deductible in a separate return context. Indeed, taxpayers challenged the validity of Reg. § 1.1502-20. See, e.g., Rite Aid Corp. v. United States, 46 Fed. Cl. 500 (2000), rev’d, 255 F.3d 1357 (Fed. Cir. 2001); Salina Partnership LP v. Commissioner, 80 T.C.M. (CCH) 686 (2000); FPL Group, Inc. v. Commissioner, T.C. Docket No. 10811-00.

b. The Court of Appeals for the Federal Circuit held that the government’s attempt to disallow losses capable of duplication was invalid. Rite Aid Corp. v. United States, 255 F.3d 1357 (Fed. Cir. 2001), rev’g, 46 Fed. Cl. 500 (2000). The Federal Circuit concluded that disallowing a loss that would otherwise be deductible under section 165 amounts to an imposition of tax on income that would otherwise not be taxed, which the government is not authorized to do under section 1502.

2. Facts of Rite Aid

a. Rite Aid Corporation (“Rite Aid”) is the common parent of an affiliated group of corporations that files a consolidated return. In 1984, Rite Aid acquired 80 percent of the stock of Penn Encore, Inc. (“Encore”) for $3 million. A section 338 election was made with respect to the acquisition. In 1988, Rite Aid purchased the remaining 20 percent of Encore stock for $1.5 million. From 1984 through 1994, Encore experienced net negative earnings and profits of approximately $10.9 million and borrowed approximately $44.9 million from Rite Aid.

b. In January 1994, Rite Aid adopted a restructuring plan, which included the sale of Encore. Rite Aid asked prospective bidders whether they would join in making an election under section

- 14 -

338(h)(10). The only bidder for Encore refused to join in a section 338(h)(10) election.

c. On November 23, 1994, Rite Aid sold all of the Encore stock, claiming a loss of approximately $22.1 million. However, the loss was disallowed under Reg. § 1.1502-20. Rite Aid determined that Encore’s duplicated loss factor was approximately $28.5 million, its extraordinary gain disposition factor was $9,624, and its positive investment adjustment factor was approximately $6.2 million. Because the sum of the loss disallowance factors, or approximately $34.7 million, exceeded Rite Aid’s investment loss, Rite Aid’s entire loss was disallowed under Reg. § 1.1502-20.

d. Rite Aid paid the tax and filed a claim for refund in the United States Court of Federal Claims, claiming that Reg. § 1.1502-20 was invalid.

3. Court of Federal Claims Decision

a. The Court of Federal Claims held that Reg. § 1.1502-20 was not arbitrary, capricious, or manifestly contrary to law, and that it served the purpose of clearly reflecting income tax liability of both the parent and the subsidiary in a consolidated group.

b. The court rejected the taxpayer’s argument that the regulation was in derogation of section 165(a), which permits a deduction for losses sustained during the taxable year, and thus exceeded the Treasury’s authority. The court noted that the taxpayer had an opportunity, which it did not take, to structure the sale of the subsidiary in a way that would have allowed the taxpayer to recognize the loss (i.e., as an asset sale instead of a stock sale).

c. The court also rejected the taxpayer’s argument that the Supreme Court’s decision in Illfield Co. v. Hernandez stands for the proposition that a duplicated loss is a loss twice enjoyed by the group – not by unrelated parties. The court noted that section 1502 permits the Treasury to implement regulations that clearly reflect income tax liability with respect to consolidated groups and their members “both during and after the period of affiliation,” which is served by prohibiting group losses that otherwise may be taken both by the group and its former member.

4. The Federal Circuit Decision

a. The Court of Appeals for the Federal Circuit reversed the Court of Federal Claims’ decision that Reg. § 1.1502-20 was a proper exercise of Treasury’s regulatory authority. The Federal Circuit

- 15 -

held that duplicated loss factor of the loss disallowance rules “distorts rather than reflects the tax liability of consolidated groups” and, therefore, the regulation is manifestly contrary to the statute.

b. The Federal Circuit held that “in the absence of a problem created from the filing of consolidated returns, the Secretary is without authority to change the application of other tax code provisions to a group of affiliated corporations filing a consolidated return.” Because the ability of a former consolidated subsidiary to realize a loss on its assets after the consolidated group realizes a loss on the subsidiary’s stock is not limited to the consolidated return context, the Secretary is without authority to change the application of section 165 to the sale of the subsidiary’s stock.

(1) This standard could have broad implications with respect to other consolidated return regulations, because the consolidated return regulations change the application of many tax code provisions. See, e.g., Reg. §§ 1.1502- 13(f)(2); -13(f)(6); -13(g)(3)(ii)(B)(2); -19; -32; -80.

(2) The Federal Circuit did not offer much guidance as to when a problem is “created” by the filing of consolidated returns.

c. The Federal Circuit noted that Congress has already addressed the problem of duplicated losses by limiting the subsidiary’s potential future deduction under sections 382 and 383 – not by disallowing the parent’s loss on the subsidiary stock.

d. The Federal Circuit rejected the government’s argument that if an affiliated group elects to take advantage of the benefits of filing a consolidated return, “it must take the bitter with the sweet,” noting that the “‘bitter with the sweet’ does not include the invalid.”

e. It is not clear from the court’s opinion whether it invalidated all of Reg. § 1.1502-20 or just the duplicated loss provision. The court stated that the “regulation” was manifestly contrary to the statute. Nonetheless, the court’s analysis clearly focuses on only the duplicated loss provision, and the Service has interpreted the opinion to invalidate only the duplicated loss provision. See Notice 2002-11, 2002-7 I.R.B. 1 (Jan. 31, 2002).

5. Notice 2002-11

a. On January 31, 2002, the Service issued Notice 2002-11, announcing that it would not seek certiorari from the Supreme Court in Rite Aid. The Notice stated that the government would

- 16 -

not continue to litigate the validity of the loss duplication factor “in the interests of sound tax administration.”

b. The Notice further provided that because of the interrelationship between all of the loss disallowance factors, Reg. § 1.1502-20 would be replaced in its entirety with interim regulations based on Reg. § 1.337(d)-2.

c. Since it issued this Notice, the Service pursued General Utilities repeal concerns separately from loss duplication concerns. However, the finalized Unified Loss Rules comprehensively address both General Utilities repeal and loss duplication concerns. Reg. § 1.1502-36, T.D. 9424 (Sept. 17, 2007), 73 Fed. Reg. 53,933 (Sept. 17, 2008).

6. Legislative Response – Section 646 of the American Jobs Creation Act of 2004 (H.R. 4520) amended section 1502 to limit the decision in Rite Aid by adding the following sentence to section 1502:

(a) IN GENERAL - In carrying out the preceding sentence, the Secretary may prescribe rules that are different from the provisions of chapter 1 that would apply if such corporations filed separate returns.

(b) RESULT NOT OVERTURNED – Notwithstanding the amendment made by subsection (a), the Internal Revenue Code of 1986 shall be construed by treating Treasury Regulation Sec. 1.1502-20(c)(1)(iii) (as in effect on January 1, 2001) as being inapplicable to the factual situation in Rite Aid Corporation and Subsidiary Corporations v. United States, 255 F.3d 1357 (Fed. Cir. 2001).

a. The Conference Report (H.R. Conf. Rep. No. 108-755) explains that the reason for the provision is the Finance Committee’s concern that the language of the Rite Aid opinion may lead taxpayers to challenge other consolidated return regulations that prescribe a result that is different from the separate return result. It also states that the provision in no way prevents or invalidates the approaches Treasury has announced it will apply in lieu of Reg. § 1.1502-20. See also H.R. Rep. No. 108-548; S. Rep. No. 108- 192.

- 17 -

E. Loss Disallowance Regulations for Dispositions Between 3/7/02 and 9/17/08 – Treasury Regulation Section 1.337(d)-2

1. On March 7, 2002, Temp. Reg. §§ 1.337(d)-2T, 1.1502-20T(i), and 1.1502-32T(b)(4)(v) were promulgated pursuant to Notice 2002-11. See 67 Fed. Reg. 11,034 (Mar. 12, 2002). These regulations were further amended in May 2002, May 2003, and March 2004. See 67 Fed. Reg. 37,998 (May 31, 2002); 68 Fed. Reg. 24,351 (May 7, 2003); 69 Fed. Reg. 12,799 (Mar. 18, 2004). The regulations were adopted as final regulations, without substantive change, on March 3, 2005.

2. The regulations address only the son of mirrors problem discussed above. At the same time the temporary regulations were issued, the Service issued Notice 2002-18, 2002-1 C.B. 644, in which it announced its intention to issue regulations to address loss duplication concerns.

3. Reg. § 1.337(d)-2 is generally applicable for dispositions of stock occurring on or after March 3, 2005.

a. Reg. § 1.337(d)-2 largely employs the rules that were in former Reg. § 1.337(d)-2. Losses on subsidiary stock are disallowed except to the extent that the parent establishes that the loss is not attributable to built-in gain on the disposition of an asset. Reg. § 1.337(d)-2(c)(2).

b. However, Reg. § 1.337(d)-2 differs from former Reg. § 1.337(d)-2 in that the selling group is no longer required to dispose of its entire interest in the subsidiary. Reg. § 1.337(d)-2.

4. For dispositions of stock occurring before March 7, 2002, or for dispositions or deconsolidation of stock of a subsidiary after March 7, 2002 effected pursuant to a binding written contract entered into before March 7, 2002 that was in continuous effect, Temp. Reg. § 1.1502- 20T(i)(2) allows a parent to choose one of three regulatory schemes for each separate disposition of subsidiary stock. These regulations were adopted as final regulations, without substantive change, on March 3, 2005.

a. Reg. § 1.1502-20 in its entirety;

b. Reg. § 1.1502-20 without regard to the loss duplication factor; or

c. Reg. § 1.337(d)-2.

5. Notice 2004-58. On August 25, 2004, the Service issued Notice 2004-58 announcing a method that it will accept for determining the extent to which loss or basis is attributable to the recognition of built-in gain on the

- 18 -

disposition of an asset for purposes of applying the exception to the loss disallowance rule in Reg. § 1.337(d)-2.

a. This method, the basis disconformity method, disallows loss in an amount equal to the least of: (i) the gain amount, which is the sum of all gains recognized on asset dispositions of the subsidiary while a member of the group; (ii) the disconformity amount, which is the excess of the share’s basis over the share’s proportionate interest in the subsidiary’s net asset basis; and (iii) the positive investment adjustment amount, which is the excess of the sum of all positive investment adjustments over the sum of all negative investment adjustments (excluding distributions) made to the share.

6. At the same time Notice 2004-58 was issued, Treasury and the Service issued temporary regulations permitting taxpayers to make, amend, or revoke elections under Temp. Reg. § 1.1502-20T(i). See Temp. Reg. § 1.1502-20T(i)(6).

7. The final Unified Loss Rules modify Reg. §§ 1.337(d)-1 and 1.337(d)-2 to state explicitly that they do not apply to transactions subject to the Unified Loss Rules. T.D. 9424, 73 Fed. Reg. at 53,944; Reg. §§ 1.337(d)-1(a)(1), 1.337(d)-2(a)(1).

F. Loss Duplication Regulations for Dispositions Between 3/7/02 and 9/17/08 – Treasury Regulation Section 1.1502-35

1. Notice 2002-18 - On March 7, 2002, the Service issued Notice 2002-18 announcing its intention to issue regulations that will prevent a consolidated group from obtaining more than one tax benefit from a single economic loss. The Notice stated that such regulations will apply to dispositions of stock occurring on or after March 7, 2002.

a. The Notice cited the following example of the type of duplicated loss to be targeted by the forthcoming regulations: A member of the consolidated group (the “transferor”) contributes a built-in loss asset to another member of the group (the “transferee”) in exchange for stock of the transferee in a transaction in which the basis of such stock is determined by reference to the basis of the transferred asset. The transferor then sells the transferee stock without causing a deconsolidation of the transferee, thus permitting the group to benefit from the built-in loss in the asset twice.

b. The Notice appears to have been triggered in part by press surrounding a similar transaction undertaken by Bank of America. See, e.g., Sheppard, Bank of America’s Tax Plan for Bad Loans, 2002 TNT 38-5 (Feb. 11, 2002); Mollenkamp, Rare Use of Tax

- 19 -

Law Helps Lift Bank of America to Hefty Profit, Wall St. J., at A2 (Jan. 24, 2002).

c. Thus, the Notice appears to target the stuffing-type loss duplication transactions illustrated in Examples 6 and 7, above. The loss duplication illustrated in Example 5, above, where the subsidiary’s inside loss can be used by it after it leaves the group, was the fact scenario that led to the invalidation of the loss duplication factor in Rite Aid. Such duplication outside the consolidated group was not the apparent target of the Notice.

2. Proposed Treasury Regulation Section 1.1502-35

a. On October 23, 2002, Treasury and the Service issued proposed regulations to implement Notice 2002-18. Prop. Reg. § 1.1502-35. Consistent with the Notice, the stated purpose of the proposed regulations was to prevent a consolidated group from obtaining more than one tax benefit from a single economic loss.

b. The proposed regulations contained a complicated set of rules:

(1) Basis Redetermination Rule - If a member of a consolidated group disposes of stock of a subsidiary member, or a share of subsidiary member stock is deconsolidated, at a loss, then all members’ bases in the subsidiary stock are aggregated and reallocated among the common and preferred stock of the subsidiary.

(2) Loss Suspension Rule - If after applying the basis redetermination rule, a member of a consolidated group still recognizes a loss on the disposition of stock of a subsidiary, then the selling member’s loss is suspended to the extent of any duplicated loss.

(3) Basis Reduction Rule - If a subsidiary’s stock becomes worthless or the subsidiary disappears in a transaction in which gain or loss is recognized, then any consolidated net operating loss allocable to the subsidiary is treated as absorbed, which results in a basis reduction under the investment adjustment rules of Reg. § 1.1502-32.

(4) The proposed regulations also contained some anti-abuse rules.

c. The regulations were generally proposed to apply retroactively to transactions that occur on or after March 7, 2002, the date of Notice 2002-18.

- 20 -

3. Temporary Treasury Regulation Section 1.1502-35T

a. Notwithstanding the criticism of the proposed regulations, see American Bar Association Section of Taxation, Comments on Consolidated Group Basis Redetermination and Loss Suspension (Feb. 20, 2003); New York State Bar Association Tax Section, Report on Temporary Regulation § 1.337(d)-2T and Proposed Regulation § 1.1502-35 (Feb. 28, 2003), on March 14, 2003, Treasury and the Service issued temporary regulations that were substantially similar to the proposed regulations, but reflected certain revisions based on the comments received.

b. Treasury and the Service made it clear in the preamble to the temporary regulations that they were continuing to study the comments they received and specifically requested comments on alternative regimes that they were considering.

4. Final Treasury Regulation § 1.1502-35

a. On March 9, 2006, the temporary regulations were issued as final regulations without substantive change. The preamble to the final regulations states that Treasury and the Service are continuing to study the issues raised by both Reg. § 1.337-2 and Reg. § 1.1502- 35. The preamble states that Treasury and the Service intend to publish proposed regulations “in the near term” addressing both circumvention of General Utilities repeal and loss duplication in a single integrated regulation.

b. The final Unified Loss Rules modify Reg. § 1.1502-35 to state explicitly that it does not apply to transactions subject to the Unified Loss Rules. T.D. 9424, 73 Fed. Reg. at 53,944; Reg. § 1.1502-35(a)(2)(iii).

G. Current Unified Loss Rules for Dispositions After 9/17/08 – Treasury Regulation Section 1.1502-36

1. On September 17, 2008, Treasury and the Service issued final Unified Loss Regulations. The final regulations adopted regulations that were proposed in January 2007 with modest changes. The final Unified Loss Rules address both the General Utilities repeal and loss duplication with an integrated set of rules.

2. As a result, the regulations modify Reg. §§ 1.337(d)-1, 1.337(d)-2, and 1.1502-35 to state explicitly that they do not apply to transactions subject to the Unified Loss Rules. T.D. 9424, 73 Fed. Reg. at 53,944; Reg. §§ 1.337(d)-1(a)(1), 1.337(d)-2(a)(1), 1.1502-35(a)(2)(iii). Additionally, the final Unified Loss Rules modify the loss suspension rule to provide

- 21 -

that it applies only to losses allowed within ten years of the date that they are recognized and that it ceases to apply ten years after the stock disposition that gave rise to the suspended loss. 73 Fed. Reg. at 53,944.

H. Avoiding the Unified Loss Rules or Prior Loss Disallowance and Duplication Rules

1. The regulations disallow losses on the sale of subsidiary stock, not assets. As such, the selling group should consider structuring the transaction as an asset sale or as a stock sale with a section 338(h)(10) election. If the subsidiary sells its assets at a loss, the loss will be recognized.

2. If a subsidiary and buyer jointly make a section 338(h)(10) election, a stock sale will be treated as a deemed asset sale, with the following consequences:

a. The subsidiary is treated as selling its assets while it was a member of the group;

b. The subsidiary is deemed to be completely liquidated under section 332, and all of the subsidiary’s tax attributes move up to the parent;

c. The buyer acquires none of the tax attributes of the subsidiary and takes a fair market value basis in the subsidiary’s assets; and

d. Gain or loss on the sale of stock is ignored.

3. For example, assume that P owns all of the stock of S. P’s basis in the S stock is $550 and its value is $500. S holds one asset, which has a basis of $550 and a value of $500. X wants to purchase the S stock for $500. If the S stock is sold, P would recognize a $50 loss, which would be disallowed. Instead, P and X make an election under section 338(h)(10) to treat the stock sale as a deemed asset sale.

a. S recognizes a $50 loss that is includable on P’s consolidated return.

b. S’s tax attributes move up to P.

c. X takes the asset with a step-down in basis to its value of $500, and X loses all of the tax attributes of S.

d. If a section 338(h)(10) election is not made, X acquires S with its tax attributes, including its built-in loss assets, preserved. As a result, even though section 382, SRLY, etc. apply, X may not be

- 22 -

willing to join in a section 338(h)(10) election or may pay less for the S stock before X will agree to a section 338(h)(10) election.

4. As an alternative to a section 338(h)(10) election, P could merge S into a single-member limited liability company (“LLC”) formed by P and sell all of the LLC interests to X. Because the LLC is disregarded as an entity separate from P, P would be treated as selling the assets of the LLC to X. See Reg. §301.7701-2(a).

5. As another alternative, P could cause S to distribute its asset to P before selling the S stock to X. Under Reg. § 1.1502-13(f)(2)(iii), the principles of section 311(b) apply to intercompany distributions. Thus, S’s $50 loss would be deferred under Reg. § 1.1502-13(c) and triggered when S leaves the group under Reg. § 1.1502-13(d). P would reduce its basis in the S stock by the $50 loss on the distribution and by the $500 value of the asset. Reg. § 1.1502-32(b)(2)(i), (iv). Because the value of S is reduced by the value of the property no longer owned by S, P would recognize no gain or loss on the sale of the S stock to X. See F.S.A. 200012046 (Dec. 9, 1999); George White, Loss Disallowance Regulations Flanked?, 41 Tax Mgmt. Memo. 248 (June 19, 2000).

II. CURRENT UNIFIED LOSS RULES FOR LOSS ON SUBSIDIARY STOCK

A. Background

1. After Rite Aid and the Service’s withdrawal of the loss duplication factor of Prior Reg. § 1.1502-20, Treasury and the Service have been studying ways to deal with (i) noneconomic losses that circumvent the repeal of the General Utilities doctrine and (ii) duplicated losses. Until Reg. § 1.1502- 36, efforts to deal with the two problems have followed separate but parallel tracks – Prior Reg.§ 1.337(d)-2 addressed noneconomic losses and Prior Reg. § 1.1502-35 addressed duplicated losses.

2. Reg. § 1.1502-36 adopts a single integrated approach to the two problems. The regulations containing these Unified Loss Rules were proposed in January 2007 and then finalized with a few modest changes in September 2008. Reg. § 1.1502-36, T.D. 9424; Prop. Reg. § 1.1502-36, 72 Fed. Reg. 2964, 2965-66 (Jan. 23, 2007). The Unified Loss Rules apply to transfers of subsidiary stock on or after September 17, 2008. Reg. § 1.1502-36(h).

3. Purposes of the Unified Loss Rules

a. The Unified Loss Rules have two principal purposes:

(1) The first is to prevent the consolidated return provisions from reducing a group’s consolidated taxable income

- 23 -

through the creation of a non-economic loss on S stock. Reg. § 1.1502-36(a)(2).

(2) The second is to prevent members (including former members) of the group from collectively obtaining more than one tax benefit from a single economic loss. Reg. § 1.1502-36(a)(2).

b. The investment adjustment system of Reg. § 1.1502-32 generally prevents noneconomic losses. However, because that system is a presumptive one based on certain operating assumptions, it may result in noneconomic gain or loss where the assumptions do not correspond to the facts. 72 Fed. Reg. at 2966.

(1) For example, the investment adjustment rules assume that all of a subsidiary’s items taken into account represent economic gain or loss, which may not be the case where a purchased subsidiary holds appreciated assets.

(2) Another example is that the investment adjustment rules assume that items accrue economically to all shares equally within a class, which may not be the case where shares have disparate bases.

c. In addition, the investment adjustment system prevents duplicated losses, but only if S’s inside loss in its assets is recognized before P’s outside loss in S’s stock. If the outside loss is recognized first, S may still benefit from the inside loss.

4. Considerations to Address General Utilities Repeal

a. The Service and Treasury rejected a tracing approach similar to Prior Reg. § 1.337(d)-2 because such an approach was determined to be inadministrable. 72 Fed. Reg. at 2970.

(1) Tracing is further complicated by redetermination events, or events that alter the relationship between the basis of a share and the interest it represents. Such events include intragroup spin-offs or section 351 exchanges. 72 Fed. Reg. at 2970.

(2) Additional difficulties are presented when tainted appreciation is recognized through wasting or consumption rather than gain on the disposition of the asset. 72 Fed. Reg. at 2970.

- 24 -

(3) The Service and Treasury also wanted to avoid valuation issues. 72 Fed. Reg. at 2970.

b. The Service and Treasury also rejected a hybrid-tracing presumptive model that would identify all assets held when a share is acquired and on each redetermination date and presume all items traced to those assets to be tainted. This approach was rejected because it required identification of all redetermination events and could potentially be easily manipulated. 72 Fed. Reg. at 2972.

c. The Service and Treasury rejected the basis disconformity approach of Notice 2004-58 because it was under inclusive in that it addresses only noneconomic stock losses to the extent of net appreciation reflected in basis, which by its nature is reduced by unrecognized depreciation reflected in basis. 72 Fed. Reg. at 2972.

(1) Instead of modifying the basis disconformity approach, the Service and Treasury incorporated elements of it into its new, unified approach discussed below. 72 Fed. Reg. at 2972-73.

5. Consideration to Address Loss Duplication Concern

a. The Service and Treasury reconsidered the appropriateness of Prior Reg. § 1.1502-35 in allowing subsidiaries to duplicate group losses after the period of consolidation. 72 Fed. Reg. at 2975-76.

b. The Service and Treasury rejected a pure loss disallowance rule because it would violate Rite Aid in deconsolidating transfers. 72 Fed. Reg. at 2976.

c. The Service and Treasury rejected the use of loss duplication suspense accounts because they would present tracing issues. 72 Fed. Reg. at 2976.

6. The Unified Approach Incorporates the Following Conventions

a. Irrebuttable presumptions. 72 Fed. Reg. at 2975.

b. Loss limitation model which disallows loss duplication and also reduces gain duplication. 72 Fed. Reg. at 2973.

c. Apply basis reduction immediately before disposition which avoids the need for separate rules for disposition and deconsolidation. 72 Fed. Reg. at 2973.

- 25 -

d. Limit basis reduction to the amount of basis disconformity. 72 Fed. Reg. at 2975-75.

e. Allow netting of all investment adjustments made to a share for all periods. 72 Fed. Reg. at 2975.

f. Reduce subsidiary’s attributes to extent of duplicated losses. 72 Fed. Reg. at 2977.

g. Reallocation of disparate investment adjustments of only transferred shares, rather than full blending as required by Prior Reg. § 1.1502-35. 72 Fed. Reg. at 2978.

B. General Rule: Reg. § 1.1502-36

1. Treasury regulation § 1.1502-36 provides rules for adjusting members’ bases in stock of a subsidiary (S) and for reducing S’s attributes when a member (M) transfers a loss share of S stock. See Reg. § 1.1502-36(a)(1). § 1.1502-36 replaces Prior Reg. § 1.337(d)-2 (including Notice 2004-58) and § 1.1502-35. However, the loss suspension and reimportation rules in Prior Reg. §§ 1.1502-35 and 1.337(d)(2) continue to apply to transactions prior to September 17, 2008. These rules are discussed in more detail below.

2. The Unified Loss Rules apply when a member “transfers” a share of subsidiary stock and, after taking into account the effects of all applicable rules of law, including those that would not be effective until after the transfer, the share is a loss share. Reg. § 1.1502-36(a)(1), (a)(3)(i).

3. A “transfer” of stock includes any event in which:

a. gain or loss would be recognized (Reg. § 1.1502-36(f)(10)(i)(A));

(1) A transfer does not generally include a nonrecognition transaction. However, a section 332 liquidation of a subsidiary into multiple members is treated as a transfer. Reg. § 1.1502-36(f)(10)(ii).

b. the holder of a share and the subsidiary cease to be members of the same group (Reg. § 1.1502-36(f)(10)(i)(B));

c. a nonmember acquires an outstanding share from a member (Reg. § 1.1502-36(f)(10)(i)(C)); or

d. the share is treated as worthless (Reg. § 1.1502-36(f)(10)(i)(D)).

- 26 -

4. Generally, the three rules apply in the order described: (i) the basis redetermination rule; (ii) the basis reduction rule; and (iii) the attribute reduction rule.

5. Reg. § 1.1502-36 applies and has effect immediately upon the transfer of a loss share even if the loss is deferred, disallowed, or otherwise not taken into account under any other applicable rules of law. Reg. § 1.1502- 36(a)(4).

a. However, as discussed below, Reg. § 1.1502-36(e)(3) defers the application of the Unified Loss Rules until the triggering of an intercompany loss under Reg. § 1.1502-13.

6. If members transfer stock of multiple subsidiaries in one transaction, the basis redetermination and basis reduction rules apply first with respect to transfers of loss shares of stock of the subsidiaries at the lowest tier and then successively to transferred shares at each next higher tier. Reg. § 1.1502-36(a)(3)(ii).

a. These rules are not applied at any tier until any gain or loss recognized (even if disallowed) on lower-tier transfers and any items resulting from lower-tier adjustments (whether required by the basis redetermination or basis reduction rule or otherwise) are taken into account and reflected in stock basis. Id.

b. After the basis redetermination and reduction rules have applied with respect to all transferred loss shares, the attribute reduction rule applies with respect to the highest-tier transferred loss shares. The attribute reduction rule then applies successively with respect to transferred loss shares at each next lower tier. Id.

7. General Application of Reg. § 1.1502-36:

a. The Unified Loss Rules in Reg. § 1.1502-36 consist of three principal rules that apply when a member transfers a loss share of subsidiary stock:

(1) The first rule, Reg. § 1.1502-36(b), the basis redetermination rule, redetermines members’ bases in subsidiary stock by reallocating Reg. § 1.1502–32 adjustments (to adjust for disproportionate reflection of gains and losses in the bases of members’ shares). Reg. § 1.1502-36(a)(3)(ii)(A).

(2) The second rule, Reg. § 1.1502-36(c), the basis reduction rule, reduces members’ bases in transferred loss shares (but not below value) by the net positive amount of all

- 27 -

investment adjustments applied to the bases of those shares, but only to the extent of the share’s disconformity amount (to address non-economic stock loss). Reg. § 1.1502- 36(a)(3)(ii)(A).

(3) The third rule, Reg. § 1.1502-36(d), the attribute reduction rule, reduces the subsidiary’s attributes to prevent the duplication of a loss recognized on, or preserved in the basis of, transferred stock. Reg. § 1.1502-36(a)(3)(ii)(A).

(4) Reg. §§ 1.1502-36(e), (f), and (g), provide general operating rules,6 definitions, and an anti-abuse rule, respectively.

b. Example 8 – Overview of Unified Loss Rules:

S2 Stock S Stock B M A Cash Cash BIL S2 Stock S Cash 1 share BIL BIG S1

1 share BIG S2

(1) Facts: M owns all the outstanding shares of S stock and one of the two outstanding shares of S2 stock, S owns all the outstanding shares of S1 stock, and S1 owns the other outstanding share of S2 stock. The S and S1 shares are loss shares and the S2 shares are gain shares. As part of one transaction, M sells all the S shares and its S2 share, and S1 sells its S2 share. The sales are to unrelated individuals, and S and S1 do not elect to file a consolidated return after

6 Specifically, predecessor/successor rules, effects of prior §362(e)(2) transactions, application of the Unified Loss Rules to intercompany transactions and section 332 liquidations, and election procedures.

- 28 -

the transaction. Each share is transferred under Reg. § 1.1502-36 (the S and S2 shares because S and S2 cease to be owned by M, and M and S1, respectively, as a result of taxable dispositions, and the S1 shares because S and S1 cease to be members of the same group). Reg. § 1.1502-36 applies to the transfer of the S and S1 (loss) shares, but not to the transfer of the S2 (gain) shares.

(2) The gain recognized on the transferred S2 shares tiers up to adjust members’ bases in all upper-tier subsidiary shares under the principles of § 1.1502–32. Then, if S’s transferred S1 shares are still loss shares, § 1.1502-36(b) and (c) apply to those shares. The loss on the S1 shares is not recognized in the transfer (because there is no taxable disposition of the shares) and so the adjustments to the basis of the S1 shares required by § 1.1502-36(b) and (c) only tier up to adjust M’s basis in the S stock. Then, if M’s transferred shares of S stock are still loss shares, Reg. § 1.1502-36(b) and (c) apply with respect to those shares.

(3) If, after giving effect to any adjustments under Reg. § 1.1502-36(b) and (c), any of the S shares are still loss shares, Reg. § 1.1502-36(d) applies with respect to the transfer of those shares. If any transferred S1 shares are still loss shares after the application of Reg. § 1.1502-36(d) with respect to the transfer of S shares, Reg. § 1.1502-36(d) applies with respect to the transfer of the S1 shares. Reg. § 1.1502-36(a)(3)(ii)(C).

8. The regulations also address section 362(e)(2) transactions. The Service and Treasury concluded that section 362(e)(2) should generally not apply to intercompany transactions. Reg. § 1.502-80(h).

a. The purpose of Reg. § 1.502-80(h) is to allow the consolidated return provisions to address loss duplication. 73 Fed. Reg. at 53,945.

b. The proposed regulations, which suspended the application of section 362(e)(2) for intercompany transactions, were not adopted because they were determined to be too complex and administratively burdensome. 73 Fed. Reg. at 53,935.

(1) The final regulations contain an anti-abuse rule to protect against any concern on the part of the Service and Treasury that the inapplicability of section 362(e)(2) could be used to reach inappropriate results. Id.

- 29 -

c. The Service and Treasury recognized that, because section 362(e)(2) continues to apply prior to September 17, 2008, distortions could result if the taxpayer does not elect to apply the rule in the final regulations. Thus, the final regulations retain the rule in Prop. Reg. § 1.1502-36(e)(2) that provided for adjustments to offset the effects of basis reductions under section 362(e)(2). 73 Fed. Reg. at 53,937.

(1) To adjust for distortions resulting from basis reduction under section 362(e)(2)(A) and similar cases, the regulations adjust the disconformity amount of the shares received in the transaction, and the attribute reduction amount upon the transfer of such shares, by the amount the basis of such shares would have been reduced had an election under section 362(e)(2)(C) been made. Reg. § 1.1502-36(e)(2)(i), (iii).

(2) To adjust for distortions resulting from basis reduction under section 362(e)(2)(C) and similar cases, the regulations reduce S’s net inside attribute amount by the amount S’s attributes would have been reduced under section 362(e)(2)(A) had no election under section 362(e)(2)(C) been made for purposes of computing the basis disconformity amount or the aggregate inside loss. Reg. § 1.1502-36(e)(2)(ii), (iii).

9. The final Unified Loss Rules apply to transfers of subsidiary stock on or after September 17, 2008, unless the transfer is made pursuant to a binding agreement that was in effect before September 17, 2008, and at all times thereafter. Reg. § 1.1502-36(h).

C. Basis Redetermination to Reduce Disparity7

1. Purpose and scope of basis redetermination rule

a. The basis redetermination rule reduces the extent to which there is disparity in members’ bases in shares of S stock. The rule is intended to prevent the operation of the investment adjustment8

7 See Section V.A., below, for examples illustrating the basis redetermination rule.

8 The term “investment adjustment” means the adjustment for items described in § 1.1502-32(b)(2), excluding Reg. § 1.1502-32(b)(2)(iv) (distributions). The term includes all such adjustments reflected in the basis of the share, whether originally applied directly by Reg. § 1.1502-32 or otherwise. The term therefore includes investment adjustments reallocated to the share, and it does not include investment adjustments reallocated from the share, whether

- 30 -

system from creating non-economic or duplicated loss when members hold S shares with disparate bases. Reg. § 1.1502- 36(b)(1)(i).

b. The rule operates by reallocating previously applied investment adjustments. The rule does not alter the aggregate amount of basis in shares of S stock held by members or the aggregate amount of investment adjustments applied to shares of S stock. Id.

c. Exceptions to Basis Redetermination Rule

(1) No Potential for Redetermination – Basis redetermination will not be required if redetermination would not result in a change to any member’s basis in any share of S stock. Thus, the basis redetermination rule does not apply if the members’ basis in the shares of S common stock are equal and the members’ basis in the shares of S preferred stock reflect no gain or loss. Reg. § 1.1502-36(b)(1)(ii)(A).

(a) For example, if S has only one class of stock outstanding and there is no disparity in members’ bases in S shares, no member’s basis would be changed by the application of the rule. Thus, no redetermination would be required.

(b) Similarly, if S has preferred and common stock outstanding, there is no gain or loss on any member’s preferred shares, and there is no disparity in members’ bases in the common stock, no member’s basis would be changed by the application of the rule. Thus, no redetermination would be required.

(c) The effect of the reallocation of investment adjustments in such cases is only an increase, not a decrease, in basis disparity. 73 Fed. Reg. at 53,938.

(2) Disposition of Entire Interest – Basis redetermination will not be required if, within the group’s taxable year in which the transfer occurs, every share of S stock held by a member is transferred to a nonmember in one or more fully

pursuant to this section or any other provision of law. It also includes the proportionate amount of investment adjustments reflected in the basis of a share after the basis is apportioned among shares, for example in a transaction qualifying under section 355. Reg. § 1.1502-36(b)(1)(iii).

- 31 -

taxable transactions, becomes worthless, or a combination thereof. Reg. § 1.1502-36(b)(1)(ii)(B).

(a) A taxpayer that qualifies for this exception may nonetheless elect to have the basis redetermination rule apply. Reg. § 1.1502-36(b)(ii)(B). Taxpayers may choose to do this, for example, if it would result in a reduced gain or avoid the application of the Unified Loss Rules with respect to upper-tier shares. 73 Fed. Reg. at 53,939.

(b) Because of this exception, the basis redetermination rule should not be widely applicable, as consolidated groups most often will dispose of their entire interest in the loss subsidiary.

2. Operation of Basis Redetermination Rule

When a member transfers a share of subsidiary (S) stock and, after the application of all other provisions of the Code and regulations, the share is a loss share, the basis redetermination rule subjects all members’ shares of S stock to redetermination. Reg. § 1.1502-36(b)(2).

a. Under the basis redetermination rule, investment adjustments (exclusive of distributions) that were previously applied to members’ bases in S stock are generally reallocated in a manner that, to the greatest extent possible, first eliminates loss on preferred shares and then eliminates basis disparity on all shares. Reg. § 1.1502-36(b)(2)(iii). The rule removes both positive and negative adjustments, and so addresses both non-economic and duplicated losses. 72 Fed. Reg. at 2978.

b. The basis redetermination rule operates by first removing positive investment adjustments (but not in excess of the share’s loss) from the bases of transferred loss shares of common stock. Reg. § 1.1502-36(b)(2)(i)(A).

(1) This represents a change from the proposed regulations, which also removed positive investment adjustments from preferred stock. The rule was changed, because the positive investment adjustments allocated to preferred stock account solely for the right to receive distributions and do not reflect unrecognized loss. 73 Fed. Reg. at 53,938.

(2) Nonetheless, representatives from the Service have informally indicated that they are considering going back to

- 32 -

the rule in the proposed regulations, because there can be recognized built-in gain attributable to the preferred shares if there is no value in the rest of the company.

c. Then, to the extent of any remaining loss on the transferred shares, negative investment adjustments are removed from shares of common stock that are not transferred loss shares and applied to reduce the loss on transferred loss shares. Reg. § 1.1502- 36(b)(2)(i)(B). The negative adjustments are reallocated to reduce preferred stock first and then to common stock. Reg. § 1.1502- 36(b)(2)(i)(B).

d. Finally, the positive adjustments removed from the transferred common loss shares are allocated and applied to increase basis in the other S shares, without regard to whether such shares are transferred. The positive investment adjustments are applied first to the preferred stock (up to the value) and then to the common stock. Reg. § 1.1502-36(b)(2)(ii).

e. Thus, the basis redetermination rule effectively removes basis from transferred loss shares and uses it to reduce disparity in members’ bases in S shares.

3. Limits to Basis Redetermination

a. First, because the premise of the basis redetermination rule is that the original allocation of an item did not represent the most economically appropriate allocation of the item, redeterminations under the rule are limited to allocations of investment adjustments that could have been made at the time an item was taken into account. Accordingly, no adjustments can be reallocated to shares that were not held by members in the year taken into account, as members’ shares would not have been able to receive those adjustments in the original allocation. Reg. § 1.1502- 36(b)(2)(iii)(B)(2)(i).

b. The primary purpose of the rule is to reduce loss on transferred shares. However, because its secondary purpose is to decrease disconformity to the greatest extent possible, in certain fact patterns, the application of the rule will actually increase loss on some shares. See Example 28, below. The application of the rule will not, however, create gain on shares. Overall, the rule has no effect on the aggregate amount of gain or loss on members’ bases in subsidiary stock. 72 Fed. Reg. at 2979.

- 33 -

c. In the basis reallocation rule, and in several other provisions of the regulations, there is a direction to allocate items in a manner that reduces disparity to the greatest extent possible. The regulations do not, however, prescribe the manner in which such determinations are to be made. According to the Unified Loss Rules, taxpayers are allowed flexibility in choosing the methods and formulas to be employed in making these determinations and the Service will respect any reasonable method or formula so employed. Reg. § 1.1502-36(b)(2)(iii)(A).

d. Another limitation on reallocation is that an investment adjustment cannot be reallocated except to the extent that the full effect of the reallocation can be accomplished. Thus, an investment adjustment cannot be reallocated to the extent the resulting basis has previously been taken into account (including at a higher-tier). This rule guards against double benefits from an adjustment (for example, by not allowing positive adjustments to be moved from, or negative adjustments be moved to, shares after the item would have affected basis that was taken into account in recognizing gain or loss). It also guards against the loss of a benefit (for example, by not allocating positive adjustments to previously transferred shares that can no longer benefit from the basis). Reg. § 1.1502- 36(b)(2)(iii)(B)(2).

D. Stock Basis Reduction to Prevent Non-Economic Loss9

1. Background: The basis reduction rule reduces M’s basis in a transferred share of S stock in order to prevent non-economic stock loss and thereby promote the clear reflection of the group’s income. The effect of the basis reduction rule is to limit the reduction of M’s basis in the S share to the amount of net unrealized appreciation reflected in the share’s basis immediately before the transfer. The rule also limits the reduction of M’s basis in the S share to the portion of the share’s basis that is attributable to investment adjustments made pursuant to the consolidated return regulations. Reg. § 1.1502-36(c)(1). The rule is intended to eliminate stock loss that is presumed non-economic.

2. Basis Reduction Rule:

If, after basis redetermination, any member’s transferred share is a loss share (even if the share only became a loss share as a result of the application of the basis redetermination rule), the basis of that share is

9 See Section V.B., below, for examples illustrating the basis reduction rule.

- 34 -

subject to reduction under the basis reduction rule. Reg. § 1.1502- 36(c)(2). a. The rule operates by reducing the basis of each transferred loss share (but not below value) by the lesser of the share’s net positive adjustment and its disconformity amount. Reg. § 1.1502- 36(c)(2)(i)-(ii).

(1) A share’s net positive adjustment is computed as the greater of zero and the sum of all investment adjustments (excluding distributions) applied to the basis of the transferred loss share, including by reason of prior basis reallocations (hereinafter referred to as the “PIA amount”). Reg. § 1.1502-36(c)(3).

(a) This rule identifies the extent to which basis has been increased by the investment adjustment provisions for items of income, gain, deduction, and loss (whether taxable or not) that have been taken into account by the group. 72 Fed. Reg. at 2979.

(2) A share’s disconformity amount is the excess of its basis over its allocable portion of S’s net inside attributes, determined at the time of the transfer. Reg. § 1.1502- 36(c)(4). This amount identifies the net amount of unrealized appreciation reflected in the basis of the share.

(a) The term “net inside attributes” is defined as the sum of S’s loss carryovers, deferred deductions, cash, and asset basis, reduced by S’s liabilities. Reg. § 1.1502-36(c)(5).

(i) The term “loss carryovers” is defined as losses that are attributable to S, including any losses that would be apportioned to S under the principles of § 1.1502-21(b)(2) if S had a separate return year.

(ii) Note that this differs from the definition of loss carryovers for purposes of the attribution reduction rule, which do not include any amount of losses waived under § 1.1502-32(b)(4). Reg. § 1.1502-36(f)(6). Waived losses are included in the disconformity computation, because excluding them would have the effect of

- 35 -

increasing disconformity under circumstances unrelated to the existence of built-in gain (which is what the disconformity amount is trying to measure). In contrast, waived losses are excluded from the computation of net inside attributes to prevent attributes that cannot be duplicated from being taken into account in reducing attributes. See 73 Fed. Reg. at 53,940.

(b) Because the disconformity amount is computed at the time of the transfer, the disconformity amount reflects the effects of all prior redetermination events. Reg. § 1.1502-36(c)(5). b. Lower-Tier Subsidiaries – For purposes of computing the disconformity amount, if S holds stock of a lower-tier subsidiary (S1) that was not transferred in the transaction, S’s net inside attribute amount is computed by treating S’s basis in S1 stock as ‘‘tentatively reduced’’ by the lesser of the S1 share’s PIA amount and its disconformity amount. Reg. § 1.1502-36(c)(6)(i)-(ii).

(1) This reduction is made only for purposes of determining basis reduction to the S share, and has no other effect. 72 Fed. Reg. at 2979.

(2) The purpose of this adjustment is to prevent S1’s recognized items from giving rise to non-economic loss in S stock, for example, when S1 recognizes gain that is already reflected (indirectly) in P’s basis in S shares. Reg. § 1.1502-36(c)(6)(i).

(3) If there are multiple tiers of subsidiaries, then the tentative reduction occurs at the lowest tier first and then to each successively higher tier. Reg. § 1.1502-36(c)(6)(iii).

(4) When determining the disconformity amount of a share of subsidiary stock, no tentative reduction is made to the basis of lower-tier shares that were transferred in the transaction (without regard to whether S retained the shares after the transaction, such as when S1 is transferred because S and S1 cease to be members of the same group but S continues to hold S1 stock). Reg. § 1.1502-36(c)(6)(iv).

(a) The basis reduction rule applies directly to each transfer, starting with the lowest-tier transfer, so any

- 36 -

non-economic loss in S stock that was attributable to S1’s items will have been eliminated by the time that the basis reduction rule applies to the S stock.

(b) In addition, the tentative basis reduction rule does not apply to shares that are lower-tier to any shares that were transferred in the transaction. Reg. § 1.1502-36(c)(6)(iv). The application of the rule to those shares is unnecessary because, when the basis reduction rule applied to S1, it eliminated any inappropriate effects from items that tiered up from subsidiaries that were lower tier to S1. Id.; see also 72 Fed. Reg. at 2979-80.

3. Netting of Gains and Losses

a. For purposes of computing the basis reduction required under Reg. § 1.1502-36(c), the basis of each transferred loss share of S stock is treated as reduced proportionately (as to loss) by the amount of gain taken into account by members with respect to all transferred gain shares of S stock, provided that:

(1) The gain and loss shares are transferred in the same transaction; and

(2) The gain is taken into account in the year of the transaction. Reg. § 1.1502-36(c)(7)(i).

b. Example 9 – Netting Rule Application (Allocation of Gain Amount to Determine Net Loss)

$60/share X P

Sh. A Sh. B Sh. C $60 Value $60 Value $60 Value $54 Basis $100 Basis $80 Basis

S

(1) Facts: P owns three outstanding shares of S common stock. Share A has a basis of $54, Share B has a basis of $100, and Share C has a basis if $80. P sells all three shares of S

- 37 -

stock to X for $60 each. P’s sales of Share B and Share C are transfers of loss shares subject to Reg. § 1.1502-36(c).

(2) For this purpose, P can net its $6 gain on Share A against its bases in Share B and Share C. The gain is allocated to Share B and Share C proportionately based on the amount of loss in each share. Thus, $4 of gain ($40/$60 × $6) is treated as allocated to Share B and $2 of gain ($20/$60 × $6) is treated as allocated to Share C. Accordingly, P computes the basis reduction required under Reg. § 1.1502- 36(c) by treating its basis in Share B as $96 ($100 less $4) and its basis in Share C as $78 ($80 less $2). If, after the application of Reg. § 1.1502-36(c), the sales of Share B and Share C are still transfers of loss shares, then the transfers are subject to Reg. § 1.1502-36(d). Although the bases of Share B and Share C are not reduced by gain for purposes of Reg. § 1.1502-36(d), Reg. § 1.1502-36(d)(3)(i)(A) applies netting principles to limit adjustments under Reg. § 1.1502-36(d). See example at Reg. § 1.1502-36(c)(7)(ii).

(3) If P sold the gain share to another member of the consolidated group, the gain would be deferred under Reg. § 1.1502-13. Because it is not taken into account in the year of the transfer, it cannot be used to reduce P’s loss on Share B. Id.

E. Attribute Reduction to Prevent Duplication of Loss10

1. Background: Reg. § 1.1502-36(d) reduces S’s attributes to the extent they duplicate a net loss on shares of S stock transferred by members in a single transaction. Reg. § 1.1502-36(d).

a. The rule is intended to insure that the group does not recognize more than one loss with respect to a single economic loss regardless of whether the group chooses to dispose of the subsidiary stock before or after the subsidiary recognizes the loss with respect to its assets or operations. 73 Fed. Reg. at 53,940.

b. The rule furthers single entity principles by preventing S from using deductions and losses to the extent that the group or its members (including former members) have either used, or preserved for later use, a corresponding loss in S shares. Reg. § 1.1502-36(d)(1).

10 See Section V.C., below, for examples illustrating the attribute reduction rule.

- 38 -

c. The rule applies without regard to whether S ceases to be a member after the transfer of its shares. Reg. § 1.1502-36(d)(1).

(1) This represents a change from the prior loss disallowance rules, which contained different rules depending upon whether S was or was not deconsolidated as a result of the transfer. See Prior Reg. § 1.1502-35(b)(1) & (b)(2).

2. Attribute Reduction Rule: If any transferred share remains a loss share after application of Reg. § 1.1502-36(c), the subsidiary’s attributes (including the consolidated attributes attributable to the subsidiary) are reduced by the attribute reduction amount immediately before the transfer. Reg. § 1.1502-36(d)(2)(i).

a. This rule differs from the prior loss duplication rules under Reg. § 1.1502-35, which suspended outside losses in favor of inside losses. Specifically, Reg. § 1.1502-35 suspended the outside loss if the subsidiary remained consolidated, reduced the suspended loss for any inside losses used by the subsidiary, and only allowed any remaining outside loss when the subsidiary left the group.

b. The Unified Loss Rules, in contrast, generally allow the outside loss but immediately eliminate the duplicate inside loss.

3. Attribute Reduction Amount: Under Reg. § 1.1502-36(d), the attribute reduction amount is computed as the lesser of the net stock loss and the aggregate inside loss. Reg. § 1.1502-36(d)(3)(i).

a. Net stock loss is the excess of the sum of the bases (after application of the basis reduction rule) of all S shares transferred by members in the same transaction over the value of such shares. Reg. § 1.1502-36(d)(3)(ii).

b. S’s aggregate inside loss is the excess of S’s net inside attributes over the value of all of the S shares. Reg. § 1.1502- 36(d)(3)(iii)(A).

(1) S’s net inside attribute amount is the sum of S’s net operating and capital loss carryovers, deferred deductions, money, and basis in assets other than money reduced by the amount of S’s liabilities. Reg. § 1.1502-36(d)(3)(iii)(B).

(2) Unlike comparable rules in the old regulations, the attribute reduction amount is not limited to the share’s proportionate interest in the subsidiary’s inside loss.

- 39 -

d. Exception: If the aggregate attribution reduction amount is less than five percent of the aggregate value of the shares transferred by members in the transaction, the attribution reduction rule does not apply to the transfer. Reg. § 1.1502-36(d)(2)(ii).

(1) If the exception applies, the taxpayer may nonetheless elect to apply the attribute reduction rule. Reg. § 1.1502- 36(d)(2)(ii). A taxpayer may want to do so, for example, to take advantage of the election to reattribute subsidiary losses, as discussed more fully below. See 73 Fed. Reg. at 53,941.

4. Application of Attribute Reduction

a. Attributes Available for Reduction:

(1) Category A – Capital loss carryovers

(2) Category B – Net operating loss carryovers

(3) Category C – Deferred deductions

(4) Category D – Basis of assets, other than Class I assets in Reg. § 1.338-6(b)(1) (i.e., cash and general deposit accounts). Reg. § 1.1502-36(d)(4)(i).

b. Category A, B, and C Attributes

(1) After S’s attribute reduction amount is determined, it is first applied to reduce or eliminate items that represent actual realized losses, such as capital loss carryovers, net operating loss carryovers, and deferred deductions. Reg. § 1.1502-36(d)(4)(ii)(A).

(2) The taxpayer may specify the allocation of S’s attribute reduction amount among the attributes in Category A, Category B and Category C. If no specification is made, the attributes will first be used to reduce Category A capital loss carryovers (oldest to newest), then Category B NOL carryovers (oldest to newest), and then Category C deferred deductions (proportionately).

(a) The final regulations reversed the order of Category A and B attributes, because capital loss carryovers have a shorter expiration period and are thus more likely than net operating losses to expire unused

- 40 -

and, therefore, less likely to duplicate loss. 73 Fed. Reg. at 53,941.

(b) Attributes in Category A, Category B, and Category C must be reduced in full before any reduction is made to Category D (asset basis). Reg. § 1.1502- 36(d)(4)(ii)(A)(1). c. Category D Attributes

(1) If S’s attribute reduction amount exceeds the Category A, B, and C items, the excess is then applied to reduce or eliminate Category D attributes (i.e., asset basis, including the stock of lower tier subsidiaries). Reg. § 1.1502- 36(d)(4)(ii)(B)(1).

(2) S’s attribution reduction amount is allocated to the non- stock Category D assets in reverse section 1060 order. Thus, it is first applied proportionately to reduce S’s bases in Class VII assets, as defined by Reg. § 1.338-6(b)(2)(vii), then in the same manner successively to Class VI, Class V, Class IV, Class III, and Class II assets, as defined by Reg. § 1.338-6(b)(2). Reg. § 1.1502-36(d)(4)(ii)(B)(2).

(3) Lower-Tier Subsidiaries – If S holds stock of another subsidiary (S1), the basis of such stock is treated as a Category D attribute. The reduction amount must be allocated proportionately (by basis) between the stock of each of S’s lower-tier subsidiaries (treating all shares of each subsidiary as a single share) and the non-stock Category D assets. Reg. § 1.1502-36(d)(4)(ii)(B)(1), (d)(5)(i)(A), (d)(5)(ii).

(a) For this purpose, S’s basis in the deemed single share of S1 stock is its deemed basis, which is the greater of the sum of S’s basis in each share of S1’s stock and the portion of S1’s net inside attribute amount allocable to S’s shares of the S1 stock. Reg. § 1.1502-36(d)(5)(i)(B), (d)(5)(ii).

(i) The deemed basis is reduced by (i) the value of S’s transferred shares of S1 stock, and (ii) the nontransferred shares’ allocable portion of the excess of S1’s non-loss assets over S1’s liabilities. Reg. § 1.1502- 36(d)(5)(ii).

- 41 -

(ii) For example, P owns all the stock of S with a basis of $150, S owns all the stock of S1 with a basis of $100, and S1 owns an asset with a basis of $150. S’s deemed basis in S1 stock is $150, the greater of $100 (S’s actual basis in S1 stock) and $150 (the S1 shares’ allocable portion of S1’s net inside attribute amount), which is the maximum amount of inside loss that S can recognize. 72 Fed. Reg. at 2980.

(b) If S1 has lower-tier subsidiaries, S’s deemed basis is determined first with respect to the stock of the lowest tier subsidiary and then for each next highest tier. Reg. § 1.1502-36(d)(5)(i)(C).

(c) The attribute reduction amount allocated to S’s block of S1 stock is then apportioned and applied to reduce the bases of S’s individual shares of S1 stock in a manner that, to the greatest extent possible, reduces loss in preferred stock and disparity in both common and preferred stock. Reg. § 1.1502- 36(d)(5)(iii)(B).

(i) However, no allocated amount is apportioned to any transferred S1 share if gain or loss is recognized on the transfer of that share. Reg. § 1.1502-36(d)(5)(iii)(A). Recognition of gain or loss establishes that the basis of that share no longer reflects unrecognized loss.

(ii) In addition, no allocated amount that is apportioned to any transferred S1 share is to be applied to reduce the basis of the share below its value. Reg. § 1.1502- 36(d)(5)(iii)(C). This prevents attribute reduction from creating gain on such shares.

(d) Any portion of S’s attribute reduction amount that is allocated to S1 stock, tiers down and becomes an attribute reduction amount of S1. The attribute reduction rules then apply to reduce S1’s attributes in the same manner that they apply S’s attribute reduction amount to reduce S’s attributes. Reg. § 1.1502-36(d)(5)(v)(A).

- 42 -

(e) Because the attribute reduction amount represents the maximum potential amount of duplication in the lower-tier subsidiary, the Unified Loss Rules include two modifications to prevent the reduction of attributes beyond the amount necessary to eliminate duplicated loss.

(i) Conforming Limit Rule: This rule prevents the tier down of attribute reduction from reducing S1’s net inside attributes below the sum of the value of the S1 shares transferred by members and the aggregate bases that members have in nontransferred S1 stock (after any reduction to those shares by the direct application of S’s attribute reduction amount). Reg. § 1.1502-36(d)(5)(v)(B).

(ii) Basis Restoration Rule: This rule reverses stock basis reductions made by the attribute reduction rule, but only to the extent necessary to conform net inside attributes and outside stock basis at each tier. Basis restoration adjustments are made at each tier, but they do not give rise to any upper- tier adjustments. Reg. § 1.1502- 36(d)(5)(vi).

(f) Commentators complained about the complexity of the tier-down rule and suggested adopting a look- through approach, which would apply the rules solely on the basis of the lower tier subsidiary’s net inside attributes. Treasury and the Service rejected this approach. Although they recognized that it would simplify the rules, they were concerned that it would produce inappropriate results by either disallowing economic loss where lower tier subsidiary stock reflected unrecognized appreciation or permitting son-of-mirrors transactions at lower tiers. d. If the attribute reduction amount exceeds attributes available for reduction, then, if the subsidiary has a liability that has not been taken into account, the excess attribute reduction amount is suspended and applied to prevent the deduction or capitalization of payments later made by S or another person with respect to the liability. Reg. § 1.1502-36(d)(4)(ii)(C)(1).

- 43 -

e. If the attribute reduction amount exceeds the amount of S’s liabilities, that excess attribute reduction amount has no further effect. Reg. § 1.1502-36(d)(4)(ii)(C)(2).

5. Elections to Reduce the Potential for Loss Duplication: The common parent of a group can elect to reduce stock basis, reattribute Category A, B, or C attributes, or do some combination of basis reduction and attribute reattribution in order to prevent the reduction of attributes otherwise required under the proposed rules. Reg. § 1.1502-36(d)(6)(i).

a. The total amount that can be the subject of the election is limited to the amount that S’s attributes would otherwise be subject to reduction. Reg. § 1.1502-36(d)(6)(i).

b. Taxpayers may make a protective election to reattribute attributes and/or to reduce stock basis to avoid attribute reduction. 73 Fed. Reg. at 53,942.

c. Special Rules for Elections to Reattribute Attributes:

(1) The election to reattribute attributes can only be made if S ceases to be a member of the P group as a result of the transfer. This is because the election is not intended to be merely a mechanism for changing location of items within a group. Reg. § 1.1502-36(d)(6)(iv)(A).

(2) Similar to the rule regarding attribute reduction in Reg. § 1.1502-36(d)(4)(ii)(A)(1), P may specify the amount of attributes in Category A, B, and C to reallocate. In the absence of any specification, they will be reattributed in order. Reg. § 1.1502-36(d)(6)(iv)(A).

(3) When the election to reattribute attributes is made, P is treated as succeeding to the attributes as though it had acquired them in a section 381(a) transaction, and it is treated as a noncapital, nondeductible expense under Reg. § 1.1502-32(b)(2)(iii). Reg. § 1.1502-36(d)(6)(iv)(A).

(4) Limitation on Reattribution from Lower-Tier Subsidiaries: P may only reattribute attributes of lower-tier subsidiaries that would otherwise be reduced as a result of tier-down attribute reduction to the extent that the reattribution does not create an excess loss account in the stock of any lower- tier subsidiary. This prevents circular computations of the attribute reduction amount. Reg. § 1.1502-36(d)(6)(iv)(C).

- 44 -

(5) To the extent loss duplication has not been eliminated by the election, the attribute reduction rules apply in their general manner. 72 Fed. Reg. at 2981.

d. Special Rules for Election to Reduce Stock Basis:

(1) The election is made with respect to all transferred S shares in proportion to the amount of loss in each share. Reg. § 1.1502-36(d)(6)(v)(A).

(2) If there is still a stock loss after taking into account elections under Reg. § 1.1502-36(d)(6), and such loss would be permanently disallowed (e.g., under section 311(a)), P will be deemed to have made the stock basis reduction election with respect to such loss. Reg. § 1.1502- 36(d)(6)(v)(C). This is intended to protect against inadvertent attribute reduction. 73 Fed. Reg. at 53,942.

6. Attribute Reduction in the Case of Worthlessness and Dissolution: All of S’s attributes are eliminated if:

a. A member’s S stock becomes worthless within the meaning of Reg. § 1.1502-80(c),11 the member recognizes a net deduction or loss, and S remains a member of the group; or

b. A member recognizes a deduction or loss on the S stock in a transaction in which S ceases to exist and does not become a nonmember within the meaning of Reg. § 1.1502-19(c)(2).

11 Reg. § 1.1502-80(c) provides that stock of a member is not treated as worthless under section 165 before the stock is treated as disposed of under Reg. § 1.1502-19(c)(1)(iii). Taxpayers raised concerns that Reg. § 1.1502-80(c) could prevent a group from claiming a worthless stock deduction if the subsidiary stock is worthless but the subsidiary ceases to be a member of the group before it has disposed of substantially all of its assets as required by Reg. § 1.1502-19(c)(1)(iii). The Service and Treasury amended Reg. § 1.1502-80(c) to clarify that the deferral of an otherwise allowable loss under section 165 terminates immediately prior to the time that the subsidiary ceases to be a member of the group. Reg. § 1.1502-80(c). The preamble to the temporary regulations noted that Reg. § 1.1502-80(c) is intended to defer, not disallow, worthless stock deductions with respect to subsidiary stock. 69 Fed. Reg. 12,799, 12,800 (Mar. 18, 2004).

- 45 -

c. Example 10 – Worthlessness Where S Continues as a Member

P

$0 Value $75 Basis

Assets S Creditors

CNOL = $100

(1) Facts: P owns the sole share of S stock. The share is worthless under section 165. S has disposed of all its assets in payment of claims to creditors. P claims a worthless securities deduction with respect to the share. The worthlessness is a transfer of a loss share. After the application of the basis redetermination rule and the basis reduction rule, P’s basis in the share is $75. The portion of the consolidated net operating loss attributable to S is $100.

(2) Under the attribute reduction rule, S’s attribute reduction amount is $75, the lesser of P’s $75 net stock loss and S’s $100 aggregate inside loss ($100 net inside attribution amount over $0 value of the S share). S’s attributes are reduced by $75, from $100 to $25. Because S remains a member of the group, the remaining $25 of the consolidated net operating loss is eliminated because the S shares is worthless and P recognizes a deduction with respect to the share. P recognizes a $75 worthless stock deduction. S has $0 net inside attributes. The consolidated net operating loss is reduced by $100. See Reg. § 1.1502- 36(d)(7)(iii).

F. Coordination with Loss Deferral and Other Loss Disallowance Rules

1. Reg. § 1.1502-36 applies and has effect immediately upon the transfer of a loss share even if the loss is deferred, disallowed, or otherwise not taken into account under any other applicable rules of law.

a. For example, if M sells loss shares of S stock to another member of the same controlled group but not consolidated group, every member’s bases in shares of S stock and all of S’s attributes may

- 46 -

be adjusted under this section even though M’s loss is deferred under Reg. § 1.267(f)-1. Reg. § 1.1502-36(a)(4).

2. However, Reg. § 1.1502-36(e)(3) defers the application of the Unified Loss Rules until the triggering of an intercompany loss under Reg. § 1.1502-13.12

a. In determining the application of the Unified Loss Rules to an intercompany loss, all transferor-members are treated as divisions of a single corporation. Reg. § 1.1502-36(e)(3)(i).

(1) Reg. § 1.1502-36(e)(3) appears to incorporate the single entity approach that applies for purposes of Reg. § 1.1502- 13, which provides that the transferor and transferee members are treated as engaging in their actual transaction, rather than treating the transaction as not occurring. Reg. § 1.1502-13(c)(3). Thus, changes with respect to the subsidiary shares resulting from the intercompany transaction should not be taken into account, but changes after the intercompany transfer but before the loss is triggered should be taken into account, in determining whether and how the Unified Loss Rules apply to the intercompany item.

(2) For example, in an intercompany sale, the buying member would be treated as receiving the sold property but would succeed to the selling member’s basis in the property. Id.

b. The regulations further provide that “appropriate adjustments” will be made to the intercompany item(s), any member’s basis in a share of a subsidiary, to the subsidiary’s attributes, or any combination thereof, to further the purposes of Reg. §§ 1.1502-36 and 1.1502-13. Id.

c. If the transfer occurred before September 17, 2008, and the transferor-member’s intercompany item is taken into account after September 17, 2008, the parent may elect to apply the current Unified Loss Rules to the transfer. Reg. § 1.1502-36(e)(3)(ii).

G. Anti-Abuse Rule - Reg. § 1.1502-36(g)

1. If a taxpayer acts with a view to avoid the purposes of Reg. § 1.1502-36, or to apply the rules of Reg. § 1.1502-36 to avoid the purposes of any other rule of law, appropriate adjustments will be made to carry out the

12 See Section V.D., below, for examples of intercompany transfers of subsidiary stock.

- 47 -

purposes of Reg. § 1.1502-36 or such other rule of law. Reg. § 1.1502- 36(g).

2. Transactions Triggering Anti-Abuse Rule

a. Loss Trafficking – S acquires and liquidates a corporation with an NOL to utilize its NOL to minimize the basis disconformity (and therefore the basis reduction). Reg. § 1.1502-36(g), Ex. 1.

b. Preventing Attribute Reduction

(1) S contributes an asset with a high basis to a partnership in order to avoid reducing its basis under Reg. § 1.1502-36(d). Reg. § 1.1502-36(g), Ex. 2.

(2) S loans cash to another member to create an intercompany receivable that would be reduced under the attribute reduction rule (thereby minimizing reduction to S’s other attributes). Reg. § 1.1502-36(g), Ex. 3.

(3) P contributes stock with a low basis to a partnership and sells stock with a high basis at a loss to reduce net stock loss under Reg. § 1.1502-36(d). Reg. § 1.1502-36(g), Ex. 4.

b. Gain Stuffing – P contributes a gain asset to S to avoid recognition of net loss (and therefore application of Reg. § 1.1502-36) upon the sale of the stock. Reg. § 1.1502-36(g), Ex. 5.

III. OLD LOSS DISALLOWANCE RULES – ADDRESSING CONCERNS RELATING TO THE REPEAL OF GENERAL UTILITIES

A. General Rule

1. The old regulations governing the disallowance of losses on the disposition of stock of a member of a consolidated group were issued as temporary regulations on March 7, 2002 and were adopted, without substantive change, as final regulations on March 3, 2005. See Prior Reg. § 1.337(d)-2.

2. In general, the old regulations provided that no deduction was allowed for any loss recognized by a member of a consolidated group with respect to the “disposition” of stock of a subsidiary. Prior Reg. § 1.337(d)-2(a)(1). This was the same as the general rule of the pre-2002 loss disallowance regulations. See Prior Reg. § 1.1502-20(a)(1).

- 48 -

3. “Disposition” was defined in the same manner as the pre-2002 regulations as any event in which gain or loss is recognized, in whole or in part. Prior Reg. § 1.337(d)-2(a)(2)(ii); cf. Prior Reg. § 1.1502-20(a)(2). A worthless stock deduction under section 165(g)(3) thus constituted a “disposition” that triggered the general loss disallowance rule. See Prior Reg. § 1.1502- 20(a)(1).

B. Deconsolidations

1. Under the old regulations, if a member’s basis in its subsidiary stock exceeded the stock’s fair market value immediately before the stock was deconsolidated, the member’s basis in the subsidiary stock was reduced to the stock’s fair market value. Prior Reg. § 1.337(d)-2(b).

2. “Deconsolidation” was defined as “any event that causes a share of stock of a subsidiary that remains outstanding to be no longer owned by a member of any consolidated group of which the subsidiary is also a member.” Prior Reg. § 1.337(d)-2(b)(2).

3. If both a disposition and a deconsolidation occurred with respect to a share in the same transaction, then the general disallowance rule of Prior Reg. § 1.337(d)-2(a)(1) applied first, and the deconsolidation rule of Prior Reg. § 1.337(d)-2(b)(1) applied next to the extent necessary to effectuate the purposes of the regulation. Prior Reg. § 1.337(d)-2(b)(1).

4. The pre-2002 regulations provided an identical rule for the reduction of basis upon deconsolidation. Prior Reg. § 1.1502-20(b)(1). However, the pre-2002 regulations contained an additional requirement if a share of stock retained by a group were deconsolidated and then disposed of within two years after the date of deconsolidation, which is not contained in Prior Reg. § 1.337(d)-2.

a. In such cases, the group was required to file a statement with its return for the year of the disposition setting forth the amount of any prior basis reduction, the basis of the subsidiary’s stock immediately before the disposition, the amount realized on the disposition, and the amount of loss recognized on the disposition. Prior Reg. § 1.1502-20(b)(5).

b. Failure to file the statement resulted in the disallowance of a deduction for the loss.

c. The preamble to the final 1991 regulations clarified that the statement must be filed regardless of whether there was a reduction in the basis of stock under the deconsolidation rule. 56 Fed. Reg. at 47,384.

- 49 -

5. The current Unified Loss Rules in Reg. § 1.1502-36 apply to transferred loss shares, which are defined to include deconsolidations. Reg. § 1.1502- 36(f)(10)(i)(B). They do not contain a separate deconsolidation rule.

6. Example 11 – Definition of Deconsolidation

P P

$200 basis $100 basis (3) S1 Stock S S Z $100 (1) $100 basis $200 basis T Stock X S1 S1 $100 $200 basis (2) Asset T T Y $100 Asset $100 Value $0 Basis

a. Facts: P owns all of the stock of S, and S owns all of the stock of S1. P has a basis of $100 in the S stock, and S has a basis of $100 in the S1 stock. In Year 1, S1 purchases all of the stock of T from X for $100. T has an asset with a value of $100 and a basis of $0. In Year 2, T sells the asset to Y for $100. In Year 3, S sells all of its S1 stock to Z, an unrelated investor, for $100.

b. T recognizes a gain of $100 on the sale of its built-in gain asset to Y in Year 2. As a result, S1’s basis in the T stock, S’s basis in the S1 stock, and P’s basis in the S stock are increased from $100 to $200. Reg. § 1.1502-32(b)(2)(i). Thus, when S sells its S1 stock to Z in Year 3, it recognizes a $100 loss.

c. Under the general loss disallowance rule of the old regulations, S’s $100 loss on the sale of S1 stock is disallowed. Prior Reg. § 1.337(d)-2(a)(1); see also Prior Reg. § 1.1502-20(a)(1).

- 50 - d. If S1 and T are not members of a consolidated group immediately after the sale of the stock of S1, then the T stock will be considered deconsolidated. S1 must, therefore, reduce its basis of the T stock to its $100 value immediately before the sale. If, however, S1 and T become members of the Z consolidated group immediately after the sale of S1, then the T stock will not be considered deconsolidated and no reduction is required. Prior Reg. § 1.337(d)-2(b)(1)-(2); see also Prior Reg. § 1.1502-20(b)(6), Ex. 3. Z may expect post-acquisition appreciation to be sheltered by the loss with respect to the T stock. Alternatively, Z may be willing to “stuff and wait” two years and avoid the application of the anti-stuffing rule of the old regulations.

(1) In anticipation of a possible sale, consolidated groups may want to preserve a lower tier subsidiary stock’s built-in loss by arranging for a member holding company to own the stock.

(2) However, if a holding company is formed in a section 351 transaction, and the holding company’s stock is sold shortly thereafter, the step-transaction doctrine may be applied to disqualify the stock contribution as tax free under section 351. See, e.g., Intermountain Lumber Co. v. Commissioner, 65 T.C. 1025 (1976); Rev. Rul. 70-140, 1970-1 C.B. 73.

(3) Additionally, the contribution of a lower tier subsidiary’s built-in loss stock to a holding company with a view to selling the holding company stock and avoiding the loss disallowance rule may trigger the anti-stuffing rule when the holding company’s stock is sold.

- 51 -

7. Example 12 – Using the Deconsolidation Rule to Avoid Gain Recognition

(2) $400 P X S stock $21 (1) $400 (3) 21% of T1 basis $79% of T1 $21 S S $300 Value $100 Value $300 Value $79 Value $100 Basis $200 Basis $100 Basis $79 Basis

T1 T2 T1 T2

a. Facts: P owns all of the stock of S, which has two assets – the stock of subsidiaries T1 and T2. The T1 stock has a value of $100 and a basis of $200, and the T2 stock has a value of $300 and a basis of $100. X is willing to acquire S for $400, but does not want the T2 stock.

b. If X purchases S and sells T2 immediately, the X group would recognize a $200 gain. If X instead immediately causes S to distribute T1 to X, S would recognize a deferred intercompany loss equal to $100. Reg. § 1.1502-13(f)(2)(iii). X’s basis in its S stock would ultimately be reduced by $200 (i.e., $100 value of T1 upon distribution, plus $100 loss when absorbed). Thus, a sale by X of S would result in $100 of gain.

c. Alternatively, S sells 21% of the T1 stock to X, producing a $21 loss which is disallowed under the loss disallowance rules of Prior Reg. § 1.337(d)-2. Because the sale results in a deconsolidation of Tl from S, S’s basis in its remaining 79% of Tl’s stock is reduced to the stock’s fair market value – $79. Prior Reg. § 1.337(d)-2(b); see also Prior Reg. § 1.1502-20(b).

(1) X then purchases S for $400. X causes S to distribute its remaining 79% of Tl’s stock and the $21 of cash to X. As a result of the distribution, X’s basis in S is reduced only by $100, the adjusted basis of the Tl stock and the cash. See Reg. § 1.1502-13(f)(2).

(2) X’s basis in the S stock is now $300. X sells S for $300 (the value of T2) and recognizes no gain or loss.

- 52 -

(3) The $21 disallowed loss on S’s sale of 21% of the Tl stock and the $79 basis reduction to the remaining 79% of the T1 stock resulting from the deconsolidation are treated as noncapital, nondeductible expenses for purposes of computing investment adjustments. Reg. § 1.1502- 32(b)(3)(iii). The deemed $100 loss produces a $100 reduction in P’s basis in S. Therefore, P’s gain on the sale of S is increased by $100 (the amount of gain eliminated by X).

B. Allowable Loss

1. General Rules - Under Prior Reg. § 1.337(d)-2(c), a loss would not be disallowed under the loss disallowance rule of Prior Reg. § 1.337(d)-2(a), and basis would not be reduced under the deconsolidation rule of Prior Reg. § 1.337(d)-2(b), “to the extent the taxpayer establishes that the loss or basis is not attributable to the recognition of built-in gain, net of directly related expenses, on the disposition of an asset” (emphasis added).

a. The regulations provided that gain recognized on the disposition of an asset was built-in gain “to the extent attributable, directly or indirectly, in whole or in part, to any excess of value over basis that is reflected, before the disposition of the asset, in the basis of the share.” Prior Reg. § 1.337(d)-2(c)(2).

(1) Practitioners, and it appears the Service, had assumed that Prior Reg. § 1.337(d)-2 required a pure tracing approach in determining whether gain is built-in.

(2) However, at an American Bar Association Tax Section meeting in May 2004, the Service announced a “paradigm shift” in its interpretation of Prior Reg. § 1.337(d)-2 based on its interpretation of the time for testing whether gain is “built-in.”

(a) Notice 87-14 defined built-in gain as the excess of value over basis at the time the subsidiary stock was acquired, regardless of whether the acquisition of stock resulted in the subsidiary’s becoming a member of the group.

(b) The transitional rule of Prior Reg. § 1.337(d)-1 defined built-in gain as the excess of value over basis determined immediately before the transitional subsidiary became a member of the group.

- 53 -

(c) Prior Reg. § 1.337(d)-2 referred to the excess of value over basis before the disposition of the asset. Practitioners assumed that this rule simply incorporated the tracing approach of Notice 87-14 and Prior Reg. § 1.337(d)-1. But the Service believed that the language difference relating to the timing of the built-in gain determination required a different approach.

(d) The Service developed two approaches for tracking built-in gain—the modified tracing approach and the basis disconformity approach, which were later enumerated in Notice 2004-58 issued on August 25, 2004.

b. Loss or basis could be attributable to the recognition of built-in gain on the disposition of an asset by a prior group. Id.

c. The taxpayer had to file a statement with its return in order to claim an allowed loss. Prior Reg. § 1.337(d)-2(c)(1). The statement had to set forth the name and EIN of the subsidiary and the amount of the loss allowed or the amount of basis not reduced.

d. The language “net of directly related expenses” was added as a clarification on March 18, 2004. See T.D. 9118, 69 Fed. Reg. 12,799 (Mar. 18, 2004). Taxpayers had questioned whether built- in gain recognized on the disposition of an asset could be reduced by expenses attributable to the recognition of that gain. The preamble to the amended regulations stated that “[t]he IRS and Treasury Department believe that, because expenses attributable to the recognition of built-in gain reduce the basis of the subsidiary’s stock, the computation of the amount of stock loss that is attributable to the recognition of built-in gain should take such expenses into account.” Id.

2. Determination of Built-In Gain

a. In order to be considered built-in gain, the gain must be attributable to an excess of value over basis that was reflected in the basis of the share before the disposition of the asset.

b. Pure Tracing Approach - Under a pure tracing approach, gain was considered built-in to the extent of the appreciation in an asset owned by the subsidiary at the time the basis of the subsidiary stock was determined.

- 54 -

c. Modified Tracing Approach

(1) Under the modified tracing approach, as announced by the Service at the May 2004 ABA meeting, gain was considered built-in (i) if the asset (or a predecessor asset with the same basis characteristics) was owned by the subsidiary when the basis of the share on which loss was being recognized (or basis was being reduced) was determined or redetermined, (ii) to the extent of the share’s proportionate interest in appreciation in the asset at that time.

(a) Redetermination was required any time the subsidiary engaged in a transaction that could alter the share’s interest in unrealized appreciation, such as acquisitions of assets with built-in gain, mergers, or intragroup spin-offs.

(2) In Notice 2004-58, the Service was much less specific about the modified tracing approach, simply stating that “[u]nder a tracing approach, events subsequent to the acquisition of a share of subsidiary stock that create or alter the disconformity between the basis of the share and the share’s interest in the aggregate basis of assets the disposition of which would adjust the basis of the share (for example, the acquisition by a subsidiary of stock of another corporation that joins the consolidated group, an intra- group spin-off under section 355, or a contribution of property to a subsidiary under section 351) may need to be taken into account to determine the extent to which stock loss or basis is attributable to the recognition of built-in gain on the disposition of an asset.” (Hereinafter such subsequent events are referred to as “alteration events.”)

(3) The Unified Loss Rules reject a tracing approach, even the modified one, as inadministrable. The Service and Treasury believed that the potentially large number of alteration events would greatly increase complexity. 72 Fed. Reg. at 2972. d. Basis Disconformity Approach

(1) The basis disconformity approach permitted by Notice 2004-58 looked at outside basis at the time the subsidiary’s

- 55 -

asset was disposed of13 to determine whether the inside gain was already accounted for, or sheltered, by outside basis.

(2) The focus of this approach was on items that increase stock basis without increasing the value of the subsidiary.

(3) Specifically, gain was treated as built-in under the basis disconformity approach to the extent of the least of:

(a) Gain Amount - The sum of all gains (net of directly related expenses) recognized on asset dispositions while the subsidiary was a member of the group;

(b) Basis Disconformity Amount - The excess of the share’s basis over the share’s proportionate interest in the subsidiary’s “net asset basis;”14 or

(c) Positive Investment Amount (PIA Amount) - The excess of the sum of all positive investment adjustments made to the share under Reg. § 1.1502- 32 over negative adjustments (excluding distributions). Thus, a netting concept was incorporated.

(4) The concept of basis disconformity was incorporated into the final Unified Loss Rules. The final regulations delete the gain amount from the formula, thus expanding the rule beyond disconformity attributable to asset dispositions to reach all events triggering PIAs (e.g., wasting assets).

3. Disposition of an Asset

a. Although the burden of proof was on the taxpayer, the regulations allowed losses that were not attributable to the recognition of built- in gain on the disposition of an asset.

13 The Service took the position that because recognition of gain moves asset and stock basis in tandem, the time for testing basis disconformity may be before either the disposition of the asset or the disposition of the subsidiary’s stock. As discussed below, certain transactions that alter the inside and outside basis conformity required special treatment.

14 Net asset basis was defined similar to the way it was defined for purposes of determining duplicated loss under prior Reg. § 1.1502-20(c)—the sum of the subsidiary’s money, asset basis other than stock of lower tier subsidiaries, net operating loss carryforwards and deductions but that have been recognized but deferred, over the subsidiary’s liabilities.

- 56 -

(1) The loss disallowance rule of Prior Reg. § 1.337(d)-2 was thus aimed at the classic son-of-mirrors transaction illustrated by Examples 2 and 3, above.

(2) It did not reach the generation of income from wasting assets illustrated by Example 4, above. Assume, for example, that P acquires S with a $100 built-in gain asset that generates $10 income each year and declines in value $10 each year. P increases its basis in S $10 each year. In Year 5, P sells the S stock for $100, recognizing a $50 loss. The loss was not disallowed under Prior Reg. § 1.337(d)-2, because it was not attributable to the disposition of a built- in gain asset.

(3) The basis disconformity approach announced by the Service in Notice 2004-58 likewise required the disposition of an asset by inclusion of the gain amount in its formula.

(4) Nonetheless, the government remained concerned about the avoidance of the General Utilities repeal through the use of wasting assets. As a result, the Unified Loss Rules delete the gain amount from the basis reduction formula, thus permitting them to reach wasting assets. b. The old regulations did not define the phrase “disposition of an asset.”

(1) It was clear that such phrase included the actual sale or exchange of an asset that is treated as such for tax purposes.

(2) However, the tax law treats a number of transactions that are not sales or exchanges as a sale or exchange. Conversely, it treats many actual sales or exchanges as something else.

(a) For example, a distribution on stock that exceeds earnings and profits and the basis of the stock is treated as “gain from the sale or exchange of property” under section 301(c)(3)(A).

(b) Similarly, section 631(a) permits taxpayers to elect to treat the cutting of timber as a sale or exchange of such timber.

(c) On the other hand, gain from the sale or exchange of stock of a controlled foreign corporation is treated as a dividend under section 1248(a).

- 57 -

(3) Perhaps the definition of “extraordinary gain dispositions” contained in Prior Reg. § 1.1502-20(c)(2)(i) could provide some guidance. Extraordinary gain dispositions included actual or deemed dispositions of capital assets, property used in a trade or business, and certain bulk asset dispositions.

c. For purposes of the old regulations, disposition of “an asset” included the disposition of stock or securities. Prior Reg. § 1.337(d)-2(c)(2).

4. Recognition of Built-In Gain

a. Reg. § 1.337(d)-2(b) provided that a loss would not be disallowed under the loss disallowance rule to the extent the taxpayer established that it was not attributable to the recognition of built-in gain. Thus, built-in gain must actually be recognized by the subsidiary before it would trigger the loss disallowance rules.

b. The basis disconformity approach likewise required the recognition of built-in gain by inclusion of the gain and PIA amounts in its formula.

c. Gain could be built-in gain if it was attributable directly or indirectly to an excess of value over basis. Prior Reg. § 1.337(d)- 2(c)(2). Thus, for example, if S had a built-in gain in Asset A and exchanged it in a tax-free section 1031 exchange for Asset B, the sale of Asset B for a gain constituted a built-in gain.

5. Offsetting Gains and Losses

a. The old regulations permitted the recognition of built-in gains to be offset by the recognition of built-in losses or by the absorption of net operating loss carryovers. See Prior Reg. § 1.337(d)-2(c)(4), Example.

b. This was considerably more favorable than the result under pre- 2002 Reg. § 1.1502-20(c)(1)(i), which disallowed a loss to the extent of any recognized extraordinary gain, without any offset for built-in losses.

- 58 -

c. Example 13 - Built-In Loss Offsets Built-In Gain

(1) S Stock S Stock X P P W $40 $50 $50 basis $50 basis

S S S (2) Sell Gain Gain Asset $40 Asset for $50 $50 Value Y Cash/New $0 Basis Asset (3) Loss Asset Sell Loss Asset for $0 $0 Value Z $50 Basis

(1) Facts: In Year 1, P acquires the stock of S for $50. S has two assets: Gain Asset with a value of $50 and a basis of $0 and Loss Asset with a value of $0 and a basis of $50. In Year 2, S sells Gain Asset to Y for $50, and reinvests the proceeds in New Asset. In Year 3, S sells Loss Asset to Z for $0. The value of New Asset declines to $40. P then sells all of the S stock to W for $40.

(2) Under the investment adjustment rules, P’s basis in the S stock is increased by S’s $50 gain in Year 2 and is decreased by S’s $50 loss in Year 3. Reg. § 1.1502- 32(b)(2)(i). Thus, when P sells its S stock in Year 3, it recognizes a $10 loss.

(3) Amount of Loss Disallowed

(a) Under a pure tracing approach, for purposes of determining whether P’s $10 loss is attributable to a built-in gain, S’s recognized built-in gain is offset by S’s recognized built-in loss. Thus, none of P’s $10 loss is attributable to built-in gain and is therefore allowed. Prior Reg. § 1.337(d)-2(c)(4),

- 59 -

Example; see also Prior Reg. § 1.337(d)-1(a)(5), Ex. 4.15

(b) The result should be the same under the modified tracing approach. There has been no alteration event. S’s reinvestment in New Asset and its subsequent decline in value did not create any inside/outside basis disparity because the value of S declined by the same amount.

(c) Similarly, under the basis disconformity approach, P’s loss should be allowed. The PIA amount is $0 because the $50 built-in loss offsets the $50 built-in gain. Accordingly, none of P’s loss is disallowed.

(4) The result would be the same if, instead of having a built-in loss in Loss Asset, S had a net operating loss carryover when P purchases the S stock, and the net operating loss is used to offset the built-in gain. Id.

(5) What if the P group were unable to use the loss that S recognized on the sale of Loss Asset? In that case, P’s loss on the sale of the S stock would be $60 instead of $10. Is the $50 loss attributable to S’s built-in gain disallowed? Because P’s basis in S reflected both S’s unrealized built-in gain and unrealized built-in loss, arguably the result should not be different.

(6) What if S’s loss was not built-in but arose after the acquisition as a result of the depreciation in value of Loss Asset?

(a) Under a pure tracing approach, because the loss arose only after S’s acquisition, it was not reflected in P’s basis in S and thus should not be available to offset S’s built-in gain. Only built-in losses may be netted against built-in gains under a pure tracing approach.

(b) The result should be the same under the modified tracing approach, because there has been no alteration event.

15 The examples in Reg. § 1.337(d)-1(a)(5) and Prior Reg. § 1.1502-20(a)(5) (other than examples 3, 4, and 5) are incorporated by reference in Reg. § 1.337(d)-2(c)(4).

- 60 -

(c) Under the basis disconformity approach, however, P’s loss should be allowed. The basis disconformity approach looks at net investment adjustments rather than built-in losses. Thus, recognized losses may offset built-in gains, even if attributable to post-acquisition depreciation in value.

(7) The new Unified Loss Rules should reach the same result as the basis disconformity approach, because positive and negative investment adjustments are netted in determining the PIA factor.

6. “Safe Harbors” – Circumstances Under Which Gain Should Not Be Considered “Built-In”

a. Pure Tracing Approach – In some circumstances, it should be relatively straightforward to show that a gain was not “built-in.” For example, there could be no potential for the recognition of built-in gain in the following circumstances:

(1) If a subsidiary was purchased in a transaction in which the parties made a section 338(h)(10) election to treat the stock purchase as a deemed asset purchase, so that the basis of the assets reflect their fair market value at the time of the acquisition.

(2) If a subsidiary was a member of the group since its inception.

(3) If a subsidiary continued to hold all of the assets it held on the date of its acquisition.

b. When the Service first announced the modified tracing and basis disconformity approaches, it stated that it would recognize the following safe harbors. However, Notice 2004-58 was silent on this issue.

(1) If a subsidiary was purchased in a transaction in which the parties made a section 338(h)(10) election, then there would be no built-in gain, as long as there had not been any redetermination transactions.

(2) If P acquired all of the S stock in a single section 351 transaction where P was the only transferor, then there would be no built-in gain, as long as there had not been any redetermination transactions.

- 61 -

c. The new Unified Loss Rules did not adopt any of these safe harbors, because the Service and Treasury viewed them as unnecessary. Because basis disconformity is measured immediately before the transfer of loss shares, the rule automatically excludes situations where there is basis conformity. See 72 Fed. Reg. at 2982.

7. Allowable Loss Under Pre-2002 Treasury Regulation Section 1.1502-20 Rules

a. Similar to Prior Reg. § 1.337(d)-2, Prior Reg. § 1.1502-20(a) contained a blanket loss disallowance rule and then carved out situations where the loss would be allowed. Under Prior Reg. § 1.1502-20(c), loss was permitted to the extent it exceeded a share’s allocable part of the sum of the following factors:

(1) Extraordinary gain dispositions – Income or gain, net of directly related expenses, allocated to the share from extraordinary gain dispositions;

(2) Positive investment adjustments (PIAs) – The amount of the positive adjustment with respect to the share under Reg. § 1.1502-32 for each consolidated return year, but only to the extent the amount exceeded the extraordinary gain amount; and

(3) Loss duplication – The amount of any duplicated loss.

b. As was the case with Prior Reg. § 1.337(d)-2(c), the limited loss allowance rule in Prior Reg. § 1.1502-20(c) also applied to the deconsolidation rule. Thus, the amount of the basis reduction was limited to the sum of the extraordinary gain dispositions, PIAs, and duplicated loss factors allocable to P’s remaining S stock. Prior Reg. § 1.1502-20(c)(1).

c. Extraordinary Gain Disposition Factor

(1) Extraordinary gain dispositions were defined as dispositions after November 18, 1990 resulting in gain from the disposition of capital assets, property used in a trade or business, and certain bulk asset dispositions. Prior Reg. § 1.1502-20(c)(2)(i).

(2) Income from discharge of indebtedness and a change in method of accounting resulting in positive section 481 adjustments were also treated as extraordinary gain dispositions.

- 62 -

(3) Note that Prior Reg. § 1.337(d)-2, with its focus on built-in gain items, essentially retained the extraordinary gain disposition factor. However, the new Unified Loss Rules eliminate it.

(4) The amount of extraordinary gain could be reduced by directly related expenses from the disposition. Prior Reg. § 1.1502-20(c)(1)(i).

(5) Unlike Prior Reg. § 1.337(d)-2, loss from extraordinary dispositions could not be used to offset gain from extraordinary gain dispositions under Prior Reg. § 1.1502- 20. See Prior Reg. § 1.337(d)-2(c)(4), Example.

(6) To illustrate the extraordinary gain disposition factor, assume that P acquires all of the stock of S for $100. S has assets it uses in its trade or business with a value of $100 and a basis of $0. S sells the assets, resulting in a $100 gain, which increases P’s basis in its S stock to $200. S uses the cash to buy a capital asset for $100. P sells its S stock for $100, resulting in a $100 loss ($200 basis less $100 value). The amount of extraordinary gain is $100. Because the $100 loss does not exceed the $100 extraordinary gain, all of the $100 loss would be disallowed under the prior rules. Note that the result was the same under Prior Reg. § 1.337(d)-2 and is the same under the new Unified Loss Rules. d. PIA Factor

(1) A “positive investment adjustment” was defined as the sum of the amounts under Reg. § 1.1502-32(b)(2)(i) through (iii) for the consolidated return year. Thus, the basis adjustments attributable to taxable income, tax loss, tax- exempt income, and noncapital, nondeductible expenses were netted for the taxable year to determine the amount (if any) of the PIA.

(a) This amount was not reduced by negative adjustments attributable to distributions.

(b) In addition, losses were taken into account as they arose, not when they were absorbed. Prior Reg. § 1.1502-20(c)(2)(ii).

(c) Netting of profits and losses arising in the same year was permitted, but netting positive and

- 63 -

negative investment adjustments from different years was not permitted. See Prior Reg. § 1.1502- 20(c)(4), Ex. 3.

(2) Thus, all net PIAs were presumed to be attributable to built-in gain. The reason for the PIA factor was the Service’s concern over “wasting” assets. Prior Reg. § 1.337(d)-2 did not reach wasting assets.16 The new Unified Loss Rules return to the historic presumption.

(3) To illustrate the PIA factor, assume that P acquires all of the stock of S for $100. S has an asset with a $100 value and a $0 basis. In Year 1, S earns $100 of operating income, and P’s basis in S is increased by $100 to $200. In Year 2, S’s asset declines in value to $0. S invests the $100 from operating income in another asset, which loses $25 during the year. The loss is absorbed by the P group, thus reducing P’s basis in S to $175. In Year 3, P sells S for $75, realizing a $100 loss. Under the limited loss allowance rule of Prior Reg. § 1.1502-20, the loss disallowed was equal to the amount of the PIAs – $100 from Year 1. This amount was not reduced by the $25 loss from Year 2. Thus, all of P’s $100 loss was disallowed.

e. Loss Duplication Factor

(1) Duplicated loss was defined as the excess of (A) the sum of the aggregate asset basis, loss carryovers, and deferred deductions (such as suspended losses under the passive activity loss rules) of the subsidiary, over (B) the value of the subsidiary’s stock and any liabilities of the subsidiary. Prior Reg. § 1.1502-20(c)(2)(vi).

(a) The amounts computed in the loss duplication formula included the subsidiary’s share of corresponding amounts with respect to lower tier subsidiaries. Id.

(b) Aggregate asset basis did not include any stock or securities in another subsidiary. Id. Note that the

16 After the Federal Circuit invalidated the loss duplication factor in Rite Aid, the Service conceded the loss disallowance issue in the context of a PIA factor in Square D Co. v. Commissioner, 118 T.C. 299 (2002). See Michael L. Schler, Consolidated Return Loss Disallowance: Conceptual Issues, 2002 Tax Notes Today 88-30 n.10 (May 3, 2002).

- 64 -

Unified Loss Rules do include the basis of lower- tier subsidiaries in computing both basis reduction and attribute reduction. See Reg. § 1.1502-36(c)(6), (d)(5).

(2) The loss duplication factor was intended to ensure that a stock loss allowed under the regulations was not duplicated at a later time. Prior Reg. § 1.337(d)-2 does not contain a loss duplication factor, but the Service issued Prior Reg. § 1.1502-35 to address its duplicated loss concern.

(3) To illustrate the duplicated loss factor, assume that P forms S with a contribution of $100. S buys a manufacturing plant for $100. S has an operating loss of $60, which the P group is unable to use on its consolidated return and which cannot be carried back. S’s basis in its assets is reduced to $40. P’s basis in its S stock remains at $100. Reg. § 1.1502-32(b)(2) and (3). P sells S to X for $40, recognizing a $60 loss. S is apportioned its $60 net operating loss carryover when it leaves the P group. Reg. §§ 1.1502-79(a), 1.1502-21(b)(2).

Under Prior Reg. § 1.1502-20, S’s duplicated loss was $60, determined as follows:

Sum of S’s asset bases ($40) and loss carryovers ($60): $100 Less value of S’s stock: $ 40 Duplicated Loss: $ 60

Because P’s $60 loss on the sale of S does not exceed the $60 duplicated loss, all of P’s loss would have been disallowed. Importantly, P would be permitted to reattribute to itself S’s $60 loss carryover. However, if the duplicated loss were attributable to a built-in loss asset, no such relief was available to P.

C. Netting Rule

1. There was a narrow exception to the loss disallowance rule of Prior Reg. § 1.337(d)-2: Loss could be recognized to the extent that gain was taken into account by group members on the sale of the subsidiary stock having the same material terms as a consequence of the same plan or arrangement. Prior Reg. § 1.337(d)-2(a)(4).

- 65 -

a. A similar netting rule applied to the deconsolidation rule. Thus, basis was not required to be reduced upon deconsolidation to the extent that gain was taken into account by group members with respect to subsidiary stock having the same material terms as a consequence of the same plan or arrangement. Prior Reg. § 1.337(d)-2(b)(4).

b. The subsidiary stock must have the same material terms for the netting rule to apply.

(1) If, for example, the subsidiary stock sold at a loss was common stock and the subsidiary stock sold at a gain was preferred stock, netting would not be available, because the preferred stock does not have the same material terms as the common stock.

(2) However, if both blocks of subsidiary stock sold were preferred stock, netting would be permitted if the loss on the sale of preferred stock were offset by the gain on the sale of preferred stock having the same material terms.

c. This netting rule was the same as the netting rule contained in pre- 2002 Reg. § 1.1502-20(a)(4), (b)(4).17

d. The Unified Loss Rules also contain a netting rule. However, the rule is not limited to stock having the same material terms, and the stock must be transferred in the same transaction rather than the same plan or arrangement. Reg. § 1.1502-36(c)(7). See Section II,D,3, above.

17 The temporary regulations, as originally promulgated, did not contain a netting rule, but they were amended on May 31, 2002 to provide such a rule. See Preamble to Prior Temp. Reg. § 1.337(d)-2T, 67 Fed. Reg. at 37,998 (indicating that Temp. Reg. § 1.337(d)-2T was amended to provide a netting rule similar to that of Prior Reg. § 1.1502-20(a)(4)). The final regulations contained the same netting rule found in the temporary regulations.

- 66 -

2. Example 14 - Netting Gains and Losses

Sell T Stock P

X 50 Shares $100 Value Sell T Stock $150 Basis S

50 Shares $100 Value $50 Basis T

a. Facts: P owns all of the stock of S and 50 shares of T common stock. S owns the remaining 50 shares of T common stock. P has a $150 basis in its T stock, which is worth $100. S has a $50 basis in its T stock, which is also worth $100. P and S sell all of their T stock outside the group to X, an unrelated buyer.

b. S’s $50 gain is permitted to offset P’s $50 loss under the netting rule.

(1) If P and S sell their T stock to the public, the public would likely be treated as one purchaser for purposes of applying the netting rule. Cf. Prior Reg. § 1.1502-20(a)(5), Ex. 5.

(2) In addition, netting is permitted if stock is sold to more than one purchaser, provided it is pursuant to the same plan or arrangement.

c. Assume that S distributes its T stock to P, resulting in $50 of section 311(b) gain that is deferred. Reg. §§ 1.1502-13(c); -13(f)(2)(iii). If P sells all of the T stock outside the group, the deferred gain is taken into account. Reg. § 1.1502-13(d)(2)(i). Loss will be permitted to be recognized under Prior Reg. § 1.337(d)-2(a)(4) – but only to the extent of the $50 deferred gain.

d. The netting rule may also have the effect of sheltering son-of- mirror type losses. Assume in Example 14 that S purchases 50 percent of the T stock for $50 and P thereafter purchases the remaining 50 percent for $100. T sells a built-in gain asset and recognizes $50 gain, which results in a basis increase of $25 to each of P’s and S’s 50-percent interest. Reg. § 1.1502-32(b)(2)(i).

- 67 -

Thus, P’s basis in its T shares is increased to $125 and S’s basis is increased to $75. When P and S sell their stock to X, P recognizes a loss of $25 and S recognizes gain of $25. Nonetheless, under the netting rule, S’s gain may offset P’s loss.

3. Example 15 - Netting Under the Deconsolidation Rule

P

X 50 Shares $100 Value Sell T Stock $150 Basis S

50 Shares $100 Value $50 Basis T

a. Facts: Same as Example 14, except that P does not sell its T stock.

b. S’s sale of T stock to X for $100 results in a $50 gain. Under the deconsolidation rule, P must reduce the basis of its T stock by $50 from $150 to $100, the value of the T stock immediately before deconsolidation. However, under the netting rule, because S’s $50 gain is recognized as a consequence of the same plan or arrangement as that giving rise to the deconsolidation, P’s $50 basis reduction in its T stock is eliminated. P’s basis in its T stock thus remains at $150. Prior Reg. § 1.337(d)-2(b)(4); see also Prior Reg. § 1.1502-20(b)(4), (b)(6), Ex. 6.

c. Of course, the P group could have avoided a gain if P had sold its T stock instead of S.

d. The netting rules also provide for an anti-duplication rule in the event the loss disallowance rule and deconsolidation rule apply to the same transaction. If gain from the sale of subsidiary stock could be used to allow the use of a loss that would otherwise be disallowed and also to prevent a reduction in basis that would be required under the deconsolidation rule, the gain may be taken into account only once. Prior Reg. § 1.337(d)-2(a)(4); see also Prior Reg. §§ 1.1502-20(a)(4), 1.1502-20(b)(6), Ex. 7. In this

- 68 -

circumstance, the group has the option to use the netting rule under either the loss disallowance rule or the deconsolidation rule.

(1) Assume in the above example that P owns 100 shares of T stock with a $200 value and a $300 basis. At the same time S sells its T stock for a $50 gain, P sells 50 shares of T stock for a $50 loss ($100 value less $150 basis). The $50 loss would be disallowed under Prior Reg. § 1.337(d)- 2(a)(1).

(2) The $50 gain on S’s sale of the T stock can be used under the netting rule in the loss disallowance rule or the deconsolidation rule, but may be taken into account only once. If the P group chooses to use the $50 gain to offset the $50 loss, P must reduce the basis of its remaining T stock by $50 to its $100 value.

D. Coordination with Loss Deferral and Other Loss Disallowance Rules

1. The old regulations expressly incorporated the rule of Prior Reg. § 1.1502- 20(a)(3) (“with appropriate adjustments to reflect differences between the approach of this section and that of § 1.1502-20”), which applied special rules in the case where a loss on the sale of subsidiary stock would be deferred or disallowed under a Code provision or regulation other than Prior Reg. § 1.1502-20. Prior Reg. § 1.337(d)-2(a)(3).

2. Prior Reg. § 1.1502-20(a)(3)(i) provided that any other provision of the Code or regulations that disallowed or deferred a loss on the disposition of subsidiary stock applied before the general loss disallowance rule. If the loss was deferred under another provision, then the loss was subject to the loss disallowance rule when the loss was ultimately taken into account.

3. However, if a so-called “overriding event” occurred prior to the time that a deferred loss was taken into account, the loss was subject to the loss disallowance rule at that time. Prior Reg. § 1.1502-20(a)(3)(i), (ii). The overriding events included the following:

a. The subsidiary stock ceased to be owned by a member of the consolidated group;

b. The subsidiary stock was canceled or redeemed (regardless of whether it is retired or held as treasury stock); and

c. The subsidiary stock was treated as disposed of under Reg. § 1.1502-19(c)(1)(ii)(B) or (c)(1)(iii). Under Reg. § 1.1502- 19(c)(1)(ii)(B) and (c)(1)(iii), a disposition is deemed to occur as follows:

- 69 -

(1) At the time S becomes a nonmember, or P’s basis in the stock is reflected, directly or indirectly, in whole or in part, in the basis of any asset other than member stock;

(2) At the time substantially all of S’s assets are treated as disposed of, abandoned, or destroyed for federal income tax purposes;

(3) At the time an indebtedness of S is discharged, if any part of the amount discharged is not included in gross income and is not treated as tax-exempt income under Reg. § 1.1502-32(b)(3)(ii)(C); or

(4) At the time a member takes into account a deduction or loss for the uncollectible indebtedness of S, and the deduction or loss is not matched in the same tax year by S’s taking into account a corresponding amount of income.

4. Compare the Unified Loss Rules, which apply immediately upon the transfer of a loss share even if the loss is deferred, disallowed, or otherwise not taken into account under any other applicable rules of law. The only exception is with respect to intercompany losses under Reg. § 1.1502-13, in which case the Unified Loss Rules apply upon the triggering of the loss. Reg. § 1.1502-36(e)(3).

- 70 -

5. Example 16 - Coordination With Loss Deferral Rules

(3) T Stock P Y $100

(2) Sell T Stock

S

$100 Value $100 Basis (1) Asset T X $100 Asset: $100 Value $40 Basis

a. Facts: P owns all of the stock of S, and S owns all of the stock of recently purchased T. S has a $100 basis in its T stock, which is worth $100. T owns one asset with a value of $100 and a basis of $40. T sells the asset to X for $100. S then sells its T stock to P for $100. P later sells all of the T stock for $100 to Y, a member of the same controlled group (as defined in section 267(f)) as P but not a member of the P consolidated group.

b. T recognizes a $60 gain on the sale of its asset, which increases S’s basis in the T stock from $100 to $160. When S sells the T stock to P, it recognizes a $60 loss that is deferred under section 267(f) and Reg. § 1.1502-13(c). (This sale is not subject to section 304(a)(1). See Rev. Rul. 74-605, 1974-2 C.B. 97; Reg. § 1.1502- 80(b)).

c. Pursuant to Prior Reg. § 1.1502-20(a)(3), the loss disallowance rule would not ordinarily apply to S’s $60 loss, because the loss is deferred under section 267(f) and Reg. § 1.1502-13(c). Although P’s sale of the T stock to Y would cause S’s deferred loss to be taken into account under Reg. § 1.1502-13(d), Reg. § 1.267(f)- 1(b)(1) and -1(c)(1)(i) provide that the loss is not taken into account, because Y is a member of the same controlled group as P and S. Nevertheless, the sale of the T stock to Y is an “overriding

- 71 -

event,” since T ceases to be a member of the P consolidated group. Therefore, S’s $60 loss is disallowed under Prior Reg. § 1.337(d)- 2(a)(1) and is never taken into account under section 267(f). See Prior Reg. § 1.1502-20(a)(5), Ex. 6(ii).

E. Successor Rule

1. Prior Reg. § 1.337(d)-2(d) incorporated the successor rules (and examples) of Prior Reg. § 1.1502-20(d). Thus, Prior Reg. § 1.337(d)-2 applied, to the extent necessary to effectuate the purposes of the regulation, to any property the basis of which is determined, directly or indirectly, in whole or in part, by reference to the basis of a subsidiary’s stock. See Prior Reg. § 1.1502-20(d)(1).

2. Example 17 – Successor Rule

P X

LP interest $100 Value

S stock $100 Value GP Interest $200 Basis

LP S

a. Facts: P owns all of the stock of S, which has a value of $100 and a basis of $200. P contributes its S stock to LP partnership in exchange for a limited partnership interest worth $100.

b. Pursuant to the deconsolidation rule, P is required to reduce the basis of its S stock to its $100 fair market value immediately before the transfer of the S stock to LP. As a result, P takes a $100 basis in its partnership interest under section 722, and the partnership takes a $100 basis in the S stock under section 723.

c. P’s partnership interest is a successor interest to the S stock. If P sells the partnership interest at a loss within two years, P’s loss will be disallowed. Prior Reg. § 1.1502-20(d)(2), Ex. 2.

3. The Unified Loss Rules apply generally to predecessor or successor persons, groups, and assets to the extent necessary to effectuate the purposes of the regulations. Reg. § 1.1502-36(e)(1).

- 72 -

F. Anti-Avoidance Rules

1. Prior Rules Apply – Prior Reg. § 1.337(d)-2(e) incorporated the anti- avoidance rules (and examples) of pre-2002 Reg. § 1.1502-20(e).

2. General Anti-Avoidance Rule

a. Prior Reg. § 1.1502-20(e)(1) contained a catch-all “anti- avoidance” rule: “The rules of section 1.1502-20 must be applied in a manner that is consistent with and reasonably carries out their purposes. If a taxpayer acts with a view to avoid the effect of the rules of this section, adjustments will be made as necessary to carry out their purposes.” See also Reg. § 1.1502-36(g).

b. Example 18 – Shifting of Value

(3) S Common S Stock P Z X P $100 $100 $100 basis (1) S Preferred $100 Land stock

(2) Asset S S S Y $100 Asset $100 Value $0 Basis

(1) Facts: In Year 1, P buys all of the stock of S for $100, and S becomes a member of the P group. S holds an asset with a value of $100 and a basis of $0. In Year 2, with the proscribed view, P transfers land with a value of $100 to S in exchange for preferred stock with a $200 redemption price. S sells the built-in gain asset to Y for $100. In Year 3, P sells the S common stock to Z for $100.

(2) The $100 redemption premium increases the value of the preferred stock to $200 and decreases the value of the common stock to $0. When S sells the built-in gain asset for $100, P’s basis in the common and preferred S stock is increased to $300. As a result of the cumulative redetermination rule of Reg. § 1.1502-32(c)(4), P’s basis in

- 73 -

the S preferred stock increases from $100 to $200 and P’s basis in the common stock remains $100. Thus, P recognizes a $100 loss when it sells the S common stock.

(3) Under section 305, the redemption premium on the preferred stock is treated as a section 301 distribution. As a result, P’s bases in the preferred and common stock are unaffected.

(4) P’s loss on the sale of the S common stock is disallowed under Prior Reg. §§ 1.1502-20(e)(1) and 1.337(d)-2(e). See Prior Reg. § 1.1502-20(e)(3), Ex. 1. The disallowance prevents the preferred stock from shifting value and stock basis adjustments from the common stock to avoid disallowance of the loss.

3. Anti-Stuffing Rule

a. The anti-stuffing rule was triggered when (i) any asset was transferred to a subsidiary, (ii) the stock of the subsidiary was disposed of within two years of the transfer, and (iii) the transfer was “with a view” to avoiding, directly or indirectly, the loss disallowance rule, deconsolidation rule, or gain recognition on the transferred asset. Prior Reg. § 1.1502-20(e)(2)(i).

b. If the anti-stuffing rule was triggered, the basis of the subsidiary stock that was disposed of was reduced immediately prior to the disposition so as to cause gain recognition in the amount equal to the avoided loss disallowance, basis reduction, or gain recognition. Prior Reg. § 1.1502-20(e)(2)(ii).

c. The anti-stuffing rule also applied to a transfer of a consolidated subsidiary’s stock to another consolidated subsidiary. Prior Reg. § 1.1502-20(e)(3), Ex. 4.

- 74 -

d. Example 19 – Basic Stuffing Case

(4) (1) $200 S Stock P Z X P S Stock $100 (3) Asset 2: $200 basis $100 basis $100 Value (2) $0 Basis Asset 1 S S S Y $100 Cash Asset 1 $100 cash $100 Value $0 Basis

(1) Facts: In Year 1, P acquires S for $100 and S becomes a member of the P group. S has an asset with a $100 value and $0 basis, which it sells outside the group to Y. In Year 5, P transfers to S in a section 351 transaction an asset with a $100 value and a $0 basis with a view to avoiding the loss disallowance rule. In Year 6, P sells its S stock to Z for $200.

(2) Under the investment adjustment rules, P’s basis in the S stock is increased by S’s $100 gain to $200. Reg. § 1.1502-32(b)(2)(i). When P sells its S stock in Year 6, it recognizes no gain or loss.

(3) Under the anti-stuffing rule, P must reduce its basis in S to $100 immediately before the sale of the S stock, resulting in a $100 gain to P. This gain is equal to the $100 gain P avoided by making the asset transfer. Prior Reg. § 1.1502- 20(e)(3), Ex. 2.

(4) If P would have waited for more than two years after the asset transfer to sell S, provided the sale was not pursuant to a prearranged plan, the anti-stuffing rule would not have applied.

(5) If the S stock were deconsolidated in Year 6 rather than sold, P would still be required to reduce the basis in its S stock by $100 immediately before the deconsolidation. Prior Reg. § 1.1502-20(e)(3), Ex. 2(iii).

- 75 -

(6) If the P stock were acquired by another group in Year 6, the $100 basis reduction would still be required, even though the asset transfer took place outside the acquiring group. According to the regulations, the anti-stuffing rule “requires only that the transferor have the view at the time of the transfer.” Prior Reg. § 1.1502-20(e)(3), Ex. 2(iv).

e. The Unified Loss Rules include anti-stuffing in the general anti- abuse rule but it is not subject to the 2-year limitation. Reg. § 1.1502-36(g).

G. No Tiering Up of Certain Adjustments

1. Prior Reg. § 1.337(d)-2(f) incorporates the rules (and examples) of pre- 2002 Reg. § 1.1502-20(f), which limited the tiering up of certain investment adjustments.

2. A loss that was recognized but disallowed was treated as a noncapital, nondeductible expense, and under Reg. § 1.1502-32(b)(3)(iii), caused a negative investment adjustment.18 Under Reg. § 1.1502-33, a subsidiary’s earnings and profits are similarly reduced (for earnings and profits purposes) by the amount of the subsidiary’s disallowed loss. Under Reg. § 1.1502-33(b), the reduction in earnings and profits is reflected in the earnings and profits of the parent of such a subsidiary.

3. In the case of a deconsolidation of a subsidiary, the basis of the stock of the subsidiary would be reduced to reflect its fair market value. Prior Reg. § 1.337(d)-2 (b)(1). Under Reg. § 1.1502-32(b)(3)(iii), the reduction of the basis of the subsidiary’s stock will be treated as a noncapital, nondeductible expense and will cause a negative investment adjustment. In that case, the earnings and profits of the parent will be reduced by a similar amount immediately prior to the deconsolidation.

4. Prior Reg. § 1.1502-20(f)(1) provided that if the basis of a subsidiary’s stock was reduced upon the deconsolidation of such subsidiary’s stock, then no corresponding adjustment was made under the investment adjustment system to the basis of the stock of the subsidiary’s parent, if

18 Note that if a member’s basis in the subsidiary stock was reduced under Reg. § 1.1502- 32 by reason of a disallowed loss, and such disallowed loss is reduced by reason of an election under Prior Reg. § 1.1502-20(i), but would have expired or been absorbed in a closed year, then the member’s basis in the subsidiary stock could be increased for purposes of determining the group’s or member’s federal income tax liability for open years. Prior Reg. § 1.1502- 20(i)(3)(v)(B).

- 76 -

there is a deconsolidation or disposition of the parent’s stock in the same transaction.

5. The Unified Loss Rules in Reg. § 1.1502-36 apply to transferred loss shares, which are defined to include deconsolidations. Reg. § 1.1502- 36(f)(10)(i)(B). They do not contain a separate deconsolidation rule.

6. Example 20 – Deconsolidation of Parent in Same Transaction as Subsidiary P

$100 Basis

S (3) S-1 Stock A $100 Basis

S-1

$100 Basis (1) T Stock $100 X S-2 $100

(2) Sell Asset T Y $100 Asset $100 Value $0 Basis

a. Facts: P owns all of the stock of S, S owns all the stock of S-1, and S-1 owns all the stock of S-2. P’s basis in the S stock is $100, S’s basis in the S-1 stock is $100, and S-1’s basis in the S-2 stock is $100. In Year 1, S-2 buys the stock of T for $100. T holds an asset with a value of $100 and a basis of $0. In Year 2, T sells the asset to Y for $100. In Year 6, S sells the S-1 stock to Individual A for $100. The new S-1 group does not file a consolidated return.

b. Under the investment adjustment system, in Year 2, when T recognizes $100 gain, the basis of each subsidiary’s stock increases from $100 to $200. Thus, when S sells the stock of S-1, it recognizes a $100 loss.

- 77 -

c. The $100 loss resulting from the sale of the S-1 stock is disallowed. Prior Reg. § 1.337(d)-2(a)(1). Under Reg. § 1.1502- 32(b)(3)(iii), S’s disallowed loss is treated as a noncapital, nondeductible expense that reduces P’s basis in the S stock. Under Reg. § 1.1502-33, S’s earnings and profits are reduced, and this reduction is also reflected in P’s earnings and profits.

d. Under Prior Reg. § 1.337(d)-2(b)(1), the basis of the stock of T and S-2 must be reduced immediately before the sale from $200 to $100, because their stock is deconsolidated as a result of S’s sale of the S-1 stock. However, under Prior Reg. § 1.1502-20(f)(1) and Prior Reg. § 1.337(d)-2(f), the basis reduction to neither the T stock nor the S-2 stock tiers up, because the S-2 stock is deconsolidated and the S-1 stock is disposed of in the same transaction.

e. Similar treatment applies for purposes of the tiering up of earnings and profits under Reg. § 1.1502-33.

H. Prior Reg. § 1.1502-20(i) – Transition Rules

1. General Rule – For dispositions of stock occurring before March 7, 2002, or for dispositions or deconsolidation of stock of a subsidiary after March 7, 2002 effected pursuant to a binding written contract entered into before March 7, 2002 that was in continuous effect, Prior Reg. § 1.1502-20(i)(2) allowed a parent to choose one of three regulatory schemes for each separate disposition of subsidiary stock:

a. Prior Reg. § 1.1502-20 in its entirety;

b. Prior Reg. § 1.1502-20 without regard to the loss duplication factor (“-20 Lite”); or

c. Prior Reg. § 1.337(d)-2.

2. Election – The parent’s election under Prior Reg. § 1.1502-20(i)(2) must have been filed with a timely filed (including extensions)19 original return for the taxable year that included any date on or before March 7, 2002. Prior Reg. § 1.1502-20(i)(3)(v)(4).

a. If the date of the disposition or deconsolidation of the stock was after March 7, 2002 pursuant to a binding contract, the election could either be filed (i) with or as part of a timely filed original

19 Note that the election provisions were amended by the Service on May 31, 2002 to clarify that a timely filed return included extensions. See 67 Fed. Reg. at 37,998.

- 78 -

return for the tax year that includes the disposition or deconsolidation date, or (ii) with or as part of an amended return filed before the date that the original return for the tax year that includes March 7, 2002 is due. Id.20

b. If no election was filed, Prior Reg. § 1.1502-20 applied in its entirety, including the loss duplication factor invalidated by Rite Aid. See Prior Reg. § 1.1502-20(i)(2). Thus, a parent need not file an election under Prior Reg. § 1.1502-20(i)(2) if it was going to apply § 1.1502-20 in its entirety.

c. In conjunction with the issuance of Notice 2004-58, the Service issued temporary regulations to permit taxpayers to make, amend, or revoke elections under Prior Reg. § 1.1502-20(i). See Prior Reg. § 1.1502-20(i)(6); Temporary Regulations on Extension of Time to Elect Method for Determining Allowable Loss, 69 Fed. Reg. 52,419 (August 26, 2004).

(1) A taxpayer that was permitted to make an election under § 1.1502-20(i), but did not previously make such an election, could make an election to apply either -20 Lite or Prior Reg. § 1.337(d)-2. The regulations also permitted a taxpayer that previously made an election to apply -20 Lite to revoke the election and apply Prior Reg. § 1.1502-20 in its entirety, or to amend the election in order to apply Prior Reg. § 1.337(d)-2. In addition, the regulations permitted a taxpayer that previously made an election to apply Prior Reg. § 1.337(d)-2 to revoke the election and apply Prior Reg. § 1.1502-20 in its entirety or to amend the election in order to apply -20 Lite.

(2) To revoke or amend an election, the taxpayer had to include a statement with or as part of any timely filed (including any extensions) original return for a taxable year that included any date on or before August 26, 2004, or with or as part of an amended return filed before the date the original return for the taxable year that included August 26, 2004 is due (including any extensions).

20 The May 31, 2002 amendments to the temporary regulations also clarified that elections could be made on an original return filed for the tax year 2001, even though it does not include March 7, 2002. See 67 Fed. Reg. at 37,998.

- 79 -

3. Cascading Losses

a. Allowance of Loss in Open Years – If an election to apply -20 Lite or Prior Reg. § 1.337(d)-2 increased the loss allowed on the disposition of subsidiary stock, but the year of the disposition is closed, and the absorption of such extra loss would have affected the tax treatment of another item that has an effect in an open year, then the taxpayer could adjust the other item. Prior Reg. § 1.1502- 20(i)(3)(v)(A).21

b. Corresponding Basis Adjustment – If a member’s basis in stock of a subsidiary was reduced pursuant to Reg. § 1.1502-32 because a loss with respect to such stock was disallowed under Prior Reg. § 1.1502-20, then to the extent such disallowed loss is allowed as a result of an election to apply -20 Lite or Prior Reg. § 1.337(d)-2 but would have been absorbed or expired in a closed year, the member’s basis in the subsidiary stock may be increased for purposes of determining the group’s or the shareholder-member’s federal income tax liability for an open year. Prior Reg. § 1.1502- 20(i)(3)(v)(B).22

c. For example, assume that P owns all of the stock of S and they file a consolidated return. P’s S stock becomes worthless in 1995, but P’s $20 loss is disallowed under Prior Reg. § 1.1502-20. P’s worthless stock loss would not be disallowed under Prior Reg. § 1.337(d)-2. The 1995, 1996, and 1997 tax years are closed. Assume that the P group has the following consolidated net income for 1995-1998:

1995 1996 1997 1998 Consolidated income $25 ($20) $10 $6 Net income after § 172 $ 5 $ 0 $10 $6

P’s allowed loss would have reduced the P group’s consolidated income in 1995 to $5, and the $20 operating loss in 1996 could have been carried back to offset the $5 in 1995 and carried forward to offset the $10 in 1997 and $5 of the $6 in 1998. Although the 1995-1997 tax years are closed, the P group may utilize the loss carryforward that would have otherwise been available to it in

21 Note that this provision was added by the amendments to the temporary regulations issued on May 31, 2002. See 67 Fed. Reg. at 37,999.

22 This provision was also added by the amendments to the temporary regulations issued on May 31, 2002. See 67 Fed. Reg. at 37,999.

- 80 -

1998 to reduce its net income after § 172 to $1. In addition, if P’s basis in S is relevant to a determination of federal income tax liability of P or the P group, then its basis may be increased by the $20 loss now allowed.

4. Reattribution Rule

a. Prior Reg. § 1.337(d)-2 did not provide for the reattribution of losses.

b. Pre-2002 Treasury Regulation Section 1.1502-20

(1) Under Prior Reg. § 1.1502-20(g)(1), upon the disposition of subsidiary stock, a parent could elect to reattribute to itself the subsidiary’s net operating and net capital losses (including SRLY losses) to the extent of the disallowed loss.

(2) The parent could elect to retain any or all such losses, and could specify the particular year and the character of the loss that is subject to reattribution.

(3) The parent could also reattribute to itself losses of a lower tier subsidiary.

(4) There were a number of limitations on a parent’s ability to reattribute losses to itself:

(a) Losses of a subsidiary could not be reattributed to the extent that the subsidiary (and all higher tier subsidiaries) was insolvent. Prior Reg. § 1.1502- 20(g)(2).

(b) A SRLY loss reattributed to a parent retained its SRLY taint in the parent’s hands.

(c) Reattributed losses could not be carried back to a parent’s taxable year.

(d) The reattribution election was not available if the anti-stuffing rule applied or for stock whose basis was reduced because of the deconsolidation rule.

(e) In the event of bankruptcy, a judge could enjoin a parent from making the reattribution election on the ground that the loss is an asset of the bankrupt subsidiary’s estate. See In re Prudential Lines Inc.,

- 81 -

928 F.2d 565 (2d Cir. 1991), aff’g, 107 Bankr. 832 (S.D.N.Y. 1989) (enjoining parent from claiming worthless stock deduction which would have effectively eliminated subsidiary’s loss carryover under section 382).

(f) If a parent reattributed to itself the losses of a subsidiary, usually the subsidiary would have less value and the parent would receive less consideration for the subsidiary stock. If the subsidiary had minority shareholders, they could complain that the parent as majority shareholder breached its fiduciary duty to them. See e.g., Meyerson v. El Paso Natural Gas Co., 246 A.2d 789 (Del. Ch. 1967) (business judgment rule protects parent’s use of subsidiary’s losses).

(5) Example 21 – Reattribution Rule

$60 P X S stock $100 Basis

S

$40 loss carryover

(a) Facts: P owns the stock of S, which has a basis of $100. S has an unused net operating loss of $40. P sells S for $60, producing a $40 loss, which is disallowed under the loss disallowance rule. P elects to reattribute to itself S’s $40 loss carryover.

(b) The reattribution is treated as a reduction of S’s loss carryover, which creates a negative investment adjustment under Reg. § 1.1502-32(b)(3)(iii). Thus, P’s basis in S’s stock is reduced by $40, and P has no gain or loss on the sale of S’s stock.

(c) Although P’s sale of the S stock may result in an ownership change under section 382, the reattributed losses are not subject to the section 382

- 82 -

limitation on the use of losses. Prior Reg. § 1.1502- 20(g)(1). c. Prior Treasury Regulation Section 1.337(d)-2

(1) Prior Reg. § 1.337(d)-2, effective for dispositions or deconsolidations on or after March 3, 2005, did not contain a reattribution rule. Nor did Prior Temp. Reg. § 1.337(d)- 2T (which is not substantively different from the final regulations), which was effective for dispositions or deconsolidations on or after March 7, 2002, contain a reattribution rule.

(2) However, for dispositions and deconsolidations before March 7, 2002, Prior Reg. § 1.1502-20(i) permitted taxpayers to elect to apply (i) Prior Reg. § 1.1502-20 in its entirety, (ii) -20 Lite, or (iii) Prior Reg. § 1.337(d)-2.

(3) The prior regulations contained a special rule where an election to reattribute losses under Reg. § 1.1502-20(g) was in place and the amount of such losses were reduced by reason of an election under Prior Reg. § 1.1502-20(i).

(a) If the parent elected to apply -20 Lite pursuant to Prior Reg. § 1.1502-20(i)(2)(i), the amount of reattributed loss had to be reduced to the extent that it exceeded the greater of (i) the amount of loss disallowed under -20 Lite; and (ii) the amount of reattributed losses that the consolidated group absorbed in the closed years. Prior Reg. § 1.1502- 20(i)(3)(i). However, in order to reattribute losses under Prior Reg. § 1.1502-20(g), the parent must have made a valid election under Prior Reg. § 1.1502-20(g). The transition rules did not extend the time for filing this election. Further, if the parent already made an election under Prior Reg. § 1.1502-20(g), it may not revoke the election. Prior Reg. § 1.1502-20(i)(3)(i).

(b) A parent could not reattribute any losses if it elected to apply Prior Reg. § 1.337(d)-2. Prior Reg. § 1.1502-20(i)(3)(ii). However, if the parent had already elected under Prior Reg. § 1.1502-20(g) with respect to the disposition of subsidiary stock, the parent could reattribute losses equal to the greater of zero or the amount of reattributed losses

- 83 -

that the consolidated group absorbed in the closed years. Prior Reg. § 1.1502-20(i)(3)(ii).

(c) If any losses were reattributed under Prior Reg. § 1.1502-20(g), such reattribution was binding on the subsidiary and any group of which the subsidiary was or became a member. Prior Reg. § 1.1502-20(i)(3)(vii).

(i) Indeed, even if the subsidiary subsequently ceased to be a member of the group, the subsidiary could not take advantage of the reattributed losses.

(ii) On the other hand, if the election to apply -20 Lite or Prior Reg. § 1.337(d)-2 resulted in the reduction in the losses reattributed to the parent pursuant to a Prior Reg. § 1.1502- 20(g) election, the subsidiary, or any group of which the subsidiary is a member, could use such losses. Prior Reg. § 1.1502- 20(i)(3)(vii).

(iii) As such, the parent had to notify the subsidiary prior to the date that the consolidated group filed its income tax return for the year that includes March 7, 2002. Prior Reg. § 1.1502-20(i)(3)(D)(iv). If the acquiror of the subsidiary stock was a member of a consolidated group at the time of the disposition, the parent also had to notify the parent of the acquiror’s group. Id.

(iv) The temporary regulations also added Prior Temp. Reg. § 1.1502-32T(b)(4)(v) to provide for the waiver of loss carryovers that revert to a subsidiary as a result of an election under Prior Temp. Reg. § 1.1502- 20T(i) (Reg. § 1.1502-32(b)(4)(v) and Prior Reg. § 1.1502-20(i) have since been made final). Absent such rules, the expiration of loss carryovers would result in a negative basis adjustment to the buyer under Reg. § 1.1502-32(b).

- 84 -

a) Under the temporary regulations, if reattributed losses reverted to the subsidiary and such loss carryovers expired or would have been used in a closed year, the buyer would be deemed to have waived the loss carryovers. Prior Temp. Reg. § 1.1502-32T(b)(4)(v)(A).

b) The temporary regulations were later amended to make the deemed waiver rule optional in order to provide relief where the deemed waiver rule operated to deny the use of excess losses. Id.; 68 Fed. Reg. 24,351.

c) If reattributed losses reverted to the subsidiary in open years, the buyer could make an election to waive those loss carryovers. Reg. § 1.1502-32(b)(4)(v)(B).

(d) Thus, in Example 21 above, if P elected to apply -20 Lite, its $40 loss would not be disallowed, and S would reacquire its net operating losses to the extent not used by the P group in the interim. X may elect to waive such carryovers if the year is open, or to deem such a waiver if the year is closed.

(e) Prior Reg. § 1.1502-20 provided special rules for apportioning a section 382 limitation when the reattributed losses were subject to such limitation.

(i) Reduction of section 382 limitation by parent – A parent could reduce the amount of section 382 limitation apportioned to itself, if, as a result of the application of the reattribution rules of Prior Reg. § 1.1502- 20(i)(3)(i) or (ii), and Prior Reg. § 1.1502- 20(i)(3)(vii), pre-change attributes subject to a 382 limitation were treated as the subsidiary’s losses and the parent previously elected to apportion all or part of such limitation to itself under Reg. § 1.1502- 96(d). This applies to separate, subgroup,

- 85 -

and consolidated section 382 limitations. See Prior Reg. § 1.1502-20(i)(3)(iii)(A)-(C).

(ii) Subsidiary no longer member of group or subgroup – If the subsidiary was no longer a member of the loss group or subgroup to which the pre-change attributes relate, the parent could increase the total amount of the section 382 limitation apportioned to such subsidiary (or loss subgroup that includes the subsidiary) under Reg. § 1.1502-95(c). Prior Reg. § 1.1502-20(i)(3)(iii)(B)-(C).

a) Subgroup section 382 limitation – The amount by which the parent could increase the subgroup 382 limitation apportioned to the subsidiary was limited to the amount by which the section 382 limitation apportioned to the parent is reduced under Prior Reg. § 1.1502- 20(i)(3)(iii)(B).

b) Consolidated section 382 limitation – The amount by which the parent could increase the consolidated 382 limitation (or subgroup section 382 limitation where the common parent was a member of the loss subgroup) apportioned to the subsidiary was limited to the product of the element (described in Reg. § 1.1502-95(c)) and the percentage of the total consolidated (or loss subgroup) pre- change attributes in the year that the subsidiary left the group that were treated as the subsidiary’s losses. Prior Reg. § 1.1502- 20(i)(3)(iii)(C).23

23 Specifically, the formula in Prior Reg. § 1.1502-20(i)(3)(III)(C) was as follows: the element x (prechange attributes subject to the 382 limitation treated as losses of the subsidiary or loss subgroup due to Prior Reg. § 1.1502-20(i)(3)(i) or (ii) and § 1.1502-20(i)(3)(vii)) ÷ (total pre- change attributes subject to the limitation determined as of the close of the taxable year in which the subsidiary ceases to be a member of the group (or loss subgroup)).

- 86 -

(iii) P’s reduction under Prior Reg. § 1.1502-20 (i)(3)(iii)(A) or (B) of its section 382 limitation was effective as of the date on which the previous apportionment was effective. Prior Reg. § 1.1502- 20(i)(3)(iii)(D)(ii). Increases in a subgroup or consolidated section 382 limitation apportioned to a departing subsidiary (or loss subgroup that includes such subsidiary) under Prior Reg. § 1.1502-20(i)(3)(iii)(B) or (C) were effective for all years ending after the date that the subsidiary ceases to be a member of the group or loss subgroup.

(iv) Prior Reg. § 1.1502-20 prescribed several limitations on the adjustments to the section 382 limitation:

a) In adjusting the consolidated or subgroup’s section 382 limitation, the parent could not include section 382 limitations that had been previously apportioned to another subsidiary or loss subgroup prior to the date of the Prior Reg. § 1.1502- 20(i)(2) election. Prior Reg. § 1.1502-20(i)(3)(iii)(D)(i).

b) Any adjustment had to be “consistent with the principles of § 1.1502- 95(c).” Prior Reg. § 1.1502- 20(i)(3)(iii)(D)(ii). For example, if apportionment of a separate section 382 limitation to a parent was reduced under the 382 limitation rules, the amount of such limitation available to the subsidiary was increased.

c) A parent could only make adjustments under Prior Reg. § 1.1502-20(i)(3)(iii)(A), (B), and (C). Prior Reg. § 1.1502- 20(i)(3)(iii)(D)(iv). These adjustments had to be made as part

- 87 -

of the election to apply -20 Lite or to apply Prior Reg. § 1.337(d)-2.

5. Waiver of Loss Carryovers

a. Waiver Election In General

(1) Reg. § 1.1502-32(b)(4) provided that, if a subsidiary had a loss carryover from a separate return limitation year when it became a member of a consolidated group, the group could make an election to treat all or any portion of the loss carryover as expiring immediately before the subsidiary became a member of the consolidated group.

(2) This election permitted an acquiring group to avoid the loss of stock basis that otherwise would result if the subsidiary’s loss carryovers were to expire before the group could absorb them. See Reg. § 1.1502-32(b)(2)(iii).

(3) The election may be made by identifying either the amount of each loss carryover deemed to expire or the amount of each loss carryover deemed not to expire.

(4) Any loss waived under Reg. § 1.1502-32(b)(4) could be excluded from the selling group’s computation of duplicated losses. Thus, the waiver could have the effect under the prior loss disallowance regulations of increasing the amount of stock loss allowed on the disposition of subsidiary stock.

(5) Under the Unified Loss Rules, losses waived under Reg. § 1.1502-32(b)(4) are excluded from the computation of net inside attributes for purposes of measuring the attribute reduction amount. This prevents attributes that cannot be duplicated from being taken into account in reducing attributes. However, for purposes of computing the basis disconformity amount, such waived losses are counted. This is because excluding them would have the effect of increasing disconformity under circumstances unrelated to the existence of built-in gain (which is what the disconformity amount is trying to measure). See 73 Fed. Reg. at 53,940.

b. Effect of Election Under Prior Reg. § 1.1502-20(i)

(1) Acquiring and selling groups could have negotiated to have the acquiring group waive loss carryovers in an effort to

- 88 -

increase the amount of loss allowed to the selling group. Thus, Treasury and the Service believed that in cases where a selling group elected to apply -20 Lite or Prior Reg. § 1.337(d)-2, it was appropriate to permit an acquiring group to amend prior waivers of loss carryovers. On May 7, 2003, Treasury and the Service amended the temporary regulations to provide for the amendment of prior waivers. See 68 Fed. Reg. at 24,352. This amendment was also reflected in the final regulations.

(2) Amendment of Prior Waivers

(a) If a selling group elected to apply -20T Lite or Prior Reg. § 1.337(d)-2, which had the effect of increasing the loss allowed on the disposition of the subsidiary stock, then the acquiring group could reduce the amount of any loss carryover deemed to expire (or increase the amount of any loss carryover deemed not to expire) as a result of Reg. § 1.1502- 32(b)(4). Reg. § 1.1502-32(b)(4)(vii)(A).

(b) The aggregate amount of loss carryovers that could be treated as not expiring as a result of such an amendment was limited to the amount of the duplicated loss with respect to the subsidiary’s stock. This limitation was intended to ensure that the loss carryovers subject to the amendment did, in fact, increase the amount of the allowed loss. 68 Fed. Reg. at 24,352.

(c) In addition, to enable the acquiring group’s use of loss carryovers that were not deemed to expire as a result of an amendment made under this provision, the regulations permitted a selling group to reapportion separate, subgroup, and consolidated section 382 limitations.

(d) If a loss previously deemed to expire was deemed not to expire as a result of an election under this provision, but the year to which such loss would have been carried was closed, then to the extent that the absorption of such loss would have affected the tax treatment of another item that has an effect in an open year, then the amendment of the waiver under Reg. § 1.1502-32(b)(4) will affect the treatment of

- 89 -

such other item. Prior Reg. § 1.1502- 20(i)(3)(v)(D).

(3) Avoiding Inadvertent Waivers – If the acquiring group made its original waiver election by identifying those losses that were deemed not to expire, it could have inadvertently waived those losses that were reattributed to the selling group but reverted to the subsidiary as a result of an election to apply -20 Lite or Prior Reg. § 1.337(d)-2. In such cases, the regulations permitted the acquiring group to amend its waiver election to provide that the additional losses were deemed not to expire. Reg. § 1.1502- 32(b)(4)(vii)(B).

6. Determining Whether and Which Election to Make – A parent’s choice of regulatory framework for the disposition of subsidiary stock could significantly affect the consequences of the disposition. For example, Prior Reg. § 1.1502-20 adopted certain presumptions that could be easier to prove than the tracing approach under Prior Reg. § 1.337(d)-2. On the other hand, Prior Reg. § 1.337(d)-2 did not provide for loss reattribution, whereas Prior Reg. § 1.1502-20 did.

a. Circumstances where taxpayer may want to apply Prior Reg. § 1.1502-20 in its entirety:

(1) The seller reattributed net operating losses, and a capital loss would not reduce the seller’s taxes.

(2) The seller reattributed net operating losses and had PIAs that it could not prove were not attributable to built-in gain.

(3) The seller benefited from netting positive and negative adjustments during the year and could not prove that they are not attributable to built-in gain.24

(4) The seller’s loss was not attributable to duplicated loss, and the seller would prefer reattributing losses over Prior Reg. § 1.337(d)-2.

(5) The failure to file a timely election under Prior Reg. § 1.1502-20(i) (Prior Reg. § 1.1502-20 in its entirety is the default rule).

24 If the taxpayer fell in this category, it could be possible to achieve the same result under the basis disconformity approach, since the PIA amount permitted netting of positive and negative adjustments.

- 90 -

b. Circumstances where taxpayer may want to elect to apply -20 Lite:

(1) The seller’s loss was attributable to duplicated loss, and a capital loss would reduce the seller’s taxes.

(2) The seller’s loss was attributable to duplicated loss and it reattributed net operating losses. Even though the seller could use the capital loss, it is paid by the buyer to make the election.

(3) The seller’s loss was attributable to duplicated loss, but the seller benefited from netting positive and negative investment adjustments within a year.25

(4) The seller was eligible to claim some of its loss, and it reattributed some net operating losses. However, it was uncertain whether it had duplicated losses.

c. Circumstances where taxpayer may want to elect to apply Prior Reg. § 1.337(d)-2:

(1) The seller had PIAs or extraordinary gain, which it could prove were not attributable to built-in gain, and the seller did not reattribute net operating losses but could use a capital loss.

(2) The seller had PIAs or extraordinary gain, which it could prove were not attributable to built-in gain, and although the seller reattributed net operating losses, the buyer would pay the seller to make the election.

(3) The seller had no loss disallowance factors, but failed to file the statement of allowed loss required by Prior Reg. § 1.1502-20(c)(3) and could not or did not want to seek 9100 relief.

IV. OLD LOSS DISALLOWANCE RULES – ADDRESSING CONCERNS RELATING TO LOSS DUPLICATION

A. Background

1. On March 7, 2002, at the same time it issued Prior Temp. Reg. § 1.337(d)- 2T, the Service issued Notice 2002-18 announcing its intention to issue

25 See supra note 18.

- 91 -

regulations addressing loss duplication concerns. See Examples 6 & 7, above for an illustration of the loss duplication concerns.

2. On October 23, 2002, Treasury and the Service issued proposed regulations to implement Notice 2002-18. Prior Prop. Reg. § 1.1502-35. On March 14, 2003, Treasury and the Service issued the proposed regulations in temporary form without significant modification. The regulations generally applied retroactively to transactions that occurred on or after March 7, 2002, the date of Notice 2002-18 (but only if such transactions occurred during a taxable year the original return for which was due after March 14, 2003, see Code § 1503(a); Prior Temp. Reg. § 1.1502-35T(i)). These regulations were made final without significant modification on March 9, 2006.

3. Treasury and the Service made it clear in the preamble to the temporary regulations that they were continuing to study the comments they received and specifically requested comments on alternative regimes that they were considering. Treasury and the Service also stated in the preamble to the final regulations on March 9, 2006 that they “intend to publish proposed regulations ‘in the near term’ addressing both circumvention of General Utilities repeal and loss duplication in a single integrated regulation.”

a. Unified Loss Rules – As described above, Treasury and the Service issued an integrated regulation addressing both the circumvention of the General Utilities repeal and loss duplication. Reg. § 1.1502-36.

b. Treasury and the Service believed that a subsidiary’s use of a group loss in a separate return year after the group has recognized the benefit of the loss distorts the subsidiary’s separate year income. However, to preserve the result in Rite Aid, stock loss may not be disallowed in deconsolidating transfers. Thus, the Unified Loss Rules permit the stock loss but reduce the subsidiary’s attributes to the extent of any duplicated loss. 72 Fed. Reg. at 2975.

4. Consistent with Notice 2002-18, Treasury and the Service identified the purpose of the temporary regulations as being the prevention of a consolidated group from obtaining more than one tax benefit from a single economic loss in a manner that does not permanently disallow the economic loss once. Prior Reg. § 1.1502-35T(a), (c)(8). However, the regulations are not limited to stuffing transactions illustrated by Examples 6 and 7, above.

5. Treasury and the Service believed that the basis redetermination and loss suspension rules in the regulations, which are discussed in detail below,

- 92 -

would not apply frequently. These rules only applied when a member sold less than all of the stock of a subsidiary member to a nonmember.

a. The government’s belief was based on the assumption that when a group seeks to raise capital, the parent or the subsidiary will typically issue stock directly, or the parent will sell all of the stock of the subsidiary member. 67 Fed. Reg. 65,060, 65,064 (Oct. 23, 2002).

b. If a loss were not subject to the loss duplication rules of Prior Reg. § 1.1502-35, the Service and Treasury would apparently rely on Reg. § 1.1502-32(e) and Charles Ilfeld Co. v. Hernandez, 292 U.S. 62 (1934), to disallow a duplicative loss. See I.L.M. 200423027 (May 17, 2004).

B. Basis Redetermination Rule

1. Effect of Investment Adjustment Rules

a. The investment adjustment rules of Reg. § 1.1502-32 are based on certain assumptions regarding shareholders’ interests in the subsidiary. One assumption is that each share within a class is entitled to an equal portion of the subsidiary’s items of income and gain. Another assumption is that the subsidiary’s losses are borne by the holders of the common stock before the holders of the preferred stock.

b. The preamble to the proposed loss duplication regulations stated that these assumptions result in an allocation of basis adjustments without regard to differences in members’ bases in their shares of stock of the subsidiary and without regard to whether a basis adjustment reflects a built-in item with respect to contributed property. Treasury and the Service thus felt that a basis redetermination rule was necessary to revise certain basis adjustments in an effort to mitigate the effect of the assumptions. 67 Fed. Reg. at 65,062.

c. The basis redetermination rule under Prior Reg. § 1.1502-35 applied differently depending on whether the subsidiary remained a member of the consolidated group.

2. Basis Redetermination Where Subsidiary Remained Member of the Group

a. If a member transferred a share of stock of a subsidiary member that had a basis in excess of its value (i.e., a loss share), and immediately after the transfer, the subsidiary remained a member of the group, then the basis of each share of the subsidiary stock

- 93 -

held by each member of the group was redetermined immediately before such transfer as follows (Prior Reg. § 1.1502-35(b)(1)):26

(1) First, the basis of all of the members of the group in the subsidiary member’s stock were aggregated.

(2) Second, the aggregated basis was first allocated to the subsidiary member’s preferred stock held by members of the group, in proportion to, but not in excess of, the value of those shares on the date of the transfer.

(3) Third, any remaining basis was allocated among all of the common shares of subsidiary member stock held by members of the group in proportion to the value of such shares on the date of the transfer.

b. The effect of the basis redetermination rule was to eliminate gain or loss on preferred shares and equalize gain or loss on each common share. Note that even though these regulations were promulgated in response to the government’s concern about duplicating economic losses, the basis redetermination rule was not limited to situations where duplicated losses exist.

c. It is not clear how the basis redetermination rules applied in the situation where there was an aggregate excess loss account (“ELA”) in the subsidiary stock. There appear to be two ways to interpret the basis redetermination rule. See Example 31, below.

(1) First, since an ELA is treated as negative basis for all federal income tax purposes, Reg. § 1.1502-19(a)(2)(ii), the preferred stock could take a proportionate share of the ELA. Arguably, because the aggregate ELA will always be less than the fair market value of the preferred stock, the entire ELA would be allocated to the preferred stock, and the common stock would take a zero basis. The entire ELA would then be triggered upon the sale of the preferred stock.

26 The final regulations, which adopted the temporary regulations without significant modification, included a somewhat simpler approach than the proposed regulations and triggered the basis redetermination rule upon the “transfer” of loss stock. Under the proposed regulations, the basis redetermination rule was triggered upon the “disposition” or “deconsolidation” of any share of loss stock. See Prop. Reg. § 1.1502-35(b)(1), (d)(1) & (d)(2). The reference to deconsolidation of a share could be easily confused with deconsolidation of a subsidiary.

- 94 -

(2) Second, one could argue that the basis redetermination rule was intended only to reallocate positive basis. Under this interpretation, only the aggregate positive basis would be reallocated. This would have the effect of reducing the basis of the preferred stock to its fair market value and reducing the ELA in the common stock.

(3) Triggering gain on the sale of the preferred stock seems like the wrong answer. Thus, the second approach seems to be the preferable one. Nonetheless, the literal language of the regulations seems to favor the first interpretation.

d. Exceptions – The basis redetermination rule of Prior Reg. § 1.1502-35(b)(1) did not apply to a transfer of subsidiary member stock if (Prior Reg. § 1.1502-35(b)(3)(i)):

(1) During the taxable year of such transfer, in one or more fully taxable transactions, the members of the group disposed of all of the shares of the subsidiary member stock to a nonmember;

(2) During the taxable year of such transfer, members of the group were allowed a worthless stock deduction under section 165(g) with respect to all of the shares of the subsidiary member stock (other than the shares that would otherwise trigger the application of Prior Reg. § 1.1502- 35(b)(1)); or

(3) Such transfer was to a member of the group, and section 332, 351, or 361 applies to such transfer.

3. Basis Redetermination Where Subsidiary Was Deconsolidated

a. Where a subsidiary was deconsolidated as a result of the transfer of subsidiary shares, the basis redetermination was more limited.

b. If, immediately before a deconsolidation of a subsidiary member, any share of stock of a subsidiary member owned by a member had a basis in excess of its value (i.e., a loss share), then the basis of each share of the subsidiary stock held by each member of the group was redetermined to the extent of the “reallocable basis amount” immediately before the deconsolidation. Prior Reg. § 1.1502-35(b)(2).

c. The reallocable basis amount was equal to the lesser of:

- 95 -

(1) The aggregate of the loss in the subsidiary’s loss shares held by members immediately before the deconsolidation; and

(2) The total of the subsidiary’s items of deduction and loss, and the subsidiary’s allocable share of items of deduction and loss of lower tier subsidiary members, that were taken into account in computing basis adjustments under Reg. § 1.1502-32 allocable to non-loss shares held by members immediately before the deconsolidation.

(3) The regulations thus presumed that items allocated to non- loss shares resulted in a duplicated loss and therefore tainted only those shares. d. The basis of the subsidiary’s shares held by members of the group were adjusted immediately before the deconsolidation as follows:

(1) First, the basis of every loss share held by members of the group was reduced, but not below its fair market value, by the reallocable basis amount in a manner that caused the ratio of the basis to the value of each such share to be the same.

(2) Second, the basis of any preferred shares of the subsidiary held by members of the group was increased, but not above its fair market value, by the reallocable basis amount in a manner that caused the ratio of the basis to the value of each such share to be the same.

(3) Third, any remaining reallocable basis amount increased the basis of all common shares of the subsidiary held by members of the group in a manner that caused the ratio of the basis to the value of each such share to be the same.

(4) Note that the problem identified above regarding reallocation of basis when there is an aggregate ELA did not appear to be present in the context of a deconsolidated subsidiary. Because the basis of a loss share could not be reduced below its fair market value, it would seem to preclude allocation of a proportionate amount of an ELA. e. Exceptions – Under Prior Reg. § 1.1502-35(b)(3)(ii), the basis redetermination rule of Prior Reg. § 1.1502-35(b)(2) did not apply to a deconsolidation of a subsidiary member if:

- 96 -

(1) During the taxable year of such deconsolidation, in one or more fully taxable transactions, the members of the group disposed of all of the shares of the subsidiary member stock to a nonmember;

(2) Such deconsolidation resulted from a fully taxable disposition of some of the shares of the subsidiary member to a nonmember, and during the taxable year of such deconsolidation, members of the group were allowed a worthless stock deduction under section 165(g) with respect to all of the shares of the subsidiary member stock that they own immediately after the deconsolidation; or

(3) The deconsolidation of the subsidiary member resulted from the deconsolidation of a higher tier member and, immediately after the deconsolidation of the subsidiary member, none of the stock of the subsidiary member was owned by a group member.

4. Lower Tier Subsidiaries – If, immediately after the transfer or deconsolidation of a subsidiary member, a lower tier subsidiary member, some of the stock of which was owned by the subsidiary member, was a member of the group, then for purposes of applying the basis redetermination rules, the subsidiary member was treated as having transferred its stock of the lower tier member. Prior Reg. § 1.1502- 35(b)(4).

5. Basis Adjustments for Higher Tier Stock – The basis adjustments made as a result of the basis redetermination rule resulted in basis adjustments to higher tier member stock. The adjustment was made from the lowest tier to the highest. Prior Reg. § 1.1502-35(b)(5).

6. Ordering Rules – The investment adjustment rules of Reg. § 1.1502-32 applied first; then the basis redetermination rules of Prior Reg. § 1.1502- 35(b) applied (from lowest tier to highest tier, if applicable); then the loss disallowance rules of Prior Reg. § 1.337(d)-2 applied. Prior Reg. § 1.1502-35(b)(6).

7. As discussed in section II above, the final Unified Loss Rules adopt a much more limited basis redetermination rule. Reg. § 1.1502-36(b). Unlike the basis redetermination rule of Prior Reg. § 1.1502-35, which fully blended basis, the final basis redetermination rule in Reg. § 1.1502- 36(b) only reallocates investment adjustments previously made to stock basis. In addition, it only reduces the basis of loss shares that were transferred, thus allowing the other shares to benefit fully from future appreciation. 72 Fed. Reg. at 2978-79.

- 97 -

C. Loss Suspension Rule

1. General Rule – If, after application of the basis redetermination rule, a member of a consolidated group recognized a loss on the disposition of a share of a subsidiary member, then such loss was suspended to the extent of the duplicated loss with respect to such share of stock. Prior Reg. § 1.1502-35(c)(1). The loss suspension rule applied only if, immediately after the disposition of such share, the subsidiary remained a member of the consolidated group. Id.

2. Duplicated Loss – Duplicated loss was determined immediately after a disposition and equaled the excess, if any, of (Prior Reg. § 1.1502- 35(d)(4)(i)):

a. The sum of:

(1) The aggregate adjusted basis of the subsidiary member’s assets, other than stock that a subsidiary member owned in another subsidiary member, and

(2) Any losses attributable to the subsidiary member and carried to the subsidiary member’s first taxable year following the disposition, and

(3) Any deductions of the subsidiary member that had been recognized but were deferred under a provision of the Code; over

b. The sum of:

(1) The value of the subsidiary member’s stock, and

(2) Any liabilities of the subsidiary member that had been taken into account for tax purposes.

c. The amounts computed in the loss duplication formula included the subsidiary member’s share of corresponding amounts with respect to lower-tier subsidiaries. Prior Reg. § 1.1502- 35(d)(4)(ii)(A).

d. The duplicated loss formula was substantially identical to the one contained in Prior Reg. § 1.1502-20(c)(2)(vi), except that securities of other members of the group were not excluded from the computation of the subsidiary’s aggregate asset basis.

- 98 -

e. Similarly, the Unified Loss Rules do not exclude securities of other members of the group form the computation of net inside asset basis.

3. Lower Tier Subsidiaries – A special rule applied if a loss was recognized on the disposition of a share of stock of a subsidiary member, but the loss suspension rule otherwise would not apply because the subsidiary member left the group. In that case, if the departing subsidiary member owned stock of a lower-tier subsidiary member that remained a member of the group after the disposition, then the loss was suspended to the extent the duplicated loss of the departing member was attributable to the remaining member. Prior Reg. § 1.1502-35(c)(2).

4. Treatment of Suspended Loss – A suspended loss was treated as a noncapital, nondeductible expense of the member that disposed of subsidiary member stock incurred during the taxable year that included the date of the disposition of stock for purposes of Reg. § 1.1502-32. As a result, the basis of a higher tier member’s stock was reduced by the suspended loss in the year it was suspended. Prior Reg. § 1.1502-35(c)(3).

5. Reduction of Suspended Loss

a. The amount of suspended loss was reduced as the subsidiary member subsequently recognized the duplicate deduction and loss items. Specifically, the suspended loss was reduced, but not below zero, by the subsidiary member’s items of deduction and loss, and the subsidiary member’s allocable share of items of deduction and loss of lower tier members, that were taken into account in determining consolidated taxable income and were allocable to the period beginning on the date of the disposition that gave rise to the suspended loss and ending on the day the subsidiary ceases to be a member of the consolidated group. Prior Reg. § 1.1502- 35(c)(4)(i).

(1) The prior regulations presumed that all deductions and losses were attributable to the duplicated loss that gave rise to the suspended loss. However, the presumption was rebuttable. If the taxpayer could establish that the item of deduction or loss was not part of the duplicated loss, then the taxpayer would not have to reduce its suspended loss. Id.

(2) A suspended loss was also reduced by items of deduction and loss of any successor to the subsidiary member. See Prior Reg. § 1.1502-35(c)(4). For this purpose, a successor was defined as a transferee of assets in a transaction (i) to

- 99 -

which section 381(a) applied, (ii) in which substantially all of the assets of the transferor were transferred to members in a complete liquidation, (iii) in which the successor’s basis in assets was determined (directly or indirectly, in whole or in part) by reference to the transferor’s basis in such assets, or (iv) which was an intercompany transaction, but only with respect to assets that were being accounted for by the transferor in a prior intercompany transaction. Prior Reg. § 1.1502-35(d)(5).

b. The prior temporary regulations also added a limitation on the reduction of a suspended loss that was not contained in the proposed regulations. This limitation was included in the prior final regulations as well.

(1) The amount of the reduction could not exceed the excess of the amount of the subsidiary member’s items of deduction and loss over the amount of such items that were taken into account in determining the basis adjustments made to the subsidiary member’s stock under Reg. § 1.1502-32. Id.

(2) The reason for this limitation was to prevent the disallowance of a tax loss for an economic loss. 68 Fed. Reg. at 12,288. To address this concern, a general statement was also added to clarify that the loss suspension rule is not to be applied in a manner that permanently disallows an otherwise allowable deduction for an economic loss and permitting a “proper adjustment” in such cases. Prior Reg. § 1.1502-35(c)(8).

6. Allowance of Loss

a. To the extent not reduced, a suspended loss was allowed as a deduction to the group when the subsidiary member (or any successor) left the group or a worthless stock loss under section 165(g) was taken with respect to all of the subsidiary member stock owned by members. Prior Reg. § 1.1502-35(c)(5)(i)(A).

b. The Unified Loss Rules amended Prior Reg. § 1.1502-35(c)(5) to provide that, in any event, a suspended loss would be allowed after 10 years. Prior Reg. § 1.1502-35(c)(5)(i)(B).

c. However, no adjustments could be made to the basis of the subsidiary member’s stock under Reg. § 1.1502-32 for a suspended loss that was taken into account. Prior Reg. § 1.1502-35(c)(5)(ii).

- 100 -

Such basis adjustments would already have been made at the time the loss was suspended. See Prior Reg. § 1.1502-35(c)(3).

d. The suspended loss was allowed only if the taxpayer filed a statement of allowed loss with its tax return. Prior Reg. § 1.1502- 35(c)(5)(iii).

7. Special Rule for Successor Assets – If a member acquires an asset and the basis of such asset was determined, directly or indirectly, in whole or in part, by reference to the basis of stock of a subsidiary member, and at the time of the acquisition there was duplicated loss in the stock of the subsidiary member, then any loss recognized on the disposition of such asset was suspended. Prior Reg. § 1.1502-35(c)(6)(i), (ii). This rule did not apply if the subsidiary member was not a member of the group immediately after the disposition of the asset. Prior Reg. § 1.1502- 35(c)(6)(iii).

8. Coordination With Other Deferral or Disallowance Rules

a. The loss suspension rules did not apply to a loss that was disallowed under any other provision of the Code or regulations. Prior Reg. § 1.1502-35(c)(7)(i).

b. If a loss was deferred under another provision, the loss suspension rules applied when the loss would otherwise be taken into account under such other provision. However, if an overriding event occurred before the deferred loss is taken into account, then the loss suspension rules applied immediately before the event occurred. Id. An overriding event occurred if the stock ceased to be owned by a member of the consolidated group, was canceled or redeemed, or was treated as disposed of under Reg. § 1.1502- 19(c)(1)(ii)(B) (subsidiary became a nonmember) or (c)(1)(iii) (worthlessness). Prior Reg. § 1.1502-35(c)(6)(ii).

9. Ordering Rules – The loss suspension rules applied only after the investment adjustment rules of Reg. § 1.1502-32, the basis redetermination rules of Prior Reg. § 1.1502-35(b), and the loss disallowance rules of Prior Reg. § 1.337(d)-2 applied. Prior Reg. § 1.1502-35(c)(9).

D. Worthlessness and Dispositions Not Followed by Separate Return Years

1. General Rule – Under Prior Reg. § 1.1502-35(f), if stock of a subsidiary member was treated as worthless under section 165 (taking into account Reg. § 1.1502-80(c)), or if a member of a group disposed of subsidiary member stock and, on the following day, the subsidiary was not a member

- 101 - of the group and did not have a separate return year (e.g., dissolution of an insolvent subsidiary to which section 332 does not apply), then: a. All losses treated as attributable to the subsidiary under Reg. § 1.1502-21(b)(2)(iv) were taken into account in computing the taxable income of the group, the subsidiary, and any carryback group of which the subsidiary was previously a member for the taxable year that includes the determination of worthlessness or the disposition and any prior taxable year. b. Any remaining losses not utilized were treated as expired, but not absorbed by the group as of the beginning of the group’s taxable year that includes the determination of worthlessness or the disposition. Thus, the losses deemed expired did not reduce basis under Reg. § 1.1502-32(b)(3)(iii), so a worthless stock deduction was available with respect to the remaining basis. c. Taxpayers expressed a concern that Prior Temp. Reg. § 1.1502- 35T(f) could eliminate a subsidiary’s losses even if the subsidiary has a separate return year following the year the group claims the worthless stock deduction. The Service and Treasury amended Prior Temp. Reg. § 1.1502-35T(f) to provide that the subsidiary’s losses were treated as expired only if a member claims a worthless stock deduction and the subsidiary was a member of a group that includes the member claiming the worthless stock deduction. Prior Temp. Reg. § 1.1502-35T(f)(1). This amendment was included in the prior final regulations. d. Example 22 – Worthless Stock Deduction

P P

S shares (1) $50

S Bank (2) $100 Bank Assets S $70 Value $70 Basis

(1) Facts: In Year 1, P forms S by transferring $50 in exchange for all of the S stock. S borrows $100 from

- 102 -

Bank. S loses $80, which is not utilized by the group. In Year 2, S dissolves and transfers its assets (with a $70 value and basis) to Bank.

(2) S has $30 of cancellation of indebtedness income and reduces its $80 NOL by $30. As a result, there is no net adjustment to P’s basis in its S stock.

(3) Because S has dissolved, P may claim a worthless stock deduction with respect to the S stock. S’s remaining NOL disappears under Prior Reg. § 1.1502-35(f).

(4) What if S remains in existence and simply ceases doing business?

(a) P would be precluded from claiming a worthless stock deduction because S has not yet disposed of substantially all of its assets. See Reg. § 1.1502- 80(c).

(b) Reg. § 1.1502-80(c) was promulgated in part to prevent the duplicated loss rule of Prior Reg. § 1.1502-20(c)(2)(vi) from eliminating the benefit of the worthless stock deduction. Given the elimination of the Prior Reg. § 1.1502-20(c) loss duplication rule, taxpayers have questioned whether Reg. § 1.1502-80(c) remains necessary. In the preamble to amendments to Reg. § 1.1502-80(c), the Service and Treasury indicated that they are evaluating this issue. 69 Fed. Reg. at 12,800.

2. Proposed Regulations – This rule differed from the proposed regulations, which required the reduction of the basis of the subsidiary stock by the amount of any loss carryforwards attributable to the subsidiary under Reg. § 1.1502-21.

a. The reason for the rule was to prevent taxpayers from taking the position that a group is entitled to a subsidiary member’s loss carryforwards even after the group has enjoyed full basis recovery through a worthless stock or other deduction.

b. Commentators contended that the basis reduction rule could deny a group a single tax loss for its economic loss. The revision made in the temporary regulations (and retained in the final regulations) was intended to address this. See 68 Fed. Reg. at 12,288.

- 103 -

3. Special Transition Election in Prior Temporary Regulations

a. Because of the change in the rule, the temporary regulations provided a special transition rule. If stock of a subsidiary member was treated as worthless between March 7, 2002 and March 14, 2003, or if a member of the group disposed of a subsidiary member during this period and, on the following day, the subsidiary was not a member of the group and did not have a separate return year, then the common parent may make an irrevocable election to reattribute to itself all or any portion of the losses treated as attributable to the subsidiary member under Reg. § 1.1502- 21(b)(2)(iv). Prior Temp. Reg. § 1.1502-35T(f)(2).

b. The reattributed losses were treated as absorbed by the group immediately before the allowance of any loss or inclusion of any income or gain with respect to the determination of worthlessness or the disposition. Prior Temp. Reg. § 1.1502-35T(f)(2).

4. The Unified Loss Rules treat worthlessness under Reg. § 1.1502-80(c) as a transfer of a loss share. Accordingly, the general rules of the Unified Loss Rules apply.

a. The basis redetermination rule does not apply if all of the subsidiary’s shares held by members become worthless under Reg. § 1.1502-80(c) in one taxable transaction. Reg. § 1.1502- 36(b)(1)(ii)(B).

b. The basis reduction rule applies normally; there are no special rules regarding worthless stock deductions. See Reg. § 1.1502- 36(c).

c. The attribute reduction rule contains a special worthless stock rule. If any attributes remain after application of the attribute reduction rule, they are eliminated if:

(1) A member transfers a share of subsidiary stock solely by reason of its worthlessness and the provisions of Reg. § 1.1502-80(c) are satisfied; the member recognizes a net deduction or loss; and the subsidiary continues to be a member of the group; or

(2) The member recognizes a net deduction or loss in a transaction in which the subsidiary ceases to be a member and does not become a nonmember (i.e., does not have a separate return year).

- 104 -

E. Anti-Avoidance Rules

1. Transfer of Share Without Loss in Avoidance – If a non-loss share of subsidiary member stock was transferred with a view to avoiding the application of the basis redetermination rules prior to the transfer of loss stock, or a deconsolidation, of such subsidiary, then the basis redetermination rule applied immediately prior to the transfer of the non- loss stock. Prior Reg. § 1.1502-35(g)(1).

2. Transfer of Loss Property in Avoidance

a. If a member of a consolidated group contributed a built-in loss asset to a partnership in a section 721 transaction or to a nonmember in a section 351 transaction, and such partnership or corporation contributed such asset to a subsidiary member in a section 351 transaction, and such contributions were undertaken with a view to avoiding the basis redetermination or loss suspension rule, then adjustments had to be made to carry out the purposes of the regulations. Prior Reg. § 1.1502-35(g)(2).

b. Example 23 – Transfer of Property to Avoid Basis Redetermination Rule

(6) PS Interest P P P P Y (3) $40 (4) $80 PS Interest 100 $100 (2) shares 20 PS Interest Asset A shares $20 Value S stock $50 Basis S S S PS PS PS S (5) $20 Asset A (4) Asset A X

(1) Facts: In Year 1, P forms S by transferring $100 in exchange for 100 shares of S stock, which is all of the outstanding stock of S. In Year 2, P contributes the 20 shares of S common stock to a partnership, PS, in exchange for a 20-percent partnership interest. S remains a member of the P group. In Year 3, P transfers Asset A, with a value of $20 and a basis of $50, to PS in exchange for an additional partnership interest. Also in Year 3, PS

- 105 -

contributes Asset A to S, and P contributes an additional $80 to S. In Year 4, S sells Asset A to X for $20, recognizing a loss of $30, and P sells its interest in PS to Y for $40, recognizing a loss of $30.

(2) If P’s contributions of S stock and Asset A to PS were undertaken with a view to avoiding the basis redetermination rule or the loss suspension rule, then adjustments must be made such that the group does not obtain more than one tax benefit from the $30 loss inherent in Asset A. Prior Reg. § 1.1502-35(g)(5), Ex. 1.

3. Anti-Loss Reimportation

a. If the consolidated group was allowed a loss from the sale of subsidiary stock, and the subsidiary was deconsolidated, the subsidiary’s duplicate inside loss may not be reimported within 10 years of the deconsolidation. Prior Reg. § 1.1502-35(g)(3)(i).

b. Loss reimportation could occur in a number of ways:

(1) The subsidiary member (or any successor) re-joined the consolidated group while it still owned the loss asset or any asset whose basis reflects the basis of the loss asset. Prior Reg. § 1.1502-35(g)(3)(i)(B)(1), (2).

(2) A member of the consolidated group acquired the loss asset, or any asset whose basis reflected the basis of the loss asset, from the subsidiary member (or any successor) in a section 381 or 351 transaction. Prior Reg. § 1.1502- 35(g)(3)(i)(B)(3).

(3) The subsidiary member (or any successor) re-joined the consolidated group while it had a liability that it had on the date of the disposition and such liability would give rise to a deduction. Prior Reg. § 1.1502-35(g)(3)(i)(B)(4).

(4) A member of the consolidated group assumed a liability that was a liability of the subsidiary member (or any successor) on the date of the disposition in a section 381 or 351 transaction. Prior Reg. § 1.1502-35(g)(3)(i)(B)(5).

(5) The subsidiary member (or any successor) re-joined the consolidated group while it had losses or deferred deductions that (i) it had on the date of the disposition, (ii) were attributable to an asset owned on the date of the disposition or an asset whose basis is reflected in the basis

- 106 -

of such asset, or (iii) were attributable to a liability (within the meaning of section 358(h)(3)) that it had on the date of the disposition. Prior Reg. § 1.1502-35(g)(3)(i)(B)(6)-(9).

(a) For this purpose, any losses and deductions or assets of the subsidiary were presumed to have existed on the date of the disposition. However, such presumption was rebuttable. Prior Reg. § 1.1502-35(g)(3)(ii)(B), (C).

(6) A member of the consolidated group succeeded to any losses or deferred deductions described in (5). Prior Reg. § 1.1502-35(g)(3)(i)(B)(10)

(7) Any losses or deferred deductions described in (5) were carried back to a pre-disposition taxable year of the subsidiary. Prior Reg. § 1.1502-35(g)(3)(i)(B)(11). c. If a loss was reimported, then the group is denied the use of:

(1) Any loss recognized that was attributable to a built-in loss asset, or any asset whose basis reflected the basis of the loss asset, that was owned by the subsidiary (or any successor) on the date of the disposition to the extent of the lesser of (i) the loss inherent in such asset on the date of the disposition, or (ii) the loss inherent in such asset on the date of the reimportation. Prior Reg. § 1.1502-35(g)(3)(iii)(A), (B).

(2) Any loss or deduction described in paragraphs (3) through (7), above. Prior Reg. § 1.1502-35(g)(3)(iii)(C), (D). However a loss or deduction described in paragraph (6), above, could be carried forward to a post-disposition taxable year of the subsidiary. Prior Reg. § 1.1502- 35(g)(3)(iii)(D). d. A loss or deduction that was disallowed under the anti-loss reimportation rule was treated as a noncapital, nondeductible expense incurred during the taxable year that such loss would otherwise be absorbed for purposes of Reg. § 1.1502-32(b)(3)(iii) and, thus, resulted in a downward basis adjustment. Prior Reg. § 1.1502-35(g)(3)(iv). e. The effective date of the anti-loss reimportation rule was different from the effective date of the prior regulation as a whole. The anti- loss reimportation rule applied to losses reimported as a result of an event that occurred on or after October 18, 2002.

- 107 -

f. Example 24 – Loss Reimportation

(2) (5) S Stock M/S Stock P X X P X $240 $300 (1) Asset A S Stock $100 Value (3) (5) (4) $120 Basis Asset A Asset D Asset B Merge S $50 Value M S Y M/S Z $70 Basis $100 $20 Asset C $10 NOL $90 Value Asset C $100 Basis $80 Value $100 Basis Asset D $60 Value $70 Basis

(1) Facts: In Year 1, P forms S by transferring Asset A, with a value of $100 and a basis of $120, Asset B, with a value of $50 and a basis of $70, and Asset C, with a value of $90 and a basis of $100, in exchange for 100 shares of S1 common stock. In Year 2, P sells the stock of S to X for $240, recognizing a $50 loss. In Year 3, S sells Asset A to Y, recognizing a $20 loss. Also in Year 3, S merges into M in a section 368(a)(1)(A) reorganization.

In Year 8, P purchases all of the stock of M for $300. At that time, M has a $10 NOL. In addition, M owns Asset D, which was acquired in exchange for Asset B in a section 1031 exchange. Asset C has a value of $80 and a basis of $100, and Asset D has a value of $60 and a basis of $70. In Year 9, P has operating income of $50, and M recognizes $20 loss on the sale of Asset C. In Year 10, P has operating income of $50, and M recognizes a $50 loss on the sale of Asset D.

(2) P’s $50 loss was attributable to a duplicated loss and, thus, may not be reimported for 10 years. M is a successor to S, and its $10 NOL is presumed to be attributable to assets owned by S on the date of P’s disposition of S. Provided that P cannot rebut the presumption, the P group will not be

- 108 -

able to use M’s $10 NOL. Such loss will, however, result in a reduction in P’s basis in its M stock during the taxable year that it would otherwise be absorbed (i.e., year 9). Reg. § 1.1502-32(b)(3)(iii)(D).

(3) In addition, the P group will be denied $10 of the loss recognized on the sale of Asset C (i.e., the lesser of the $10 built-in loss on the date of P’s disposition of S and the $20 built-in loss on the date of the reimportation). The P group will also be denied $10 of the loss recognized on the sale of Asset D (i.e., the lesser of the $20 built-in loss in Asset B, the predecessor to Asset D, on the date of P’s disposition of S and the $10 built-in loss on the date of the reimportation). Each disallowed loss will result in a reduction in P’s basis in its M stock during the taxable year that includes the date of the disposition of the asset with respect to which the loss was recognized. Prior Reg. § 1.1502-35(g)(5), Ex. 2.

4. Revised Anti-Loss Reimportation Rule

a. On April 10, 2007, the Service and Treasury released temporary regulations that revised the anti-loss reimportation rule that applied following a disposition of a stock of a subsidiary at a loss for corporations filing consolidated returns. 72 Fed. Reg. 17804 (Apr. 10, 2007).

b. The temporary regulations revised Prior Reg. § 1.1502-35(g)(3) to clarify that losses reflected in the basis of subsidiary stock at the time of deconsolidation could not be recognized and reimported into the group, regardless of whether the stock losses were recognized when the subsidiary is a member of the group. 72 Fed. Reg. 17804, 17805 (Apr. 10, 2007). Specifically, under Prior Temp. Reg. § 1.1502-35T(g)(3), immediately before the time that a reimported item (or any portion of a reimported item) would be properly taken into account, such item (or such portion of the item) was reduced to zero and no deduction or loss was allowed, directly or indirectly, with respect to that item. See Prior Temp. Reg. § 1.1502-35T(g)(3)(ii).

c. The anti-loss reimportation rule was also revised to replace the list of events that caused the application of the rule with a list of criteria that identify reimportation transactions that would be treated as subject to the rule. 72 Fed. Reg. 17804 (Apr. 10, 2007). Specifically, Prior Temp Reg. § 1.1502-35T(g)(3) applied when:

- 109 -

(1) a member of a group (the selling group) recognized and was allowed a loss with respect to a share of stock of S, a subsidiary or former subsidiary of the selling group;

(2) the stock loss was duplicated (in whole or in part) in S’s attributes (duplicating items) at the earlier of the time that the loss was recognized or that S ceased to be a member;

(3) within ten years of the date that S ceased to be a member, there was a reimportation event (defined for this purpose as any event after which a duplicating item is a reimported item). A reimported item was any duplicating item that was reflected in the attributes of any member of the selling group, including S, or, if not reflected in the attributes, would be properly taken into account by any member of the selling group.

(4) The temporary regulations that revised the anti-loss reimportation rule applied to reimportation events that occurred on or after April 10, 2007 if they occurred with respect to stock of a subsidiary sold on or after March 7, 2002, or with respect to stock of a subsidiary or former subsidiary sold on or after April 10, 2007. See Prior Temp. Reg. § 1.1502-35T(g)(3)(i).

5. Avoidance of Gain Recognition

a. If a transaction was structured with a view to, and had the effect of, deferring or avoiding the recognition of gain on a disposition of stock by invoking the basis redetermination rule, and the stock loss that gave rise to the application of the basis redetermination rule was “not significant,” then the basis redetermination and loss suspension rules applied. Prior Reg. § 1.1502-35(g)(4). No definition was provided for the phrase “not significant.”

b. This anti-abuse rule was added in the prior temporary regulations in response to comments that the basis redetermination rule could be used to shift the location of gain and loss within the consolidated group in a manner unintended by the proposed regulations. See 68 Fed. Reg. at 12,289. The prior final regulations maintained the same anti-abuse rule.

- 110 -

c. Example 25 – Transfers to Avoid Gain Recognition

(2) S2 Stock P X

$500 $500 Value $400 Basis

S1 (1) S2 S3 Preferred 50% Common $200 Value 50% Common $100 Basis $150 Basis Preferred S3 S3 Common $9 Value $10 Basis

(1) Facts: P owns all of the stock of S1 and S2. The S2 stock has a value of $500 and a basis of $400. S1 owns 50 percent of the S3 common stock, with a basis of $150. S2 owns the remaining S3 common stock, with a value of $200 and a basis of $100 and one share of S3 preferred stock, with a value of $9 and a basis of $10. P, intending to sell the S2 stock without recognizing a substantial portion of the built-in gain, causes a recapitalization of S3 in which S2’s common stock in S3 is exchanged for new S3 preferred stock. P then sells the S2 stock.

(2) Because S2 owns stock of S3, which remains a member of the P group, S2 is deemed to have transferred the S3 stock, including the one share of built-in loss stock. As a result, the basis redetermination rule applies, and the aggregate basis of S3 stock is allocated first to the S3 preferred shares held by S2 up to their value of $209 and then to the S3 common share held by S1. Thus, S2’s basis in the S3 preferred stock is increased from $110 to $209. This tiers up and increases P’s basis in the S2 stock from $400 to $499. Accordingly, P will recognize gain of only $1 on the sale of S2.

(3) However, because the recapitalization of S3 was structured with a view to, and has the effect of, avoiding the recognition of gain by invoking the basis redetermination rule, Prior Temp. Reg. § 1.1502-35T(g)(4) applies to turn

- 111 -

off the basis redetermination rule. Thus, P recognizes $100 gain on the disposition of S2 stock. Prior Reg. § 1.1502- 35(g)(5), Ex. 3.

(1) What if the stock loss giving rise to the application of the basis redetermination rule were not merely $1? How much loss is necessary before it becomes “significant”?

6. Other Anti-Abuse Rules – The rules of Prior Reg. § 1.1502-35 did not preclude the application of anti-abuse rules under other provisions of the Code and Regulations thereunder. Prior Reg. § 1.1502-35(h).

7. General Anti-Avoidance Rule

a. The prior temporary regulations that revised the anti-loss reimportation rule also added a general anti-avoidance rule under Prior Temp. Reg. § 1.1502-35T(g)(6), which provided that appropriate adjustments would be made if a taxpayer acted with a view to avoid the purposes of Prior Reg. § 1.1502-35. 72 Fed. Reg. 17804, 17805 (Apr. 10, 2007).

b. The general anti-avoidance rule under Prior Temp. Reg. § 1.1502- 35T(g)(6) applied on or after April 10, 2007. See Prior Temp. Reg. § 1.1502-35T(j)(2)(i).

- 112 -

V. EXAMPLES APPLYING THE UNIFIED LOSS RULES AND LOSS DISALLOWANCE AND DUPLICATION RULES

A. Basis Redetermination Examples

1. Example 26 – Basis Redetermination To Prevent Non-Economic Loss

(2) 1 Block 1 Share Asset 1 Asset 2 Block 1 Asset 2 P Z $80 Basis P $0 Basis 4 Shares P $0 Basis $80 Value $20 Value $20/sh Basis $20 Value $20

Block 1 Block 2

Asset 1 Block 1 Block 2 Asset 2 (1) 4 shares 1 share Asset 2 S X $20 S S Asset 1 Asset 2 $80 Basis $0 Basis $80 Value $20 Value Asset 1 $80 Basis $80 Value

a. Facts: P owns two assets, Asset 1 and Asset 2. On January 1, Year 1, P receives four shares of S common stock (the Block 1 shares) in exchange for Asset 1, which has a basis and value of $80. The exchange qualifies under section 351 and, therefore, under section 358, P’s aggregate basis in the Block 1 shares is $80 ($20 per share). On July 1, Year 1, P receives another share of S common stock (the Block 2 share) in exchange for Asset 2, which has a basis of $0 and value of $20. This exchange also qualifies as a section 351 exchange and, under section 358, P’s basis in the Block 2 share is $0. P’s Block 1 and Block 2 shares are the only outstanding shares of S stock. On October 1, Year 1, S sells Asset 2 for $20. On December 31, Year 1, P sells one of its Block 1 shares for $20.

b. P’s basis in each Block 1 share is $24 (P’s original $20 basis increased under Reg. § 1.1502-32 by $4, the share’s allocable portion of the $20 gain recognized on the sale of Asset 2). In addition, P’s basis in its Block 2 share is $4 (P’s original $0 basis increased under Reg. § 1.1502-32 by $4 (the share’s allocable portion of the $20 gain recognized on the sale of Asset 2)). P’s sale of the Block 1 share is a transfer of a loss share.

- 113 -

c. Application of Unified Loss Rules:

(1) P’s bases in all its shares of S stock are subject to redetermination. Under Reg. § 1.1502-36(b)(2)(i)(A), P’s basis in the transferred loss share is reduced, but not below value, by removing PIAs applied to the basis of the share. Accordingly, P’s basis in the transferred Block 1 share is reduced by $4 (the amount of the PIA applied to the share), from $24 to $20. No further reduction to the basis of the share is required because the basis of the share is then equal to value. Reg. § 1.1502-36(b)(3), Ex. 1(i).

(2) The PIA removed from the transferred loss share is reallocated and applied to increase P’s bases in its S shares in a manner that reduces basis disparity to the greatest extent possible. Reg. § 1.1502-36(b)(2)(ii)(B). Accordingly, the $4 PIA removed from the Block 1 share is reallocated and applied to the basis of the Block 2 share, increasing it from $4 to $8. Id.

(3) After the application of the basis redetermination rule, the Block 1 share is no longer a loss share, so the basis reduction rule of Reg. § 1.1502-36(c) and the attribute reduction rule of Reg. § 1.1502-36(d) do not apply. Id. d. Application of Prior Rules:

(1) Because the sale of the Block 1 share is the transfer of a loss share that does not result in a deconsolidation, the basis redetermination rule of Prior Reg. § 1.1502-35(b)(1) would apply.

(2) The group members’ aggregate bases of $100 is allocated in proportion to the fair market value of the common shares, or $20 per share. Thus, the sale results in no gain or loss.

- 114 -

2. Example 27 – Basis Redetermination To Prevent Duplicated Loss

(2) 1 Block 1 Share Asset 1 Asset 3 Block 1 Asset 3 P Z $80 Basis P $5 Basis 4 Shares P $5 Basis $5 $80 Value $5 Value $20/sh Basis $5 Value

Block 1 Block 2

Asset 1 Block 1 Block 2 Asset 3 (1) 4 shares 1 share Asset 1 S X $20 S S Asset 1 Asset 3 $80 Basis $5 Basis $20 Value $5 Value Asset 1 $80 Basis $20 Value

a. Facts: The facts are the same as above, except that, at the time of the second contribution, the value of Asset 1 had declined to $20 and so, instead of contributing Asset 2, P contributed Asset 3 to S in exchange for the Block 2 share. At the time of that exchange, Asset 3 had a basis and value of $5. On October 1, Year 1, S sells Asset 1 for $20, recognizing a $60 loss that is absorbed by the group. On December 31, Year 1, P sells one of its Block 1 shares for $5.

b. P’s basis in each Block 1 share is $8 (P’s original $20 basis decreased under Reg. § 1.1502-32 by $12, the share’s allocable portion of the $60 loss recognized on the sale of Asset 1). P’s basis in its Block 2 share is an excess loss account of $7 (its original basis of $5 reduced by $12, the share’s portion of the loss recognized on Asset 1). P’s sale of the Block 1 share is a transfer of a loss share.

c. Application of Unified Loss Rules:

(1) P’s bases in all its shares of S stock are subject to redetermination. P’s basis in the transferred Block 1 share is reduced, but not below value, by reallocating negative investment adjustments from shares that are not transferred loss shares. See Reg. § 1.1502-36(b)(2)(i)(B). In total, there were $48 of negative investment adjustments applied to shares that are not transferred loss shares. Accordingly,

- 115 -

P’s basis in the Block 1 share is reduced by $3, from $8 to its value of $5. Reg. § 1.1502-36(b)(2), Ex. 1(ii).

(2) The negative investment adjustments applied to the transferred share are reallocated from (and therefore cause an increase in the basis of) S shares that are not transferred loss shares in a manner that reduces basis disparity to the greatest extent possible. See Reg. § 1.1502-36(b)(2)(i)(B). Thus, the $3 negative investment adjustment reallocated and applied to the transferred Block 1 share is reallocated entirely from the Block 2 share, increasing the basis in the Block 2 share from an excess loss account of $7 to an excess loss account of $4. Id.

(3) Because the Block 1 share is no longer a loss share, the basis and attribute reduction rules of Reg. § 1.1502-36(c) and (d) are not applicable. Id.

d. Application of Prior Rules:

(1) Because the sale of the Block 1 share is the transfer of a loss share that does not result in a deconsolidation, the basis redetermination rule of Prior Reg. § 1.1502-35(b)(1) would apply.

(2) The group members’ aggregate bases of $25 is allocated in proportion to the fair market value of the common shares, or $5 per share. Thus, the sale results in no gain or loss.

3. Example 28 – Increase In Basis of Transferred Loss Share

(2) 1 Block 1 Share 1 Block 2 Share P X $10 / share Block 1 5 Shares $20 Basis/Share

Block 2 5 shares $10 Basis/Share (1) Asset 1 S Z

$100 Asset 1 $50 Basis

- 116 -

a. Facts: On January 1, Year 1, P owns all 10 outstanding shares of S common stock. Five of the shares have a basis of $20 per share (the Block 1 shares) and five of the shares have a basis of $10 per share (the Block 2 shares). S’s only asset, Asset 1, has a basis of $50. S has no other attributes. On October 1, Year 1, S sells Asset 1 for $100. On December 31, Year 2, S sells one Block 1 share and one Block 2 share to X for $10 per share. b. P’s basis in each Block 1 share is $25 (P’s original $20 basis increased under Reg. § 1.1502-32 by $5 (the share’s allocable portion of the $50 gain recognized on the sale of Asset 1)), and P’s basis in each Block 2 share is $15 (P’s original $10 basis increased by $5). P’s sale of the Block 1 and Block 2 shares is a transfer of loss shares. c. Application of Unified Loss Rules:

(1) P’s bases in all its shares of S stock are subject to redetermination. P’s basis in the transferred Block 1 and Block 2 shares is reduced, but not below value, by removing the PIAs applied to the bases of the transferred loss shares. See Reg. § 1.1502-36(b)(2)(i)(A). Accordingly, the basis of the Block 1 share is reduced by $5, from $25 to $20. The basis of the Block 2 share is also reduced by $5, from $15 to $10. (Although the Block 1 share is still a loss share, there is no reduction to its basis under Reg. § 1.1502-36(b)(2)(i)(B) because there were no negative investment adjustments to shares that are not transferred loss shares). Reg. § 1.1502-36(b)(2), Ex. 2.

(2) The $10 of PIAs removed from the transferred loss shares are reallocated and applied to increase P’s bases in its S shares in a manner that reduces basis disparity to the greatest extent possible. See Reg. § 1.1502-36(b)(2)(ii)(B). Accordingly, of the $10 PIAs to be reallocated, $6 is reallocated and applied to the basis of the Block 2 share (increasing it from $10 to $16) and $4 is reallocated and applied equally to the basis of each of the four retained Block 2 shares (increasing the basis of each from $15 to $16). P’s basis in each retained Block 1 share is $25, P’s basis in the transferred Block 1 share is $20, and P’s basis in each Block 2 share, including the transferred Block 2 share, is $16. Reg. § 1.1502-36(b)(2), Ex. 2.

(3) Because the Block 1 and Block 2 shares are still loss shares after application of the basis redetermination rule, they are

- 117 -

subject to the basis reduction rule of Reg. § 1.1502-36(c). Under Reg. § 1.1502-36(c), there is no adjustment to the Block 1 share because the net PIA is $0. However, the basis of the Block 2 share is reduced by $6 (the lesser of its net PIA and its disconformity amount). Id.

(4) Because the Block 1 share is still a loss share after application of the basis reduction rule, it is subject to the attribute reduction rule of Reg. § 1.1502-36(d). Under Reg. § 1.1502-36(d), no adjustment is required because there is no aggregate inside loss. Id.

d. Application of Prior Rules:

(1) Because the sale of the Block 1 and Block 2 shares are transfers of loss shares that do not result in a deconsolidation, the basis redetermination rule of Prior Reg. § 1.1502-35(b)(1) would apply.

(2) The group members’ aggregate bases of $200 is allocated in proportion to the fair market value of the common shares, or $20 per share. Thus, the sale results in no gain or loss.

4. Example 29 – No Investment Adjustments; Basis Redetermination Under Prior Rules But Not Unified Loss Rules

P $100 Basis $120 Basis

(2) S3 Preferred S1 S2 X S3 (1) S3 Common Preferred $20 $100 Asset A $20 Value S3 $50 Basis

a. Facts: P owns all of the stock of S1, with a value of $130 and a basis of $100, and S2, with a value of $90 and a basis of $120. In Year 1, S1 and S2 form S3. S1 contributes $100 cash to S3 in

- 118 -

exchange for all of the S3 common stock. S2 contributes Asset A, with a value of $20 and a basis of $50 in exchange for all of the preferred stock of S3. In Year 3, S2 sells the S3 preferred stock to X for $20, and S3 remains a member of the P group. b. Application of Unified Loss Rules:

(1) Neither the basis redetermination rule of Reg. § 1.1502- 36(b) nor the basis reduction rule of Reg. § 1.1502-36(c) would apply because there have been no positive or negative investment adjustments.

(2) However, the attribute reduction rule of Reg. § 1.1502- 36(d) would result in a $30 reduction in S3’s attributes. Thus, the final Unified Loss Rules permit S2’s loss upfront at the expense of S3’s later loss on its assets. Note that P may make an election to further reduce the S2 stock basis to avoid the application of the attribute reduction rule. c. Application of Prior Rules:

(1) Because S2’s basis in the S3 preferred stock exceeds its value, the basis redetermination rule applies. Of the group members’ total bases of $150 in the S3 stock, $20 is allocated to the preferred stock (i.e., the fair market value of the preferred stock on the date of the sale), and the remaining $130 is allocated to the common stock. Thus, S2’s sale results in the recognition of no gain or loss. Prior Reg. § 1.1502-35(b)(1) & (e), Ex. 1.

(2) The redetermination of S1 and S2’s bases in the stock of S3 results in adjustments to P’s basis in the S1 and S2 stock. Specifically, P’s basis in the S1 stock is increased by $30 to $130, and its basis in the S2 stock is decreased by $30 to $90. Prior Reg. § 1.1502-35(b)(5) & (e), Ex. 1. d. What if, prior to S2’s sale of the S3 preferred stock, S3 had borrowed $500 and suffered an operating loss of $600? The loss would be allocated to the S3 common stock held by S1, see Reg. § 1.1502-32(c)(1), which would result in a ($500) ELA in the S3 common stock held by S1.

(1) Under the Unified Loss Rules, S1’s negative $600 investment adjustment would be reallocated to the transferred loss shares to the extent necessary to eliminate S2’s loss. Reg. § 1.1502-36(b).

- 119 -

(2) Under the prior loss duplication rules, it is not clear how the basis redetermination rules would apply in the situation where the aggregate basis is negative.

(a) Applying the language of the regulations, the aggregate ($450) ELA must be allocated first to the preferred shares in proportion to, but not in excess of, their value. Any remaining basis is allocated to the common shares. Prior Reg. § 1.1502-35(b)(1).

(b) There appear to be two ways to interpret this language.

(i) First, since an ELA is treated as negative basis for all federal income tax purposes, Reg. § 1.1502-19(a)(2)(ii), the preferred stock could take a proportionate share of the ELA. Arguably, the entire ELA is allocated to the preferred stock, because the ELA will always be less than the fair market value of the stock. The entire ELA would be triggered upon the sale of the preferred stock.

(ii) Second, one could argue that the basis redetermination rule was intended only to reallocate positive basis. Under this interpretation, only S2’s positive basis would be reallocated $20 to the preferred stock (i.e., equal to its value) and the remaining $30 to the common stock. Thus, S2 would recognize no gain or loss on the sale of the preferred stock, and S1’s ELA in the common stock would be reduced to ($470).

- 120 -

5. Example 30 – Partial Duplicated Loss Allowed Under Prior Rules But Not Unified Loss Rules

Year 1 Year 3 1 Share

X P P $20 4 Shares 1 Share Asset A $20 Value $20 Value $20 Value $20 Basis $50 Basis S share $50 Basis

S S

Asset A $20 Value $50 Basis

a. Facts: P owns four shares of the common stock of S with a value and basis in each of $20. In Year 1, P contributes Asset A, with a value of $20 and a basis of $50, to S in exchange for one additional share of S common stock. In Year 3, P sells the new share to X for $20, claiming a loss of $30.

b. Application of Unified Loss Rules:

(1) Neither the basis redetermination rule of Reg. § 1.1502- 36(b) nor the basis reduction rule of Reg. § 1.1502-36(c) would apply because there have been no positive or negative investment adjustments.

(2) However, the attribute reduction rule of Reg. § 1.1502- 36(d) would result in a $30 reduction in S’s attributes. Thus, the final Unified Loss Rules permit P’s loss upfront at the expense of S’s later loss on Asset A. Note that P may make an election to further reduce the S stock basis to avoid the application of the attribute reduction rule.

c. Application of Prior Rules:

(1) Because P’s basis in the new share exceeds its value, the basis redetermination rule applies. The total basis of $130 is allocated $26 to each share of S stock. As a result, P recognizes loss of $6 on the sale of the new share.

- 121 -

(2) S can still sell Asset A at a $30 loss. P has thus been able to duplicate $6 of loss.27

6. Example 31 – No Deconsolidation; Economic Loss Disallowed Under Prior Rules But Not Under Unified Loss Rules

1 Share P X $20

4 Shares 1 Share $20 Value $20 Value $0 Basis $50 Basis

S

Assets $100 Value $0 Basis

a. Facts: P owns five shares of common stock of S, four of which have a value of $20 and a basis of $0 and one of which has a value of $20 and a basis of $50. S’s assets have a value of $100 and a basis of $0. P sells the loss share to X for $20, claiming a loss of $30.

b. Application of Unified Loss Rules:

(1) Neither the basis redetermination rule of Reg. § 1.1502- 36(b) nor the basis reduction rule of Reg. § 1.1502-36(c) would apply because there have been no positive or negative investment adjustments.

(2) In addition, the attribute reduction rule of Reg. § 1.1502- 36(d) does not apply because S’s net inside attribute amount is $0. Thus, P’s economic loss is allowed without any reduction in S’s attributes.

27 Note that P’s $6 loss would not be suspended under the loss suspension rules, because S has no overall duplicated loss.

- 122 -

c. Application of Prior Rules:

(1) Because P’s basis in the sold share exceeds its value, the basis redetermination rule applies. The total basis of $50 is allocated $10 to each share of S stock.

(2) As a result, P recognizes gain of $10 on the sale of the new share, notwithstanding the fact that S has no duplicated loss because its inside asset basis is $0.

7. Example 32 – Deconsolidation; Economic Loss Disallowed Under Prior Rules But Not Under Unified Loss Rules

Year 1 Year 3 Year 4

Old Share P P P Y New $50 (1) Old New $100 $150 Basis Share $200 Share Old $50 Basis (2) S S Asset A S X

Asset A Asset A

$200 $100 Value $100 Value $200 Basis $200 Basis

a. Facts: In Year 1, P forms S by contributing $200 for one share of common stock. S buys Asset A for $200, which subsequently declines in value to $100. In Year 3, P contributes $100 to S in exchange for one new share of S common stock. S loses the $100, which results in a reduction of basis of $50 each for the old S share and new S share. In Year 4, P sells the old S share to Y for $50, recognizing a $100 loss.

b. Application of Unified Loss Rules:

(1) The transfer of S’s old share results in a deconsolidation of S; accordingly, all of S’s shares are treated as transferred. See Reg. § 1.1502-36(f)(10)(i)(B).

(2) The basis redetermination rule of Reg. § 1.1502-36(b) does not apply because there have been no PIAs, and there are

- 123 -

no non-transferred shares from which negative adjustments may be reallocated.

(3) In addition, the basis reduction rule of Reg. § 1.1502-36(c) does not apply because there have been no PIAs.

(4) However, the attribute reduction rule of Reg. § 1.1502- 36(d) would result in a $100 reduction in S’s attributes, which is the lesser of S’s aggregate stock loss of $100 and S’s net inside attribute amount of $100. Thus, the final Unified Loss Rules permit P’s loss upfront at the expense of S’s later loss on Asset A. Note that P may make an election to further reduce the S stock basis to avoid the application of the attribute reduction rule.

c. Application of Prior Rules:

(1) Immediately before S’s deconsolidation, the reallocable basis amount is $50 (the lesser of $100, the gross loss inherent in the loss share, and $50, the aggregate amount of S’s items of deduction and loss that were previously taken into account in adjusting the basis of the non-loss shares). Thus, P’s basis in the old S share is reduced by $50 and its basis in the new share is increased by $50.

(2) As a result, P recognizes a $50 loss on the sale of the old share, notwithstanding the fact that P suffered an economic loss of $100 resulting from the decline in the value of Asset A.

8. Example 33 – Basis Redetermination to Eliminate an ELA

P

(2) S3 Preferred S1 S2 X (1) S3 $20 $100 Value Preferred ($10) Basis Asset A

S3 $20 Value $50 Basis

- 124 -

a. Facts: P owns all of the stock of S1 and S2. In Year 1, S1 contributes an asset with a basis and value of $50 to S3. The asset increases in value to $100, and S3 generates operating losses of $60, which are used by the P group, resulting in a ($10) ELA. In Year 4, S2 contributes Asset A, with a value of $20 and a basis of $50 to S3, in exchange for S3 preferred stock. In Year 5, S2 sells the S3 preferred stock to X for $20, recognizing a $30 loss.

b. Application of Unified Loss Rules:

(1) $30 of the negative investment adjustments resulting in S1’s ELA would be reallocated to S2’s stock in S3 under the basis redetermination rule, thus reducing S2’s basis to $20 and resulting in no gain or loss on the sale. As a result of the reallocation, S1’s ($10) ELA is increased to $20 positive basis. See Reg. § 1.1502-36(b).

c. Application of Prior Rules:

(1) Because S2’s basis in the S3 preferred stock exceeds its value, the basis redetermination rule applies. Of the group members’ total bases of $40 in the S3 stock, $20 is allocated to the preferred stock (i.e., the fair market value of the preferred stock on the date of the sale), and the remaining $20 is allocated to the common stock. Similar to the Unified Loss Rules, although S2’s sale results in the recognition of no gain or loss, S1’s ELA has been eliminated.

B. Basis Reduction Examples

Note that the following examples involve no basis redetermination under Reg. § 1.1502-36(b) either because members hold only one share of S stock, so redetermination would not change any members’ basis, or because P transfers its entire interest in S to a non-member in a fully taxable transaction. See Reg. § 1.1502-36(b)(1)(ii).

- 125 -

1. Example 34 – Son-of-Mirrors Transaction

(1) S Stock S Stock

X P P Z $100 $100 $100 basis $140 basis

(2) Asset 1 S S Y S $40 Asset 1 $40 Cash $40 Value Asset 2 $0 Basis $60 Value Asset 2 $60 Basis $60 Value $60 Basis a. Facts: On January 1, Year 1, P purchases the sole outstanding share of S stock for $100. At that time, S owns two assets, Asset 1 with a basis of $0 and a value of $40, and Asset 2 with a basis and value of $60. In Year 1, S sells Asset 1 for $40. On December 31, Year 1, P sells its S share for $100. P’s basis in the S share is $140 (P’s original $100 basis increased under § 1.1502-32 to reflect the $40 gain recognized on the sale of Asset 1). P’s sale of the S share is a transfer of a loss share.

b. Application of Unified Loss Rules:

(1) P’s basis in the S share is reduced, but not below value, by the lesser of the share’s net PIA and disconformity amount. The share’s net PIA is the greater of zero and the sum of all investment adjustments applied to the basis of the share. The only investment adjustment to the share is the $40 adjustment attributable to the gain recognized on the sale of Asset 1. Thus, the share’s net PIA is $40.

(2) The share’s disconformity amount is the excess, if any, of its basis ($140) over its allocable portion of S’s net inside attribute amount. S’s net inside attribute amount is the sum of S’s money ($40 from the sale of Asset 1) and S’s basis in Asset 2 ($60), or $100. The share is the only outstanding S share and so its allocable portion of the $100 net inside attribute amount is the entire $100. Thus, the share’s disconformity amount is $40, the excess of $140 over $100.

- 126 -

(3) The lesser of the net PIA ($40) and the share’s disconformity amount ($40) is $40. Accordingly, the basis in the share is reduced by $40, from $140 to $100, immediately before the sale and no gain or loss is recognized. Reg. § 1.1502-36(c)(8), Ex. 1(i).

c. Application of Prior Rules:

(1) Because P’s $40 loss is attributable to the recognition of built-in gain on the disposition of an asset by S, the entire $40 loss is disallowed under Prior Reg. § 1.337(d)-2. See Prior Reg. §§ 1.337(d)-1(a)(5), Ex. 1, 1.1502-20(a)(5), Ex. 1.

(2) The result is the same under the basis disconformity approach of Notice 2004-58 because the loss is disallowed to the extent of the least of (i) the gain amount, or $40, (ii) the disconformity amount, or $40 (i.e., $140 S stock basis less $100 S net asset basis), and (iii) the PIA amount, or $40.

2. Example 35 – Wasting Asset

a. Facts: Same facts as Example 34 above, except that, instead of selling Asset 1, the value of Asset 1 is consumed in the production of $40 of income in year 1 (reducing the value of Asset 1 to $0).

b. Application of Unified Loss Rules: Because the net PIA includes items of income as well as items of gain, the result is the same as Example 34 above. Reg. § 1.1502-36(c)(8), Ex. 1(ii).

c. Application of Prior Rules: Because the loss on the S stock was not attributable to the recognition of built-in gain on the disposition of an asset, the loss is not disallowed under Prior Reg. § 1.337(d)- 2.

3. Example 36 - Post-Acquisition Appreciation Eliminates Stock Loss

a. Facts: Same facts as Example 34 above, except that, in addition, the value of Asset 2 increases to $100 before the stock is sold. As a result, P sells the S share for $140.

b. Application of Unified Loss Rules: Because P’s sale of the S share is not a transfer of a loss share, the basis reduction rule does not apply to the transfer, notwithstanding that P’s basis in the S share was increased by the gain recognized on Asset 1. Reg. § 1.1502- 36(c)(8), Ex. 1(iii).

- 127 -

c. Application of Prior Rules: Similarly, Prior Reg. § 1.337(d)-2 only applies to a disposition of S stock at a loss.

4. Example 37 – Distributions

a. Facts: Same facts as Example 34 above, except that, in addition, S distributes a $10 dividend before the end of year 1. As a result, the value of the share decreases and P sells the share for $90. P’s basis in the S share is $130 (P’s original $100 basis increased by $30 under Reg. § 1.1502-32 (the net of the $40 gain recognized on the sale of Asset 1 and the $10 dividend)). P’s sale of the S share is a transfer of a loss share.

b. Application of Unified Loss Rules:

(1) P’s basis in the S share is reduced, but not below value, by the lesser of the share’s net PIA and disconformity amount. The share’s net PIA is $40 (the sum of all investment adjustments applied to the basis of the share, computed without taking distributions into account).

(2) The share’s disconformity amount is the excess of its basis ($130) over its allocable portion of S’s net inside attribute amount. S’s net inside attribute amount is the sum of S’s money ($30, the $40 sale proceeds minus the $10 distribution) and S’s basis in Asset 2 ($60), or $90. The share is the only outstanding S share and so its allocable portion of the $90 net inside attribute amount is the entire $90.

(3) The lesser of the share’s net PIA ($40) and its disconformity amount ($90) is $40. Accordingly, the basis in the share is reduced by $40, from $130 to $90, immediately before the sale and no gain or loss is recognized. Reg. § 1.1502-36(c)(8), Ex. 1(iv).

c. Application of Prior Rules:

(1) Because the entire $40 loss is attributable to the disposition of S’s built-in gain asset, the loss is disallowed under Prior Reg. § 1.337(d)-2.

(2) The result is the same under the basis disconformity approach of Notice 2004-58 because the loss is disallowed to the extent of the least of (i) the gain amount, or $40, (ii) the disconformity amount, or $90, and (iii) the PIA amount (excluding distributions), or $40.

- 128 -

5. Example 38 – Loss Attributable to Post-Acquisition Loss

(1) S Stock S Stock

X P P Y $100 $60 $100 basis $100 basis

S S S

Asset 1 Asset 1 $40 Value $0 Value $0 Basis $0 Basis Asset 2 Asset 2 $60 Value $60 Value $60 Basis $60 Basis

a. Facts: On January 1, Year 1, P purchases the sole outstanding share of S stock for $100. At that time, S owns two assets, Asset 1 with a basis of $0 and a value of $40, and Asset 2 with a basis and value of $60. The value of Asset 1 declines to $0 and P sells its S share for $60. P’s basis in the S share remains $100. P’s sale of the S share is a transfer of a loss share.

b. Application of Unified Loss Rules:

(1) P’s basis in the S share ($100) is reduced immediately before the sale, but not below value ($60), by the lesser of the share’s net PIA and disconformity amount. There were no adjustments to P’s basis in the share and so the share’s net PIA is $0.

(2) Thus, although the share’s disconformity amount is $40 (the excess of P’s basis in the share ($100) over the share’s allocable portion of S’s net inside attribute amount ($60)), no basis reduction is required under Reg. § 1.1502-36(c) (because the PIA amount is less), and the $40 loss is allowed. Reg. § 1.1502-36(c)(8), Ex. 2.

(3) Note that the attribute reduction rule would not apply because S does not have an aggregate inside loss, so there is no duplicated loss.

- 129 -

c. Application of Prior Rules:

(1) Because S did not dispose of its built-in gain asset, P’s $40 loss is allowed under Prior Reg. § 1.337(d)-2(c)(2). See Prior Reg. §§ 1.337(d)-1(a)(5), Ex. 2, 1.1502-20(a)(5), Ex. 2.

(2) The result is the same under the basis disconformity approach of Notice 2004-58 because the gain and PIA amounts are both $0.

6. Example 39 – Built-In Gain Asset Depreciates in Value

(1) S Stock S Stock X P P Z $50 $100 $100 basis $100 basis

(2) Asset S S Y S $50 Asset $50 Cash $100 Value $50 Basis

a. Facts: In Year 1, P acquires the stock of S for $100. S has a built- in gain asset with a value of $100 and a basis of $50. The asset depreciates in value to $50, and in Year 2, S sells the asset to Y for $50. P subsequently sells all of the S stock to Z for $50. Because S recognized no gain or loss on the sale of the asset, P’s basis in the S stock remains $100. Thus, when P sells its S stock, it recognizes a $50 loss.

b. Application of Unified Loss Rules:

(1) P’s basis in the S share ($100) is reduced immediately before the sale, but not below value ($50), by the lesser of the share’s net PIA and disconformity amount. There were no adjustments to P’s basis in the share and so the share’s net PIA is $0.

(2) Thus, although the share’s disconformity amount is $50 (the excess of P’s basis in the share ($100) over the share’s allocable portion of S’s net inside attribute amount ($50)),

- 130 -

no basis reduction is required under Reg. § 1.1502-36(c) (because the PIA amount is less), and the $50 loss should be allowed.

(3) Note that the attribute reduction rule would not apply because S does not have an aggregate inside loss, so there is no duplicated loss.

c. Application of Prior Rules:

(1) Even though S had a built-in gain asset when P acquired the S stock, and S disposed of that asset, none of the built- in gain was recognized because of the depreciation in value. Thus, P’s $50 loss should be allowed under Prior Reg. § 1.337(d)-2(c).

(2) The result is the same under the basis disconformity approach because both the gain and PIA amounts are $0.

7. Example 40 – Built-In Gain Asset Appreciates in Value

(1) S Stock S Stock X P P Z $150 $100 $100 basis $200 basis (2) Asset S S S Y $150 Asset $150 Cash $100 Value $50 Basis

a. Facts: In Year 1, P acquires the stock of S for $100. S has a built- in gain asset with a value of $100 and a basis of $50. The asset appreciates in value to $150, and in Year 2, S sells the asset to Y for $150. P subsequently sells all of the S stock to Z for $150. Under the investment adjustment rules, P’s basis in the S stock is increased by S’s $100 gain in Year 2. Reg. § 1.1502-32(b)(2)(i). Thus, when P sells its S stock, it recognizes a $50 loss.

- 131 -

b. Application of Unified Loss Rules:

(1) P’s basis in the S share ($200) is reduced immediately before the sale, but not below value ($150), by the lesser of the share’s net PIA and disconformity amount.

(2) The only investment adjustment to the share is the $100 adjustment attributable to the gain recognized on the sale of the asset. Thus, the share’s net PIA is $100.

(3) The share’s disconformity amount is the excess, if any, of its basis ($200) over its allocable portion of S’s net inside attribute amount ($150 from the sale of the asset), or $50.

(4) The lesser of the net PIA ($100) and the share’s disconformity amount ($50) is $50. Accordingly, the basis in the share is reduced by $50, from $200 to $150, immediately before the sale, and no gain or loss is recognized. c. Application of Prior Rules:

(1) Loss is generally treated as first attributable to recognized built-in gain. See Prior Reg. § 1.337(d)-1(a)(5), Ex. 3. Thus, under a pure tracing approach, because S had a built- in gain of $50 at the time P acquired the S stock, P’s $50 loss is attributable to the recognition of built-in gain on the disposition of an asset by S, the entire $50 loss is disallowed under Prior Reg. § 1.337(d)-2.

(2) The result is the same under the basis disconformity approach of Notice 2004-58, because the disconformity amount is $50 (i.e., $100 S stock basis less $50 S net asset basis), which is less than the $100 gain and PIA amounts.

- 132 -

8. Example 41 – Built-In Gain in After-Acquired Asset

S Stock S Stock W P P Z $100 $100 $100 basis $140 basis

Asset 3 S S S Y $40 Asset 1 Asset 1 Asset 3 $40 Value $0 Value $40 Value $0 Basis $0 Basis $0 Basis

Asset 2 Asset 2 $60 Value $60 Value $60 Basis $60 Basis

a. Facts: On January 1, Year 1, P purchases the sole outstanding share of S stock for $100. At that time, S owns two assets, Asset 1, with a basis of $0 and a value of $40, and Asset 2, with a basis and value of $60. In Year 1, the value of Asset 2 declines to $20. In year 2, the value of Asset 1 declines to $0, the value of Asset 2 returns to $60, and S creates Asset 3 (with a basis of $0). In Year 3, S sells Asset 3 for $40. On December 31, year 3, P sells its S share for $100. P’s basis in the S share is $140 (P’s original $100 basis increased under § 1.1502-32 to reflect the $40 gain recognized on the sale of Asset 3 in Year 3).

b. Application of Unified Loss Rules:

(1) P’s basis in the S share ($140) is reduced immediately before the sale, but not below value ($100), by the lesser of the share’s net PIA and disconformity amount. The share’s net PIA is $40 (the Year 3 investment adjustment). The share’s disconformity amount is the excess of its basis ($140) over its allocable portion of S’s net inside attribute amount. S’s net inside attribute amount is $100, the sum of S’s money ($40 from the sale of Asset 3) and its basis in its assets ($60 (the sum of Asset 1’s basis of $0 and Asset 2’s

- 133 -

basis of $60)). S’s $100 net inside attribute amount is allocable entirely to the sole outstanding S share.

(2) Thus, the share’s disconformity amount is the excess of $140 over $100, or $40. The lesser of the share’s net PIA ($40) and its disconformity amount ($40) is $40. Accordingly, the basis in the share is reduced by $40, from $140 to $100, immediately before the sale and no gain or loss is recognized. Reg. § 1.1502-36(c)(8), Ex. 3(ii). c. Application of Prior Rules:

(1) Gain is built-in if it is attributable to an “excess of value over basis that is reflected, before the disposition of the asset, in the basis of the share.” Prior Reg. § 1.337(d)- 2(c)(2). The reference to the disposition of “the” asset appears to require that the built-in gain reflected in the stock basis be with respect to the same asset that is disposed of at a gain. Before S disposed of the first asset, the original amount of built-in gain was reflected in P’s basis in S. However, S never recognized the built-in gain in that asset because it depreciated in value. The excess of value over basis recognized by S on the disposition of the new asset was not the same gain reflected in P’s basis in S. P’s loss thus should not be disallowed under a pure tracing approach.

(2) Under the basis disconformity approach of Notice 2004-58, however, P’s $40 loss would be disallowed, since the gain, disconformity, and PIA amounts are all $40.

(3) Thus, the original built-in gain taint is preserved under the basis disconformity approach to attach to any after- acquired asset. This result seems inconsistent with the language of Prior Reg. § 1.337(d)-2(c)(2), and arguably disallows an economic loss for the loss in value of the original asset. The built-in gain taint is similarly preserved under the Unified Loss Rules.

- 134 -

9. Example 42 – Lower-Tier Subsidiary/No Transfer of Lower-Tier Stock

(2) S Share P Y $160 $160 Basis

Asset A $100 Value S $100 Basis

$60 Basis (1) Asset 1 S1 X

Asset 1 $60 $60 Value $20 Basis

a. Facts: P owns the sole outstanding share of S stock with a basis of $160. S owns two assets, Asset A with a basis and value of $100, and the sole outstanding share of S1 stock with a basis of $60. S1 owns one asset, Asset 1, with a basis of $20 and value of $60. In Year 1, S1 sells Asset 1 to X for $60, recognizing $40 of gain. On December 31, Year 1, P sells its S share to Y, a member of another consolidated group, for $160. P’s basis in the S share is $200 (P’s original $160 basis increased under Reg. § 1.1502-32 by $40 (to reflect the tiering up of the increase to S’s basis in S1 under Reg. § 1.1502-32 by $40 (to reflect the gain recognized on S1’s sale of Asset 1)). P’s sale of the S share is a transfer of a loss share. S does not transfer the S1 share because S and S1 are members of the same group following the transfer.

b. Application of Unified Loss Rules:

(1) P’s basis in the S share ($200) is reduced immediately before the sale, but not below value ($160), by the lesser of the share’s net PIA and disconformity amount. The S share’s net PIA is $40. The share’s disconformity amount is the excess, if any, of the basis of the share ($200) over the share’s allocable portion of S’s net inside attribute amount. S’s net inside attribute amount is the sum of S’s basis in Asset A ($100) plus S’s basis in the S1 share.

(2) Although S’s actual basis in the S1 share is $100 (S’s original $60 basis increased by S1’s year 1 $40 PIA), for

- 135 -

purposes of computing the S share’s disconformity amount, S’s basis in the S1 share is tentatively reduced by the lesser of the S1 share’s net PIA and its disconformity amount. The S1 share’s net PIA is $40 (the year 1 PIA). The S1 share’s disconformity amount is the excess, if any, of its basis ($100) over its allocable portion of S1’s net inside attribute amount. S1’s net inside attribute amount is $60 (its cash received on the sale of Asset 1) and it is entirely attributable to S’s S1 share. The S1 share’s disconformity amount is therefore the excess of $100 over $60, or $40. The lesser of the S1 share’s net PIA ($40) and its disconformity amount ($40) is $40. Accordingly, for purposes of computing the disconformity amount of the S share, S’s basis in its S1 share is tentatively reduced by $40, from $100 to $60.

(3) Disconformity Amount of P’s S Share: S’s net inside attribute amount is treated as the sum of its basis in Asset A ($100) and its (tentatively reduced) basis in its S1 share ($60), or $160. S’s net inside attribute amount is allocable entirely to P’s S share. Thus, the S share’s disconformity amount is the excess of $200 over $160, or $40.

(4) Amount of Reduction: P’s basis in its S share is reduced by the lesser of the S share’s net PIA ($40) and disconformity amount ($40), or $40. Accordingly, P’s basis in the S share is reduced by $40, from $200 to $160, immediately before the sale, and no gain or loss is recognized.

(5) Effect on S’s Basis in its S1 Share: The transaction has no effect on S’s basis in the S1 share—the tentative reduction is only for purposes of computing S’s net inside attribute amount. Thus, S owns the S1 share with a basis of $100, S’s original $60 basis in the share plus the $40 adjustment for the gain recognized on the sale of Asset 1 in Year 1. Reg. § 1.1502-36(c)(8), Ex. 7. c. Application of Prior Law:

(1) The built-in gain is reflected in P’s basis and must be taken into account in determining whether any loss realized by P on the sale of its S stock is allowed. See Prior Reg. § 1.337(d)-2(c)(4), Example. The result was the same

- 136 -

under the prior regulations. See Prior Reg. § 1.1502- 20(c)(5), Ex. 5.28

10. Example 43 – Carryover Basis in Stock

A A (1) P Stock S Stock $60 basis S Stock P Y S P $90

$60 basis (2) $160 basis Asset S X S $100 Asset $100 Value New Asset

$0 Basis $90 Value $100 Basis

a. Facts: Individual A owns all of the stock of S and has a $60 basis in such stock. S has a built-in gain asset with a value of $100 and a basis of $0. In Year 1, P acquires the stock of S from A in a section 351 exchange. Section 362(a)(1). In Year 2, S sells the asset to X for $100. S reinvests the proceeds in New Asset, and New Asset declines in value to $90. In Year 3, P sells the stock of S to Y for $90. When P acquires the S stock from A, P takes a $60 carryover basis in such stock. Under the investment adjustment rules, P’s basis in the S stock is increased by S’s $100 gain in Year 2, to $160. Reg. § 1.1502-32(b)(2)(i). Thus, when P sells its S stock in Year 3, it recognizes a $70 loss.

b. Application of Unified Loss Rules:

(1) P’s basis in the S stock ($160) is reduced immediately before the sale, but not below its value ($90) by the lesser of the net PIA and disconformity amount. The net PIA is $100. The disconformity amount is the excess of the basis

28 Similar to Prior Reg. § 1.337(d)-2(c)(2), Prior Reg. § 1.1502-20(c)(2)(iii) provided that extraordinary gain and positive investment adjustments were taken into account “only to the extent they are reflected in the basis of the share, directly or indirectly, immediately before the disposition or deconsolidation.”

- 137 -

of the share ($160) over the share’s alocable portion of S’s net inside attribute amount. S’s net inside attribute amount is $100, its basis in the New Asset. Thus, the share’s disconformity amount is the excess of $160 over $100, or $60.

(2) The lesser of the share’s net PIA ($100) and its disconformity amount ($60) is $60. Accordingly, P must reduce its basis from $160 to $100, and P recognizes a $10 loss on the transfer of the S share.

(3) Because the S share is still a loss share, the attribute reduction rule applies. S must reduce its attributes by the lesser of the net stock loss and the aggregate inside loss. In this case, the net stock loss (after basis reduction) and the aggregate inside loss are both $10. Because S has no other attributes, S must reduce its basis in the New Asset from $100 to $90. Note that P may make an election to further reduce its basis in the S stock basis to avoid the application of the attribute reduction rule. c. Application of Prior Rules:

(1) Even though P’s basis in the S stock was increased by $100 as a result of S’s built-in gain, only $60 of the $70 loss on the sale of S stock is reflected in P’s basis (and $60 is the amount of the basis disconformity, which is less than the $100 gain and PIA amounts). Accordingly, only $60 of the $70 loss is disallowed under Prior Reg. § 1.337(d)-2. See Prior Reg. § 1.337(d)-1(a)(5), Ex. 5.

- 138 -

11. Example 44 – Unrecognized Loss Reflected in Stock Basis

(2) S Share P Y $100

$100 Basis

(1) Asset 1 S X

$60 Asset 1 $60 Value $20 Basis

Asset 2 $40 Value $60 Basis

a. Facts: P owns the sole outstanding share of S stock with a basis of $100. S owns two assets, Asset 1 with a basis of $20 and a value of $60, and Asset 2 with a basis of $60 and a value of $40. In Year 1, S sells Asset 1 for $60. On December 31, Year 1, P sells the S share for $100. P’s basis in the S share is $140 (P’s original $100 basis increased under Reg. § 1.1502-32 to reflect the $40 gain recognized on the sale of Asset 1). P’s basis is unaffected by the unrealized post-acquisition decline in the value of Asset 2. Reg. § 1.1502-32(b)(2)(i). When P sells its S stock, it recognizes a $40 loss. P’s sale of the S share is a transfer of a loss share.

b. Application of Unified Loss Rules:

(1) P’s basis in the S share ($140) is reduced immediately before the sale, but not below value ($100), by the lesser of the share’s net PIA and disconformity amount. The share’s net PIA is $40 (the Year 1 investment adjustment). The share’s disconformity amount is the excess of its basis ($140) over its allocable portion of S’s net inside attribute amount. S’s net inside attribute amount is the sum of S’s money ($60 from the sale of Asset 1) and S’s basis in Asset 2 ($60), or $120. S’s net inside attribute amount is allocable entirely to the sole outstanding S share.

(2) Thus, the share’s disconformity amount is the excess of $140 over $120, or $20. The lesser of the share’s net PIA

- 139 -

($40) and its disconformity amount ($20) is $20. Accordingly, the basis in the share is reduced by $20, from $140 to $120, immediately before the sale and P recognizes a $20 loss. Reg. § 1.1502-36(c)(8), Ex. 4.

(3) Because the S share is still a loss share, the attribute reduction rule applies. S must reduce its attributes by the lesser of the net stock loss ($20) and the aggregate inside loss ($20), or $20. Because S has no other attributes, it must reduce its basis in Asset 2 fom $60 to $40. Note that P may make an election to further reduce its basis in the S stock basis to avoid the application of the attribute reduction rule. c. Application of Prior Rules:

(1) Even though S had a $20 built-in loss, which was reflected in P’s basis, P’s loss on the sale of the S stock is treated first as attributable to the recognized built-in gain on Asset 1. Thus, under a pure tracing P’s $40 loss is disallowed. See Prior Reg. § 1.337(d)-1(a)(5), Ex. 3.

(2) However, under the basis disconformity approach of Notice 2004-58, only $20 of P’s loss should be disallowed because the disconformity amount is $20. Thus, the unrecognized built-in loss effectively offsets the recognized built-in gain under the disconformity approach, but not tracing.

- 140 -

12. Example 45 – Post-Acquisition Appreciation Removes Taint of Built-In Gain

(1) S Stock S Stock X P P Z $180 $100 $100 basis $200 basis (2) Asset S S Y S

$100 Asset New Asset $100 Value $180 Value $0 Basis $100 Basis

a. Facts: In Year 1, P acquires the stock of S for $100. S has a built- in gain asset with a value of $100 and a basis of $0. In Year 2, S sells the asset to Y for $100, and reinvests the proceeds in New Asset. New Asset appreciates in value to $180. In Year 7, P sells all of the S stock to Z for $180. Under the investment adjustment rules, P’s basis in the S stock is increased by S’s $100 gain in Year 2. Reg. § 1.1502-32(b)(2)(i). As a result, when P sells its S stock in Year 7 for $180, it recognizes a $20 loss.

b. Application of Unified Loss Rules:

(1) P’s basis in the S share ($200) is reduced immediately before the sale, but not below value ($180), by the lesser of the share’s net PIA and disconformity amount. The share’s net PIA is $100 (the Year 2 investment adjustment). The share’s disconformity amount is the excess of its basis ($200) over its allocable portion of S’s net inside attribute amount. S’s net inside attribute amount is S’s basis in New Asset, or $100. S’s net inside attribute amount is allocable entirely to the sole outstanding S share.

(2) Thus, the share’s disconformity amount is the excess of $200 over $100, or $100. The lesser of the share’s net PIA ($100) and its disconformity amount ($100) is $100. Accordingly, the basis in the share is reduced to its value, from $200 to $180, immediately before the sale.

- 141 -

c. Application of Prior Rules:

(1) S had a $100 built-in gain reflected in P’s basis, all of which was recognized. However, the post-acquisition appreciation of New Asset has the effect of reducing P’s disallowed loss. So, under the prior rules, the entire $20 loss would be disallowed under the tracing approach. See Prior Reg. §§ 1.337(d)-1(a)(5), Ex. 7, 1.1502-20(a)(5), Ex. 2.

(2) Similarly, under the basis disconformity approach of Notice 2004-58, the $20 loss would be disallowed, because all of the gain, PIA, and disconformity amounts are $100. d. The Service has permitted this taxpayer-favorable result since the original loss disallowance regulations, because it determined that administratively burdensome tracing would be required to reverse the PIA attributable to the sold built-in gain asset. e. As a result, a consolidated group could transfer assets that are expected to appreciate to a subsidiary with built-in gain assets subject to the anti-stuffing rule, anti-avoidance rule, and section 269.

(1) If a subsidiary is “stuffed” with assets with a high value and low basis, the anti-abuse rule of the Unified Loss Rules would likely apply. Reg. § 1.1502-36(g). Under the prior rules, the anti-stuffing rule could apply if the subsidiary stock is sold or deconsolidated within two years of the assets’ transfer. Prior Reg. § 1.1502-20(e)(2).

(2) If the subsidiary is transferred an asset with a basis equal to its value, which it expects to appreciate, on its face the anti- abuse (or anti-stuffing) rule may apply. However, the mere hope that an asset will appreciate seems to be a slender reed on which to apply the anti-abuse (or anti-stuffing) rule.

- 142 -

13. Example 46 - Post-Acquisition Appreciation Removes Taint of Built-In Gain – Lower Tier Subsidiary

(4)

S Stock P P Y $110

$75 basis $120 basis (3) (1) T Stock Purchase T, S S X U, and V $70 Stock for $25 Each $45 basis T U V T U V

Asset 1 $5 Value $20 Cash $0 Basis (2)

Sell Asset 1 for Asset 3 Asset 2 $5 and Asset 2 $50 Value $15 Value $0 Basis for $15 $5 Basis

Asset 3 $5 Value $5 Basis

a. Facts: In Year 1, P forms S with $75 cash and thus has an original basis of $75 in S. S purchases all of the stock of three corporations, T, U, and V, each at a price of $25. T owns three assets: Asset 1 with a value of $5 and basis of $0, Asset 2 with a value of $15 and a basis of $0, and Asset 3 with a value of $5 and a basis of $5. In Year 2, T sells Asset 1 for $5 and Asset 2 for $15. Asset 3 then appreciates in value to $50. In Year 3, S sells all of the stock of T to X for $70. The value of U then depreciates to $15. In Year 4, P sells all of the stock of S to Y for $110.

T recognizes gain of $5 and $15 on Asset 1 and Asset 2, respectively. Under the investment adjustment rules, S increases its basis in T by $20 to $45. Reg. § 1.1502-32(b)(2)(i). P, in turn, increases its basis in S by $20 to $95. Reg. § 1.1502-32(a)(3)(iii). The value of T increases to $70 when Asset 3 appreciates (i.e., $20 cash + $50 Asset 3). When S sells its T stock in Year 3, it recognizes a $25 gain. P increases its basis in S by the $25 gain to $120. Reg. § 1.1502-32(b)(2)(i). The value of S decreases to $110 when the value of U depreciates (i.e., $15 U + $25 V + $70 cash from sale of T). When P sells its S stock in Year 4, it is transfer of a loss share. P’s basis is $120 and the value of the S stock is $110.

- 143 -

b. Application of Unified Loss Rules:

(1) Under the basis reduction rule, P’s basis in the S share ($120) is reduced immediately before the sale, but not below value ($110), by the lesser of the share’s net PIA and disconformity amount. The share’s net PIA is $45 (the Year 2 investment adjustment plus the Year 3 investment adjustment). The share’s disconformity amount is the excess of its basis ($120) over its allocable portion of S’s net inside attribute amount. S’s net inside attribute amount is the sum of S’s money ($70 from the sale of T) and S’s basis in the lower-tier subsidiary shares.

(2) Under Reg. § 1.1502-32(c)(6), S’s net inside attribute amount is determined by treating the lower tier subsidiary shares as tentatively reduced by the lesser of their net PIA and the disconformity amount. U and V have no net PIAs, so there is no tentative reduction. Thus, S’s basis in its lower tier subsidiaries is $50 - U ($25) and V ($25). As a result, S’s net inside attribute amount is $120, which is allocable entirely to the sole outstanding S share. Thus, the share’s disconformity amount is the excess of $120 over $120, or $0. The lesser of the share’s net PIA ($45) and its disconformity amount ($0) is $0. Accordingly, the basis in the share is not reduced immediately before the sale.

(3) Under the attribute reduction rule, S’s attribution reduction amount is the lesser of P’s net stock loss and S’s aggregate inside loss. P’s net stock loss is $10 ($120 basis over $110 value). For purposes of computing S’s aggregate inside loss, the Unified Loss Rules look through S to determine S’s deemed basis in the U and V stock. See § 1.1502- 36(d)(3)(iii)(B) and (5)(i)(B). The regulations then treat the stock of S’s subsidiaries as Category D attributes. Since S has no other Category A, B, C, or D attributes, the attribution reduction amount would first be applied to the bases of the U and V stock, then is tiered down to reduce U and V’s attributes. Reg. § 1.1502-36(d)(5)(iii), (v). c. Application of Prior Rules:

(1) Even though T sold built-in gain assets, and the PIAs tiered up to S and P, S’s sale of T at a gain due to an appreciated asset removes the taint of the built-in gain. See T.A.M. 200138005 (May 4, 2001). This is because P’s $10 loss was attributable to the appreciation of T and not the

- 144 -

disposition of T’s built-in gain assets. This can be illustrated with a modification of the facts of Example 46.

(2) Assume the same facts as Example 46, except that T did not dispose of Asset 1 and Asset 2. Asset 3 appreciates in value to $50, and S sells the T stock for $70, recognizing a gain of $45 (i.e., $70 value - $25 basis). P thus increases its basis in S by $45 to $120. When the value of U depreciates, S is worth $110, and P still recognizes a $10 loss on the sale of S. Id.

C. Duplicated Loss Examples

Note that the following examples involve no basis redetermination under Reg. § 1.1502-36(b) because there is no basis disparity and involve no basis reduction under Reg. § 1.1502-36(c) because there are no net PIAs.

1. Example 47 – Computation of Attribute Reduction Amount/Transfer of All S Shares

30 S Shares P X

$30 100 Shares $2 Basis/Share $1 Value/Share

S

100 Basis in Land $120 Loss Carryover

a. Facts: P owns all 100 of the outstanding shares of S stock with a basis of $2 per share. S owns land with a basis of $100, has a $120 loss carryover, and has no liabilities. Each share has a value of $1. P sells 30 of the S shares to X for $30. As a result of the sale, P and S cease to be members of the same group. Accordingly, P is treated as transferring all 100 S shares. P’s transfer of the S shares is a transfer of loss shares.

b. Application of Unified Loss Rules:

(1) S’s attributes are reduced by S’s attribute reduction amount. Under Reg. § 1.1502-36(d)(3), S’s attribute

- 145 -

reduction amount is the lesser of the net stock loss and S’s aggregate inside loss. The net stock loss is the excess of the aggregate bases of the transferred shares ($200) over the aggregate value of the transferred shares ($100), or $100. S’s aggregate inside loss is the excess of its net inside attribute amount ($220, the sum of the $100 basis of the land and the $120 loss carryover) over the value of all outstanding S shares ($100), or $120.

(2) The attribute reduction amount is therefore the lesser of the net stock loss ($100) and the aggregate inside loss ($120), or $100. Under Reg. § 1.1502-36(d)(4), S’s $100 attribute reduction amount is allocated and applied to reduce S’s $120 loss carryover to $20. Reg. § 1.1502-36(d)(8), Ex. 1(i).

c. Application of Prior Rules: The loss suspension rule does not apply because P and S are deconsolidated. Prior Reg. § 1.1502- 35(c)(1).

2. Example 48 – Transfer of less than all S shares

a. Facts: Same facts as above, except that P only sells 20 S shares to X. P’s sale of the 20 S shares is a transfer of loss shares, but there is no transfer of S’s remaining shares because they remain members of a consolidated group.

b. Application of Unified Loss Rules:

(1) Under Reg. § 1.1502-36(d), S’s attributes are reduced by S’s attribute reduction amount. Under Reg. § 1.1502- 36(d)(3), S’s attribute reduction amount is the lesser of the net stock loss and S’s aggregate inside loss. The net stock loss is the excess of the aggregate bases of the transferred shares ($40) over the aggregate value of the transferred shares ($20), or $20. S’s aggregate inside loss is the excess of its net inside attribute amount ($220) over the value of all outstanding S shares ($100), or $120. The attribute reduction amount is therefore the lesser of the net stock loss ($20) and the aggregate inside loss ($120), or $20.

(2) Under Reg. § 1.1502-36(d)(4), S’s $20 attribute reduction amount is allocated and applied to reduce S’s $120 loss carryover to $100. Reg. § 1.1502-36(d)(8), Ex. 1(ii).

- 146 -

c. Application of Prior Rules:

(1) The loss suspension rule applies at the time of P’s sale of S stock. The duplicated loss amount with respect to the sold shares is $24 (i.e., 20 percent of $220 inside attributes - $20 value of S stock). Thus, the entire $20 loss is suspended. The suspended loss is reduced by losses and deductions absorbed by the P group, and any remaining loss may be recognized upon the sale of the remaining S stock or 10 years, whichever is earlier. Prior Reg. § 1.1502-35(c).

3. Example 49 – Multiple Dispositions of S Stock

(1) (4) 20 shares 80 shares

W P Z P P $4 $48

100 Asset A $140 $132 $20 Value shares Basis Basis (3) $100 Basis Asset B (2) $50 $40 Value S S X S $50 Basis Y Asset B $40 Asset A $50 Value Asset A $20 Value $50 Basis $20 Value $100 Basis $100 Basis $40 Cash

a. Facts: In Year 1, P forms S by transferring Asset A, with a value of $20 and a basis of $100, in exchange for 100 shares of S stock. In Year 3, P sells 20 shares of S common stock to W for $4, recognizing a $16 loss. In Year 5, S earns $50 and purchases Asset B from X. In Year 7, S sells Asset B to Y for $40, recognizing a $10 loss. Under the investment adjustment rules, P increases its basis in the S shares by 80% of S’s $50 income in Year 5 and $10 loss in Year 7. In Year 8, P sells the remaining 80 shares of S stock to Z for $48, recognizing a $84 loss.

b. Application of Unified Loss Rules:

(1) Under Reg. § 1.1502-36(d), S’s attributes are reduced by S’s attribute reduction amount. Under Reg. § 1.1502- 36(d)(3), S’s attribute reduction amount is the lesser of the net stock loss and S’s aggregate inside loss.

- 147 -

(2) In Year 3, P would recognize a $16 loss. However, S would be required to reduce its attributes by $16 (i.e. the lesser of P’s loss or S’s duplicated loss of $80 ($100 asset basis less $20 value)).

(3) In Year 8, P would recognize a $84 loss. However, S would be required to reduce its attributes by $84 (i.e. the lesser of P’s loss or S’s duplicated loss of $92 ($40 cash plus $100 asset basis less $48 value)). c. Application of Prior Rules:

(1) Although the basis redetermination rule technically applies when P sells the 20 S shares at a loss in Year 3, its application does not result in any shift in basis.

(2) However, the loss suspension rule applies at that time. The duplicated loss amount with respect to the sold shares is $16 (i.e., 20 percent of $100 basis in Asset A - $20 value of S stock). Thus, the entire $16 loss is suspended. The suspended loss is reduced by losses and deductions absorbed by the P group. However, P should be able to establish that the $10 loss on Asset B, which was not acquired until after the sale of the S stock, was not part of the duplicated loss. The entire $16 suspended loss should thus be allowed in Year 8 when S leaves the group.

(3) Assume that S purchased Asset B before P sold the 20 shares of S stock, and Asset B increased in value to $60. The duplicated loss with respect to the sold shares would be $14 (i.e., 20 percent of $100 basis in Asset A + $50 basis in Asset B - $80 value of S stock). Thus, only $14 of the $16 loss would be suspended. The duplicated loss computation essentially permits S’s unrealized gains to offset its unrealized losses and limits the duplicated loss to the net loss.

- 148 -

4. Example 50 – Allocation of Attribute Reduction Amount Among Category D Assets

S Shares P X

$90

$150 Basis

S

$60 Basis in land $30 Basis in factory $30 Basis in publicly traded securities $30 Basis in goodwill

a. Facts: P owns the sole outstanding share of S stock with a basis of $150. S owns land with a basis of $60, a factory with a basis of $30, publicly traded securities with a basis of $30, and goodwill with a basis of $30. P sells its S share for $90. P’s sale of the S share is a transfer of a loss share.

b. Application of Unified Loss Rules:

(1) Under Reg. § 1.1502-36(d)(3), S’s attribute reduction amount is determined to be $60, the lesser of the net stock loss ($150 basis over $90 value) and S’s aggregate inside loss ($60, the excess of S’s $150 net inside attribute amount (the $60 basis of the land, plus the $30 basis of the factory, plus the $30 basis of the publicly traded securities, plus the $30 basis of the goodwill) over the $90 value of the S share).

(2) Under Reg. § 1.1502-36(d)(4), the $60 attribute reduction amount is allocated and applied proportionately to reduce S’s attributes (all Category D) in reverse section 1060 order as follows (Reg. § 1.1502-36(d)(8), Ex. 2):

- 149 -

Available Attributes Attribute Amount Allocable Portion Adjusted of Attribute Attributes Reduction Amount Amount Class VII $30 $30 $0 Basis of Goodwill Class V Basis of Land $60 (60/90 x $60) $40 $20 Basis of Factory $30 (30/90 x $60) $20 $10 Total Class V $90 $60 $30 Class II $30 $0 $30 Basis of Publicly Traded Securities Totals $150 $60 $90

c. Application of Prior Rules: The loss suspension rule does not apply because P and S are deconsolidated. Prior Reg. § 1.1502- 35(c)(1).

5. Example 51 – Allocation of Attribute Reduction Amount to Category A, Category B, and Category C Assets

P

$1,000 Basis S Share M X

$100 $210 Basis

S

$10 Capital Loss Carryover $200 NOL Carryover $40 Deferred Deduction $50 Basis in land

- 150 -

a. Facts: P owns the sole outstanding share of M stock with a basis of $1,000 and M owns the sole outstanding share of S stock with a basis of $210. M sells its S share to X for $100. M’s sale of the S share is a transfer of a loss share. At the time of the sale, S has no liabilities and the following attributes:

Category Attribute Attribute Amount Category A Capital Loss Carryover $10 Category B NOL Carryover $200 Category C Deferred Deductions $40 Category D, Class V Basis in Land $50 Total Attributes $300

b. Application of Unified Loss Rules:

(1) Under Reg. § 1.1502-36(d)(3), S’s attribute reduction amount is $110, the lesser of the net stock loss ($210 basis over $100 value) and S’s aggregate inside loss ($200, the excess of S’s $300 net inside attribute amount (the $10 capital loss carryover, plus the $200 NOL carryover, plus the $40 deferred deductions, plus the $50 basis in land) less the $100 value of all outstanding S shares).

(2) Under Reg. § 1.1502-36(d)(4)(ii)(A)(1), the $110 attribute reduction amount is allocated and applied to reduce S’s attributes as follows (Reg. § 1.1502-36(d)(8), Ex. 3(i)):

Category Available Attribute Application of Adjusted Attributes Amount Attribute Attribute Reduction Amount Amount A Capital Loss $10 $10 $0 Carryover B NOL Carryover $200 $100 $100 C Deferred $40 $0 $40 Deductions D Basis in Land $50 $0 $50 Totals $300 $110 $190

- 151 -

(3) P could also elect to allocate the attribute reduction amount to, for example, eliminate the $40 deferred deductions, preserve the capital loss carryover, and reduce NOLs to $130. Reg. § 1.1502-36(d)(8), Ex. 3(ii).

c. Application of Prior Rules: The loss suspension rule does not apply because M and S are deconsolidated. Prior Reg. § 1.1502- 35(c)(1).

6. Example 52 – Wholly Owned Lower-Tier Subsidiary

S Share

P P1 $50 $250 Basis

S

Share A Asset $40 Basis

$100 Basis Share B $60 Basis

S1

Asset 1 $50 Basis

a. Facts: P owns the sole outstanding share of S stock with a basis of $250. S owns Asset with a basis of $100 and the only two outstanding shares of S1 stock (Share A has a basis of $40 and Share B has a basis of $60). S1 owns Asset 1 with a basis of $50. P sells its S share to P1, the common parent of another consolidated group, for $50. The sale is a transfer of a loss share.

b. Application of Unified Loss Rules:

(1) Computation of Attribute Reduction Amount:

(a) Under Reg. § 1.1502-36(d)(3), S’s attribute reduction amount is the lesser of P’s net stock loss and S’s aggregate inside loss. P’s net stock loss is $200 ($250 basis minus $50 value). S’s aggregate inside loss is the excess of S’s net inside attribute amount over the value of the S share. Under Reg. §§ 1.1502-36(d)(3)(iii)(B) and (d)(5)(i)(B), S’s net

- 152 -

inside attribute amount is $200, computed as the sum of S’s basis in Asset ($100) and its deemed basis in the S1 stock (treated as a single share) ($100, computed as the greater of S’s $100 total basis in the S1 shares and S1’s $50 basis in Asset 1).

(b) S’s aggregate inside loss is therefore $150 ($200 net inside attribute amount minus $50 value of the S share). Accordingly, S’s attribute reduction amount is $150, the lesser of the net stock loss ($200) and the aggregate inside loss ($150). Reg. § 1.1502- 36(d)(8), Ex. 5(i).

(2) Allocation of S’s Attribute Reduction Amount:

(a) Under Reg. §§ 1.1502-36(d)(4) and (d)(5)(ii), S’s $150 attribute reduction amount is allocated proportionately (by basis) between Asset (basis $100) and the S1 stock (treated as a single share) (deemed basis $100). Accordingly, $75 of the attribute reduction amount ($100/$200 × $150) is allocated to Asset and $75 of the attribute reduction amount ($100/$200 × $150) is allocated to the S1 stock.

(b) The $75 allocated to Asset is applied to reduce S’s basis in Asset to $25.

(c) The $75 allocated to the S1 stock is first apportioned between the shares in a manner that reduces disparity to the greatest extent possible. Thus, of the total $75 allocated to the S1 stock, $27.50 is apportioned to Share A and $47.50 is apportioned to Share B. The application of the apportioned amounts reduces the basis of each share to $12.50.

(d) As a result of the application of S’s attribute reduction amount, S’s basis in Asset is $25 and S’s basis in each of the S1 shares is $12.50. Id.

(3) Tier Down of S’s Attribute Reduction Amount (Application of Conforming Limitation):

(a) Under Reg. § 1.1502-36(d)(5)(v)(A), any portion of S’s attribute reduction amount allocated to S1 stock

- 153 -

is an attribute reduction amount of S1 (regardless of the extent to which it is applied to reduce the basis of any shares of S1 stock). Under Reg. § 1.1502- 36(d), the $75 allocated to the S1 stock would be applied to reduce S1’s basis in Asset 1 to $0.

(b) However, under Reg. § 1.1502-36(d)(5)(v)(B), the reduction of S1’s attributes as a result of tier down attribute reduction is limited to $25. This represents the excess of the portion of S1’s net inside attribute amount that is allocable to all S1 shares held by members immediately before the transaction ($50) over the sum of aggregate value of S1 shares transferred by members in the transaction (none) and the aggregate amount of members’ bases in nontransferred S1 shares after reduction under Reg. § 1.1502-36(d)(5)(v)(A) ($25). Therefore, of S1’s $75 tier down attribute reduction amount, only $25 is applied to reduce S1’s basis in Asset, from $50 to $25. The remaining $50 of attribute reduction amount has no further effect. Id.

(4) Basis Restoration:

(a) Assume the same facts as Example 52, except that S’s basis in Share A is $15, S’s basis in Share B is $35, and S1’s basis in Asset 1 is $100.

(b) Reg. § 1.1502-36(d) applies in the same manner as in Example 52, except that the application of the apportioned amounts ($27.50 to Share A and $47.50 to Share B) result in an ELA in each share of ($12.50).

(c) Under Reg. § 1.1502-36(d)(5)(iv)(A), after Reg. § 1.1502-36(d) has been applied with respect to all transfers of subsidiary stock, any reduction made to the basis of a share of subsidiary stock under Reg. § 1.1502-36(d)(5)(iii) is reversed to the extent necessary to conform the basis of that share to the share’s allocable portion of the subsidiary’s net inside attribute amount. S1’s net inside attribute amount after the application of Reg. § 1.1502-36(d) is $25 and thus each of the two S1 share’s allocable portion of S1’s net inside attribute amount is $12.50.

- 154 -

(d) Accordingly, the basis of each share (as reduced by Reg. § 1.1502-36(d)(5)(iii)) are reversed to the extent necessary to restore the basis of each share to $12.50. Thus, $25 of the $27.50 of reduction to the basis of share A, and $25 of the $47.50 of reduction to the basis of share B is reversed, restoring the basis of each share to $12.50. Reg. § 1.1502- 36(d)(8), Ex. 5(ii).

c. Application of Prior Rules: The loss suspension rule does not apply because P and S are deconsolidated. Prior Reg. § 1.1502- 35(c)(1).

D. Intercompany Transfers of Subsidiary Stock Examples

1. Example 53 – Intercompany Sale with Duplicated Loss

P P

S share X M M1 S share $80 M M1 $70 $60 $100 Basis Basis

S S

Asset Asset $100 Basis $90 Basis

a. Facts: M owns the sole outstanding share of stock of S with a basis of $100. S has one asset with a basis of $100. M sells the S share to M1 for $70, recognizing a loss of $30. While owned by M1, S recognizes $10 of depreciation deductions that are absorbed by the group. S's basis in the asset is reduced by $10, from $100 to $90. M1's basis in the S stock is therefore reduced by $10, from $70 to $60. Later, M1 sells the S share to X, an unrelated person, for $80.

- 155 -

b. M's sale of its S share to M1 is a transfer of a loss share, so the Unified Loss Rules would apply when M's intercompany item is taken into account under Reg. § 1.1502-13, as if M and M1 were divisions of a single corporation. If M and M1 were divisions of a single corporation, the S share's basis would be $90, $100 reduced by $10 for the depreciation deductions absorbed by the group, and the group would recognize a $10 loss on the sale of the share that is potentially subject to this section. Thus, the sale would be a transfer of a loss share, subject to the Unified Loss Rules, to the extent of $10.

(1) There is no adjustment under the basis redetermination rule because S has only one share outstanding and so there is no disparity in bases of common shares and no unrecognized gain or loss with respect to preferred shares.

(2) There is no adjustment under the basis reduction rule because S has no net PIAs.

(3) Under the attribute reduction rule, S would be subject to $10 of attribute reduction (the lesser of the $10 net stock loss and S's $10 aggregate inside loss), allocable to the basis in S's asset.

(4) S's basis in its asset is reduced by $10, from $90 to $80, M takes its $30 intercompany stock loss into account, and M1 recognizes a $20 stock gain. See Reg. § 1.1502- 36(e)(3)(iii), Ex. 1(i). c. If M ceases to be a member of the group before M1 sells the S share, M’s intercompany loss must be taken into account, and the Unified Loss Rules apply, at that time. Assuming that the value of the S share is $80 when M deconsolidates, S's basis in its asset is reduced from $90 to $80, M takes its $30 intercompany stock loss into account, and M1 holds the S stock with a basis of $60 (and an unrecognized gain of $20). See Reg. § 1.1502-36(e)(3)(iii), Ex. 1(ii).

- 156 -

2. Example 54 – Intercompany Sale of Built-in Gain Stock

P P

S share X (2) M M1 S share $120 M M1 $100 $100 $200 Basis Basis (1) Asset S S Z $100 Asset $0 Basis

a. Facts: M owns the sole outstanding share of stock of S with a basis of $100. S's sole asset has a basis of $0. S sells its asset for $100 and recognizes a $100 gain that increases M's basis to $200. M sells the S share to M1 for $100 and recognizes a $100 intercompany loss. Later, M1 sells the S share to X, an unrelated person, for $120.

b. M's sale of its S share to M1 is a transfer of a loss share, so the Unified Loss Rules would apply when M's intercompany item is taken into account under Reg. § 1.1502-13, as if M and M1 were divisions of a single corporation. If M and M1 were divisions of a single corporation, the S share's basis would be $200 ($100 increased by $100 for the gain recognized on the sale of the asset) and the group would recognize an $80 loss on the sale of the share. Thus, the sale would be a transfer of a loss share and would be subject to the Unified Loss Rules to the extent of the $80 loss.

(1) There is no adjustment under the basis redetermination rule because S has only one share outstanding and so there is no disparity in bases of common shares and no unrecognized gain or loss with respect to preferred shares.

(2) Under the basis reduction rule, the basis in the S share would be reduced, but not below its $120 value, by the

- 157 -

lesser of the $100 disconformity amount and the $100 net PIA that was applied to the share when held by M. The basis in the S share would be reduced by $80, to $120.

(3) Because the S share would not be a loss share after the application of the basis reduction rule, the attribute reduction rule would not apply.

(4) M's intercompany item is adjusted to reflect what it would have been had M's basis in its S share been reduced by $80 immediately before its sale to M1. Thus, M's intercompany loss is reduced to $20 and M takes this loss into account, and M1 recognizes a gain of $20. See Reg. § 1.1502- 36(e)(3)(iii), Ex. 2.

3. Example 55 – Intercompany Sale Creates Built-in Gain Stock

P P

(2) S Share S Share

M M1 M M1 X $100 $120

$0 Basis $200 Basis (1) Asset S S Z $100 Asset Asset $0 Basis $0 Basis

a. Facts: M owns the sole outstanding share of stock of S with a basis of $0. S's sole asset has a basis of $0. M sells the S share to M1 for $100 and recognizes a $100 intercompany gain. While owned by M1, S sells its asset for $100, recognizing a $100 gain that increases M1's basis in the S share to $200. Later, M1 sells the S share to X for $120.

b. M's sale of its S share to M1 is a transfer of a share, so the Unified Loss Rules would apply when M's intercompany item is taken into account under Reg. § 1.1502-13, as if M and M1 were divisions of

- 158 -

a single corporation. If M and M1 were divisions of a single corporation, the S share's basis would be $100 ($0 increased by $100 for the gain recognized on the sale of the asset) and the group would recognize a $20 gain on the sale of the share. Thus, the sale would not be a transfer of a loss share and the Unified Loss Rules would not apply. As a result, no portion of M1’s $80 loss is subject to the Unified Loss Rules. M takes its $100 intercompany stock gain into account, and M1 recognizes an $80 loss. See Reg. § 1.1502-36(e)(3)(iii), Ex. 3.

4. Example 56 – Subsidiary with Built-in Gain and Built-in Loss Assets

P P

S Share S Share M M1 M M1 X $90 $90

$100 Basis $150 Basis (1) Asset 1 S S Z $60 Asset 1 Asset 2 Asset 2 $0 Basis $80 Basis $80 Basis

a. Facts: M owns the sole outstanding share of stock of S with a basis of $100. S has two assets—Asset 1 with a basis of $0 and Asset 2 with a basis of $80. M sells the S share to M1 for $90 and recognizes a $10 intercompany loss. While owned by M1, S sells Asset 1 for $60, recognizing a $60 gain that increases M1's basis in the S share to $150. Later, M1 sells the S share to X for $90.

b. M's sale of the S share to M1 is a transfer of a loss share, so the Unified Loss Rules would apply when M's intercompany item is taken into account under Reg. § 1.1502-13, as if M and M1 were divisions of a single corporation. If M and M1 were divisions of a single corporation, the S share's basis would be $160 ($100 increased by $60 for the gain recognized on the sale of Asset 1) and the group would recognize a $70 loss on the sale of the share. Thus, the sale would be a transfer of a loss share, and would be subject to the Unified Loss Rules to the extent of the $70 loss.

- 159 -

(1) There is no adjustment under the basis redetermination rule because S has only one share outstanding and so there is no disparity in bases of common shares and no unrecognized gain or loss with respect to preferred shares.

(2) Under the basis reduction rule, the basis in the S share would be reduced, but not below its $90 value, by the lesser of the $20 disconformity amount ($160 stock basis over $140 net inside attribute amount) and the $60 net PIA that was applied to the share when held by M1. The basis in the S share would be reduced by $20, to $140.

(3) Under the attribution reduction rule, S would have an attribute reduction amount of $50, the lesser of the $50 net stock loss ($140 basis over $90 value) and S's $50 aggregate inside loss (the excess of the sum of S's $80 basis in Asset 2 and S's $60 cash from the sale of Asset 1, over the $90 value of the S share).

(4) Because the positive adjustment was applied to the share when held by M1, the $20 basis reduction required under the attribution reduction rule is applied to M1's basis in its S share immediately before its sale to X, reducing it from $150 to $130. In addition, S's basis in Asset 2 is reduced by $50, from $80 to $30. M takes its $10 intercompany stock loss into account and M1 recognizes a loss of $40. See Reg. § 1.1502-36(e)(3)(iii), Ex. 5(i).