REG-157711-02

From TaxAlmanac

You Are Here: Main Page > Tax Research Resources > Proposed Regulations > REG-157711-02

Internal Revenue Bulletin:

2007-8 February 20, 2007

REG-157711-02

Notice of Proposed Rulemaking Unified Rule for Loss on Subsidiary Stock

AGENCY:

Internal Revenue Service (IRS), Treasury.

ACTION:

Notice of proposed rulemaking.

SUMMARY:

This document contains proposed regulations under sections 358, 362(e)(2) and 1502 of the Internal Revenue Code (Code). The regulations apply to corporations filing consolidated returns. The regulations implement aspects of the repeal of the General Utilities doctrine by redetermining members’ bases in subsidiary stock and requiring certain reductions in subsidiary stock basis on a transfer of the stock. The regulations also promote the clear reflection of income by redetermining members’ bases in subsidiary stock and reducing the subsidiary’s attributes to prevent the duplication of loss. Additionally, the regulations provide guidance limiting the application of section 362(e)(2) with respect to transactions between members of a consolidated group.

DATES:

Written or electronic comments or a request for a public hearing must be received by April 23, 2007.

ADDRESSES:

Send submissions to: CC:PA:LPD:PR (REG-157711-02), room 5203, Internal Revenue Service, PO Box 7604, Ben Franklin Station, Washington, DC 20044. Submissions may be hand-delivered Monday through Friday between the hours of 8 a.m. and 4 p.m. to CC:PA:LPD:PR (REG-157711-02), Courier’s Desk, Internal Revenue Service, 1111 Constitution Avenue, N.W., Washington, DC, or sent electronically, via the IRS Internet site at www.irs.gov/regs or via the Federal eRulemaking Portal at www.regulations.gov (IRS/REG-157711-02).

FOR FURTHER INFORMATION CONTACT:

Concerning the proposed regulations, Theresa Abell (202) 622-7700 or Phoebe Bennett (202) 622-7770; concerning submissions of comments, Richard Hurst, Richard.A.Hurst@irscounsel.treas.gov, (202) 622-7180 (not toll-free numbers).

SUPPLEMENTARY INFORMATION:

Paperwork Reduction Act

The collection of information contained in this notice of proposed rulemaking has been submitted to the Office of Management and Budget for review in accordance with the Paperwork Reduction Act of 1995 (44 U.S.C. 3507(d)). Comments on the collection of information should be sent to the Office of Management and Budget, Attn: Desk Officer for the Department of the Treasury, Office of Information and Regulatory Affairs, Washington, DC 20503, with copies to the Internal Revenue Service, Attn: IRS Reports Clearance Officer, SE:W:CAR:MP:T:T:SP, Washington, DC 20224. Comments on the collection of information should be received by March 26, 2007.

Comments are specifically requested concerning:

Whether the proposed collection of information is necessary for the proper performance of the functions of the IRS, including whether the information will have practical utility;

The accuracy of the estimated burden associated with the proposed collection of information;

How the quality, utility, and clarity of the information to be collected may be enhanced;

How the burden of complying with the proposed collection of information may be minimized, including through the application of automated collection techniques or other forms of information technology; and

Estimates of capital or start-up costs and costs of operation, maintenance, and purchase of services to provide information.

The collection of information in these proposed regulations is in §§1.1502-13(e)(4)(v) and 1.1502-36(d)(6). The respondents are corporations filing consolidated returns. The collection of information is required to allow a corporation to preserve a subsidiary’s attributes by foregoing a stock loss. The collection of information is required to obtain a benefit.

Estimated total annual reporting and/or recordkeeping burden: 25 hours.

Estimated average annual burden per respondent and/or recordkeeper: 15 minutes.

Estimated number of respondents and/or recordkeepers: 100.

Estimated annual frequency of responses: Once.

An agency may not conduct or sponsor, and a person is not required to respond to, a collection of information unless it displays a valid control number assigned by the Office of Management and Budget.

Books or records relating to the collection of information must be retained as long as their contents may become material in the administration of any Internal Revenue law. Generally, tax returns and tax return information are confidential, as required by 26 U.S.C. 6103.

Background

The discussion in this preamble begins with an overview of the history of the regulatory attempts to address both the circumvention of General Utilities repeal and the duplication of loss by consolidated groups, in particular, in §1.1502-20 (the Loss Disallowance Rule, or LDR). The discussion then turns to Rite Aid Corp. v. United States, 255 F.3d 1357 (Fed. Cir. 2001), which rejected the loss duplication rule in the LDR. Section A.4 of this preamble discusses the immediate administrative responses to Rite Aid. Section A.5 of this preamble discusses the legislative response to Rite Aid. Following the Rite Aid decision, the IRS and Treasury Department undertook a study to reconsider the issues addressed by §1.1502-20. Section B of this preamble discusses the various issues considered in that study, including both the original noneconomic and duplicated stock loss specifically addressed by the LDR and certain related issues with which the Internal Revenue Service and Treasury Department have grown concerned since the LDR was promulgated. Section C of this preamble describes the various approaches that were considered to address noneconomic stock loss and sets forth the conclusions reached regarding each. Section D of this preamble describes the various approaches that were considered to address loss duplication and sets forth the conclusions reached regarding each. Section E of this preamble describes the various approaches that were considered to address the noneconomic and duplicated loss that can arise from the general operation of the investment adjustment system and sets forth the conclusions reached regarding each. Section F of this preamble describes the specific provisions of this proposed regulation §1.1502-36. Section G of this preamble discusses the proposed removal of §§1.337(d)-1, 1.337(d)-2, and 1.1502-35.

The IRS and Treasury Department are also proposing regulations to address the application of section 362(e)(2) to members of consolidated groups. These proposed regulations are described in section H of this preamble.

Finally, the IRS and Treasury Department are proposing various technical and administrative revisions to the consolidated return regulations. These proposed regulations are described in section I of this preamble.

The IRS and Treasury Department request comments on the proposed regulations and other approaches that could be adopted, as well as other issues currently under study. See section J of this preamble for further discussion of comments requested.

A. History of General Utilities Repeal and Loss Disallowance under §1.1502-20.

1. The repeal of the General Utilities doctrine. In 1986, Congress enacted section 337(d), which directs the Secretary to prescribe such regulations as may be necessary or appropriate to carry out the repeal of the General Utilities doctrine (GU repeal). See Tax Reform Act of 1986, Public Law 99-514 (100 Stat. 2085 (1986)). The legislative history states that Congress was concerned that the General Utilities doctrine allowed “assets to leave corporate solution and to take a stepped-up basis in the hands of the transferee without the imposition of a corporate-level tax” and thus “tend[ed] to undermine the corporate income tax.” H.R. Rep. No. 99-426, 99^th Cong., 1^st Sess. 282 (1985). The General Utilities doctrine and GU repeal are discussed extensively in the Treasury Decisions referenced in this preamble; in addition, see generally, H.R. Rep. No. 99-426 at 274-282 for a discussion of the history of the General Utilities doctrine; see also General Utilities & Operating Co. v. Helvering, 296 U.S. 200 (1935).

2. The administrative response to GU repeal: §1.1502-20. The IRS and Treasury Department first responded to GU repeal by issuing Notice 87-14, 1987-1 C.B. 445, which set forth the intent to promulgate regulations affecting adjustments to members’ bases in stock of any subsidiary acquired when the subsidiary held an appreciated asset. Notice 87-14 indicated that, in general, adjustments to subsidiary stock basis would not reflect gains on such assets. Thus, Notice 87-14 implied that a tracing-based regime would be adopted to determine adjustments to members’ bases in shares of subsidiary stock.

After several years of study, the IRS and Treasury Department concluded that any approach relying on the identification and tracing of appreciation on particular assets, while theoretically accurate, would impose substantial administrative burdens on taxpayers and on the government. See T.D. 8294, 1990-1 C.B. 66 [55 FR 9426, 9428] (March 14, 1990). As a result, the tracing-based approach envisioned in Notice 87-14 was implemented only in regulations promulgated under section 337(d). Those regulations applied only for the period of time between the issuance of Notice 87-14 and the effective date of final regulations under §1.1502-20 (February 1, 1991). See T.D. 8364, 1991-2 C.B. 43 [56 FR 47379] (September 19, 1991), §§1.337(d)-1 and 1.337(d)-2 (as contained in 26 CFR part 1 revised as of April 1, 1991).

In lieu of tracing, the LDR used certain operating presumptions to determine the extent to which investment adjustments would be permitted to give rise to allowable stock loss. Because the LDR only disallowed loss, noneconomic investment adjustments were able to increase stock basis and thus reduce gain without limitation. As a result, the LDR reduced the duplication of gain in the tax system. The IRS and Treasury Department considered the reduction of gain duplication an important balance to the imprecision inherent in the LDR’s use of irrebuttable presumptions.

The study following the issuance of Notice 87-14 led the IRS and Treasury Department to consider the issue of loss duplication by members of consolidated groups. Their conclusion was that loss duplication was inappropriate in the consolidated setting. Further, the IRS and Treasury Department recognized that there were administrative advantages to addressing both issues in a single integrated rule. Thus, unlike the regulations under section 337(d), the LDR was at once directed at both the circumvention of GU repeal through the use of noneconomic stock loss and the duplication of loss. See T.D. 8294 and T.D. 8364.

3. The Rite Aid opinion. Ten years after the promulgation of the LDR, the validity of the duplicated loss component of the LDR was considered in Rite Aid, supra. Under the duplicated loss component of the LDR, Rite Aid had been disallowed a deduction for an economic loss on subsidiary stock solely because the stock loss could be duplicated by the subsidiary after it left the group. The Federal Circuit stated that the Secretary’s authority to change the application of a Code provision to a consolidated group was limited to situations in which the change was necessary to address a problem created by the filing of a consolidated return. Because duplicated stock loss occurs and is allowable in the separate return setting, the court concluded that the duplicated loss component of the LDR was not addressing a problem arising from the filing of a consolidated return. Accordingly, the court held that the Secretary did not have the authority to change the Code rule allowing a deduction for the stock loss.

4. The administrative response to Rite Aid. In response to the Rite Aid decision, on February 19, 2002, the IRS announced that it would not continue to litigate the validity of the duplicated loss rule in §1.1502-20. See Notice 2002-11, 2002-1 C.B. 526. On March 7, 2002, the IRS and Treasury Department promulgated §1.1502-20T(i) (to suspend the application of the LDR) and §1.337(d)-2T (to provide an interim rule addressing noneconomic stock loss). See T.D. 8984, 2002-1 C.B. 668 [67 FR 11034] (March 12, 2002). Concurrently with the promulgation of §§1.337(d)-2T and 1.1502-20T(i), the IRS issued Notice 2002-18, 2002-1 C.B. 644, announcing that loss duplication regulations would also be promulgated. Following the publication of T.D. 8984, the IRS and Treasury Department undertook a study of the issues underlying both noneconomic and duplicated loss on subsidiary stock.

In general, §1.337(d)-2T disallowed stock loss and reduced stock basis (to value) upon the disposition or deconsolidation of subsidiary stock by a member of a consolidated group. However, under §1.337(d)-2T(c)(2), loss disallowance and basis reduction were avoided to the extent the taxpayer could establish that the loss or basis “is not attributable to the recognition of built-in gain on the disposition of an asset.” Section 1.337(d)-2T(c)(2) defined the term “built-in gain” as gain that is “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, directly or indirectly, in whole or in part.”

On March 14, 2003, the IRS and Treasury Department promulgated §1.1502-35T as an interim measure to address the problem of loss duplication in consolidated groups. See T.D. 9048, 2003-1 C.B. 644 [68 FR 12287] (March 14, 2003). In the preamble to T.D. 9048, the IRS and Treasury Department announced that the issues addressed in §1.1502-35T were still under study. The provisions of §1.1502-35 are discussed in more detail in section D.1 of this preamble.

Further guidance on the interim rules was issued August 25, 2004, in the form of Notice 2004-58, 2004-2 C.B. 520. In Notice 2004-58, the IRS announced that it would accept the “basis disconformity” method as an alternative approach to determining whether stock loss or basis was attributable to “built-in gain” within the meaning of §1.337(d)-2T.

Under the basis disconformity method described in Notice 2004-58, stock loss or basis is treated as attributable to built-in gain to the extent of the least of (i) the net positive investment adjustment applied to the stock basis (disregarding distributions), (ii) the aggregate gain (net of directly related expenses) recognized on asset dispositions by the subsidiary, and (iii) the disconformity amount (generally, the amount by which the basis of the share exceeds the share’s proportionate interest in the subsidiary’s net inside asset basis; for this purpose, net inside asset basis is defined as the excess of the sum of the subsidiary’s money, asset basis, loss carryforwards, and deferred deductions over its liabilities). Notice 2004-58 also requested comments on the general scope of GU repeal and on other approaches that could be adopted to safeguard the purposes of GU repeal in the consolidated return context.

5. The legislative response to Rite Aid. Congress responded to the Rite Aid opinion on October 22, 2004, in the American Jobs Creation Act (the AJCA), Public Law No. 108-357 (118 Stat. 1418 (2004)). In the AJCA, Congress added a sentence at the end of section 1502 of the Code, so that the section now reads:

The Secretary shall prescribe such regulations as he may deem necessary in order that the tax liability of any affiliated group of corporations making a consolidated return and of the each corporation in the group, both during and after the period of affiliation, may be returned, determined, computed, assessed, collected, and adjusted, in such manner as clearly to reflect the income tax liability and the various factors necessary for the determination of such liability, and in order to prevent avoidance of such tax liability. 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.

In the legislative history to the AJCA, Congress stated that the Secretary is authorized to change the application of a Code provision when the Secretary determines it is necessary to clearly reflect the income tax liability of the group and each corporation in the group, both during and after the period of affiliation. See H.R. Conf. Rep. No. 108-755, 108th Cong., 2d Sess. 653 (2004). Congress thus rejected the suggestion in the Rite Aid opinion that the Secretary’s authority to change the general application of the Code is limited to promulgating regulations that address problems created by the filing of a consolidated return.

In the AJCA legislative history, Congress also spoke to the proper scope of future regulations. Regarding the promulgation of regulations addressing noneconomic stock loss, Congress stated that “presumptions and other simplifying conventions” could be used to prevent the circumvention of GU repeal. See H.R. Conf. Rep. No. 108-755, fn. 595. In addition, Congress indicated two acceptable methods for addressing loss duplication by group members. The first would disallow subsidiary stock loss to the extent it duplicates losses that remain available to the group. The second would reduce the subsidiary’s attributes in order to prevent the subsidiary from using losses outside the group, to the extent the losses duplicate stock loss. But Congress also stated its intention that the result of the Rite Aid decision is to be preserved. The IRS and Treasury Department interpret this statement to mean that regulations addressing loss duplication by consolidated groups must not disallow a deduction for an economic loss on subsidiary stock solely because the stock loss duplicates unrecognized or unabsorbed losses that later could be used outside the group.

6. Further administrative response to Rite Aid. On March 3, 2005, the IRS and Treasury Department finalized §1.337(d)-2. See T.D. 9187, 2005-1 C.B. 778 [70 FR 10319] (March 3, 2005). In T.D. 9187, the IRS and Treasury Department stated that the issues addressed in §1.337(d)-2 were still under study and that an alternative approach would be proposed. On March 14, 2006, the IRS and Treasury Department finalized §1.1502-35. See T.D. 9254, 2006-13 I.R.B. 662 [71 FR 13008] (March 14, 2006). In T.D. 9254, the IRS and Treasury Department stated that both noneconomic and duplicated loss were still under study, and that regulations would be proposed adopting a singe integrated approach to addressing both issues. The results of that study and the proposed integrated approach are described below in sections D through H of this preamble.

B. Issues Considered in the Post-Rite Aid Study.

1. GU repeal and noneconomic investment adjustments under the LDR. Section 337(d) generally directs the Secretary to prescribe regulations to prevent the circumvention of GU repeal and, in particular, section 337(d)(1) directs the Secretary to promulgate regulations to prevent the circumvention of GU repeal through the use of the consolidated return regulations. Congress’ concern stems from the general operation of the investment adjustment system of §1.1502-32.

The purpose of the investment adjustment system is to promote the clear reflection of the group’s income. See §1.1502-32(a)(1). One of the principal ways that the investment adjustment system promotes clear reflection is by preventing a subsidiary’s items of income, gain, deduction and loss from giving rise to duplicative gain or loss on the subsidiary’s stock. To that end, the investment adjustment system adjusts members’ bases in shares of subsidiary stock to reflect such items once they have been taken into account by the group. See T.D. 8560, 1994-2 C.B. 200 [59 FR 41666] (August 15, 1994).

Example 1. Economic adjustment to stock basis prevents duplication. P, the common parent of a consolidated group, purchases all 100 outstanding shares of S common stock for $100 cash, taking a basis of $1 in each share. At the time, S owns one asset, A1, with a basis and value of $100. Later, the value of A1 increases to $150. S sells A1 to a nonmember for $150 and recognizes a $50 gain, which the P group takes into account. Under the investment adjustment system, P increases its basis in its S stock to reflect the $50 taken into account by the group. As a result, the basis of each share increases to $1.50, its fair market value. P can then sell all or any portion of its S stock for its fair market value without recognizing duplicative gain on the disposition.

The result in Example 1 is that the group takes its economic gain into account only once, on the disposition of S’s asset, and not again on the subsequent disposition of the S stock. Thus the group’s income is clearly reflected and there is no circumvention of GU repeal.

The investment adjustment system is not a tracing regime. Rather, it is a presumptive regime based on certain operating assumptions. A principal assumption is that all of a subsidiary’s items taken into account represent economic accruals (of gain or loss) to the group. Another principal assumption is that all such items accrue equally to all outstanding shares, at least within a class. When these assumptions correspond to the facts of a particular situation, as in Example 1, the investment adjustment system produces appropriate results: stock basis, which reflects only the investment in the stock, increases to reflect economic accrual (the group’s return on its stock investment), and, as a result, stock basis can then shelter that return on the group’s investment, protecting it from being taken into account again when the stock is sold.

The assumptions, however, do not correspond to the facts of all situations. For example, if stock of a subsidiary is purchased for its fair market value when the subsidiary holds appreciated assets, the items of income or gain generated when that appreciation is recognized do not represent an economic accrual on the group’s investment (because the appreciation was already reflected in the basis of the stock). Nevertheless, the presumptive rules of the investment adjustment system treat such items as economic accruals and include them in the investment adjustment to be applied to the basis of the stock.

Example 2. Noneconomic adjustment to stock basis creates noneconomic stock loss. Assume the same facts as in Example 1 except that P does not purchase the stock of S until the value of A1 has increased to $150. Accordingly, P purchases the stock for $150, taking a basis of $1.50 in each share. As in Example 1, when S sells A1, the investment adjustment system again increases P’s basis in its S stock to reflect the $50 taken into account by the group. As a result, P’s basis in each of its shares increases to $2, even though the fair market value of each share remains $1.50. If P were then to sell all or some portion of the S stock for its fair market value, P would recognize a $.50 loss on each share ($50 loss in the aggregate).

In this situation, a deduction for the stock loss would be inappropriate because neither the group nor its members have suffered any economic loss. If P were allowed to deduct that noneconomic loss, the deduction would offset the gain recognized on S’s asset and, effectively, eliminate the corporate-level tax on the gain on S’s asset. This is the circumvention of GU repeal that concerned Congress in 1986.

At the time Notice 87-14 was issued, the IRS and Treasury Department had identified the creation of noneconomic stock loss in situations similar to those illustrated in Example 2. Thus, Notice 87-14 referred specifically to investment adjustments attributable to the disposition of assets that, at the time of the acquisition of the subsidiary stock, had a fair market value in excess of adjusted basis. For that reason, §1.337(d)-1, which implemented Notice 87-14, disallowed subsidiary stock loss unless the taxpayer could show that the loss was not attributable to the recognition of appreciation on assets owned, directly or indirectly, by a subsidiary when it became a member.

2. Duplicated loss and the clear reflection of group income under the LDR. In the study that followed the issuance of Notice 87-14, the IRS and Treasury Department also considered the issue of loss duplication by members of a consolidated group. The specific concern of the IRS and Treasury Department was the loss duplication that occurs when an economic loss is reflected in both a member’s basis in subsidiary stock and in the subsidiary’s assets or operations, and the loss is first recognized with respect to the stock.

Example 3. Duplication of loss. P forms S by contributing $110 to S in exchange for all 100 outstanding shares of S stock. S uses the cash to purchase an asset, A1. The value of A1 later declines to $10. If P were then to sell all or some portion of the S stock for its fair market value, P would recognize a $1 loss on each share.

In this situation, even though P would have recognized the group’s economic loss on its disposition of the S stock, the loss continues to be reflected in the basis of A1. As a result, that loss would remain available for use by P (if the stock sale did not deconsolidate S) or S (if the stock sale deconsolidated S). Upon the disposition of A1, the group’s single economic loss would thus be recognized and taken into account more than once by the group and its members or former members.

In contrast, if the duplicated loss had first been taken into account with respect to A1, the investment adjustment system would have prevented a duplicative benefit to the group and its members by reducing P’s basis in S stock by the amount of the loss. In that case, the group would have enjoyed the tax benefit attributable to the loss, but that benefit would not remain available for another use by the group and its members or former members.

The IRS and Treasury Department concluded that the duplication of a group’s tax benefit (represented by a single economic loss) distorts income without regard to whether the duplicated loss is taken into account first with respect to the subsidiary’s stock or first with respect to the subsidiary’s assets and operations. The IRS and Treasury Department further concluded that, even if the duplicated loss is used by a former member outside the group, that duplicative use distorts the income of the group and its members. Accordingly, the IRS and Treasury Department decided to promulgate regulations that would complement the investment adjustment system by addressing the stock-first recognition of a duplicated loss and that such regulations would apply to both deconsolidating and nondeconsolidating dispositions. Recognizing the administrative benefits of addressing both noneconomic and duplicated stock loss in a single integrated rule, the IRS and Treasury Department promulgated the LDR as a single rule with components directed at both.

The method adopted by the LDR to address loss duplication was the disallowance of stock loss (or reduction of stock basis) that duplicated unrecognized inside loss, such as that illustrated in Example 3. However, groups had several mechanisms available to recognize or preserve the inside loss and thereby avoid loss disallowance (by eliminating loss duplication). Inside losses could be recognized through an actual asset sale or a deemed asset sale under section 338(h)(10), and, following the sale, the subsidiary’s unabsorbed losses would be available to the group. In addition, the LDR allowed the common parent to elect to reattribute the subsidiary’s losses (to itself) under §1.1502-20(g). If the group chose not to exercise those options, then the stock loss was denied, but the inside loss was preserved for a nonduplicative use by the subsidiary, in or out of the group.

At the time the LDR was promulgated, the duplication potential illustrated in Example 3 was the principal form of loss duplication with which the IRS and Treasury Department were concerned. Thus it is the only form of loss duplication specifically addressed by the LDR. The anti-abuse rule in the LDR did, however, provide a limited mechanism for expanding the scope of that provision.

3. Noneconomic and duplicated loss resulting from investment adjustments allocated to shares with disparate bases. Since the promulgation of the LDR, the IRS and Treasury Department have become increasingly concerned with the noneconomic and duplicated loss potential arising from the interaction of §1.1502-32 and the disparate reflection of gain or loss in members’ bases in individual shares of subsidiary stock.

As discussed in section B.1 of this preamble, the investment adjustment system is a presumptive regime that allocates a subsidiary’s items of income, gain, deduction, and loss taken into account by the group. It operates in accordance with the assumption that all such items reflect economic accruals to all shares equally within each class. When its underlying assumptions correspond to the facts of a particular situation, the investment adjustment system produces appropriate results, as illustrated in Example 1. But when its underlying assumptions do not correspond to the facts of a situation because shares held by members have disparate bases, the general operation of the investment adjustment system can give rise to both noneconomic and duplicated loss on individual shares of subsidiary stock.

Example 4. Noneconomic loss. P and M (a member of the P group) form S by contributing property to S in exchange for all 100 outstanding shares of S stock. P contributes A1, with a basis and value of $80, in exchange for 80 shares of S stock. M contributes A2, with a basis of $0 and a value of $20, to S in exchange for 20 shares of S stock. S then sells A2 for $20 and recognizes a $20 gain that is taken into account by the group. As a result, the basis of each share increases by $.20. P’s basis in each of its shares is then $1.20 (or, $96 in the aggregate), and M’s basis in each of its shares is then $.20 (or, $4 in the aggregate), even though the value of each share remains $1. P then sells all or some portion of its shares to X, a nonmember, and, under general principles of tax law, recognizes a $.20 noneconomic loss on each share, effectively eliminating up to $16 of the gain on A2.

Example 5(a). Duplicated loss, inside recognition precedes stock disposition. P forms S with $100 and receives all 50 shares of S common stock. S uses the $100 to buy A1, which then declines in value to $50. P contributes another $50 for a second 50 shares of common stock. S then sells A1 and recognizes a loss of $50 that is taken into account on the P group return. The absorption of the $50 loss results in a $.50 reduction to the basis of each share (original and newly issued). P then sells all or some portion of the original shares to X for $1 each (each with a basis of $1.50) and recognizes a $.50 loss on each share (up to $25 total). Although the $50 asset loss and the $25 stock loss both reflect an economic loss of the group, they are both reflecting the same loss. The group has actually experienced only $50 of economic loss. Therefore, the $.50 loss recognized on each of the original shares (up to $25 total) is duplicative.

Example 5(b). Duplicated loss, stock disposition precedes inside recognition. The facts are the same as in Example 5(a), except that, before S sells A1, P sells 20 of its original 50 shares to X for $20 (aggregate basis $40), recognizing a $20 loss that is taken into account on the P group return, and S remains a member of the group. S then sells A1, recognizing a $50 loss that is taken into account on the P group return. Although the $50 asset loss and the $20 stock loss both reflect an economic loss of the group, they are both reflecting the same loss. As in Example 5(a), the group has actually experienced only $50 of economic loss. Therefore, $20 of the recognized loss is duplicative. Alternatively, if P sold all its original 50 shares, P would recognize a $50 loss even though the entire $50 group loss would remain available to S for a duplicative use against its separate year income.

The IRS and Treasury Department recognize that, in each case where the disproportionate reflection of an item in a particular share causes an inappropriate stock loss, whether noneconomic or duplicated, that loss is offset by unrecognized gain in other shares. However, that gain can be deferred indefinitely or even eliminated by the group. Accordingly, the IRS and Treasury Department do not believe that the system is appropriately balanced in such cases.

The IRS and Treasury Department further recognize that these issues could be addressed by adopting a tracing-based approach to the allocation of investment adjustments. However, the complexity and burden of a tracing-based approach would render such an approach generally inadministrable for consolidated taxpayers and for the government. As a result, the system would be prone to error and, in practice, inconsistently applied. Moreover, the IRS and Treasury Department continue to believe that the assumptions on which the investment adjustment system is based are appropriate for typical commercial transactions, as the IRS and Treasury Department understand that typically subsidiaries have only common stock outstanding, that their stock is wholly owned by group members, and that members’ bases in shares of subsidiary stock are uniform, as under the facts of Example 1. See section E.2 of the preamble of CO-30-92, 1992-2 C.B. 627 [57 FR 53634, 53639] (November 12, 1992).

Because a tracing-based approach to the allocation of investment adjustments would not be administrable, the IRS and Treasury Department are not considering revising the investment adjustment system to adopt such an approach. Instead, the IRS and Treasury Department have considered various presumptive approaches that could be adopted to mitigate the creation of noneconomic and duplicated loss when members hold subsidiary stock with disparate bases. The approaches considered and decisions reached are discussed in section E of this preamble.

4. Redetermination events: changes in the extent that unrecognized gain or loss is effectively reflected in the basis of individual shares. Because the investment adjustment system adjusts the basis of each share in accordance with its proportionate interest in S’s assets and operations, the relationship between a share’s basis and its allocable portion of unrecognized appreciation or depreciation determines the extent to which such amounts are effectively reflected in the basis of the share. This relationship, however, is not fixed at the time that stock is acquired. The reason is that there are many transactions, referred to here as redetermination events, that alter either the basis of a share or the interest it represents. These events generally occur in one of three types of situations.

a. Stock basis is reallocated. The relationship between the basis of a share and the interest represented by the share can be altered whenever stock basis is reallocated among shares, including when it is allocated to shares of stock of other members.

Example 6. Intragroup spin-off. P forms S by contributing $100 to S in exchange for all the stock of S. S purchases two assets, A1 and A2, for $50 each. Subsequently, A1 appreciates to $75 and A2 depreciates to $25. In a transaction qualifying under sections 355 and 368(a)(1)(D), S transfers A2 to C in exchange for all of the C stock and S then distributes all the C stock to P. Under section 358 and §1.358-2, P’s basis in the S stock is allocated among the S and C stock in proportion to the value of the stock of S and C. As a result, P’s basis in its S stock is $75 (75/100 x $100) and P’s basis in its C stock is $25 (25/100 x $100). S sells A1 for $75, recognizing a $25 gain that is taken into account on the P group return. P’s basis in its S stock increases by $25, from $75 to $100. P then sells its S stock for $75 and recognizes a $25 loss.

In this Example 6, after the reallocation of stock basis, P’s basis in its S stock reflects the unrecognized appreciation on A1, just as P’s basis in its S stock reflected unrecognized appreciation on A1 in Example 2. As a result, P’s reallocated S stock basis protects the appreciation on A1 from being recognized as both asset gain and stock gain. Increasing P’s basis in its S stock to reflect the recognition of S’s gain on A1 is not only unnecessary, it inflates stock basis and thereby gives rise to either noneconomic loss or noneconomic reduction of gain when the stock is sold.

Basis reallocations, and the consequences described, can occur for a number of reasons, including, for example, under rules like §1.1502-32(c)(4) (cumulative redetermination of investment adjustments) and §1.1502-35(b) (basis redetermination to reduce disparity) and the corresponding provision in these proposed regulations.

b. Capital transactions expand or contract the subsidiary’s pool of assets. The relationship between the basis of a share and the nature of the interest represented by the share can also be altered by capital transactions that have no effect on the basis or value of outstanding shares, but that nevertheless alter the interest represented by those share. Some common examples arise in the context of section 351 exchanges, even though, as illustrated in Example 7(a), a section 351 exchange in its simplest form cannot give rise to stock basis that reflects unrecognized appreciation.

Example 7(a). Contribution of appreciated asset in section 351 exchange. P forms S by contributing an asset, A1, to S in exchange for all 80 outstanding shares of S stock. The basis of A1 is $40 and its value is $80. S sells A1 and recognizes a $40 gain that is taken into account by the P group. As a result, P’s aggregate basis in its S shares is increased by $40, from $40 to $80. Subsequently, P sells its S stock for $80, the stock’s fair market value, and recognizes $0 on the sale. The group is thus taxed once on its $40 economic gain.

In Example 7(a), P holds appreciated S stock and S holds an appreciated asset, but that appreciation is not reflected in either P’s basis in its S stock or S’s basis in its asset. Each share has a basis of $.50 and an interest in 1/80 of S’s asset, A1, which has $40 of unrecognized appreciation (allocable $.50 to each share). If this relationship between P’s basis in its S shares and the interest represented by the shares remains constant, as in Example 7(a), the investment adjustment system produces appropriate results. But if there is a change in that relationship, the underlying assumptions of the investment adjustment system may no longer correspond to the facts of the situation and, as a result, the general operation of the system could produce inappropriate results. Such changes can occur whenever S acquires property in exchange for additional shares of its stock.

Example 7(b). Contribution of appreciated asset in subsequent section 351 exchange creates disconformity in original shares. The facts are the same as in Example 7(a), except that, before A1 is sold, P contributes a second asset, A2, to S in exchange for an additional 20 shares of S stock. A2 has a basis of $0 and a value of $20. S sells both assets and recognizes a $60 gain that is taken into account by the P group. As a result, P’s basis in its original shares increases by $48 ($.60 per share), from $40 to $88 (or, from $.50 to $1.10 per share), and P’s basis in its new shares increases by $12, from $0 to $12 (or, from $0 to $.60 per share). P then sells 20 of its original shares (basis of $22) for $20, their fair market value, and recognizes a $2 loss.

In Example 7(b), P’s basis in the original S stock reflected no unrecognized appreciation when the stock was issued. After the second contribution, however, P’s basis in those shares reflects a portion of the unrecognized appreciation on A2. The reason is that each share represents an interest in S’s entire pool of assets. When the pool changes, the nature of the interest represented by the shares changes, even though the share’s basis and value remain constant. Thus, in Example 7(b), while each original share’s basis ($.50) and value ($1) remain constant, the interest represented by each share changed from 1/80 of an asset with unrecognized appreciation of $40 (or, $.50 per share), to 1/100 of assets with unrecognized appreciation of $60 (or, $.60 per share). This shift causes the basis of each original share to reflect $.10 of unrecognized appreciation. When the gain is recognized, $.10 of the gain allocated to each original share under the investment adjustment system is a noneconomic increase in the share’s basis. That increase will give rise to noneconomic stock loss or gain reduction. Although this (noneconomic) allocation of the (economic) item results in an offsetting stock gain on the basis of the new shares, that gain can be indefinitely deferred and even eliminated.

The principles that increase the reflection of unrecognized appreciation in the original shares in Example 7(b) can also cause the reflection of unrecognized appreciation in the basis of shares that are received in exchange for property that is not appreciated, including cash. Although such shares would have a substituted basis (which generally precludes the reflection of unrecognized appreciation, as illustrated in Example 7(a)), the reflection of unrecognized appreciation is prevented only if the shares represent, wholly and solely, the transferee’s interest in its transferred property. If there are previously issued shares outstanding, or if other shares are issued in the exchange, the shares represent an interest in a pool of assets that includes more than the transferred assets. As a result, the interest represented by each such share may be significantly different from what it would be if the subsidiary held only the transferred property.

Example 7(c). Multiple transferors in single section 351 exchange. The facts are the same as in Example 7(a), except that, when P contributes A1 to S in exchange for 80 shares of S stock, M (another member in the group), also contributes $20 cash to S in exchange for 20 shares of S stock. S sells A1 for $80 and recognizes a $40 gain that is taken into account by the group. Accordingly, P’s aggregate basis in its shares increases by $32 (80/100 x $40), from $40 to $72, and M’s aggregate basis in its shares increases by $8 (20/100 x $40), from $20 to $28. M then sells its shares for $20, their fair market value, and recognizes an $8 noneconomic loss.

Similar changes in the extent to which unrecognized amounts are reflected in basis can occur whenever the subsidiary’s pool of assets is increased or decreased by a capital transaction. The reason is that the interest represented by each share, and thus the relationship between a share’s basis and the interest represented by the share, changes whenever the subsidiary’s pool of assets changes. Such transactions include acquisitive reorganizations (if new shares are issued) and redemptions.

c. Assets are acquired with a basis that reflects unrecognized appreciation. The relationship between the basis of a share and the nature of the interest represented by the share can also be altered by transactions in which S acquires assets with a basis that reflects unrecognized appreciation, such as stock of a new member. The reason is that, after the lower-tier acquisition, the S shares have an interest in unrecognized appreciation and the investment adjustment system will increase the basis of the S shares when those lower-tier items are recognized.

Example 8. Acquisition of lower-tier subsidiary with appreciated assets. P forms S by contributing $100 to S in exchange for all the stock of S. S then purchases all the stock of T for $100 when T holds one asset, A1, with a basis of $0 and a value of $100. T sells A1, recognizing a $100 gain that is taken into account on the P group return. As a result, both S’s basis in its T stock and P’s basis in its S stock are increased by $100, from $100 to $200. P then sells its S stock, recognizing a $100 loss.

The result is the same noneconomic loss illustrated in Example 2.

d. Other redetermination events. The IRS and Treasury Department expect that other transactions and events can alter the extent to which unrecognized asset appreciation is reflected in stock basis. Accordingly, the preceding discussion is not intended to present an exhaustive list of possible redetermination events.

e. Conclusions regarding redetermination events. The IRS and Treasury Department recognize that redetermination events occur as the result of bona fide business transactions engaged in frequently and routinely throughout the time a share is held by any member of the group, and that these transactions are typically not tax-structured transactions. Still, these events generate a significant potential for noneconomic stock loss or gain reduction that facilitates the circumvention of GU repeal. Accordingly, the IRS and Treasury Department believe that all such events, whether described in this preamble or not, must be taken into account in any model that is adopted to address the circumvention of GU repeal.

Nevertheless, the IRS and Treasury Department recognize, and are concerned that, the factual analysis necessary to identify all redetermination events for all members’ shares would be an extensive, complex, difficult, and, therefore, expensive undertaking and, as such, would impose a substantial burden on both taxpayers and the government. Moreover, the nature of the undertaking would make it prone to error and, as a result, the rule would be unevenly administered and similarly situated taxpayers would not be similarly treated.

The IRS and Treasury Department recognize that redetermination events can also create or increase the extent to which the basis of an individual share duplicates an inside loss. However, because duplicated loss is measured at the time that a stock loss is either recognized or preserved for later use, loss duplication rules by their operation account for redetermination events. Accordingly, regulations addressing loss duplication do not generally require specific provisions to address redetermination events.

C. Methods Considered to Implement GU Repeal. The IRS and Treasury Department considered a number of approaches to address the circumvention of GU repeal independently from the issue of loss duplication. The approaches fall into two broad categories: tracing-based and presumptive approaches.

1. Tracing-based methods. Under a tracing-based method, the extent to which a member can enjoy the benefit of subsidiary stock basis attributable to the recognition of an item of income or gain is determined by the extent to which the recognized item is reflected in the basis of the share and thus already protected from duplicative recognition on a later disposition of the stock. The IRS and Treasury Department continue to believe that tracing is a theoretically correct method for implementing GU repeal in the consolidated return setting and so considered various tracing-based proposals.

a. Pure tracing. In general, a tracing approach would look solely to the connection between a subsidiary’s recognized items and any appreciation reflected in stock basis in order to determine the extent to which the group will be allowed the benefit of stock basis attributable to those items. However, such an approach would require taxpayers to create and maintain (and the IRS to examine) records to establish:

• the identity of every “tainted asset,” that is, every asset held by the subsidiary and any lower-tier subsidiaries on every “measuring date,” which includes the date on which the member (or its predecessor) purchased the share and all subsequent dates on which the subsidiary has a redetermination event;
• the “tainted appreciation,” that is, the appreciation on each tainted asset held by the subsidiary and any lower-tier subsidiaries on each measuring date; and
• the extent to which tainted appreciation is recognized, whether as income or gain, and included in an adjustment to the basis of the share.

In addition, to fully benefit from a tracing regime, taxpayers would need to create and maintain similar records for tainted assets with unrecognized depreciation on a measuring date, because the recognition of that depreciation would be allowed to reduce the amount of recognized appreciation treated as tainted.

These records would have to be created and maintained for each share of stock of each subsidiary and each share of lower-tier subsidiary stock held by a subsidiary on each measuring date. In addition, these records would need to be created and maintained not just for subsidiaries, but for all corporations the stock of which is acquired by a member, because the information would be necessary if the corporation becomes a member at some later date.

In administering the various temporary and final regulations promulgated as loss limitation rules under §1.337(d)-1 and §1.337(d)-2, the IRS has found that taxpayers encounter substantial difficulty in attempting to satisfy these requirements.

To begin, taxpayers are generally unable to accurately identify all of a subsidiary’s tainted assets. One reason is simply the vast number of assets implicated. Another reason is that many assets are accounted for in mass accounts and thus cannot be separately identified. Problems are exacerbated if appropriate records are not created contemporaneously; taxpayers have found this a particular concern when subsidiaries have been acquired with inadequate records.

Furthermore, the commonplace nature of many redetermination events makes it difficult to identify all such dates. For example, many taxpayers routinely issue stock when a member contributes cash or property to a subsidiary, even if the issuance of stock would not be required for section 351 to apply, and each such occurrence is a redetermination event.

Valuation also imposes significant financial and administrative burdens on both taxpayers and the government. These problems are exacerbated because the corporation’s assets are not themselves the subject of an arms-length transaction and, in most cases, the date on which the assets are actually valued is long after the stock transaction.

The most problematic aspect of tracing, however, has typically been establishing the connection, or lack thereof, between items taken into account by the group and particular amounts of tainted appreciation. If much time has elapsed between a measuring date and the disposition of a tainted asset, or if an asset is held in a mass account, this can be difficult or even impossible. If tainted appreciation is recognized as income earned through the wasting or consumption of the appreciation, instead of as gain on the disposition of the asset, there are additional difficulties. In those cases, tracing is possible only if the tainted appreciation generates an identifiable stream of income. However, this is frequently not the case. For example, intangible assets, like patents or goodwill, are the source of significant tainted appreciation and they typically do not generate identifiable income streams.

i. Conclusions regarding tracing. For all the reasons set forth in this preamble, the IRS and Treasury Department have again, as in 1990, concluded that tracing is not a viable method for preventing the circumvention of GU repeal in consolidation. This conclusion, while arguably based on theoretical concerns in 1990, is now based on several years of administering §1.337(d)-2 (in both its temporary and final form) as a tracing regime. The IRS found that the difficulties encountered, by taxpayers and the government alike, in administering §1.337(d)-2 as a tracing-based rule were overwhelmingly greater than those encountered in administering it as a presumption-based rule under the basis disconformity method permitted under Notice 2004-58. Accordingly, the IRS and Treasury Department are not proposing to adopt a tracing-based approach.

ii. Tracing in other contexts. The IRS and Treasury Department recognize that tracing-based regimes are used to implement other provisions in the Code. For example, section 382(h), which prescribes the tax treatment of built-in items recognized by a corporation that has had an ownership change, and section 1374, which prescribes the tax treatment of built-in items recognized by an S corporation that was formerly a C corporation, both use tracing-based regimes. Further, the IRS and Treasury Department are proposing regulations implementing section 362(e)(2) in a consolidated return context that require certain items to be traced. See section H of this preamble.

The tracing regimes appropriate for those sections, however, do not present compliance and administrative concerns of the scope and magnitude presented by a tracing regime appropriate for GU repeal in the consolidated setting for at least three reasons.

To begin, both sections 382(h) and 1374 apply only for a limited period of time—five years in the case of section 382(h) and ten years in the case of section 1374—and so whatever burden is imposed is more limited in nature.

More importantly, sections 382(h) and 1374 are generally concerned only with the unrecognized appreciation and depreciation in a pool of assets held by a corporation on a single date—the date the C corporation converts to an S corporation or the date the S corporation acquires assets of a C corporation in the case of section 1374, and the date a corporation has an ownership change in the case of section 382(h). Similarly, section 362(e)(2) is only concerned with net unrecognized depreciation in a pool of assets on the date of the transaction to which section 362(e)(2) applies. But the ability to circumvent GU repeal using the consolidated return provisions can be created any time the subsidiary has a redetermination event. Thus, any rule implementing GU repeal in the consolidated context, unlike rules implementing sections 362(e)(2), 382(h), and 1374, must trace the pool of assets held on all measuring dates, and not just the pool of assets held when subsidiary stock is acquired (or when assets are transferred).

Finally, unlike regulations implementing GU repeal, regulations implementing those other sections do not need to take into account the changing relationship between the basis in a particular share of stock and the unrecognized appreciation and depreciation in the corporation’s assets.

For these reasons, any tracing-based regime appropriately implementing GU repeal in the consolidated setting would be much more expansive and complex, and therefore much less administrable, than the tracing regimes appropriately implementing sections 382(h) or 1374 (or proposed to implement section 362(e)(2)).

b. Modified tracing. The IRS and Treasury Department considered several approaches that could be adopted to modify a tracing model by limiting the extent to which tracing would be required, in order to mitigate the administrative burdens of a pure tracing model.

i. Exclusion for items attributable to after-acquired assets. Several commentators have suggested an approach, generally called the “after-acquired asset exception,” which allows taxpayers to identify assets acquired after the acquisition of subsidiary stock, in order to treat any gain realized on those assets as economic to the group. In general, all other items of gain and income would be deemed to be noneconomic, that is, attributable to the recognition of appreciation that was already reflected in basis. Stock loss would be allowed only to the extent that stock basis was attributable to the amounts deemed economic to the group. In response to concerns raised by the IRS and Treasury Department about redetermination events, the proposal was modified to provide that only assets acquired after the latest measuring date would be treated as giving rise to economic amounts. The principal advantage of this approach is that it identifies some untainted items with no need for valuation.

To begin, the IRS and Treasury Department are concerned with the burden and error potential presented by the need to identify all redetermination events. Moreover, because these events can occur with considerable frequency in the ordinary course of business, it is unlikely that a great deal of time will typically elapse between the last redetermination date and the date of a stock disposition. Thus, the amount of gain recognized on an asset acquired and sold during such periods of time will not likely be significant. As a result, it appears unlikely that this approach would afford much relief to taxpayers (in terms of administrative burden or reducing the disallowance amount) or to the government (in terms of administrative burden).

Furthermore, in order to implement GU repeal appropriately, such an approach must take into account not only gains, but also losses, recognized on after-acquired assets. But the identification of such losses imposes an additional administrative burden that taxpayers have no incentive to facilitate. In any event, a requirement to take losses into account could be easily manipulated by the timing and structuring of redetermination events.

ii. Exclusion for items recognized after prescribed period of time. Several commentators also suggested a tracing-based approach that would apply to investment adjustments taken into account only during a prescribed period of time following the acquisition of a share. The chief advantage to this approach is that, regardless how burdensome the administration of the rule, it would not extend indefinitely.

Like the proposed after-acquired-asset approach, however, this approach would need to take redetermination events into account. The tracing period would then begin again on the date of each redetermination event. Thus, like the after-acquired-asset exception, this approach is unlikely to afford much relief to taxpayers (in terms of administrative or tax burden) or the government (in terms of administrative burden) because the period for tracing may never close.

Moreover, the IRS and Treasury Department are concerned that such an approach does not adequately respond to GU repeal. The reason is that noneconomic investment adjustments circumvent GU repeal whenever they are taken into account. Thus, the IRS and Treasury Department continue to believe that, in the absence of any direction from Congress, such as in the case of section 1374, imposing time limits on the implementation of GU repeal would be inappropriate. See T.D. 8294.

iii. Exclusion for basis conforming acquisitions. Commentators have also suggested adopting a tracing-based approach that excepted any stock acquired in either a section 351 exchange or a qualified stock purchase for which an election was made under section 338. The rationale for this approach is that, by operation of statute, the basis of stock acquired in these transactions can reflect no unrecognized appreciation.

The IRS and Treasury Department agree that, in certain circumstances, the structure of a stock acquisition will, by operation of law, preclude the reflection of unrecognized appreciation in stock basis. The IRS and Treasury Department are concerned, however, that many acquisitions under section 351 or section 338 actually do not preclude the reflection of unrecognized asset appreciation in stock basis. For example, if subsidiary stock is acquired in a section 351 exchange in multiple transactions or by multiple transferors, as illustrated in Example 7(b) and Example 7(c), respectively, the basis of the shares received can reflect unrecognized appreciation. Similarly, because only 80 percent of the stock of a subsidiary need be acquired to elect section 338 treatment, the basis of up to 20 percent of a subsidiary’s shares may reflect unrecognized appreciation. Moreover, even if the initial acquisition precludes the reflection of unrecognized gain, once there is a redetermination event, the form of the acquisition no longer prevents the reflection of unrecognized appreciation in stock basis. Thus, very few, if any, such transactions would ultimately qualify for this exception.

Thus, like the two previously described approaches to modified tracing, this approach has the inaccuracy and burden associated with identifying redetermination dates and a limited potential for relief to either taxpayers or the government.

iv. Conclusions regarding modified tracing. Each approach considered would increase the administrative burden significantly without significantly increasing precision or relief. Accordingly, the IRS and Treasury Department are not proposing to adopt any of these approaches.

2. Hybrid tracing-presumptive model: asset tracing. The IRS and Treasury Department also considered a hybrid tracing-presumptive approach that would identify all assets held when a share is acquired and on each redetermination date thereafter (again, the “tainted assets”) and then presume all items of income, gain, deduction, and loss traced to those assets to be tainted. The intent was to design an approach that would be more precise than either a modified tracing or purely presumptive approach, while being more administrable than a pure tracing-based approach. The chief advantages of this approach are that it may enhance precision and, like the after-acquired asset exception described in section C.1.b.i of this preamble, may eliminate any need for valuation.

However, like the modified tracing approaches described above, this approach would require the identification of all redetermination events. Furthermore, it would require the identification of all assets held at the time of each such event and the tracing of those assets to particular investment adjustments. Thus, it presents even more complexity, burden, and expense than the modified tracing regimes considered. Furthermore, the IRS and Treasury Department are concerned that this approach could be easily abused, either by the manipulation of redetermination dates or the use of intercompany transactions to make valuation elective. (That is, taxpayers could selectively engage in intercompany transactions so that, in effect, some assets would be valued and not others.)

Finally, the IRS and Treasury Department are not convinced that the approach in fact significantly enhances the precision of a pure presumptive model in light of the fact that there is no actual valuation (and therefore no actual determination that there was any gain reflected in stock basis).

For all these reasons, the IRS and Treasury Department concluded that the potential advantages of this hybrid tracing-presumptive approach are outweighed by its disadvantages. Accordingly, the IRS and Treasury Department are not proposing to adopt this approach.

3. Presumptive-based models. Recognizing that even the hybrid tracing-presumptive model would present significant burden and imprecision, the IRS and Treasury Department considered various presumptive models that, like the LDR, would eliminate all elements of tracing. A principal advantage of such approaches is that they are readily administrable by both taxpayers and the IRS. Thus, the rules can apply uniformly and consistently, with the result that similarly situated taxpayers will be similarly treated, increasing the overall fairness of the system. The elimination of any tracing element, however, increases the importance of limitations, where appropriate, on the nature and amount of items treated as noneconomic to a share. The approaches considered are discussed in this section C.3 and in section C.4 of this preamble.

a. Basis disconformity under Notice 2004-58. One model considered was the basis disconformity model described in Notice 2004-58, presently available as a method to avoid disallowance under §1.337(d)-2. As noted in section A.4 of this preamble, the basis disconformity model treats as built-in gain (within the meaning of §1.337(d)-2) the smallest of three amounts. The first is the basis disconformity amount (which identifies the minimum amount of built-in gain that could be reflected in the share), the second is the net positive adjustment amount (which identifies the actual amount of stock basis attributable to the consolidated return system), and the total gains on property dispositions (which responds to the definition of the term built-in gain in §1.337(d)-2). A significant advantage of this approach is that both taxpayers and the IRS find it readily administrable with information that taxpayers are already required to maintain.

However, the Notice 2004-58 basis disconformity model, because it is an interpretation of the current loss limitation rule in §1.337(d)-2, reflects limitations that inhibit the extent to which the rule addresses the circumvention of GU repeal and promotes the clear reflection of group income. For example, the model did not account for the consumption of unrecognized appreciation reflected in stock basis (the “wasting asset” problem). Thus, if unrealized gain reflected in stock basis was recognized as income (for example through a lease, instead of a disposition of the property), the resulting noneconomic stock loss was not disallowed under the current rule. In addition, the model did not address the problem of basis disparity. (See, for example, Example 4.)

A more significant concern, however, is that the basis disconformity approach is underinclusive in that it can only address noneconomic stock loss to the extent of net appreciation reflected in stock basis, which is, by its nature, reduced by unrecognized depreciation reflected in basis. As a result, a potentially significant amount of noneconomic stock loss remained unaddressed, particularly in deconsolidating dispositions of subsidiary stock.

Example 9. Unrecognized loss reflected in stock basis. P purchases all the outstanding stock of S for $150. At the time, S owns one asset, A1, with a basis of $25 and value of $100, and one asset, A2, with a basis of $100 and a value of $50. S sells A1 to a nonmember for $100 and recognizes a $75 gain, which the P group takes into account. Under the investment adjustment system, P increases its basis in the S stock by $75, to $225, to reflect the $75 taken into account by the group. If P then sells the S stock for $150 (its fair market value), P will recognize a $75 loss. Under the basis disconformity approach, only $25, the excess of P’s S stock basis ($225) over S’s net inside asset basis ($100 cash plus S’s $100 basis in A2, or, $200), of the $75 gain is treated as a noneconomic investment adjustment. Thus, although the entire loss is noneconomic, only $25 of that loss would be disallowed under this approach.

b. Modified basis disconformity. The IRS and Treasury Department considered several modifications to the basis disconformity model, all of which were intended to address the underinclusivity of that model. One approach suggested by commentators would mitigate the wasting assets concern by first, for a prescribed period of time, treating the sum of all property gains and, up to the disconformity amount, all income as noneconomic (and thus included in the disallowance amount). After the prescribed time, all gains and income would be treated as noneconomic, but only to the extent of the disconformity amount. Other approaches considered reflected variations on this suggestion.

The IRS and Treasury Department recognize that the model described, and any similar models, would be readily administrable, but are concerned that such a model would not adequately preserve the group’s ability to deduct economic loss sustained by the group. The reason is that stock loss could be attributable to economic investment adjustments (adjustments attributable to the recognition of items of income and gain that were not reflected in stock basis) that were followed by economic loss (attributable to a decline in the value of the subsidiary’s assets). For example, assume that P contributed an asset to S (basis and value of $10), the asset appreciated and S sold it for $100 (recognizing a $90 gain that increased P’s basis in S stock to $100), S reinvested the $100 in an asset that declined in value to $10, and P then sold the stock for $10. P would recognize a $90 loss that would be disallowed because S had a $90 gain on the disposition of an asset. Yet the entire loss was an economic loss. As a result, the IRS and Treasury Department are concerned that the result in Rite Aid (that the group receive the tax benefit of its economic loss) would not be adequately protected.

Ultimately, the IRS and Treasury Department concluded that the basis disconformity model in Notice 2004-58 would not be modified, but that elements of the model would be incorporated in a new approach.

4. The presumptions and simplifying conventions adopted in these proposed regulations.

a. Loss limitation model. As discussed in section A.2 of this preamble, when the IRS and Treasury Department rejected a tracing approach in favor of the presumptive approach in 1990, the decision was made to balance the use of irrebuttable presumptions by adopting a loss limitation model. Under a loss limitation model, losses attributable to noneconomic investment adjustments are disallowed, but gain reduction (or elimination) attributable to noneconomic investment adjustments is not. The IRS and Treasury Department believed that allowing noneconomic gain reduction not only balanced the benefits and burdens of the presumptive approach, it also provided the considerable advantage of reducing gain duplication in consolidated groups.

Example 10. Noneconomic gain reduction, elimination of gain duplication. P purchases all the stock of S for $150 when S holds one asset, A1, with a basis of $100. S sells A1 for $150, recognizing $50 of gain. S uses the $150 proceeds from the sale of A1 to purchase A2. The value of A2 appreciates to $200, and P then sells its S stock for $200.

If the investment adjustment system did not adjust stock basis for items attributable to appreciation reflected in basis, P’s basis in S stock would remain $150 and, when P sells the S stock, P would recognize a gain of $50 (reflecting the $50 appreciation in A2). When S sells A2, S would recognize the same $50 of economic gain a second time. However, because P’s basis in S is increased by the $50 gain recognized on the sale of A1, P will recognize no gain or loss on its sale of S stock. The gain on A2 is therefore taxed once, when there is a recognition event with respect to A2.

These proposed regulations adopt a loss limitation model for the same reasons such a model was adopted in 1990, in the regulations promulgated under section 337(d) and the LDR (to balance the use of a presumptive approach).

However, the LDR, as well as §§1.337(d)-1 and 1.337(d)-2, applied the loss limitation model by disallowing loss recognized on the disposition of subsidiary stock and reducing basis on the deconsolidation of subsidiary stock. The IRS and Treasury Department recognize that the effect of a loss disallowance rule can be achieved by applying a basis reduction rule immediately before the disposition of loss stock. Modifying the loss limitation model to reduce basis in all cases simplifies the structure of the rule by avoiding the need for two distinct rules.

b. Amount of basis reduction. The IRS and Treasury Department considered two basic approaches to determining the amount of basis reduction. One would be determined with reference to a share’s adjusted basis and the other would be determined with reference to the disconformity between the share’s basis and its allocable portion of the subsidiary’s attributes.

i. Adjusted purchase price cap. Under this approach, the basis of a transferred loss share would be reduced by the amount that the subsidiary’s items increased the share’s basis, but only to the extent of the adjusted purchase price. For purposes of this rule, the adjusted purchase price would be defined as the holder’s original basis in the stock, adjusted to take into account all redetermination events. The rationale for this rule is that the adjusted purchase price represents the maximum amount of unrecognized gain that could be reflected in stock basis. However, this cap does not establish that, in fact, there was any appreciation reflected in stock basis and, therefore, it could prove to be substantially overinclusive.

The IRS and Treasury Department considered several rules that could be combined with the adjusted purchase price cap in order to mitigate its potential for overinclusiveness. One approach would combine this cap with the asset tracing model described in this preamble. Another approach would combine this cap with rules that treat income items as included in the basis reduction amount under a different rate (for example, using a declining percentage over time) or amount (for example, using an annual income cap, perhaps based on a percentage of the gross items). The IRS and Treasury Department ultimately concluded that the limitations either imposed unacceptable burdens (because of the need to identify redetermination dates and trace assets) or did not significantly increase the theoretical soundness of the approach, and that the potential for overinclusiveness prevented the approach from responding adequately to the Congressional mandate to preserve the result in Rite Aid.

ii. Modified adjusted purchase price cap. To address the potential overinclusivity of the adjusted purchase price cap, the IRS and Treasury Department considered modifying the rule by reducing the cap by the basis of any tainted assets sold at a gain. The rationale for this modification is that the maximum potential amount of appreciation reflected in basis is reduced by the basis of tainted assets as they are sold. While this modification reduced the potential for overinclusiveness in a theoretically sound manner, it exacerbated the administrative difficulties by requiring not only the identification of all redetermination dates, but also of all assets held on such dates. Moreover, the IRS and Treasury Department ultimately concluded that the basic premise (that the limitation represented the maximum possible noneconomic income) remained an inadequate response to the Congressional directive that the group be allowed to deduct its economic loss.

iii. Disconformity cap. This model would also reduce basis by the amount that the subsidiary’s items increased the share’s basis, but only to the extent of the disconformity amount. For this purpose, the disconformity amount would generally be the same as the basis disconformity amount described in Notice 2004-58. The rationale for this limitation is that the disconformity amount identifies the minimum amount of unrecognized appreciation actually reflected in the basis of a share of subsidiary stock at the relevant time. Thus, although the amount of such appreciation could actually be considerably greater (as in Example 9), and could even be equal to the adjusted purchase price (assuming a subsidiary was purchased with no basis in any of its assets), it is not lower. Not only does the disconformity cap have the advantage of identifying an amount of appreciation actually reflected in stock basis, it allows for the computation of that amount with information taxpayers are already required to know. Additionally, it avoids the need to identify redetermination events because, by computing disconformity immediately before a transfer, this approach automatically takes the effect of all such events into account.

iv. Modified disconformity cap. Because the use of a disconformity cap raises significant potential for underinclusivity, as illustrated in Example 9, the IRS and Treasury Department considered increasing the disconformity cap by the amount of unrecognized loss on any tainted assets held by the subsidiary. The rationale for this increase is that those losses could prevent an equal amount of recognized tainted appreciation from being treated as noneconomic. Thus, the rule would not undermine the theoretical foundation of the disconformity cap.

However, this approach would require the identification of redetermination dates, as well as the identification and valuation of all assets held on the last such date. Recognizing the imprecision inherent in this approach, the IRS and Treasury Department considered increasing the disconformity cap by only a discounted portion of those unrecognized losses. The IRS and Treasury Department concluded that this approach would introduce burden and imprecision much greater than the potential benefit obtained by increasing the cap on basis reductions, at least in the majority of commercially typical cases.

The IRS and Treasury Department also considered implementing this modification not as a general rule, but only as an anti-abuse rule, so that it would apply only in circumstances that indicated a significant amount of tainted income or gain might be sheltered by unrecognized loss on tainted assets. For example, such a rule could require an increase to the disconformity cap if there was a significant loss in stock, if the subsidiary recognized significant gain shortly before stock sale, or if the stock was held for only a short period of time before it was sold. The IRS and Treasury Department were concerned, however, that the increased uncertainty and burden introduced by such an approach could not be justified in light of the protections against manipulation that exist in the Code and other rules of law. For example, see sections 269, 362(e)(2), and 482, as well as various anti-avoidance and anti-abuse provisions in the regulations, including these proposed regulations.

v. Disconformity cap with duplication rule. In considering the structural potential for underinclusivity in the disconformity cap, the IRS and Treasury Department observed that the recognition of noneconomic gains in excess of the disconformity amount causes the subsidiary’s unrecognized losses to be expressed in stock basis. The facts of Example 9 illustrate this point. In that example, P purchased S for $150 when S held A1 (basis $25, value $100) and A2 (basis $100, value $50). S sold A1 and recognized $75 gain, which increased P’s basis in S to $225. P then sold the S stock and recognized a $75 loss. At the time of the stock sale, S’s net asset basis was $200 (the $100 received for A1 and the basis of A2), which exceeds the value of the stock by $50. Thus, the basis disconformity amount is $25 (the excess of the $225 stock basis over the $200 net asset basis), and so (although there is a $75 recognized gain), only $25 is disallowed. However, at that point, S’s $200 net asset basis exceeds S’s $150 value by $50. The $50 of unrecognized loss on A2 is reflected in both P’s basis in S stock and S’s basis in its assets. That is, the loss on A2 has been duplicated. As a result, the underinclusivity of the disconformity cap can be measured and addressed as duplicated loss.

The IRS and Treasury Department recognize that addressing this loss as a duplicated loss allows taxpayers to accelerate the benefit of a subsidiary’s unrecognized losses (that is, obtain the benefit of the loss without a recognition event with respect to its loss assets). However, this approach allows taxpayers the benefit of their economic loss while limiting any arguably excessive benefit to the ability to accelerate inside loss. In the end, loss duplication is prevented. (The IRS and Treasury Department have long recognized that it is appropriate for a group to offset recognized built-in gains and losses, see §§1.337(d)-1 and 1.337(d)-2, as promulgated in 1990 and again as temporary and final regulations following the Rite Aid decision).

vi. Conclusion. In light of the concerns raised by any method that would reduce basis beyond the disconformity amount, the IRS and Treasury Department have concluded that the amount of basis reduction should be limited to the disconformity amount and that combining the disconformity cap with a loss duplication rule to address its underinclusivity provides the most appropriate balancing of interests. Under this approach, the group’s economic loss is appropriately protected and neither the group nor its members will receive more than one benefit for the subsidiary’s economic loss.

c. Items applied to reduce basis.

i. Character of items applied to reduce basis. In general, the IRS and Treasury Department have concluded, and commentators have generally agreed, that all gains on property dispositions, as well as various gain equivalents, should be fully available to reduce basis under a presumptive rule.

Questions arose, however, regarding whether income items should also be fully available to reduce basis. The reasons for these questions center on the general difficulty of tracing income items (which is limited in the best of circumstances) and the observation that the likelihood of a particular income item being attributable to tainted appreciation generally decreases over time. Accordingly, the IRS and Treasury Department considered several proposals to limit both the amount and the rate of inclusion for income items.

All of these approaches would segregate income that could be traced to particular appreciation reflected in stock basis and treat those amounts in the same manner as items of gain. The net income remaining would be applied to reduce basis according to prescribed limits. For example, one proposal would apply net income to reduce basis for a prescribed period of time following a measuring date, but, after that time, net income would be so applied only according to a declining percentage.

The IRS and Treasury Department are concerned, however, that the approaches considered could be readily manipulated, for example, by converting gain into income that cannot be readily traced to particular assets or by delaying the recognition of income items until after the applicable time period. Therefore, any such rule would inappropriately influence the structure of business transactions and, at the same time, fail to provide adequate protection for GU repeal. In addition, the need to account for redetermination dates would add complexity and diminish the potential relief afforded under any such approach. Moreover, the IRS and Treasury Department identified no theoretical basis for any particular rule and were concerned that the increased precision may be more perceived than real.

ii. Capital transfers. Adjustments to reflect transfers of capital, whether contributions or distributions, are not adjustments attributable to the recognition of appreciation or depreciation. Accordingly, these adjustments do not increase or decrease the extent to which stock basis is noneconomic or facilitates the circumvention of GU repeal. For that reason, such amounts are not taken into account in determining the extent to which subsidiary stock basis is subject to reduction.

Commentators have suggested that the nature of an intercompany cancellation of indebtedness is similar to that of a capital contribution and thus should not be taken into account in determining basis reduction. The IRS and Treasury Department recognize that this may often be the case, but are concerned that, under some circumstances, this may not be the case. Because it will be administratively very difficult to identify situations in which intercompany cancellation of indebtedness is not similar to a capital contribution, and to distinguish intercompany cancellation of indebtedness from other arguably similar cases, these proposed regulations treat items related to intercompany cancellation of indebtedness like all other items of income or loss. However, the IRS and Treasury Department continue to study the issue and invite further comments.

d. Netting of items from different tax periods. Under the LDR, there was no cross-year netting of investment adjustments. Positive investment adjustments were taken into account in determining the loss disallowance amount, negative investment adjustments were not. The IRS and Treasury Department have reconsidered whether items from different tax periods should be considered together in determining basis reduction.

The IRS and Treasury Department recognize that the particular circumvention of GU repeal at issue here is a product of the manner in which the investment adjustment system adjusts stock basis to reflect a subsidiary’s amounts that are taken into account by the group. Thus, the IRS and Treasury Department have concluded that the appropriate measure of the concern must take into account the net extent to which the basis of a share has been increased or decreased by the investment adjustment system. Whether a loss is taken into account in the same year in which a gain is taken into account or in a separate year does not change the net effect of the investment adjustment system. Thus, unlike the LDR, these proposed regulations allow netting of all investment adjustments made to a share for all periods.

e. Summary and conclusions. Only a presumptive approach can eliminate the substantial administrative burdens imposed by the tracing-based and hybrid regimes discussed above. As a result, only a presumptive approach can be applied consistently among taxpayers and thus achieve the overall fairness necessary to these regulations. Importantly, if presumptions are rebuttable, the administrative burdens associated with a tracing system are not avoided. In fact, they are exacerbated, because taxpayers will feel it necessary to be prepared to establish, and the government will then need to be prepared to examine, returns using both systems. Accordingly, the proposed regulations reflect a presumptive approach that does not permit the rebuttal of its operating presumptions. As noted in section A.5 of this preamble, Congress has specifically sanctioned the use of presumptions and other simplifying conventions to address the circumvention of GU repeal.

To balance the use of irrebuttable presumptions, the proposed regulations adopt several provisions that are intended to enhance their overall fairness and theoretical soundness. First, the proposed regulations adopt the disconformity amount as the maximum amount of potential stock basis reduction. The reason, as discussed, is that only the disconformity amount both establishes the fact that the taxpayer had unrealized gain reflected in stock basis and identifies the minimum amount of such gain. Second, the proposed regulations include all items taken into account, from all years, in the determination of the basis reduction amount. Thus, basis is not reduced for certain amounts (such as capital transfers) that cannot be attributable to noneconomic investment adjustments. In addition, by presuming all items of income, gain, deduction and loss as attributable to appreciation or depreciation reflected in basis, the proposed regulations avoid the administrative burden and other concerns inherent in various tracing and hybrid approaches. Moreover, by presuming all items to be reflected in basis, the benefits and burdens inherent in the use of irrebuttable presumptions are fairly balanced between taxpayers and the government. Presuming all items of income and gain are noneconomic favors the fisc, while presuming all items of deduction and loss are noneconomic favors taxpayers.

D. Loss Duplication. The IRS and Treasury Department continue to believe that a group’s income is distorted when the group enjoys more than one tax benefit from an economic loss. Further, the IRS and Treasury Department believe that a subsidiary’s use of a group loss in a separate return year, after the group has already recognized the benefit of the loss, distorts the subsidiary’s separate year income.

Moreover, the IRS and Treasury Department do not believe that the manner or order in which a group takes its losses into account affects the extent to which loss duplication is inappropriate. Thus, loss duplication is inappropriate and must be addressed whether arising in situations like that illustrated in Example 3 (loss reflected in both stock and assets) or in Example 5 (duplication attributable to disparate stock basis). In addition, loss duplication is inappropriate and must be addressed whether the group chooses to recognize loss first as an inside loss, on the subsidiary’s assets and operations (which is addressed by §1.1502-32), or as a stock loss (which is currently addressed, at least partially, by §1.1502-35).

Accordingly, the IRS and Treasury Department have returned to a fundamental premise of the LDR and again concluded that a loss duplication rule that operates without regard to members’ continued affiliation is a necessary complement to the investment adjustment system. The IRS and Treasury Department have also concluded that such a rule must also address the potential for loss duplication presented when loss is disproportionately reflected in the bases of individual shares.

Importantly, as noted in section A.5 of this preamble, Congress has indicated that it, too, views the prevention of loss duplication, including in deconsolidating stock dispositions, as an area that is appropriately addressed by regulation. See H.R. Conf. Rep. No. 108-755 at 652.

Therefore, the IRS and Treasury Department have reviewed the current rules and considered alternative approaches to address the duplication of loss.

1. Reconsideration of §1.1502-35. Loss duplication is currently addressed in §1.1502-35. That rule generally applies whenever there is a disposition of loss shares of subsidiary stock. To address the loss duplication problems arising when loss is disproportionately reflected in stock basis, the rule first redetermines members’ bases to reduce that disparity (to address the problems illustrated in Example 5). Different rules apply depending on the subsidiary’s status as a group member following the stock disposition. If the subsidiary remains a member, the full blending rule of §1.1502-35(b)(1) applies and all members’ bases in shares of the subsidiary’s stock are combined and then allocated evenly to preferred (to value) and then to common (equally). If the subsidiary ceases to be a member, the basis redetermination rule of §1.1502-35(b)(2) applies and members’ bases are redetermined to reduce loss on all members’ shares. However, this rule only redetermines basis to the extent of items of deduction and loss included in negative adjustments applied to nonloss shares. As under the full blending rule, redetermination under this rule first reduces or eliminates loss on preferred shares and then equalizes members’ bases in common shares.

The potential for loss duplication following the redetermination of members’ bases is addressed only if the subsidiary remains a member of the group. In that case, stock loss (to the extent of loss duplication) is suspended, the suspended loss is reduced as the subsidiary’s items of deduction and loss are taken into account, and any suspended loss remaining when the subsidiary ceases to be a member is allowed at that time. The regulation does not address the duplication of loss when the subsidiary ceases to be a member, other than to prevent the reimportation of duplicated losses back into the group.

The IRS and Treasury Department understand that certain administrability concerns have arisen under §1.1502-35. For example, taxpayers have commented that the rules relating to the suspension of loss in nondeconsolidating dispositions and the treatment of reimported losses present substantial compliance issues. The experience of the IRS is consistent with those comments.

Moreover, the IRS and Treasury Department have reconsidered the appropriateness of allowing subsidiaries to duplicate group losses after the period of consolidation. Under this approach, former members can use group losses (that have already been used by the group) to offset their separate year income. This duplicative use of group losses distorts the former member’s separate income. Under section 1502, consolidated return regulations are directed to promote the clear reflection of not only the income of a group, but also of its members, including former members. Accordingly, as in 1990, the IRS and Treasury Department have concluded that a group loss, once used by the group, should not be available to a former member for a second, duplicative use outside the group.

For these reasons, the IRS and Treasury Department propose to remove §1.1502-35 and replace it with a more easily administered and more comprehensive approach to addressing loss duplication among members of a consolidated group.

2. Other methods considered for addressing loss duplication. As discussed in section D of this preamble, the IRS and Treasury Department have concluded that loss duplication is an inappropriate distortion of income (of either a group or its members, including former members) regardless of the subsidiary’s status after a transfer of its stock. Accordingly, these proposed regulations address loss duplication in both nondeconsolidating and deconsolidating stock transfers. Several approaches were considered.

a. Disallowance of stock loss. As a general matter, the IRS and Treasury Department believe that disallowing duplicative stock loss better implements single entity principles because it results in the recognition of the subsidiaries’ economic gain or loss on its assets and operations, instead of on its stock. However, to preserve the result in Rite Aid, stock loss could only be disallowed for nondeconsolidating transfers and additional rules would be necessary to address both the loss remaining in the group and the duplication of loss in deconsolidating transfers (which could not be subject to the loss disallowance rule). Thus, a rule implementing this approach would need to include a provision comparable to §1.1502-35(c), which taxpayers and the IRS have found to present significant compliance issues. In addition, this approach would need to include a provision to address loss duplication in deconsolidating transfers.

b. Loss duplication accounts. The IRS and Treasury Department also considered an approach that would allow stock loss, but identify the amount of loss duplication and create a suspended account to limit the deductibility of items as they are taken into account. One advantage of this approach is that it only requires one set of rules to address both nondeconsolidating and deconsolidating transfers. This approach also has the advantage of increasing the precision in identifying (and disallowing) losses that are actually duplicated.

However, unless the rule were to use presumptions to treat items as chargeable against the loss duplication account, it would present considerable tracing issues. In addition, this approach raises administrability issues comparable to those associated with the loss suspension regime in §1.1502-35(c). These difficulties are exacerbated by the need to have the account follow the subsidiary, possibly through subsequent acquisitions, until the account is eliminated.

The IRS and Treasury Department are also concerned that, because this approach would reduce or eliminate duplication only when inside losses were recognized, taxpayers could avoid the effect of the rule by waiting until assets appreciated before disposing of them. To mitigate this concern, the rule could require the subsidiary to take into account the duplication account, either ratably over time or at some specified time, but this could give rise to income in the absence of any loss duplication.

c. Attribute reduction. The IRS and Treasury Department also considered a presumptive rule that would identify the extent of duplicated loss and then reduce the subsidiary’s attributes by that amount. This approach, like the loss duplication account, has the advantage of needing only one set of rules to govern both deconsolidating and nondeconsolidating transfers. It has the added advantage of being similar to regimes that are already familiar to taxpayers, such as the attribute reduction rules of sections 108 and 1017, and §1.1502-28. Although attribute reduction could be based on valuation, like the rule in section 362(e)(2), the IRS and Treasury Department believe that mandatory valuation would present a significant administrative burden and expense for both taxpayers and the IRS.

d. Conclusions. The IRS and Treasury Department have concluded that the complexity, administrative burden, and expense of the loss disallowance and the loss duplication account approaches outweighed their respective advantages. Accordingly, these proposed regulations adopt an attribute reduction rule. The IRS and Treasury Department recognize that the attribute reduction approach allows taxpayers to accelerate economic losses of the subsidiary, but believe that this approach best preserves the result in Rite Aid while addressing loss duplication. In general, the approach adopted operates as an irrebuttable presumption, to avoid the burden of mandatory valuation in all cases, but taxpayers continue to have several mechanisms available to structure their transactions to permit valuation (for example, by using actual or deemed asset sales).

3. Gain duplication. Notwithstanding the conclusions regarding duplication of loss, for the reasons set forth in the LDR preambles, the IRS and Treasury Department have tentatively concluded that adequate protections, and the incentive to use them, already exist to prevent the duplication of gain. See T.D. 8294, T.D. 8364 and T.D. 8984. For example, see sections 332, 336(e) (which is the subject of another current guidance project), and 338(h)(10). Accordingly, the duplication of gain is not addressed in these proposed regulations, except as a result of the adoption of a loss disallowance model. The IRS and Treasury Department continue to study the issues, however, and invite further comment. See section J of this preamble for further discussion of the issues on which comments are requested.

E. Noneconomic and Duplicated Loss from Investment Adjustment System. For all the reasons discussed in this preamble, the IRS and Treasury Department believe that the approaches to noneconomic and duplicated loss that are adopted in these proposed regulations represent the best approach to the (original) noneconomic and duplicated loss concerns described in sections B.1 and B.2 of this preamble. However, those rules alone do not adequately address the problem of noneconomic and duplicated loss attributable to investment adjustments applied to shares of stock with disparate bases. This is the concern described in section B.3 of this preamble and illustrated in Example 4 and Example 5, as well as Example 7(b) and Example 7(c).

The IRS and Treasury Department believe it is essential to address this concern. One reason is that stock basis would be inappropriately eliminated when, in cases like Example 4, there is noneconomic loss on one share because appreciated assets were contributed to a corporation in exchange for other shares. In those cases, the noneconomic loss should not be allowed, but a rule that only prevents that loss does not address the problem that there is insufficient basis on the shares received in the exchange. The result would be noneconomic gain on the sale of those shares. An equally important reason is that loss could otherwise be duplicated when, in cases like Example 5, loss is disproportionately reflected in the basis of some shares. Although regulations could prevent duplication in such cases (by eliminating inside loss to the full extent of duplicated stock loss), allowing a deduction for disproportionate stock loss in such cases permits the acceleration of a disproportionate amount of inside loss. To the extent that loss is disproportionately reflected in the basis of an individual share, acceleration is generally unwarranted and should be prevented to the extent possible. Accordingly, the IRS and Treasury Department have considered various approaches to mitigating these effects.

1. Revise investment adjustment system to adopt a tracing approach. The IRS and Treasury Department recognize that one approach to this problem would be to revise the investment adjustment system so that it would allocate subsidiaries’ items of income, gain, deduction, and loss to their shares in accordance with the actual reflection of those items in the each share’s basis. This approach would be similar to the section 704(c) regime applicable to partnerships. However, this approach is a tracing model and, as discussed in section C of this preamble, the IRS and Treasury Department do not believe that tracing is administrable in the consolidated setting.

Moreover, as noted above, the IRS and Treasury Department continue to believe that the presumptive-based rules of §1.1502-32 are not only administrable, but appropriate in the vast majority of cases because typically subsidiary stock is common stock owned entirely by members with uniform bases. Where subsidiaries have issued preferred stock, it is generally section 1504(a)(4) stock. In addition, the investment adjustment system contains some guidance for situations that do not reflect the general assumptions on which the rules are based (for example, the cumulative redetermination rule in §1.1502-32(c)(4)). In such cases, tracing would be unnecessary. Moreover, the IRS and Treasury Department do not believe that typical commercial transactions generally require groups to alter a subsidiary’s capital structure in a manner that would require tracing. Accordingly, the IRS and Treasury Department are not considering revising the investment adjustment system to implement a tracing regime.

2. Presumptive approaches to reduce basis disparity. The two presumptive approaches considered to reduce basis disparity were a full blending rule similar to that in §1.1502-35(b)(1) and a rule that would redetermine investment adjustments made under §1.1502-32, similar to the rule in §1.1502-35(b)(2).

a. Full basis blending. Under the full basis blending approach, all members’ bases are aggregated and then allocated among members’ shares in a manner that results in the elimination of loss on preferred shares and of basis disparity on all other shares, at least within each class. As a result, members’ bases are aligned with the operating premises of the investment adjustment system.

Full basis blending not only mitigates the effects of previous noneconomic investment adjustments, addressing the concern illustrated in Example 4 and Example 5(a), it also prevents the acceleration of disproportionate amounts of unrecognized loss, addressing the concern illustrated in Example 5(b).

A full basis blending rule is, however, a significant departure from the rules generally applicable under the Code. Commentators have suggested that this departure from generally applicable law may be more significant than is warranted in light of the extent to which the concerns can be addressed under the investment adjustment redetermination approach described in this preamble.

b. Redetermination of investment adjustments previously made to stock basis. The investment adjustment redetermination approach is less a departure from Code provisions as it is a departure from the general operation of §1.1502-32. In general, this approach would reallocate investment adjustments previously applied to members’ bases in subsidiary stock with the goal of reducing, to the greatest extent possible, the disparity in members’ bases in subsidiary stock. Thus, like the full blending approach, this approach would bring members’ bases closer into alignment with the assumptions underlying the investment adjustment system. However, it would do so to a more limited extent than the full blending rule and in a manner that is less of a departure from general Code rules.

i. Recomputation of individual investment adjustments. Presently, §1.1502-35(b)(2) addresses duplicated loss by redetermining investment adjustments when there is a deconsolidating disposition of subsidiary stock. To achieve the greatest reduction in basis disparity possible, §1.1502-35(b)(2) in effect deconstructs investment adjustments in order to remove negative items (that is, items of deduction and expense) from adjustments to the bases of gain shares and then apply those items to reduce members’ bases in loss shares. Taxpayers have raised concerns with the complexity and administrability of this approach. The IRS has observed compliance and audit difficulties with this approach.

Accordingly, the IRS and Treasury Department have reconsidered whether this general approach, redetermining investment adjustments, could be adopted in a simpler form. The principal method considered was a presumptive reallocation of entire investment adjustments (exclusive of distributions), instead of the individual items that comprise them. The approach is similar to that used in the cumulative redetermination rule of §1.1502-32(c)(4). A significant advantage to this simplified approach is that it is readily administered with information that taxpayers are already required to know (§1.1502-32 already requires taxpayers to determine investment adjustments exclusive of distributions).

The IRS and Treasury Department recognize that this general approach, in whichever form adopted, does not address the acceleration illustrated in Example 5(b) to the extent that full blending would. However, this approach is less disruptive to the general determination of basis.

ii. Reallocations to loss shares that are not transferred. Presently, §1.1502-35(b)(2) reallocations can result in the reduction of any member’s basis in a loss share of subsidiary stock. The IRS and Treasury Department have reconsidered whether reallocated investment adjustments should be applied to reduce loss on shares that are not transferred in the transaction.

The IRS and Treasury Department have concluded that reallocating investment adjustments to reduce the basis of only transferred loss shares better implements the loss disallowance model. The reason is that this approach allows subsidiary stock basis to remain intact until there is a taxable disposition, deconsolidation, or worthlessness of the share, thereby permitting that basis to enjoy the full protection of subsequent appreciation as long as it remains in the group and otherwise subject to the consolidated return system. This approach has the added benefit of affording the maximum potential to eliminate disparate reflection of loss on transferred shares because all the reallocations are directed to transferred shares. As a result, this approach reduces the amount of loss that can be accelerated (as illustrated in Example 5(b)).

iii. Reallocations of positive and negative investment adjustments. Under the basis redetermination rule in §1.1502-35(b)(2), only negative items are reallocated. However, the sole purpose of §1.1502-35, and thus the basis redetermination rules in §1.1502-35(b), is to address the duplication of loss. (The full blending approach of §1.1502-35(b)(1) addresses noneconomic loss attributable to basis disparity as well as loss duplication, but only incidentally as a result of its broad operation.) The IRS and Treasury Department believe that, although it is appropriate for a rule addressing only loss duplication to reallocate just negative items (or negative investment adjustments), a rule addressing both noneconomic and duplicated loss must reallocate both negative and positive items (or investment adjustments). As illustrated in Example 4 and Example 5, reallocations of both positive and negative amounts are necessary to prevent the noneconomic and duplicated stock loss that results from the disparate reflection of unrecognized gain and to do so without causing inappropriate results to taxpayers (specifically, noneconomic gain).

For the foregoing reasons, the IRS and Treasury Department have concluded that the reallocation of both positive and negative adjustments is appropriate and necessary to balance the use of a presumptive system. Accordingly, these proposed regulations provide for the reallocation of both positive and negative investment adjustments to minimize the potential over- and under-application of the noneconomic and duplicated loss rules.

Explanation of Provisions

F. Explanation of the Proposed Regulations.

1. Overview. The proposed regulation consists of three principal rules that apply when a member transfers a loss share of subsidiary stock. The first rule redetermines members’ bases in subsidiary stock by reallocating §1.1502-32 adjustments (to adjust for disproportionate reflection of gains and losses in the bases of members’ shares). The second rule reduces members’ bases in transferred loss shares (but not below value) by the net positive amount of all investment adjustments applied to the bases of those shares, but only to the extent of the share’s disconformity amount (to address noneconomic stock loss). The third rule reduces the subsidiary’s attributes to prevent the duplication of a loss recognized on, or preserved in the basis of, transferred stock.

The three rules generally apply in the order described. 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. 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. After the basis redetermination and reallocation 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.

For purposes of these proposed regulations, a transfer of stock includes any event in which gain or loss would be recognized (but for these proposed regulations), the holder of a share and the subsidiary cease to be members of the same group, a nonmember acquires an outstanding share from a member, or the share is treated as worthless. This rule allows the proposed regulations to prescribe one integrated set of rules that implements a loss limitation approach and that can be applied to all loss shares, regardless of the event giving rise to the application of the section.

2. The 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, this rule subjects all members’ shares of S stock to redetermination.

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. The rule moves both positive and negative adjustments, and so addresses both noneconomic and duplicated losses. Because it generally requires adjustments to be made to reduce disparity, it brings members’ bases closer in line with the fundamental principals underlying the investment adjustment system. As a result, there is less likelihood for later noneconomic or duplicated loss attributable to the investment adjustment system.

The rule operates by first removing positive investment adjustments (up to the amount of the loss) from the bases of transferred loss shares. Then, to the extent of any remaining loss on the transferred shares, negative investment adjustments are removed from shares that are not transferred loss shares and applied to reduce the loss on transferred loss shares. The positive adjustments removed from the transferred loss shares are allocated and applied only after the negative items have been reallocated. The reason is to preserve the most flexibility possible in reallocating positive adjustments, in order to minimize disparity to the greatest extent. Thus, the operation of these rules has the effect of removing basis from transferred loss shares and using it to reduce disparity in members’ bases in S shares.

Redetermination is limited in several respects. First, because the premise of the 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.

A related 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 can not 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).

The principle 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. Importantly, in no fact patterns will the application of the rule create gain on shares. Overall, the rule has no effect on the aggregate amount of gain or loss on members’ bases in subsidiary stock.

In the basis reallocation rule, and in several other provisions of the proposed 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. The IRS and Treasury Department intend that taxpayers have flexibility in choosing the methods and formulas to be employed in making these determinations and the IRS will respect any reasonable method or formula so employed.

The IRS and Treasury Department recognize that the redetermination of basis imposes a certain administrative burden. Thus, the rule contains two safe harbors that excuse taxpayers from reallocating basis in situations in which redetermination is deemed unnecessary. One safe harbor is for situations in which redetermination would have no ultimate effect on the basis of any share held by a member.