ACE bad debt deduction must reflect reduced loan basis

Plain-English summary

Chief Counsel considered how a corporate taxpayer should calculate a bad debt deduction for adjusted current earnings after an ownership change. The taxpayer had a net unrealized built-in loss, so IRC § 56(g)(4)(G) reduced the adjusted basis of its pre-change loans to fair market value for ACE purposes. It argued that the reduction should not affect the bad debt deduction because § 56(g) did not expressly require the deduction to be recomputed. Chief Counsel rejected that position and treated ACE as a separate but parallel computation that applies the relevant Code provisions using ACE-adjusted amounts. The reduced loan basis therefore must be used in determining the bad debt deduction for ACE. The advice relies on the statutory basis rules, the structure and history of the alternative minimum tax, and the ACE regulations.

Ruling snapshot

• Question: Must the IRC § 56(g)(4)(G) basis reduction be considered when calculating a bad debt deduction for adjusted current earnings?
• Outcome: Advice given
• Key authorities: IRC §§ 55, 56(g)(4)(G) and (H), and 382; Treas. Reg. § 1.56(g)-1(a)(5) and (k)(4)

Full text (IRS public release)

       Office of Chief Counsel
       Internal Revenue Service
       Memorandum
       Number: 201444040
       Release Date: 10/31/2014
       CC:ITA:B05:FBoone                                Third Party Communication: None
       POSTS-112688-14                                  Date of Communication: Not Applicable

UILC: 56.00-00

date: July 17, 2014

 to:   Associate Area Counsel, Philadelphia
       (Large Business & International)
       Attn: Lisa Blades

from: William A. Jackson
Chief, Branch 5
(Income Tax & Accounting)

subject: Effect of Section 56(g)(4)(G) ACE Adjustment on Bad Debt Deduction

       This Chief Counsel Advice responds to your request for assistance. This advice may
       not be used or cited as precedent.

       LEGEND

       Taxpayer = ----------------------------------------

       Bank        = ---------------------

       Date 1     = -----------------------

       Date 2     = ----------------------

       Date 3     = ---------------------------

       $A         = ----------------

       Year 1     = ------

POSTS-112688-14 2

ISSUE

Is the basis reduction required under section 56(g)(4)(G) taken into account in
determining the amount of a bad debt deduction for purposes of calculating adjusted
current earnings?

CONCLUSION

The basis reduction required under section 56(g)(4)(G) is taken into account in
determining the amount of a bad debt deduction for purposes of calculating adjusted
current earnings.

FACTS

On Date 1 Taxpayer underwent an ownership change within the meaning of section
382(g) of the Code1. At that time, the aggregate adjusted basis of its assets exceeded
the fair market value of those assets by a significant amount. Consequently, on the
date of the ownership change, Taxpayer had a net unrealized built-in loss (NUBIL)
within the meaning of section 382(h)(3)(A). A large portion of this NUBIL was
attributable to loans held by Bank, a subsidiary of Taxpayer. Taxpayer and Bank file a
consolidated return for federal income tax purposes. In the remainder of this
memorandum, references to Taxpayer shall be understood to include all the
corporations that file a consolidated return with Taxpayer.

Under Notice 2008-83, 2008-2 C.B. 905, Taxpayer was not required to treat post-
ownership change charge-offs or write-downs of loans outstanding prior to the
ownership change (pre-change loans) as recognized built-in losses within the meaning
of section 382(h)(2)(B). Furthermore, no statute required Taxpayer to reduce the basis
of those loans for regular tax and alternative minimum tax purposes (excluding the
effect of any adjusted current earnings adjustment) as a result of the ownership change.
Between Date 2 and Date 3, dates subsequent to the ownership change, Taxpayer
claimed approximately $A in bad debt deductions in computing taxable income and
alternative minimum taxable income (prior to taking into account any adjusted current
earnings adjustment) for charge-offs and write downs of pre-change loans.

Because Taxpayer underwent an ownership change under section 382, and Taxpayer
had a NUBIL, section 56(g)(4)(G) required Taxpayer to reduce the adjusted basis of the
pre-change loans to their fair market value immediately before the ownership change.
In computing the amount of its bad debt deduction attributable to the pre-change loans
for purposes of computing its adjusted current earnings (ACE) for Year 1, Taxpayer
initially took the section 56(g)(4)(G) basis reduction adjustment into account. However,
Taxpayer now contends that although it was required to reduce the adjusted basis of

1
Unless specifically provided otherwise, references to the Code refer to the Internal Revenue Code of
1986 as applicable to the taxable years at issue or under discussion.

POSTS-112688-14 3

the pre-change loans under 56(g)(4)(G), that basis reduction should be ignored in
determining the amount of its bad debt deduction for purposes of calculating ACE.

LAW AND ANALYSIS

Section 55(b)(2) generally defines the term alternative minimum taxable income (AMTI)
as the taxable income of the taxpayer for the taxable year--

(A) determined with the adjustments provided in section 56 and section 58, and
(B) increased by the amount of the items of tax preference described in section 57.

In addition, section 59 provides special rules for computing AMTI, some of which require
further modifications in addition to those listed in sections 56 through 57.

For C corporations, one of the adjustments required in determining AMTI is the ACE
adjustment. Section 56(g) provides that the AMTI of any corporation for any taxable
year shall be increased by 75 percent of the excess (if any) of--

(1) the ACE of the corporation, over
(2) the AMTI determined without regard to the ACE adjustment and the alternative

tax net operating loss deduction (pre-adjustment AMTI).

Section 56(g)(2) allows a negative ACE adjustment by providing that the corporation’s
AMTI for the taxable year shall be reduced by 75 percent of the excess (if any) of--

  (1) pre-adjustment AMTI, over
  (2) ACE.

Section 56(g)(2)(B) limits the negative ACE adjustment for any taxable year to the
excess, if any, of cumulative positive ACE adjustments for prior taxable years over
cumulative negative ACE adjustments for prior taxable years.

Section 56(g)(3) defines ACE as the AMTI for the taxable year—

   (1) determined with the adjustments provided in section 56(g)(4), and
   (2) determined without regard to this subsection and the alternative tax net

operating loss deduction (ATNOL deduction).

One of the adjustments required by section 56(g)(4) pertains to ownership changes.
Section 56(g)(4)(G) provides that If--

  (1) there is an ownership change (within the meaning of section 382) in a taxable

year beginning after 1989 with respect to any corporation, and
(2) there is a NUBIL (within the meaning of section 382(h)) with respect to such
corporation,

POSTS-112688-14 4

then the adjusted basis of each asset of such corporation (immediately after the
ownership change) shall be its proportionate share (determined on the basis of
respective fair market values) of the fair market value of the assets of such corporation
(determined under section 382(h)) immediately before the ownership change.

Section 56(g)(4)(H) provides that the adjusted basis of any property with respect to
which an adjustment under this paragraph applies shall be determined by applying the
treatment prescribed in this paragraph.

Taxpayer asserts that although section 56(g)(4)(G) requires Taxpayer to reduce the
adjusted basis of its pre-change loans for ACE purposes, there is no provision in section
56(g) that specifically requires Taxpayer to take that basis reduction into account when
determining Taxpayer’s ACE bad debt deduction. Therefore, Taxpayer contends that
the same amount of bad debt deduction that is allowable in computing taxable income
and pre-adjustment AMTI also is allowable in determining ACE. Taxpayer points out
that there are a number of specifically enumerated adjustments in section 56(g)(4)
under which the adjusted basis of an asset is relevant to the amount of the allowable
ACE deduction. For example, section 56(g)(4)(A) prescribes special rules in computing
depreciation for ACE purposes. Section 56(g)(4)(F)(i) generally provides that in
computing ACE, only section 611 cost depletion is allowable on property placed in
service in a taxable year beginning after December 31, 1989. Taxpayer asserts that
section 56(g)(4)(H) requires that the basis adjustments prescribed by section
56(g)(4)(G) are only taken into account in applying the other adjustment provisions of
section 56(g)(4).

Taxpayer contends that AMTI is computed by starting with taxable income and then
increasing or decreasing that number by the adjustments and preferences specifically
listed in sections 56 through 59. Likewise, Taxpayer contends that ACE is computed by
starting with pre-adjustment AMTI and then increasing or decreasing that number only
by the specific items listed in section 56(g)(4). Under Taxpayer’s approach, all Code
sections that apply to the computation of taxable income are taken into account once in
determining taxable income which is the starting point in computing AMTI. Once these
Code sections are taken into account in determining taxable income, they are not taken
into account again in modifying taxable income into AMTI unless such a reapplication is
expressly required by the specific language used in sections 56 through 59.

Taxpayer rejects the position that to compute AMTI one must effectively recompute
taxable income taking into account the adjustments and preferences listed in sections
56 through 59 (the separate but parallel approach). Likewise, Taxpayer rejects the
position that ACE is computed by recomputing AMTI taking into account all the
adjustments listed in section 56(g)(4) (the separate but parallel approach) under which
the basis reduction of the pre-change loans would have to be taken into account in
determining Taxpayer’s ACE bad debt deduction. Taxpayer contends that its position,
at least in the context of the computation of pre-adjustment AMTI, has been adopted by

POSTS-112688-14 5

the Tax Court, the United States Court of Federal Claims, and the United States Court
of Appeals for the Federal Circuit. Therefore, Taxpayer asserts that its bad debt
deduction allowed in computing ACE must be the same as that allowed in computing
taxable income and AMTI.

To determine if Taxpayer is correct, a brief history of the alternative minimum tax may
prove helpful.

History of the Evolution of the AMT

A. Minimum Tax

Prior to enacting the first alternative minimum tax (AMT), Congress enacted the
minimum tax (MT) in the Tax Reform Act of 1969. Congress enacted the MT to more
equitably allocate the tax burden by imposing a tax on certain tax preference items in
certain circumstances. See S. Rep. No. 552, 91st Cong., 1st Sess. 112 (1969). The
MT tax base consisted of the sum of a taxpayer's preferences, less an applicable
deduction, multiplied by a flat tax rate. It was imposed in addition to any regular tax
liability that might be imposed on the taxpayer. A variation of this tax continued to
apply to corporations until repealed in the Tax Reform Act of 1986. However, the MT
failed to achieve its goal of preventing high income taxpayers from avoiding most or all
of their tax liability through the use of tax preferences. Congress sought to improve
upon the MT by transitioning to an AMT starting with the Tax Reform Act of 1978 (the
1978 Act).

B. 1978 Act AMT

The AMT, which has now completely replaced the MT, has evolved through three
distinct stages beginning with the 1978 Act, continuing through the Tax Equity and
Fiscal Responsibility Act of 1982 (TEFRA), and finally reaching its current basic
structure in the Tax Reform Act of 1986 (the 1986 Act). For noncorporate taxpayers, in
the 1978 Act Congress supplemented the MT with a limited scope AMT. In contrast to
the MT, the first AMT was imposed on a tax base quite similar to regular taxable
income, with a few notable exceptions. The most significant differences were that no
long-term capital gain deduction was allowed in computing AMTI, and in certain cases
some of a taxpayer’s itemized deductions were effectively disallowed in computing
AMTI. The same net operating loss deduction was allowed in computing regular taxable
income and AMTI. Basis of property was the same for both regular tax and AMT
purposes.

Since its first incarnation in the 1978 Act, the AMT has functioned as an alternative to
the regular tax. From a purely technical standpoint the Code has always imposed AMT
on a taxpayer only to the extent the taxpayer’s taxable AMTI multiplied by the
appropriate AMT tax rates, less certain credits, exceeded the taxpayer’s regular tax

POSTS-112688-14 6

liability. However, a taxpayer essentially computes tax liability on AMTI and regular tax
liability on taxable income and pays the higher amount.

C. TEFRA AMT

Post-1978 Act versions of the AMT reveal a trend toward greater differences between
how AMTI and taxable income are computed. In TEFRA Congress repealed the MT for
non-corporate taxpayers and replaced it with a revised AMT. Congress generally
incorporated the old MT preferences into the computation of AMTI by causing such
amounts to increase AMTI relative to taxable income, and Congress created new
preferences either nondeductible or nonexcludable from gross income in computing
AMTI. Congress also disallowed, in computing AMTI, certain itemized deductions
allowable in computing taxable income.

The TEFRA AMT expressly took adjusted gross income, an intermediate step in the
computation of regular taxable income, as its starting point in the computation of AMTI.
To determine AMTI, adjusted gross income was generally simply increased by specified
preferences and only certain itemized deductions were allowable in computing AMTI2.

Like its 1978 Act predecessor, the differences between TEFRA AMTI and taxable
income remained permanent in nature, serving primarily to increase AMTI relative to
taxable income. This held true even for preferences attributable to accelerated
deductions. For example, in computing TEFRA AMTI the Code required a taxpayer to
treat as a preference the excess of allowable accelerated depreciation on each section
1250 property over the amount that would have been allowable on the property using
the straight-line method. However, the Code did not provide a taxpayer a later negative
adjustment in computing AMTI when the depreciation on the property that would have
been allowable under the straight-line method exceeded that allowable under the
accelerated method.

Consistent with the permanent difference scheme, TEFRA continued to use the regular
tax basis of property as the basis for computations affecting AMTI. Thus, if a taxpayer
sold a piece of section 1250 property, the basis of which had previously been reduced
by accelerated depreciation deductions, TEFRA required the taxpayer to use the
property's regular tax basis in computing gain or loss for purposes of computing AMTI.
TEFRA required this result even though, because of the preference for accelerated
depreciation deductions, some of the depreciation allowed in computing taxable income
had been disallowed in computing AMTI.

2
The primary exception to this rule was the alcohol fuel credit. Although the amount of this credit was
required to be included in gross income in determining taxable income, it was not includible in AMTI.
Also, throwback trust distributions were not taken into account in determining the tentative tax imposed on
AMTI. However, as a practical matter in almost all circumstances the modifications to adjusted gross
income and disallowance of certain itemized deductions made AMTI greater relative to taxable income.

POSTS-112688-14 7

Finally, in computing TEFRA AMTI, Congress provided for a separate alternative tax net
operating loss (ATNOL) deduction. In determining the amount of the ATNOL, the
TEFRA AMT started with the regular NOL and then decreased that number by items not
deductible in computing AMTI. Consequently, the ATNOL could not be greater than the
regular NOL but could be less, making the TEFRA ATNOL a subset of the NOL.

In summary, by greatly increasing the number of items treated differently in computing
taxable income and AMTI, Congress, in TEFRA, further delineated the separateness of
AMTI from taxable income. Nevertheless, for the most part TEFRA AMTI could be
described as regular taxable income increased by items deductible in computing taxable
income but not deductible in computing AMTI.

D. 1986 Act AMT

In the Tax Reform Act of 1986 (the 1986 Act), Congress repealed the MT for corporate
taxpayers and first subjected them to the AMT. Congress also made major
modifications to the computation of AMTI completing its transformation to a type of
taxable income “separate from but parallel to” regular taxable income.

While still providing for permanent differences between AMTI and taxable income, in the
1986 Act Congress for the first time provided for differences regarding when items of
income or deductions are taken into account in computing taxable income and AMTI
(timing differences), a difference of paramount importance in properly resolving the
issue in this case. For example, in computing AMTI the depreciation allowable on an
item of tangible property placed in service after December 31, 1986, in the early years
of the property’s recovery period generally will be less than the depreciation allowable
on such property in computing taxable income. This difference, however, will reverse in
subsequent taxable years when the depreciation allowable in computing AMTI exceeds
that allowable on the property in computing taxable income. To truly reflect differences
between when items of income or deduction are taken into account in computing AMTI
and taxable income, in the 1986 Act Congress also provided that the basis of the same
property could be different for AMT and regular tax purposes.

In addition, with regard to the portion of the AMT attributable to timing items3, Congress
provided for a minimum tax credit. This credit could be used to reduce regular tax
liability to the extent it exceeded the amount of tax tentatively imposed on a taxpayer’s
AMTI as reduced by the AMT foreign tax credit (tentative minimum tax or TMT). The
credit generally becomes allowable when, as a result of the reversal of prior timing
differences, the deductions and exclusions from gross income taken into account in
computing AMTI exceed those allowable in computing taxable income and this reversal
results in regular tax liability exceeding TMT. The net effect of the MT credit system is

3
Congress subsequently changed the law to allow corporations a MT credit for all of their AMT. Non-
corporate taxpayers for the most part are still limited to MT credits for AMT attributable to items other than
certain specified “exclusion items”. Thus, if an item is not specifically defined as an exclusion item, a MT
credit would be allowable for any AMT attributable to that item.

POSTS-112688-14 8

that AMT attributable to timing items does not result in a permanent tax increase relative
to a tax regime that just included the regular tax. Rather, AMT attributable to timing
items simply results in an acceleration of some portion of the regular tax liability.

With the enactment of the 1986 Act AMT, one may classify items of income and
deduction in one of three ways regarding how such items are taken into account in
computing taxable income and AMTI: (1) items treated the same under both systems,
(2) items with permanent different treatment under both systems, and (3) items taken
into account under both systems but with timing differences.

E. 1986 Act Bluebook

The “Bluebook” to the 1986 Act contains the following passage regarding the nature of
the 1986 Act AMT:

    Structure of minimum tax as an alternative system.--For most purposes, the tax
    base for the new alternative minimum tax is determined as though the alternative
    minimum tax were a separate and independent income tax system. Thus, for
    example, where a Code provision refers to a “loss” of the taxpayer from an
    activity, for purposes of the alternative minimum tax the existence of a loss is
    determined with regard to the items that are includable and deductible for
    [alternative] minimum tax, not regular tax, purposes.

    In certain instances, the operation of the alternative minimum tax as a separate
    and independent tax system is set forth expressly in the Code. With respect to
    the passive loss provision, for example, section 58 provides expressly that, in
    applying the limitation for [alternative] minimum tax purposes, all [alternative]
    minimum tax adjustments to income and expense are made and regular tax
    deductions that are items of tax preference are disregarded.

    In other instances, however, where no such express statement is made,
    Congress did not intend to imply that similar adjustments were not necessary.
    Thus, for example, for [alternative] minimum tax purposes it was intended that
    section 1211 (limiting capital losses) be computed using [alternative] minimum
    tax basis, that section 263A (requiring the capitalization of certain depreciation
    deductions to inventory) apply with regard to [alternative] minimum tax
    depreciation deductions, and that section 265 (relating to expenses of earning
    tax-exempt income) apply with regard only to items excludable from alternative
    minimum taxable income.

Staff of the Joint Committee on Taxation, General Explanation of the Tax Reform Act of
1986, 99th Cong., 1st Sess. 438 (Comm. Print 1987).

The phrase “separate from but parallel to” does not appear in the explanation section of
any of the official committee reports to the1986 Act. It appears twice in the “present

POSTS-112688-14 9

law” sections of the conference report to the 1986 Act. The conferees used the phrase
to explain the pre-1986 Act treatment of the carryover of ATNOLs and AMT foreign tax
credits.

Allen v. Commissioner

In Allen v. Commissioner, 118 T.C. 1 (2002), the petitioners, shareholders in an S
corporation, claimed targeted jobs credits under section 51(a) for 1994 and 1995 for
wages paid to the S corporation’s employees. As required by section 280C(a), the
amount of the wage deductions claimed in determining taxable income for each of the
taxable years was reduced by the amount of targeted jobs credit generated for that
year. Although no provision in sections 56 through 59 provided for a different treatment
in computing AMTI, the petitioners contended that under the separate but parallel
method of computing AMTI no wage deduction reduction applied. They reasoned that
the purpose of the section 280C(a) wage deduction reduction was to prevent the
taxpayers from getting a double tax benefit from the same expenditure, once as a tax
credit, and again as a deduction. Because the targeted jobs credit was not allowable in
determining AMT liability, they asserted that no double tax benefit was possible in the
context of the AMT and therefore in determining AMTI no section 280C(a) wage
deduction reduction applied.

The Commissioner agreed that the computation of the 1986 Act version of AMTI
required a separate but parallel approach. That is, rather than determining AMTI simply
by adding to or subtracting items from taxable income, determining 1986 Act AMTI
requires that taxable income be recomputed taking into account the adjustments and
preferences specified in sections 56 through 59. However, because the petitioners had
claimed targeted jobs credits, and because no provision of sections 56 through 59
allowed section 280C(a) to be applied differently in computing AMTI than in computing
taxable income, the Commissioner contended that the wage reduction limitation also
applied in determining AMTI.

The Commissioner pointed out that in another context involving a credit allowable
against the regular tax but not the AMT, Congress had provided a special rule for
computing AMTI. In section 232 of the Crude Oil Windfall Profit Tax Act of 1980 (the
1980 Act) Congress first provided tax credits for certain uses of alcohol (alcohol fuel
credits). Congress required the amount of any alcohol fuel credit earned to be included
in gross income. However, in the 1980 Act Congress amended section 55(b)(1) to
exclude the amount of alcohol fuel credit earned from inclusion in AMTI. Like the
targeted jobs credit, under the 1986 Act version of the AMT the alcohol fuel credit could
be used to reduce regular tax liability but not AMT.

Since the adoption of alcohol fuel credits in the 1980 Act, in subsequent amendments to
the Code, taking retroactive technical corrections into account, Congress had continued
to include the amount of such credits earned in gross income for purposes of computing
taxable income but not AMTI. For the taxable years at issue section 56(a)(8) had

POSTS-112688-14 10

specifically excluded the amount of alcohol fuel credit earned from gross income for
purposes of computing AMTI. Requiring the amount of a credit to be included in gross
income is quite similar to reducing the amount of a deduction by the amount of credit
earned. That Congress, for the 1986 Act version of the AMT, continued to provide a
statutory rule eliminating the amount of alcohol fuel credit earned from AMTI while
providing no special AMT rule for purposes of applying section 280C(a), in the
Commissioner’s view, provided additional evidence that Congress intended for the
section 280C(a) wage deduction reduction to also be applied in computing AMTI.4

In Allen, the petitioners essentially argued that a separate but parallel computation of
AMTI should be performed as if the AMT were the only tax that applied, that is, as if the
regular tax did not even exist (completely independent tax systems). In such a universe
it would be absurd to reduce a taxpayer’s wage deduction by a “credit” that did not exist.
The Commissioner asserted that a separate but parallel computation of AMTI did not
require such an approach. Rather, separate but parallel simply required that taxable
income be recomputed taking into account those preferences and adjustments
specifically set forth in sections 56 through 59, taking into account the actual facts.
The petitioners had actually claimed entitlement to a targeted jobs credit that would
either reduce their regular tax liability as a credit or be allowed as a deduction in the
future in computing both AMTI and taxable income. Because there was no provision in
sections 56 through 59 that allowed section 280C(a) to be applied any differently in
determining AMTI than it applied in determining taxable income, the wage deduction
reduction also applied in determining AMTI.

Despite the litigating parties’ agreement that determining 1986 Act AMTI required a
separate but parallel approach, albeit disagreeing regarding what that approach
entailed, the Tax Court, sua sponte, took a different tack to resolve the case in favor of
the Commissioner. After quoting the statutory definition of taxable income and then
AMTI, the Tax Court stated:

    From this text, we understand explicitly that the base of AMTI is “taxable
    income”, and that this base may be affected by the items described in sections
    56, 57, and 58. Sec. 55(b)(2). See generally section 59, which, although not
    specifically mentioned in section 55, provides definitions and special rules that
    apply in the setting of AMT. As to the meaning of the term “taxable income”,
    Congress has provided unambiguously and with sweeping breadth that “for
    purposes of this subtitle, the term ‘taxable income’ means gross income [see sec.

4
The Commissioner also cited Hightower v. Commissioner, T.C. Memo. 1982-559, a case involving the
issue of whether, in computing income subject to self-employment tax, section 280C requires a taxpayer
to reduce its deduction for wages by the amount of new jobs credit earned. New jobs credit reduces a
taxpayer’s regular tax but cannot be used to reduce a taxpayer’s self-employment tax. The court noted
that deductions are a matter of legislative grace, New Colonial Ice Co. v. Helvering, 292 U.S. 435, 440
(1934), and concluded that the language of the applicable statutes required the taxpayer to reduce its
deduction for wages in computing income subject to self-employment tax by the amount of new jobs
credit earned.

POSTS-112688-14 11

  61(a) for the applicable meaning of the term “gross income” ] minus the
  deductions allowed by this chapter (other than the standard deduction).” Sec.
  63(a) (emphasis added). We conclude on the basis of our plain reading of the
  unambiguous text of sections 55 and 63(a) that a computation of AMTI requires
  that a taxpayer first compute its taxable income and then alter that amount (by
  way of an adjustment or an increase) to reflect the items described in the
  remainder of part. VI, subchapter A, chapter 1, subtitle A (part VI) [sections 55-
  59]. …

  Because section 280C is a wage-expense limitation that enters into the
  computation of taxable income for purposes of section 63(a), and section
  280C(a) is not referenced in part VI, we conclude naturally that the limitation is
  reflected in the calculation of AMTI. …

  Respondent does not disagree with the parallel tax regime rationale advanced by
  petitioners. Respondent invites the Court to hold that the systems are “parallel” in
  the sense that a taxpayer who has calculated taxable income must start from
  scratch in a separate computation of AMTI. Both respondent and petitioners rely
  extensively upon the Staff of Joint Comm. on Taxation, General Explanation of
  the Tax Reform Act of 1986 (J. Comm. Print 1987) (General Explanation of the
  1986 Act), in arguing that the legislative history under the current AMT regime
  supports the treatment of that regime as a system that is parallel to the regular
  tax regime. …

  We decline to adopt the parties’ parallel system contention, however, because,
  as discussed herein, the plain and unambiguous text of the statutes (and the
  related legislative history) disproves that contention.

118 T.C. at 10-11. Subsequent to Allen, the United States Court of Federal Claims,
affirmed by the United States Court of Appeals for the Federal Circuit, also held that in
computing 1986 Act AMTI a taxpayer’s AMT wage deduction had to be reduced by the
amount of targeted jobs credit generated by such wages. See Ventas v. United States,
57 Fed. Cl. 411 (2003), aff”d, 381 F.3d 1156 (Fed. Cir. 2004).

According to Taxpayer, these cases establish that taxable income constitutes the
starting point in computing 1986 Act AMTI. That number is then mechanically increased
or decreased only as precisely specified in sections 56 through 59. This mechanical
process does not involve the reapplication of any Code section previously applied in
determining taxable income but taking into account the adjustments and preferences
specified in sections 56 through 59. If taxable income is to be increased or decreased
as a result of a provision in sections 56 through 59, that increase or decrease must be
directly required by that provision without reference to any statutory provision not
specified in sections 56 through 59. For example, if some provision of section 56
requires that the adjusted basis of an asset be determined differently for AMT and

POSTS-112688-14 12

regular tax purposes, that different basis is taken into account only for purposes of
applying the other provisions of sections 56 through 59.

Taxpayer applies a similar analysis in determining ACE. The starting point in
determining ACE is pre-adjustment AMTI. That number is then mechanically increased
or decreased as precisely specified in section 56(g)(4) without regard to any Code
sections other than section 56(g)(4).

Notwithstanding Allen and the other cases cited by Taxpayer, AMTI should be
determined by recomputing taxable income taking into account the adjustments and
preferences specified in sections 56 through 59. Likewise, the proper way to
determine ACE is to recompute pre-adjustment AMTI taking into account the section
56(g)(4) ACE adjustments. In both cases the process will involve the application of
Code sections other than those specified in the AMT or ACE provisions.

The Service’s Position

In Allen, the Tax Court relied in large part on the “unambiguous” statutory language
defining AMTI. To reiterate, section 55(b) generally defines the term AMTI as the
taxable income of the taxpayer for the taxable year--

(A) determined with the adjustments provided in section 56 and section 58, and
(B) increased by the amount of the items of tax preference described in section 57.

Contrary to the view expressed by the Tax Court in Allen, we do not regard the above
language to be free from ambiguity. The language could be interpreted to mean (1)
taxable income is computed de novo taking into account the specified preferences and
adjustments, or (2) taxable income as determined for regular tax purposes is simply
increased or decreased by the specified preferences and adjustments, much as
adjusted gross income was simply increased by the specified items in determining
TEFRA AMTI.

The 1986 Act’s introduction of differences between when items of income or deduction
are taken into account in computing AMTI versus taxable income require that
interpretation (1) be applied. Consider the 1986 Act provisions that define the 1986 Act
ATNOL. Section 56(a)(4) provides that in computing AMTI the ATNOL deduction shall
be allowed in lieu of the net operating loss deduction allowed under section 172.
Section 56(d)(2)(A) generally defines an ATNOL as follows:

(A) Post-1986 loss years. In the case of a loss year beginning after December 31,
1986, the net operating loss for such year under section 172(c) shall--

   (i) be determined with the adjustments provided in this section and section 58

and

POSTS-112688-14 13

 (ii) be reduced by the items of tax preference determined under section 57 for

such year.

An item of tax preference shall be taken into account under clause (ii) only to the extent
such item increased the amount of the net operating loss for the taxable year under
section 172(c).

With the exception of the flush language, the language used to define an ATNOL is
remarkably similar to the language used in section 55(b)(2) to define AMTI. Both
statutes provide that a well-defined item under the regular tax system, in one case the
NOL, and in the other taxable income, (1) is to be determined with the adjustments
provided in section 56 and section 58 and (2) is to be reduced (in the case of the NOL,
increased in the case of taxable income) by the items of tax preference determined
under section 57 for such year. If the “unambiguous text “ of section 55(b)(2) requires
that AMTI be computed by mechanically increasing or decreasing taxable income as
specified in sections 56 through 59, it also follows that the ATNOL is computed by
mechanically increasing or decreasing the NOL as specified in sections 56 through 59.

Consider the results this approach can produce. For example, assume that Corporation
A and Corporation B have the following gross income and deductions:

                       Corporation A            Corporation B

Gross income $100,000 $50,000
Depreciation Allowed for
Regular Tax Purposes (50,000) (50,000)

Taxable Income $ 50,000 $( 0)

Section 172(c) defines a “net operating loss” as the excess of the deductions allowed by
chapter 1 of the Code over the gross income, computed with certain modifications
specified in section 172(d). Assume that none of those modifications apply in the
example. Neither taxpayer’s regular tax deductions exceed the taxpayer’s regular tax
gross income. Thus, neither taxpayer has an NOL, or stated differently each taxpayer’s
NOL is $ 0.

Assume that under section 56(a)(1)(A) each taxpayer is entitled to an AMT depreciation
deduction of $150,000. If the ATNOL is computed by starting with the NOL (zero) and
subtracting the additional $100,000 of AMT depreciation allowed by section 56(a)(1)(A),
both A and B will have the same ATNOL ($100,000). This would be the case even
though A and B have the same deductions and A’s gross income is $50,000 greater
than B’s.

POSTS-112688-14 14

Although under the stipulated facts the result achieved for B makes sense, this method
produces a nonsensical result for A. A’s ATNOL should be $50,000 (AMT gross income
of $100,000 less AMT depreciation of $150,000). Such a result properly reflects
economic reality which includes the $50,000 difference in their gross incomes. A proper
result in all possible scenarios can only be achieved by a “start from scratch” approach
under which the ATNOL is calculated in the same manner used to calculate the NOL
modified by taking into account the adjustments and preferences set forth in sections 56
through 59.

Likewise, assume that Corporation C has the following gross income and deductions:

Gross Income $200,000
Depreciation Allowed in
Computing Taxable Income (100,000)

Taxable Income 100,000

Additional Depreciation
Allowed in Computing
AMTI (50,000)

AMTI $50,000

Once again, for regular tax purposes C’s NOL is zero because C’s regular tax
deductions do not exceed its gross income. If C’s ATNOL is computed by starting with
the NOL ($0) and subtracting from that number the additional $50,000 AMT
depreciation allowed by section 56(a)(1)(A), C with have both positive AMTI of $50,000
and an ATNOL of $50,000, an absurd result.

We recognize that the goal in interpreting statutes is to determine the true intent of
Congress and “[t]here is no invariable rule for the discovery of that intention.” United
States v. American Trucking Ass’n, 310 U.S. 534, 542 (1940). In American Trucking, the
Supreme Court made the following observations on statutory construction:

   There is, of course, no more persuasive evidence of the purpose of a statute
   than the words by which the legislature undertook to give expression to its
   wishes. Often these words are sufficient in and of themselves to determine the
   purpose of the legislation. In such cases we have followed their plain meaning.
   When that meaning has led to absurd or futile results, however, this Court has
   looked beyond the words to the purpose of the act. Frequently, however, even
   when the plain meaning did not produce absurd results but merely an
   unreasonable one “plainly at variance with the policy of the legislation as a
   whole” this Court has followed that purpose, rather than the literal words.

POSTS-112688-14 15

Id. at 543. However, there must be unequivocal evidence that Congress intended a
different result than that given by the plain meaning of words used in a statute before it
is appropriate to override the plain meaning of those words. Segel v. Commissioner, 89
T.C. 816, 841 (1987).

As the above examples illustrate, even if the statutory language defining an ATNOL
literally required the ATNOL to be determined simply by increasing or decreasing the
amount of the NOL, the absurdity of some of the results produced by following that
approach most likely would justify an interpretation that deviated from the literal wording
of the statute. Because the provisions defining an ATNOL may be fairly interpreted as
requiring a separate computation of the ATNOL, using the same method as that
employed in determining the NOL taking into account preferences and adjustments, and
because to do otherwise can produce absurd results, a separate but parallel
computation should be employed in determining the amount of the ATNOL.

The scenarios discussed above provide the clearest illustrations of how adopting
Taxpayer’s theory of statutory interpretation may lead to absurd results in determining
AMTI. We will not attempt here to illustrate every instance of how Taxpayer’s restrictive
interpretation would lead to results clearly at odds with the purposes of the AMT
provisions. However, Taxpayer has focused on the interpretation of an ACE provision
concerning adjusted basis. Therefore, we will also focus on adjusted basis.

Section 56(a)(1) provides special rules to determine AMT depreciation on tangible
property. Section 56(a)(6) provides in part that the adjusted basis of any property to
which section 56(a)(1) applies shall be determined based on the treatment prescribed in
section 56(a)(1). Taxpayer would concede that to determine the amount of AMT
depreciation on tangible property, that property’s AMT adjusted basis must be used.
This is because section 56(a)(6), in conjunction with section 56(a)(1), expressly requires
this result.

However, gain or loss on the sale or other disposition of property is determined under
section 1001(a). There is no statutory provision (assuming that the general definition of
AMTI is applied in the manner contended for by Taxpayer) in sections 56 through 59
that expressly requires section 1001(a) to be applied separately from its regular tax
application to determine AMT gain or loss on depreciable tangible property. Therefore,
if Taxpayer is correct, as was the case under the TEFRA AMT, gain or loss on the sale
or other disposition of depreciable property is the same for both regular tax and AMT
purposes. This would be true notwithstanding that prior to the sale substantially more
depreciation may have been taken on such property for regular tax purposes than that
deducted in determining AMTI.

The Senate report to the 1986 Act provides as follows:

   For all depreciable property to which minimum tax adjustments apply, adjusted
   basis is determined for minimum tax purposes with reference to the amount of

POSTS-112688-14 16

   depreciation allowed for minimum tax purposes under the alternative system.
   Thus, the amount of gain on the disposition of such property will differ for regular
   and minimum tax purposes.

S. Rep. No. 313, 99th Cong., 2d Sess. 524 (1986). The House report to the 1986 Act
provides as follows:

   For all depreciable property to which minimum tax adjustments apply, adjusted
   basis is determined for minimum tax purposes with reference to the amount of
   depreciation claimed for minimum tax purposes under the nonincentive system.
   Thus, the amount of gain on the disposition of such property will differ for regular
   and minimum tax purposes.

H.R. Rep. No. 426, 99th Cong., 1st Sess. 310 (1985). Both the 1986 Act House and
Senate reports run counter to Taxpayer’s restrictive view of the definition of AMTI. To
get the result specified in the legislative history, section 1001(a) with respect to gain or
loss on the sale or other disposition of depreciable tangible property must be applied
once for regular tax purposes and again for AMT purposes taking into account the AMT
adjusted basis of the property. The intent of Congress, as expressed in the above-cited
legislative history, is statutorily expressed in section 56(a)(6), which provides that the
adjusted basis of any property to which certain specified AMT adjustments apply, shall
be determined on the basis of the treatment prescribed by those adjustments.

Similarly, in calculating ACE, taxpayers must determine gain or loss under section
1001(a) using the ACE adjusted basis of an asset. Section 56(g)(4)(H) is worded
similarly to section 56(a)(6) and seems to require that the basis of an asset, as
determined by applying the adjustments under section 56(g)(4), be used for all
purposes in calculating ACE. For the same reasons set forth above regarding the
calculation of AMTI, Taxpayer’s restrictive view makes no sense in light of the statutory
language and legislative history of the 1986 Act.

Furthermore, Taxpayer’s restrictive interpretation also frustrates the purpose of the
minimum tax credit. That purposes is to ensure that any AMT imposed attributable to
timing differences results in a temporary rather than a permanent increase in tax liability
vis-à-vis the cumulative amount of tax imposed under the regular tax system.

The House report to the 1986 Act provides as follows:

   [T]he committee believes that the present law structure of the alternative
   minimum tax requires modification in certain respects. In particular, to the extent
   that tax preferences reflect deferral, rather than permanent avoidance of tax
   liability, some adjustment is required with respect to years after the taxpayer has
   been required to treat an item as a minimum tax preference, and potentially to
   incur minimum tax liability with respect to the item. Absent such an adjustment,
   taxpayers could lose the benefit of certain deductions altogether.

POSTS-112688-14 17

Id. at 307-08. The report goes on to make clear that the adjustment being referred to is
the minimum tax credit:

   When a taxpayer pays alternative minimum tax, the amount of such tax paid (i.e.
   the net minimum tax) is allowed as a credit against the regular tax liability of the
   taxpayer in subsequent years. However, this credit (known as the minimum tax
   credit) cannot be used to reduce tax below the tentative minimum tax in
   subsequent years. For individuals, the minimum tax credit applies only to
   minimum tax liability incurred due to deferral preferences (such as depreciation)
   i.e. preferences for which the timing, rather than the amount, of a deduction gives
   rise to its treatment as a tax preference.

Id. at 308-09.

Taxpayers generally are able to use the minimum tax credit to reduce their regular tax
liability when the net amount of deductions attributable to timing items allowed for AMT
purposes exceeds the amount of such deductions allowed in determining taxable
income. Thus, the minimum tax credit generally is allowable when the timing
differences reverse. For depreciable tangible property this reversal takes place when
the amount of depreciation allowed in determining AMTI exceeds that allowable in
determining taxable income. However, if the property is sold prior to being fully
depreciated, the depreciation timing difference reversal will be incomplete. If, consistent
with Taxpayer’s theory, the adjusted basis for gain or loss on the sale of the property for
both AMT and regular tax purposes is the regular tax adjusted basis, then gain or loss
from the sale of the property will be the same in determining both taxable income and
AMTI. In such a case, contrary to congressional intent, there will be permanent tax
increases imposed on the taxpayer as a result of a timing difference. Although such
permanent tax increases can occur if, for example, an individual taxpayer dies without
using all of the taxpayer’s minimum tax credits, in other circumstances where the
statutes imposing AMT may be fairly interpreted to avoid such a result, that
interpretation should apply.

Moreover, notwithstanding Allen, since that case was decided the Tax Court has not
interpreted the provisions defining AMTI as restrictively as advocated by Taxpayer. In
Merlo v. Commissioner, 126 T.C. 205 (2006), aff’d, 492 F.3d 618 (5th Cir 2007), the
taxpayer exercised incentive stock options (ISOs). Under section 421(a)(1) the
taxpayer was not required to recognize any income on the exercise for regular tax
purposes. However, for AMT purposes section 56(b)(3) required the taxpayer to
recognize income equal to the difference between the exercise price and the fair market
value of the stock acquired on the date of exercise. The last sentence of section
56(b)(3) provides that in determining AMTI the adjusted basis of any stock acquired by
the exercise of an ISO shall be determined taking the required income recognition into
account. So under section 56(b)(3) the AMT adjusted basis of the stock acquired upon
the exercise of the ISOs was the fair market value of the stock.

POSTS-112688-14 18

Section 165(g)(1) generally provides that if any security (including stock in a
corporation) which is a capital asset becomes worthless during the taxable year, the
loss resulting therefrom shall be treated as a loss from the sale or exchange, on the last
day of the taxable year. This loss is determined under section 1001(a) treating the
sales price as zero. The stock acquired in Merlo became worthless causing the
taxpayer to sustain a loss. One issue was whether the capital loss limitations of section
1211(b) applied for both regular tax and AMT purposes. A second issue was whether
the loss that the taxpayer realized on the worthlessness of the shares generated an
ATNOL.

There is no provision in sections 56 through 59 that expressly requires that section
1001(a) be applied separately from how it is applied in determining taxable income for
purposes of determining AMT gain or loss from the sale or exchange of stock.
However, the Tax Court took it as a given that the AMT loss on the worthlessness of the
stock was based on the much greater AMT adjusted basis of the stock. See also
Palahnuk v. Commissioner 127 T.C. 118 (2006), aff’d, 544 F.3d 471 (2d Cir. 2008)
(different gain or loss for regular tax and AMT purposes on sale of stock acquired
pursuant to ISOs); Kadillak v. Commissioner, 127 T.C. 184 (2006), aff’d, 534 F.3d 1197
(9th Cir. 2008) (nondeductible AMT capital loss). The only way to achieve this result is
to separately apply section 1001(a) (a separate but parallel approach) in determining
AMTI taking into account the AMT adjusted basis as prescribed under section 56(b)(3).

We also note that the ACE regulations specifically require a separate but parallel
approach in determining ACE. Section 1.56(g)-1(a)(5) provides as follows:

   General rule for applying Internal Revenue Code provisions in determining
   adjusted current earnings -- (i) In general. Except as otherwise provided by
   regulations or other guidance issued by the Internal Revenue Service, all Internal
   Revenue Code provisions that apply in determining the regular taxable income of
   a taxpayer also apply in determining adjusted current earnings. For example, the
   rules of part V of subchapter P (relating to original issue discount and similar
   matters) of the Code apply in determining the amount (and the timing) of any
   interest income included in adjusted current earnings under this section. In
   applying Code provisions, however, the adjustments of section 56(g) and this
   section are also taken into account. For example, in applying the capitalization
   provisions of section 263A, the amount of depreciation to be capitalized is based
   on the amount of depreciation allowed in computing adjusted current earnings.

Finally, the example of the application of section 56(g)(4)(G) contained in the ACE
regulations, Treas. Reg. 1.56(g)-1(k)(4), requires that gain or loss on the disposition of
assets for purposes of determining ACE must be based on the adjusted basis of the
assets as determined under section 56(g)(4)(G). After setting forth the facts in the
example and illustrating how the section 56(g)(4)(G) basis modification applies, the
regulation includes the following sentence: “L must use these new adjusted bases for

POSTS-112688-14 19

all purposes in determining adjusted current earnings, including computing depreciation
and any gain or loss on disposition.” Thus, notwithstanding that there is no provision in
section 56(g)(4) that expressly requires section 1001(a) to be applied separately in
determining gain or loss for ACE purposes, such a computation is required under the
regulations. Likewise, in determining the amount of its bad debt deduction in
determining ACE, a taxpayer must use the adjusted basis of the loans as determined
under section 56(g)(4)(G).

Based on the preceding analysis, we conclude that to determine the amount of its bad
debt deduction for purposes of computing ACE, Taxpayer must take the basis reduction
required under section 56(g)(4)(G) into account.

CASE DEVELOPMENT, HAZARDS AND OTHER CONSIDERATIONS

This writing may contain privileged information. Any unauthorized disclosure of this
writing may undermine our ability to protect the privileged information. If disclosure is
determined to be necessary, please contact this office for our views.

Please call (202) 317-7006 if you have any further questions.