Specified Business

Office of Chief Counsel
Internal Revenue Service memorandum
Number: 201612013
Release Date: 3/18/2016
CC:ITA:B05:RFBoone
POSTF-143803-14
Third Party Communication: None
Date of Communication: Not Applicable
UILC: 172.03-00 date: November 19, 2015 to: Associate Area Counsel
--- from: Chief, Branch 05
Income Tax & Accounting subject: Nuclear Decommissioning Costs
This Chief Counsel Advice responds to your request for assistance. This advice may not be used or cited as precedent.
LEGEND
Taxpayer = ---
State A = ---
Power Generating Subsidiary = ---
Facility 1 = ---
Facility 2 = ---
Group 1 = ---
---

$D = ---
Fund = ---
Date 1 = ---

# ISSUE

Does a yearly liability to make payments imposed by state law for the storage of spent nuclear fuel at a nuclear power plant located in the state, and that is contingent on the plant being in operation for that year, arise prior to the operation of the plant for that year?

# CONCLUSION

A yearly liability to make payments imposed by state law for the storage of spent nuclear fuel at a nuclear power plant located in the state, and that is contingent on the plant being in operation for that year, does not arise until the plant is operating for that year.

# FACTS

Taxpayer operates nuclear power generating facilities including facilities located in State
A. In the late 1970’s the federal government prohibited civilian nuclear power plant operators from reprocessing spent nuclear fuel. This increased the need to find a permanent solution to the problem of disposing of spent nuclear fuel. In section 111(a)(4) of the Nuclear Waste Policy Act of 1982 (the 1982 Act), Congress made a general finding that while the federal government has the responsibility to provide for the permanent disposal of high-level radioactive waste (“HLW”) and spent nuclear fuel ("SNF") as may be disposed of in order to protect the public health and safety and environment, the costs of such disposal should be the responsibility of the generators and owners of such waste and spent fuel.
In section 302 of the 1982 Act, Congress authorized the Department of Energy ("DOE") to enter into contracts (the standard contract) with any person who generates or holds title to HLW or SNF of domestic origin, under which the DOE would accept title to, responsibility for subsequent transportation of, and responsibility for disposal of such HLW or SNF. For these services section 302 of the 1982 Act specifies that the contracts require contracting generators or title holders of SNF or HLW to pay fees to the DOE intended to offset its costs in performing these services. Such fees are deposited into a Nuclear Waste Fund until needed to effectuate the purposes of the 1982 Act.
Section 111(a)(5) of the 1982 Act provides that the generators and owners of HLW and SNF have the primary responsibility to provide for, and the responsibility to pay the costs of, the interim storage of such waste and SNF until such items are accepted by the Secretary of Energy for permanent disposal.
Power Generating Subsidiary, which subsequently became a member of the Taxpayer consolidated group, entered into standard contracts with the DOE and paid fees to the DOE pursuant to those contracts. Under provisions of the 1982 Act, the DOE was to develop a site that would accept radioactive waste from the country's nuclear power plants beginning on January 31, 1998. However, to date no permanent waste disposal site has been approved and the DOE has not accepted any SNF from anyone for permanent disposal.
All nuclear power plants in the United States initially store SNF in spent fuel pools.
These pools are robust constructions made of reinforced concrete several feet thick, with steel liners. The pools contain water which serves both to shield the radiation and cool the rods. After the SNF has cooled sufficiently, it may be stored using means other than pool storage.
By the late Years, with a nuclear waste repository years from completion and Facility 1 running out of pool storage space, Power Generating Subsidiary started the process of seeking approval to construct a dry cask independent spent fuel storage facility (SFSF) at Facility 1. This process required multiple governmental approvals.
In Year 1, Power Generating Subsidiary successfully obtained a license from the Nuclear Regulatory Commission (“NRC”) to construct a SFSF at Facility 1 with a capacity of A# casks. However, Power Generating Facility found it difficult to obtain state approval for the project. The SFSF was objected to by various parties including Group 1. Ultimately, through litigation it was determined that construction of the facility required legislative approval by State A.
In Year 2, State A enacted legislation approving, with conditions, the use of some dry cask storage at Facility 1. The Year 2 legislation conditioned expansion of additional dry cask storage on certain future acts by Power Generating Subsidiary. In subsequent legislative acts, State A approved additional dry cask storage at Facility 1 as well as dry cask storage at Facility 2. In the original legislation and the subsequent acts, State A required either the owner or the operator of the facilities to make payments into Fund.
For the taxable years at issue State A law required the owner of Facility 1 to make a payment of $A into Fund contingent on two conditions: (1) the nuclear generating plant was in operation that year, and (2) SNF was in dry cask storage at the plant location for any part of the year (Legislative Mandate 1). In Year 4, State A enacted a law requiring the owner of Facility 2 to make payments into Fund of $B for each dry cask of SNF stored at that facility during the year (Legislative Mandate 2). The obligation to make these payments was also contingent on (1) the nuclear generating plant being in operation for the year, and (2) SNF being in dry cask storage at the plant location for any part of the year. Amounts paid into Fund are not used to pay for either the direct or overhead costs of operating the dry cask storage facilities.
In Year 3, Power Generating Subsidiary entered into a Settlement Agreement with Group 1 under which Group 1 agreed not to challenge the initial authority for dry cask storage at Facility 1 nor to intervene, advocate, or otherwise participate in any state administrative or state legislative decision-making process or state judicial proceeding related to relicensing or authorizing additional dry cask storage at Facility 1 during a specified period. In return Power Generating Subsidiary agreed to make various payments benefiting Group 1. These payments included but were not limited to the following:

(1) Beginning on Date 1, $C each year during Facility 1 plant operations (Legislative Mandate 3);

(2) Beginning on Date 1, $D each year for the placement of SNF dry storage casks at Facility 1 (Legislative Mandate 4). This liability was to terminate when all SNF generated at Facility 1 and stored there during the operational life of the plant was removed from storage at Facility 1. The obligation does not apply to SNF stored in dry casks and placed at Facility 1 after operations have ceased and preparation for decommissioning of the facility has begun.

# LAW AND ANALYSIS

A. Overview
Section 172(a)1 allows a deduction for the taxable year equal to the aggregate of (1) the net operating loss (NOL) carryovers to such year, plus (2) the NOL carrybacks to such year. With certain modifications, § 172(c) defines a NOL as the excess of the deductions allowed by Chapter 1 of the Code over the gross income. Section 172(b)(1)(A) generally provides that a NOL for any taxable year is carried back to each of the 2 taxable years preceding the taxable year of the loss and carried forward to each of the 20 taxable years following the year of the loss. However, § 172(b)(1)(C) provides a 10-year carryback period for the portion of any NOL that qualifies as a specified liability loss.
Moreover, § 172(f)(3) provides a special carryback period for a specified liability loss attributable to amounts incurred in the decommissioning of a nuclear power plant (or any unit thereof). Section 172(f)(3) provides that except as provided in regulations, the portion of a specified liability loss which is attributable to amounts incurred in the decommissioning of a nuclear power plant (or any unit thereof) may, for purposes of § 172(b)(1)(C), be carried back to each of the taxable years during the period—

(1) beginning with the taxable year in which such plant (or unit thereof) was placed in service, and

(2) ending with the taxable year preceding the loss year.
Section 172(f)(1)(B)(i) defines a specified liability loss, in part, as any amount allowable as a deduction under Chapter 1 of the Code (other than § 468(a)(1) or § 468A(a)) which is in satisfaction of a liability under a federal or state law requiring the decommissioning of a nuclear power plant (or any unit thereof) that is taken into account in computing the NOL for the taxable year. Section 172(f)(1)(B)(ii) provides that a deduction for a liability may only generate a specified liability loss if (I) the act (or failure to act) giving rise to such liability occurs at least 3 years before the beginning of the taxable year (the 3-year rule), and (II) the taxpayer used an accrual method of accounting throughout the period or periods during which such act (or failure to act) occurred.
The examining agents and LB&I Counsel (the Field) agree with Taxpayer that Legislative Mandate 4 satisfies the 3-year rule. However, the Field contends that the other legislative mandates do not meet the requirements of the 3-year rule. For taxable years Year 5, Year 6, and Year 7, Taxpayer2 paid amounts to satisfy these legislative mandates. Taxpayer contends that the portion of NOLs generated by deductions attributable to those payments generated specified liability losses, eligible, under § 172(f)(3), to be carried back to the taxable years in which the nuclear power plants were first placed in service.
1 Unless indicated otherwise, references to sections refer to sections of the Internal Revenue Code of 1954 or 1986 as applicable.
2 Henceforth, references to Taxpayer are to be understood as also including Power Generating Subsidiary.
Before addressing whether Taxpayer satisfies the 3-year rule for the liabilities at issue, we note that we are not addressing the question of whether payment of the legislative mandates result in deductions nor whether, if such amounts are deductible, such deductions would be in satisfaction of a state or federal law requiring the decommissioning of a nuclear power plant (or any unit thereof)3. This memorandum only addresses application of the 3-year rule.
By using the phrase "the act or failure to act" rather than say "an act or failure to act"
§ 172(f)(1)(B)(ii) requires identifying a particular act or failure to act giving rise to the liability. However, the occurrence of a given event, such as the creation of a liability, results from an infinite series of necessary preceding causes. Because a number of acts or failures to act may satisfy a "but for" test with regard to causation of a given liability, the phrase "act or failure to act" cannot be said to be free from ambiguity. Therefore, one must examine the legislative history of § 172(f)(1)(B)(ii) to determine which act or failure to act in the chain of causation leading to the creation of a given liability to treat as "the" act or failure to act for purposes of § 172(f)(1)(B)(ii).
B. The Legislative History
Prior to the enactment of the economic performance requirement in § 461(h), §1.461-1(a)(2) of the Income Tax Regulations generally treated an accrual method taxpayer as incurring a liability for federal income tax purposes when the following two-pronged (the all-events test) test was satisfied:

(1) all the events had occurred that established the fact of the liability, and

(2) the amount of the liability could be determined with reasonable accuracy.
The Treasury Department became concerned when courts began interpreting the two pronged all-events test in a manner that allowed accrual method taxpayers to deduct liabilities far in advance of when the liabilities had to be satisfied by payment or other performance. Because of the time value of money, the benefit to taxpayers from such accruals could be substantial. The Treasury Department's concern became particularly acute in the early 1980s with the advent of historically high United States interest rates.
3 For prior taxable years Taxpayer achieved some success in holding the DOE liable, as mitigation damages for partial breach of the standard contracts, for reimbursement of at least some portion of these payments. For the taxable years at issue, however, Taxpayer settled its partial breach of contract claims against the DOE in part by surrendering any reimbursement claims associated with payments attributable to the legislative mandates. The DOE has argued that such payments are not reimbursable as mitigation damages for partial breach of the standard contracts because such liabilities were not foreseeable as a probable result of the breach when the parties entered into the contracts. We also note that there is a legitimate question whether State A was preempted under the Atomic Energy Act of 1954, as amended, from imposing these legislative mandates if these mandates may be properly characterized as a cost imposed by State A to obtain its approval for the dry cask storage. We do not address any of these issues in this memorandum.
For example, state and/or federal laws generally require miners to restore the surface of land they strip mine to a condition comparable to its pre-mined state. A miner's legal obligation to restore arises when the miner disturbs the land, although actual restoration may not occur until much later.
If strip miners failed to reasonably estimate future costs to restore the land, the Service succeeded in preventing them from deducting estimated restoration costs for taxable years when the land was disturbed. See e.g. Patsch v. Commissioner, 208 F.2d 532, 534-535 (3d Cir. 1953); Commissioner v. Gregory Run Coal Co., 212 F.2d 52, 57-58 (4th Cir.), cert. denied, 348 U.S. 828 (1954). On the other hand, if the deductions claimed were based on reasonably accurate estimates of costs to restore, the courts generally allowed the strip miners to deduct the estimated costs for the taxable years when the land was disturbed. See e.g. Harrold v. Commissioner, 192 F.2d 1002, 1006 (4th Cir. 1951); Denise Coal Co. v. Commissioner, 271 F.2d 930, 936 (3d Cir. 1959);
Ohio River Collieries Co. v. Commissioner, 77 T.C. 1369, 1377 (1981).
Likewise, Treasury became concerned when courts concluded that the occurrence of a work-related injury satisfied the first prong of the all-events test in the case of uncontested self-insured workmen's compensation liabilities, thereby allowing taxpayers that could reasonably estimate liabilities to be paid well in the future, such as workmen's compensation disability or survivor annuities, to deduct such amounts currently rather than when actually paid. See, e.g., Crescent Wharf & Warehouse Co. v. Commissioner, 59 T.C. 751 (1973), rev’d & remanded, 518 F.2d 772 (9th Cir. 1975); Wien Consolidated Airlines, Inc. v. Commissioner, 60 T.C. 13 (1973), aff'd, 528 F.2d 735 (9th
Cir. 1976).
Another situation that concerned Treasury and involved a much greater potential for a taxpayer to deduct an amount far in excess of the present value of the legal obligation giving rise to that deduction involved the obligation to decommission a nuclear power plant. In the case of a nuclear power plant the legal obligation to decommission could arise well in advance of the time when the decommissioning was completed.4 The executive branch decided to seek a legislative solution to the problem caused by cases such as Ohio River Collieries. Specifically, the executive branch proposed adding an "economic performance" requirement to the all-events test. See Staff of the Joint Committee on Taxation, Summary of Administration's Revenue Proposals in the Fiscal Year 1985 Budget Proposal 31 (Comm. Print 1984). Under the proposed change, the 4 Decommissioning a nuclear power plant requires reducing the level of radioactivity in the plant to a level considered safe for unrestricted use. Some methods of decommissioning may take over 100 years to complete. See Timing and Measurement of Taxpayer Deductions for Obligations to be Paid in the Future:
Hearing Before the Subcommittee on Oversight on the Committee on Ways and Means House of
Representatives, 98th Cong., 2d Sess. 112 (February 24, 1984)(statement of Donald W. Kiefer, Congressional Research Service, Library of Congress). all-events test would be "clarified" so that with certain exceptions, deductions would not be permitted until services were performed, the use of property actually occurred, or in the case of workmen's compensation or similar liabilities, the liability was actually satisfied. Id.
In February 1984, the Subcommittee on Oversight of the House Ways and Means Committee held a hearing on the proposal to deal with "premature accruals" by the addition of a new economic performance test. See Timing and Measurement of
Taxpayer Deductions for Obligations to be Paid in the Future, Hearing Before the
Subcommittee on Oversight of the Committee on Ways and Means House of
Representatives, 98th Cong., 2d Sess. (February 24, 1984). Most of the taxpayers and tax practitioners who testified at the hearing objected to the proposal because in their view, it would result in a mismatching of revenue and expenses.
For example, in the case of mining reclamation if reclamation costs can only be deducted in the taxable year when the work is actually done, such deductions will not be matched with the earlier gross income they helped to generate. If the deferred deductions cannot be used to reduce tax liability when they are allowable, the economic performance requirement would not just eliminate the time value of money economic distortion caused by a premature accrual. It would effectively penalize the taxpayer from a time value of money standpoint.
To remedy this potentially unfavorable result, Treasury officials proposed liberalizing the NOL carryback provisions for certain deductions significantly deferred because of the economic performance requirement:
We recognize that requiring deductions for future expenses to be taken in the year of economic performance also requires that the net operating carryback rules be amended to insure that taxpayers are not overtaxed. Our proposals provide for extension of the carryback period in appropriate circumstances to insure that the deferred expenses will be able to be fully utilized.
Generally expenses attributable to liabilities arising more than 3 years prior to economic performance will be permitted to be carried back for a period not to exceed 10 years, subject to certain transition rules. Special carryback rules might be appropriate for certain expenses to be paid in the future such as the nuclear powerplant decommissioning costs.
Id. at 7 (statement of Ronald A. Pearlman Deputy Assistant Secretary for Tax Policy, U.S. Treasury).
Congress adopted the proposed economic performance requirement by enacting § 461(h) in § 91(a) of the Tax Reform Act of 1984 (the 1984 Act), and in § 91(d) of that act, Congress simultaneously enacted a provision allowing expanded carryback periods for NOLs generated by certain deductions. In the 1984 Act, Congress generally allowed a 10-year carryback period for the portion of an NOL that qualified as a deferred statutory or tort liability loss. However, Congress provided for an alternative special carryback period for certain losses incurred in decommissioning a nuclear power plant.
In § 91(d)(2) of the 1984 Act, Congress enacted a new § 172(k) of the 1954 Code which defined a deferred statutory or tort liability loss as follows:

(1) The term “deferred statutory or tort liability loss” means for any taxable year, the lesser of—

(A) the net operating loss for such taxable year, reduced by any portion thereof attributable to—

(i) a foreign expropriation loss, or

(ii) a product liability loss, or

(B) the sum of the amounts allowable as a deduction under this chapter (other than any deduction described in subsection (j)(1)(B)) which—

(i) is taken into account in computing the net operating loss for such taxable year, and

(ii) is for an amount incurred with respect to a liability which arises under a federal or state law or out of any tort of the taxpayer and –

(I) in the case of a liability arising out of a federal or state law, the act (or failure to act) giving rise to such liability occurs at least 3 years before the beginning of such taxable year, or

(II) in the case of a liability arising out of a tort, such liability arises out of a series of actions (or failures to act) over an extended period of time a substantial portion of which occurs at least 3 years before the beginning of such taxable year.
A liability shall not be taken into account under the preceding sentence unless the taxpayer used an accrual method of accounting throughout the period or periods during which the acts or failures to act giving rise to such liability occurred.
Section 91(d)(2) of the 1984 Act also provided for the special carryback period for NOL carrybacks attributable to deductions incurred in decommissioning a nuclear power plant.
Although the House and Senate Reports to the 1984 Act describe the operation of the proposed new 10-year NOL carryback provision, neither of these reports discuss the reason for its enactment. The Conference Report, however, provides:
The House bill provides a 10-year carryback for net operating losses attributable to certain liabilities deferred under these provisions … The provisions of the bill apply generally to expenses incurred (without regard to the economic performance requirement) after the date of enactment.
CONFERENCE AGREEMENT
The conference agreement generally follows the House bill, . . .
H.R. (Conf.) Rep. No. 861, 98th Cong., 2d Sess. 872-73 (1984). Examination of the quoted language's context makes clear that the reference to provisions deferring liabilities refers to the economic performance requirement.
The legislative history indicates that Congress' primary concern when it enacted the special NOL carryback rules in the 1984 Act was to enhance the probability that taxpayers, whose deduction of certain liabilities was deferred because of the economic performance requirement, be able to use those deductions when finally allowable to offset gross income. This could be done either in the taxable year allowable or in prior taxable years through the vehicle of an NOL carryback. Thus, Congress only meant to provide relief for existing liabilities the deduction of which is deferred for a prescribed period.
In § 11811 of the Omnibus Budget Reconciliation Act of 1990 (the 1990 Act), Congress reorganized the provisions in § 172. Congress placed the 10-year carryback for product liability losses and what had previously been called deferred statutory or tort liability losses under the same subsection of § 172, namely § 172(f), labeling both types of losses, although substantively distinct, specified liability losses. The 1984 Act statutory provisions that formerly defined a deferred statutory or tort liability loss were recodified in the 1990 Act as § 172(f)(1)(B). These amendments were essentially an exercise in statutory housecleaning and reorganization. There were no substantive changes to what had formerly been called deferred statutory or tort liability losses.
C. Prior 3-Year Rule Litigation
Following the enactment of the extended carryback period for deferred statutory or tort liability losses, some taxpayers adopted an aggressive stance regarding when the act or failure to act occurred for purposes of the 3-year rule. These taxpayers asserted that the relevant act or failure to act occurred when an event occurred that made the incurrence of the liability foreseeable, not when an act or failure to act actually made the taxpayer liable for the particular obligation. For example, some taxpayers contended that initially hiring employees constituted the relevant act regarding all future payroll tax liabilities instead of the act occurring when the employees either earned or were paid wages. Likewise, some taxpayers took the position that the act giving rise to all real property tax liabilities a taxpayer incurred with respect to a particular parcel of real property occurred when the taxpayer initially acquired the property.
This state of affairs ultimately resulted in litigation. In Sealy Corp. v. Commissioner, 107 T.C. 177 (1996), aff'd, 171 F.3d 655 (9th Cir. 1999), the petitioners asserted that the portion of NOLs generated by deductions for the following expenses constituted specified liability losses within the meaning of § 172(f)(1)(B):

(1) professional fees incurred to comply with reporting, filing, and disclosure requirements imposed by the Securities and Exchange Act of 1934,

(2) professional fees incurred to comply with ERISA reporting requirements, and

(3) professional fees incurred in connection with an IRS income tax audit.
This case might have been decided in favor of the government on the grounds that the 3-year rule was not satisfied with regard to any of the liabilities. However, the Tax Court did not focus on application of the 3-year rule. Instead, the Tax Court held that deduction of the above expenses did not result in specified liability losses because the liabilities for the expenses did not arise under a federal or state law within the meaning of § 172(f)(1)(B). The Tax Court gave three reasons for its conclusion.
First, the court noted that the federal law cited by the petitioners did not establish the petitioners' liability to pay the amounts at issue. The petitioners' liability did not arise until the services were contracted for and received and the petitioners' choice of the means of compliance, rather than the cited regulatory provisions, determined the nature and amount of their costs. If the petitioners had failed to comply with the auditing and reporting requirements or had not obtained the particular services at issue, their liability would not have been measured by the value of the services they actually contracted for and received. 107 T.C. at 184.
Second, the court read the legislative history of § 172(f)(1)(B) to suggest that Congress intended the provision to apply only to liabilities the deduction of which the economic performance requirement caused to be deferred. Because the economic performance requirement did not delay petitioner’s accrual of the deductions at issue, the court concluded that Congress did not intend for NOLs generated by those deductions to qualify as specified liability losses. Id. at 185-86.
Finally, in determining the scope of liabilities arising under either federal or state law within the meaning of § 172(f)(1)(B), the court considered the specific types of liabilities referred to in § 172(f): product liability, nuclear decommissioning liabilities, and torts.
Invoking the statutory construction rule of ejusdem generis, the court concluded that Congress intended the 10-year carryback to apply to a relatively narrow class of liabilities similar to those identified in the statute. The court thought the costs at issue in Sealy were routine costs not like those identified in the statute. Id. at 186.
On appeal, however, the Ninth Circuit did address the failure to satisfy the 3-year rule:
It is, therefore, not simply an expense incurred with respect to an obligation under federal law but an act "giving rise" to the liability that qualifies as a specified liability under the statute. The act giving rise to each of the liabilities in question was the contractual act by which Sealy engaged lawyers or accountants. In each of these instances the act did not occur at least three years before the beginning of the taxable year.
Sealy's argument essentially is that the act giving rise to the liability is the first event in a chain of causes which gives rise to the liability. The argument leads to a reductio ad absurdum. The organization of the company gave rise to an obligation to comply with all pertinent state and federal laws and thereby gave rise to the liabilities incurred in complying with these laws. According to this logic, every corporation would have a specified liability carryback for all costs the corporation incurred to comply with relevant laws. Congress did not create such a windfall.
171 F.3d 657-58.
In the wake of Sealy, the Service took the position that only a narrow class of liabilities arose under federal or state law for purposes of § 172(f)(1)(B). First, the liability had to literally be imposed by a federal or state law. Second, the liability had to be a kind whose fundamental nature meant that the deduction of the liability would be delayed because of the economic performance requirement (“inherent delay liabilities”). For example, a mining reclamation liability imposed by state or federal law qualifies as an inherent delay liability. This is because in the normal course of events there will be a significant delay between when the mine is created and the obligation to restore the land arises and when the land is actually restored. Because any restoration deduction will not be allowable until the land is restored, there is an inherent delay between the creation of the liability and the deduction attributable to the economic performance requirement.
Following this inherent delay approach, the Service asserted state income tax liabilities did not arise under state law under § 172(f)(1)(B) because such liabilities do not qualify as inherent delay liabilities. State income tax liabilities are generally paid either before or shortly after they accrue. There may be a significant delay between when such liabilities arise and when they are deductible if the taxpayer does not timely pay the liabilities. However, in the ordinary course of events the taxpayer is expected to pay its state income tax liabilities when due.
In Host Marriott Corp. v. United States, 113 F.Supp. 2d 790 (D. Md. 2000), aff”d, 267 F.3d 363 (4th Cir. 2001) the Service took the position that deductions attributable to liabilities for workers’ compensation and interest on federal income tax underpayments did not arise under either federal or state law for purposes of § 172(f)(1)(B). The district court rejected that position, noting that state and federal statutes imposed those liabilities. The district court concluded that that was all that was required to satisfy the arising under federal or state law requirement. This position was upheld on appeal.
The district court also concluded that the act or failure to act, for purposes of applying the 3-year rule to the workers’ compensation liabilities occurred when the workers were injured.
The interest on the federal income tax underpayments related to income tax returns for fiscal 1977, 1978, and 1979 tax years. The Service began auditing those returns in
1981. At the conclusion of the audit, the Service proposed income tax deficiencies for the tax years 1977-79. The taxpayer challenged the assessments and settled the deficiencies with the Service in 1991.
The Service maintained that even if the interest on the underpayments arose under federal law, the portion of the interest that economically accrued 3 or less years before the beginning of the taxable year such interest was deductible did not satisfy the 3-year rule. The district court disagreed, concluding without analysis that the initial failure to timely pay the tax when due was the relevant act to determine if the 3-year rule was satisfied for all of the interest.
The Fourth Circuit agreed with the district court on this point:
[T]he parties to this case stipulated that the taxpayer's interest liability "was imposed under § 6601” (imposing interest on unpaid taxes) and that the district court found that the taxpayer's § 6601 liability occurred upon the filing of its 1977, 1978, and 1979 tax returns. Section 6622, providing for daily compounding of interest, provides simply the rate of interest and the method of calculating the amount of liability, but does not, of its own force, impose liability. We agree with the taxpayer that the three-year requirement of § 172(f)(1)(B)(i) (1991) does not operate on the continuing calculation of a loss from the liability, but on "the act (or failure to act) giving rise to such liability." … Here, the acts giving rise to interest liability under § 6601 were the deficiencyproducing filings in 1978, 1979, and 1980 for the taxable years 1977-795.
5 The language used in both the district court and the court of appeals’ opinion literally state that the filing of tax returns reporting less than the correct amount of tax (“deficiency producing filings”) constitutes the relevant act for purposes of determining when the entire interest liability arises for purposes of § 172(f).
However, whether a taxpayer files a tax return or not, interest begins to run on the due date of the return if the taxpayer underpays the taxpayer’s tax liability. Therefore, we view the courts’ opinions as adopting the position that the initial underpayment of federal income tax constitutes the relevant act to determine whether the 3-year rule is satisfied irrespective of whether a return is filed.
267 F.3d at 365. Thus, in Host Marriot, the courts viewed the initial underpayment of tax as the act giving rise to the interest liability. It appears that they viewed the subsequent accrual of additional interest attributable to the passage of time with the continued failure to pay the amount owed as simply affecting the amount of the liability.
In Intermet Corp. v. Comm'r, 117 T.C. 133 (2001), the petitioner claimed that the portion of an NOL generated by deductions attributable to state tax deficiencies for the taxpayer’s taxable years 1986, 1987, and 1988, and interest thereon, as well as interest on a federal tax deficiency for 1987, all paid in 1992, qualified as a specified liability loss. Once again, the Service contended that these liabilities did not arise under federal or state law within the meaning of § 172(f)(1)(B) because they did not qualify as inherent delay liabilities. Additionally, the Service asserted that, assuming arguendo that the disputed federal and state interest payments constitute specified liability losses in the first instance, a portion of those interest payments did not qualify for carryback under § 172(f)(1)(B)(i) because they were not incurred at least 3 years prior to the beginning of the taxable year 1992. Specifically, the Service maintained that all interest that accrued within 3 years of January 1, 1992, did not satisfy the 3-year rule.
The Tax Court agreed with the petitioner. In distinguishing Sealy, the Tax Court only referred to the first reason (i.e., federal law did not establish the liability to pay the amounts at issue) that it gave in Sealy for concluding that the professional fee liabilities at issue in that case did not arise under federal or state law. Consistent with Host Marriot, the Tax Court stated that the liabilities at issue qualified as arising under federal or state law within the meaning of § 172(f)(1)(B) because such liabilities were directly imposed by federal or state law.
Likewise, consistent with the rationale stated in Host Marriot, the Tax Court concluded that all of the interest liabilities at issue satisfied the 3-year rule:
Respondent contends that the act giving rise to interest on a tax deficiency arises daily as the taxpayer fails to pay the underlying tax.
We hold that the act giving rise to petitioner's liability for interest on its Federal and State tax deficiencies was the act of filing erroneous tax returns, and, as a consequence, failing to pay the correct amount of tax on or before the last date prescribed for payment. See Host Marriott Corp. v. United States, supra. Simply put, respondent's position confuses the method of computing interest under section 6621, under which additional interest accrues each day that a tax liability remains unpaid, with the act giving rise to the liability for interest; i.e., failure to pay the tax on or before the prescribed date.
117 T.C. at 141.
In Notice 2005-20, 2005-1 C.B. 635, the Service abandoned its position that only inherent delay liabilities arise under federal or state law for purposes of § 172(f)(1)(B).
Consistent with the Tax Court’s first rationale for its decision in Sealy, the Service stated that to arise under federal or state law within the meaning of § 172(f)(1)(B), the liability must be directly imposed by federal or state law and must not be the result of decisions made by the taxpayer or others.
In Notice 2005-20 the Service maintained that the act or failure to act resulting in the establishment of a legal liability constitutes “the act or failure to act” for purposes of applying § 172(f)(1)(B)(i). In the case of interest on unpaid federal or state taxes, the Service reaffirmed its position that a taxpayer's use of the government's money over discrete periods, such as days, months or portions of a month, is an essential element that creates the liability. The Service stated its view that the courts in Host Marriott and Intermet incorrectly concluded that the initial failure to pay the taxes when due constituted the act or failure to act giving rise to any interest that economically accrued during the taxable year such interest was deductible and the 3-year period prior to the beginning of that taxable year. Consequently, the Service informed taxpayers that it would continue to assert that interest that economically accrues on a liability for unpaid taxes in the taxable year such interest is deductible and the 3-year period prior to the beginning of that taxable year does not satisfy the 3-year rule.
D. Legislative Enactment of a Narrow Class
In § 3004 of the Tax and Trade Relief Extension Act of 1998 (the 1998 Act), Congress restricted the types of liabilities the deduction of which could generate a specified liability loss to five enumerated liabilities (in addition to product liability losses), including federal or state law liabilities to decommission a nuclear power plant (or any unit thereof). As noted above, prior to the 1998 Act, a specified liability loss could be based on any deduction arising out of a federal or state law provided the additional requirements of the statute were satisfied. However, Congress made no substantive change to the 3-year rule when it restricted specified liability loss treatment to deductions for a narrow specified class of liabilities arising under federal or state law.
Consequently, the legislative history to the 1984 Act continues to apply in interpreting the 3-year rule in the current statute.
E. What is the Appropriate Act or Failure to Act in this Case?
1. Overview
The legislative history indicates that Congress' primary concern when it enacted the § 172(f)(1)(B) language pertinent to this case was to ensure that taxpayers, whose deduction of certain liabilities was deferred because of the economic performance requirement, be able to use those deductions when finally allowable to reduce taxable income, either in the taxable year allowable or in prior taxable years through the vehicle of the new 10-year NOL carryback. The statutory language is not limited to liabilities that satisfy the first and second prong of the all-event tests but the deduction of which is deferred solely because of the economic performance requirement. The deduction of an existing legal obligation may be deferred for other reasons and still satisfy the statutory requirements. For example, a taxpayer may be liable for an obligation that the taxpayer cannot reasonably estimate, thereby failing to satisfy the second prong of the all-events test. However, Congress only meant to provide relief for existing liabilities the deduction of which is deferred for a prescribed period.
To effectuate this intent, consistent with Notice 2005-20, we believe the final act or failure to act in the chain of causation leading to the creation of a given liability from which it can be determined that the taxpayer has a legal obligation qualifies as "the act or failure to act" within the meaning of § 172(f)(1)(B)(ii). Treating an act or failure to act occurring any earlier than this as the relevant act or failure to act for section 172(f)(1)(B)(ii) purposes could frustrate the intent of Congress by allowing an extended carryback period for deductions for liabilities involving little or no deferral between the actual creation of the liability and the allowance of the deduction therefore.
Next we must apply this principle to the problematic facts of this case. The Field raised no issue regarding whether federal or state law imposes the legislative mandates. As previously noted, in applying the 3-year rule to these facts we assume, but express no opinion thereon, that payment of those liabilities are in satisfaction of a liability requiring the decommissioning of a nuclear power plant (or any unit thereof).
Taxpayer asserts that the relevant act occurred when the operating licenses for the nuclear power plants were issued and the plants were first placed in service. Taxpayer did incur a general obligation to decommission each nuclear power plant when the plant was initially placed in service. However, at that time Taxpayer was under no legal obligation to decommission plant components that would not be added to the plant until the future. The granting of a license to operate a nuclear power plant does not create a liability to decommission nuclear fuel assemblies that have not even been fabricated when the license is issued and the plant is first placed in service. When the plant is first placed in service, it may have been reasonably foreseeable that Taxpayer might incur future obligations to decommission nuclear fuel assemblies subsequently placed in service. However, as the Ninth Circuit pointed out in Sealy, simply pointing to an act in the chain of causation leading to the creation of a liability is not enough to satisfy the 3year rule. The act or failure to act that satisfies the 3-year rule must actually result in the creation of the liability.
Next we look to obligations imposed under the 1982 Act. Section 111(a)(5) of the 1982 Act provides that the generators and owners of HLW and SNF have the primary responsibility to provide for, and the responsibility to pay the costs of, the interim storage of such waste and SNF until such items are accepted by the Secretary of Energy for permanent disposal. Section 2(23) of the 1982 Act provides that “the term ‘spent nuclear fuel’ means fuel that has been withdrawn from a nuclear reactor following irradiation, the constituent elements of which have not been separated by reprocessing.” When a nuclear fuel assembly is permanently withdrawn from a nuclear reactor, and most likely as soon as the assembly is first placed into the reactor and irradiated, §111(a)(5) of the 1982 Act imposes an obligation on the owners of the assembly to pay the interim storage costs6 for that assembly. An argument can be made that once these events have occurred, the taxpayer is liable for whatever it ultimately costs to pay for the interim storage. The relevant question here is whether the post-1982 Act State A statutes imposing the legislative mandates on Taxpayer affect the amount of Taxpayer’s liability rather than imposing new liabilities.
Certainly, there can be significant periods between the creation of a liability and the occurrence of the events that make that liability deductible. For example, in the case of a mining reclamation liability, the taxpayer becomes liable to restore the land when the land is initially disturbed. Restoration may not occur until many years later. If the taxpayer subsequently contracts with a third party who agrees to perform the restoration work, the taxpayer has no liability to the third party under state law until the third party actually performs restoration work pursuant to the contract. However, in contrast to the situation in Sealy, entering into a contract with a third party does not create the relevant liability. If the third party fails to perform under the contract, the taxpayer remains directly liable under federal or state law for what it costs to restore the land.
In Crescent Wharf & Warehouse the petitioner elected to be self-insured for a portion of its workers’ compensation liabilities. For the taxable year in which a worker was injured, the petitioner took the position that it could accrue and deduct what it expected to pay for uncontested claims for medical benefits, disability payments, and death benefits associated with an injury. The Tax Court concluded that the occurrence of the injury was just the first, albeit important step in fixing the petitioner's liability. The Tax Court concluded that all the events necessary to fix the liability for medical benefits did not occur until the actual rendering of the medical services.
On appeal the Ninth Circuit reversed the Tax Court. That court concluded that all the events giving rise to the workers’ compensation liabilities occurred when the worker was injured. The Ninth Circuit apparently believed that the fact that the exact mix of medical services and the prices that would be charged for those services were not known at the time of the injury only affected the amount of the liability. Variations in the actual services to be provided and the amount that might be charged for those services were certainly foreseeable when the injury occurred.
Granted, if the State A statutes impose decommissioning liabilities, these liabilities are of the same class as those imposed under section 111(a)(5) of the 1982 Act. Drawing a 6 For purposes of this memorandum “interim storage costs” means all storage costs incurred before permanent disposition of the SNF by the DOE. distinction between subsequent events that merely affect the amount of an existing liability, as contrasted with events that create a new liability is not without difficulty.
However, when a subsequent legislative act results in a materially different obligation that was not foreseeable when the initial liability came into existence, we believe the subsequent event should be treated as creating a new liability, even if this liability is of the same class as the previously existing liability. We agree with the DOE that when the standard contracts were entered into by Taxpayer, and when the 1982 Act was enacted, the types of obligations imposed by State A pursuant to the legislative mandates were not foreseeable. Therefore, we conclude that the legislative mandates constitute new liabilities that must independently satisfy the 3-year rule.
Analyzing the obligations imposed by the legislative mandates, the Field contends that Taxpayer has no liability in a given year to pay Legislative Mandate 3 unless the nuclear power plant at Facility 1 is in operation for that year. Likewise, the Field asserts that Taxpayer has no obligation to make payments into Fund for any of the SNF stored at a particular generating facility for a year unless the nuclear power plant at that facility is in operation for that year. Therefore, the Field contends that the final act giving rise to an obligation to make a legislative mandate payment other than one in satisfaction of Legislative Mandate 4 does not occur until the relevant nuclear power plant is operated by Taxpayer. Because Taxpayer will pay and deduct these liabilities in a taxable year beginning less than 3 years after the operation of the plant giving rise to the liability, the Field asserts that the 3-year rule is not satisfied for Legislative Mandates (1)-(3).
In Crescent Wharf & Warehouse most of the disability payments imposed by the workers’ compensation statutes were periodic in nature. These payment were to be made either "during the period of such disability" or "during the continuance" thereof.
The Tax Court concluded that the continuance of the disability was ordinarily a condition precedent to the petitioner’s liability to make the payments. Thus, the first prong of the all-events test was not satisfied when the worker first became disabled. In the Ninth Circuit’s view, the factors cited by the Tax Court just affected the amount of the liability.
The Ninth Circuit apparently viewed the possibility that an injured worker might recover from a disability causing injury as a condition subsequent that extinguished an existing legal obligation.
In Wien Consolidated Airlines three of the petitioner’s pilots were killed in airplane crashes in the course of their employment. The applicable Alaska workers’ compensation statutes required the taxpayer to make periodic payments to the widows and children of the deceased pilots. Payments had to be made to each widow until her death or remarriage. Payments had to be made to each child until the child attained age 19, subject to termination upon the child’s death. In Wien two of the widows had remarried prior to resolution of the case.
The petitioner did not contest liability to make the payments. For the taxable year of each pilot’s death, based on actuarial tables, the petitioner accrued and deducted what it anticipated it would be required to pay in total to each beneficiary. The Service asserted that the contingencies of death or remarriage were conditions precedent, making the petitioner's liability conditional.
The Tax Court, upheld by the Ninth Circuit on appeal, disagreed with the Service’s view.
The Tax Court concluded that all events fixing the workers’ compensation liabilities occurred when the pilots were killed. The contingencies of death or remarriage constituted conditions subsequent which, if they occurred, would extinguish an existing liability7.
In Rev. Rul. 80-191, 1980-2 C.B. 168, the Service announced it would not follow Crescent Wharf and Wien, and would continue to disallow the accrual of workers' compensation liabilities subject to the types of contingencies in those cases. Following the issuance of that revenue ruling, the Service continued to lose on this issue. The issue became moot on the deductibility issue for similar liabilities after the 1984 Act because the economic performance requirement required that accrual method taxpayers pay the liabilities in order to deduct them.

(2) Specific Liabilities

(A) Legislative Mandate 4
Although the Field has conceded that Legislative Mandate 4 satisfies the 3-year rule, it is useful to examine the rationale for that concession because it is relevant to the analysis required with respect to the other liabilities. First, Taxpayer and Group 1 entered into the Settlement Agreement that finally determined the nature of the liability in Year 3, a year which precedes the deductions at issue by more than 3 taxable years.
Second, Legislative Mandate 4 requires Taxpayer to make payments to Group 1 each year for the placement of dry casks containing SNF at Facility 1. This liability continues until all SNF in dry casks placed at Facility 1 during the operational life of the plant are removed from Facility 1.
Until the DOE accepts the SNF stored at Facility 1 or until Taxpayer finds an alternative interim storage location, Taxpayer has no viable alternative to storing the SNF produced 7 See also Burnham Corp. v. Commissioner, 90 T.C. 953 (1988), aff’d, 878 F.2d 86 (2d Cir. 1989). In that case the taxpayer settled a patent infringement action by agreeing to make a fixed number of periodic monthly payments to the plaintiff or her estate after which the taxpayer would make additional periodic payments for the remainder of the plaintiff’s life. On its return for the taxable year of the settlement the taxpayer deducted the present value of the payments it anticipated it would have to make to the plaintiff based on actuarial tables. In the litigation the taxpayer contended that no discount was required. The Service argued that at settlement there was no fixed liability to make some of the payments because the obligation to make those payments was contingent on the survival of the plaintiff. Both the Tax Court and the Ninth Circuit concluded that the taxpayer’s liability was fixed by the settlement agreement. The courts viewed the plaintiff’s death as a condition subsequent that would terminate an existing liability. For federal income tax purposes, a condition subsequent is only relevant in calculating the amount of a liability, not whether the liability exists. Such conditions are only relevant in determining if the liability can be estimated with reasonable accuracy. and leaving wet pool storage at Facility 1 other than at the dry cask storage facility at Facility 1. Taxpayer, to no avail, made efforts and incurred expenditures to find alternative storage sites outside of State A. As a practical matter, absent a legislative act that relieves Taxpayer of the obligation or a judicial finding that the obligation is unenforceable, Taxpayer is saddled with the obligation to make these payments until the SNF produced at Facility 1 during its operational life is removed from storage there.
Consequently, no later than entry into the Settlement Agreement and placement of casks containing SNF at Facility 1, Taxpayer had a liability to make yearly payments required by Legislative Mandate 4. Continuing liability to make these payments is subject to being extinguished by a condition subsequent, that is, removal of the casks from Facility 1.

(B) Legislative Mandate 1
Taxpayer is liable to make the payments required by Legislative Mandate 1 (A) each year the nuclear generating plant at Facility 1 is in operation, and (B) SNF is stored in a dry cask at Facility 1. State A enacted this legislative mandate in Year 3, a year that precedes the beginning of any of the taxable years at issue by more than 3 years. As noted in the discussion of Legislative Mandate 4, Taxpayer’s avoidance of this liability through removal of all the SNF in dry cask storage at Facility 1 is a condition subsequent that would extinguish an existing liability.
The Field contends that operation of the nuclear generating plant at Facility 1 for a year is a condition precedent necessary to create a Legislative Mandate 1 liability for that year. Because this event will occur less than 3 years before the beginning of the taxable year in which Taxpayer makes the payment, the Field argues that deductions for those liabilities do not satisfy the 3-year rule.
The statutes do not elaborate on what is meant by “the plant is in operation”. State A law does not impose continuing liability to make payments after discontinuation of operation of the nuclear power plant at Facility 1 unless Taxpayer fails to satisfy certain conditions. Based on this one might argue that the phrase “the plant is in operation” means during the operational life of the plant, whether the plant is generating electricity during a particular year or not. In other words, the liability continues until the plant is permanently shut down.
In our view this interpretation is unwarranted. If State A had meant to impose liability until the plant was permanently shut down they could have easily done so. Nuclear power plants may be shut down for extended periods, not just for refueling and periodic maintenance, during the operational life of the plant. For example, a nuclear accident occurred at the Unit 2 reactor in 1979 at Three Mile Island. When this happened the Unit 1 reactor at Three Mile Island was already shut down for refueling. After the accident the Unit 1 reactor, which did not suffer an accident, was kept shut down during lengthy proceedings by the NRC. During the shutdown, the plant was modified and training and operating procedures were revamped in light of the lessons learned from the accident. The Unit 1 reactor remained shut down until October 1985.
Likewise, Japan shut down all of its nuclear reactors after the 2011 Fukushima nuclear disaster. That country is now in the process of restarting those reactors after making various safety changes to the plants.
We can only speculate as to why State A enacted a statute that would excuse Taxpayer from making a payment under Legislative Mandate 1 if the plant was not in operation for a year. However, it is not unreasonable to assume that the State A legislature believed that it would be inappropriate to impose such liability on Taxpayer for the year if the plant did not generate revenue to pay the liability. Although a nuclear power plant may have various levels of operation, we believe that State A was focusing on the commercial operation of the plant when it required that the plant be in operation during the year as a precondition to imposing the liability.
Taxpayer must expend significant amounts of money and make great efforts to keep the plant at Facility 1 producing revenue generating power during any given year. We agree with the Field that operating the plant for a given year is a condition precedent to the creation of a Legislative Mandate 1 liability for that year. Therefore, we conclude that Legislative Mandate 1 does not satisfy the 3-year rule.

(C) Legislative Mandate 2
Legislative Mandate 2 fails to satisfy the 3-year rule for the same reason that Legislative Mandate 1 fails to satisfy the rule. In addition, we note that State A did not enact Legislative Mandate 2 until Year 4 which occurred less than 3 years before the beginning of some of the taxable years at issue.

(D) Legislative Mandate 3
Legislative Mandate 3 fails to satisfy the 3-year rule for the same reason that Legislative Mandate 1 fails to satisfy the rule.
CASE DEVELOPMENT, HAZARDS AND OTHER CONSIDERATIONS
Please call (202) 317-7006 if you have any further questions.