Wells Fargo & Co. And Subsidiaries v. United States , 641 F.3d 1319 ( 2011 )


Menu:
  •   United States Court of Appeals
    for the Federal Circuit
    __________________________
    WELLS FARGO & COMPANY AND SUBSIDIARIES,
    Plaintiff-Appellant,
    v.
    UNITED STATES,
    Defendant-Appellee.
    __________________________
    2010-5108
    __________________________
    Appeal from the United States Court of Federal
    Claims in case no. 06-CV-628, Judge Thomas C. Wheeler.
    __________________________
    Decided: April 15, 2011
    ___________________________
    DAVID FARRINGTON ABBOTT, Mayer Brown, LLP, of
    New York, New York, argued for plaintiff-appellant.
    With him on the brief were BRIAN W. KITTLE; and
    STEPHEN M. SHAPIRO, JOEL V. WILLIAMSON, THOMAS C.
    DURHAM, TIMOTHY S. BISHOP and MICHAEL R. EMERSON, of
    Chicago, Illinois.
    CORY A. JOHNSON, Trial Attorney, Commercial Litiga-
    tion Branch, Tax Division, United States Department of
    Justice, of Washington, DC, argued for defendant-
    appellee. On the brief were JOHN A. DICICCO, Acting
    WELLS FARGO   v. US                                        2
    Assistant Attorney General, RICHARD FARBER and JUDITH
    A. HAGLEY, Attorneys.
    __________________________
    Before NEWMAN, BRYSON, and LINN, Circuit Judges.
    BRYSON, Circuit Judge.
    This case requires us to evaluate the federal income
    tax consequences of sale-in, lease-out (“SILO”) transac-
    tions. The Court of Federal Claims denied Wells Fargo
    $115 million in claimed deductions for tax year 2002
    stemming from its participation in 26 SILO transactions
    with tax-exempt entities. We affirm.
    I
    A sale-in, lease-out transaction of the sort at issue in
    this case consists of two concurrent leases of an asset
    owned by a tax-exempt entity. In the first lease, known
    as the “head lease,” the tax-exempt entity leases the asset
    to the taxpayer for a lease term that exceeds the useful
    life of the asset. Because the asset will be returned to the
    tax-exempt entity only after its useful life has expired, the
    IRS treats the head lease as a sale of the asset. In the
    second lease, known as the “sublease,” the taxpayer leases
    the asset back to the tax-exempt entity for a term that is
    less than the asset’s remaining useful life. The sublease
    is a net lease, meaning that the tax-exempt entity is
    responsible for all expenses normally associated with
    ownership of the asset. The tax-exempt entity also re-
    tains legal title to the asset.
    The taxpayer prepays the entire “rent” of the head
    lease in one lump sum. The taxpayer funds the rent
    prepayment in part with its own funds and in part with a
    nonrecourse loan. The portion of the rent prepayment
    3                                         WELLS FARGO   v. US
    funded by the taxpayer’s own funds is known as the
    “equity portion,” and the portion funded through borrow-
    ing is known as the “debt portion.” The tax-exempt entity
    receives a small percentage of the head lease rent pre-
    payment, usually between 4 percent and 8 percent of the
    asset value, as its fee for participation in the SILO trans-
    action. The remainder of the rent prepayment, minus
    transaction costs, is placed in two restricted accounts, one
    for the equity portion and one for the debt portion. Each
    account is nominally held by the tax-exempt entity, but
    the funds are controlled by an affiliate of the taxpayer’s
    nonrecourse lender.
    The debt portion account is used to make the tax-
    exempt entity’s sublease rental payments. Money never
    actually changes hands during a rental payment; the
    lender’s affiliate simply moves funds from the tax-exempt
    entity’s debt portion account to the lender’s account in an
    amount sufficient to service the taxpayer’s nonrecourse
    loan debt. The debt portion account has sufficient funds
    to cover the tax-exempt entity’s payments for the life of
    the sublease or, equivalently, the taxpayer’s payments for
    the life of its nonrecourse loan. Because of the circular
    nature of the debt payments, the funds in the debt portion
    account are known as “loop debt.” The taxpayer’s debt is
    effectively “defeased,” or extinguished, meaning that
    dedicated funds exist for the purpose of paying off the
    debt. The taxpayer can therefore ignore its nonrecourse
    loan debt for purposes of its balance sheet.
    The lender’s affiliate invests the equity portion ac-
    count in high-grade debt, such as government bonds. The
    growth of the account is managed so that the tax-exempt
    entity has sufficient funds to repurchase its asset from
    the taxpayer at the conclusion of the sublease. The
    repurchase price, or “exercise price,” is set at the begin-
    WELLS FARGO   v. US                                       4
    ning of the SILO transaction. The tax-exempt entity can
    exercise its option to repurchase the asset simply by
    giving notice to the taxpayer. If the option is exercised,
    the funds in the equity portion account are transferred to
    the taxpayer, and a final payment of loop debt is made
    from the debt portion account to the taxpayer’s nonre-
    course lender. The debt portion account is emptied and
    the debt of the taxpayer to the lender is satisfied. The net
    result is the same as if the taxpayer had simply invested
    its equity portion account in high-grade debt, receiving a
    predictable return on that investment over the life of the
    sublease.
    If the tax-exempt entity chooses not to exercise its re-
    purchase option, the taxpayer generally has two choices.
    The taxpayer can elect either the “return option,” under
    which the taxpayer takes control of the asset immedi-
    ately, or the “service contract option,” under which the
    taxpayer postpones taking control of the asset. Under the
    “service contract option,” the tax-exempt entity is re-
    quired to satisfy several conditions before continuing to
    use the asset or arranging for its use by a third party.
    Those conditions are described below. Under either
    option, the tax-exempt entity ultimately receives the
    balance of the funds in the two accounts.
    SILOs offer three tax benefits to the taxpayer. First,
    as owner of the asset for tax purposes based on the head
    lease, the taxpayer may take depreciation deductions on
    the asset for the remainder of its useful life. See 26
    U.S.C. (“I.R.C.”) § 167(a). Second, the taxpayer may take
    deductions for interest payments made from the tax-
    exempt entity’s debt portion account to service the tax-
    payer’s nonrecourse loan. See I.R.C. § 163(a). Third, the
    taxpayer may deduct certain transaction costs associated
    with the SILO. If the tax-exempt entity exercises its
    5                                          WELLS FARGO   v. US
    repurchase option, these tax benefits are partially offset
    at the end of the sublease by taxes owed on the taxpayer’s
    receipt of funds from the equity portion account. Even
    taking those taxes into account, however, the deferral of
    tax payments during the life of the sublease has substan-
    tial economic value to the taxpayer.
    SILOs evolved in response to a long-running battle
    among Congress, the IRS, and enterprising taxpayers
    regarding the boundaries of permissible leasing of tax-
    exempt property to generate tax benefits. In 1981, Con-
    gress enacted “safe-harbor leasing rules” that allowed
    taxpayers to lease property from tax-exempt entities.
    Economic Recovery Tax Act, Pub. L. No. 97-34, 95 Stat.
    172 (1981). The safe-harbor rules, however, were quickly
    repealed in 1982. Tax Equity and Fiscal Responsibility
    Act, Pub. L. No. 97-248, 96 Stat. 324 (1982). In 1984,
    Congress enacted the so-called “Pickle Rule,” which
    provided that property leased from a tax-exempt entity
    would be depreciated at a slower rate than other property
    in order to limit the tax benefits generated by such trans-
    actions. Deficit Reduction Act, Pub. L. No. 98-369, 98
    Stat. 494 (1984).
    Taxpayers then began to employ creative strategies to
    avoid the Pickle Rule and receive greater tax benefits
    from the property of tax-exempt entities. One of those
    strategies was the use of lease-in, lease-out (“LILO”)
    transactions. A LILO is like a SILO, except that the head
    lease term is shorter than the asset’s remaining economi-
    cally useful life, so the IRS treats it as a lease rather than
    a sale. The end-of-lease options are also different for
    LILOs if the tax-exempt entity does not exercise its
    repurchase option: At the taxpayer’s discretion, the tax-
    exempt entity may be required to return the assets, renew
    its lease, or lease the assets to a third party. See Rev.
    WELLS FARGO   v. US                                       6
    Rul. 02-69, 2002-2 C.B. 760. Like the SILO options, each
    of the end-of-lease options in a LILO is structured to
    effectively eliminate risk of loss to the taxpayer. See
    BB&T Corp. v. United States, 
    523 F.3d 461
    , 464-65 (4th
    Cir. 2008). The Federal Transit Administration (“FTA”)
    has at various times promoted LILO and SILO leases as a
    means of providing infusions of cash for financially trou-
    bled public transit agencies.
    The LILO market came to an end beginning in 1999,
    when the Treasury Department issued new regulations
    requiring that prepayment of the head lease rent be
    treated as a loan for tax purposes. See Treas. Reg.
    § 1.467-4. At that point, taxpayers were “[p]resumptively
    alerted that the IRS would challenge exotic efforts to
    transfer tax deductions from tax indifferent entities.”
    AWG Leasing Trust v. United States, 
    592 F. Supp. 2d 953
    ,
    959 (N.D. Ohio 2008). In 2002, the IRS clarified that
    LILO transactions did not satisfy the substance-over-form
    doctrine, discussed below. Rev. Rul. 02-69. Taxpayers
    then began using SILOs instead of LILOs in order to
    avoid the new regulations on leases by characterizing the
    head lease as a sale. In addition, taxpayers asserted that
    the term of a service contract was not subject to the Pickle
    Rule for determining the applicable rate of depreciation.
    Finally, in 2004, Congress put an end to tax benefits
    generated from both LILO and SILO transactions by
    amending the Internal Revenue Code. American Jobs
    Creation Act, Pub. L. No. 108-357, 118 Stat. 1418 (2004);
    see I.R.S. Notice 05-13, 2005-1 C.B. 630. While the
    amendments were prospective in effect, they were not
    designed to alter the general principles of tax law that
    apply in determining the legitimacy of transactions de-
    signed to generate tax deductions. See H.R. Rep. No. 108-
    755, at 660 (2004) (Conf. Rep.).
    7                                         WELLS FARGO   v. US
    II
    From 1997 to 2003, Wells Fargo entered into several
    SILO transactions with tax-exempt entities. For tax year
    2002, Wells Fargo claimed $115 million in deductions
    based on 26 SILO transactions. Seventeen of the transac-
    tions were with domestic transit agencies, and nine were
    with owners of qualified technological equipment (“QTE”).
    When the IRS denied the deductions, Wells Fargo paid
    the disputed amount and filed a refund suit in the Court
    of Federal Claims. Before trial, the parties agreed to let
    the court’s disposition of five representative SILO trans-
    actions guide the resolution of the entire claim. Four of
    these representative transactions involved domestic
    transit agencies, including New Jersey Transit (“NJT”),
    Caltrans, Houston Metro, and the Washington Metropoli-
    tan Area Transit Authority (“WMATA”). The final repre-
    sentative transaction was a QTE transaction involving
    two lots of cellular telecommunications equipment owned
    by Belgacom Mobile (“Belgacom”), a Belgian company.
    The four transit agency transactions were typical
    SILO transactions. The assets in each case were public
    transit vehicles: light-rail vehicles (NJT), locomotives and
    passenger rail cars (Caltrans), buses (Houston Metro),
    and subway cars (WMATA). Each transit agency had the
    option to repurchase the assets at a fixed price at the end
    of the sublease term. In each case, the length of the
    combined sublease and service contract term was no more
    than 80 percent of the equipment’s remaining economi-
    cally useful life, meaning that if the transit agency chose
    not to exercise its repurchase option, the equipment
    would be returned to Wells Fargo with useful life remain-
    ing. Due to the length of the sublease terms, it is not
    known at this time whether the tax-exempt entities will
    exercise their repurchase options.
    WELLS FARGO   v. US                                       8
    To provide for the possibility that the repurchase op-
    tion might not be exercised, each of the four transit SILOs
    included a service contract option. The length of the
    service contracts varied from seven to fourteen years.
    Wells Fargo did not have to inform a transit agency
    whether it would impose a service contract until less than
    one year prior to termination of the sublease. If Wells
    Fargo chose to exercise the service contract option, the
    transit agency would need to locate a third-party operator
    for the equipment and negotiate an operating agreement
    with that third party. The agency would also be required
    to refinance the outstanding debt from the debt portion
    account, procure and pay for insurance to cover the resid-
    ual value of assets, maintain the equipment in newly
    refurbished condition, and arrange for payment of peri-
    odic fees to Wells Fargo sufficient to preserve its economic
    return on investment. If Wells Fargo chose to exercise
    the return option instead of the service contract option,
    the transit agency would receive the balance of the debt
    portion and equity portion accounts in exchange for giving
    the equipment to Wells Fargo in newly refurbished condi-
    tion.
    The end-of-sublease options were somewhat different
    for the Belgacom QTE transaction. Instead of having the
    option to repurchase the equipment at the conclusion of
    the sublease, Belgacom was given an “early buyout op-
    tion” a few years before the end of the sublease. Under
    the early buyout option, Belgacom could repurchase the
    equipment from Wells Fargo for an amount equivalent to
    the balance of the equity portion account at the time the
    option was executed. The Belgacom SILO did not include
    a service contract option. Instead, if Belgacom did not
    exercise its repurchase option, it would need to repur-
    chase the equipment at the conclusion of the sublease,
    renew the sublease for a series of one-year terms, or
    9                                       WELLS FARGO   v. US
    return the equipment to Wells Fargo. Upon return to
    Wells Fargo, the telecommunications equipment would
    need to be in “as new” condition, including any relevant
    hardware and software updates, and Belgacom would be
    required to pay for any installation of the equipment as
    directed by Wells Fargo. In 2007 and 2008, Belgacom
    terminated its SILOs for the two lots of equipment by
    exercising its early buyout option. During the sublease
    period, Belgacom claimed tax ownership and depreciation
    deductions for the QTE under Belgian law.
    Following a four-week bench trial addressing the five
    representative SILO transactions, the Court of Federal
    Claims denied Wells Fargo’s claim in its entirety. The
    court found that Wells Fargo participated in SILO trans-
    actions only when it had sufficient “tax capacity” to use
    the tax benefits of the transactions. The court further
    found that Wells Fargo expected the tax-exempt entities
    to exercise their options to repurchase their assets be-
    cause “the economic effects of the alternatives were so
    onerous and detrimental that a rational tax-exempt entity
    would do nothing other than exercise the options.” In any
    event, the court concluded, the service contract option in
    the transit agency subleases protected Wells Fargo from
    residual value risk during the assets’ remaining useful
    lives. Analyzing the economic effect of the transactions,
    the court found that, other than the tax advantages, Wells
    Fargo received no net economic benefit from entering into
    the SILO transactions. In addition, the court found that
    the SILO transactions had no effect on the tax-exempt
    entities’ use of the assets, that no funds changed hands
    during the sublease period, 1 that the assets were ap-
    1   In the Belgacom SILO, Wells Fargo was provided
    with a small return on the equity portion account during
    the sublease period.
    WELLS FARGO   v. US                                      10
    praised at an inflated value in order to maximize depre-
    ciation deductions, and that there were no arms-length
    negotiations of any of the lease terms. In summary, the
    court determined that each of the representative SILO
    transactions “essentially amount[ed] to Wells Fargo’s
    purchase of tax benefits for a fee from a tax-exempt entity
    that cannot use the deductions.” Accordingly, the court
    held that the transactions were abusive tax shelters and
    that the claimed tax benefits from those transactions
    should be disallowed. In so ruling, the court found that
    transactions had to be disregarded for tax purposes under
    both the “substance-over-form” doctrine and the “eco-
    nomic substance” doctrine.
    On appeal, Wells Fargo argues that the trial court (1)
    employed an inappropriate methodology to determine that
    Wells Fargo lacked the benefits and burdens of ownership
    in the assets that were the subject of the SILO transac-
    tions; (2) used the wrong test to measure its pretax profit;
    and (3) misapplied the “nontax business purpose” test.
    The first argument is directed to the “substance-over-
    form” doctrine. The remaining arguments relate to the
    “economic substance” doctrine. We review the characteri-
    zation of transactions for tax purposes de novo, based on
    underlying findings of fact, which we review for clear
    error. Stobie Creek Invs., LLC v. United States, 
    608 F.3d 1366
    , 1375 (Fed. Cir. 2010); see Frank Lyon Co. v. United
    States, 
    435 U.S. 561
    , 581 n.16 (1978).
    III
    Judicial anti-abuse doctrines “prevent taxpayers from
    subverting the legislative purpose of the tax code.” Coltec
    Indus., Inc. v. United States, 
    454 F.3d 1340
    , 1354 (Fed.
    Cir. 2006). The substance-over-form doctrine provides
    that the tax consequences of a transaction are determined
    11                                        WELLS FARGO   v. US
    based on the underlying substance of the transaction
    rather than its legal form. See Griffiths v. Helvering, 
    308 U.S. 355
    , 357 (1939) (looking to “the crux” of transaction
    by imagining it in its simplest form); Minn. Tea Co. v.
    United States, 
    302 U.S. 609
    , 613 (1938) (“A given result at
    the end of a straight path is not made a different result
    because reached following a devious path.”); Holiday Vill.
    Shopping Ctr. v. United States, 
    773 F.2d 276
    , 280 (Fed.
    Cir. 1985).
    In order to be entitled to deductions for depreciation
    of assets and associated interest and transaction ex-
    penses, Wells Fargo had to show that it owned the SILO
    equipment. See 
    Coltec, 454 F.3d at 1355
    . Ownership for
    tax purposes is not determined by legal title. Instead, in
    order to qualify as an “owner” for tax purposes, the tax-
    payer must bear the benefits and burdens of property
    ownership. Frank 
    Lyon, 435 U.S. at 572-73
    ; Corliss v.
    Bowers, 
    281 U.S. 376
    , 378 (1930). Here, the parties agree
    that the clearest indicator of ownership is the allocation of
    risk associated with the value of the leased assets.
    Wells Fargo argues that it acquired the benefits and
    burdens of ownership in the leased assets because there
    was a possibility that it would regain possession of the
    leased assets at a time when they still retained some
    economically useful life. Wells Fargo notes that the
    sublease term, even when added to the period of a subse-
    quent service contract, would still leave Wells Fargo with
    economically useful assets, and that it would be subject to
    financial risk based on the actual value remaining in
    those assets after the SILO transaction ended. Wells
    Fargo’s argument is predicated on uncertainty regarding
    whether the tax-exempt entities would exercise their
    options to repurchase the assets. The trial court noted
    that for tax purposes Wells Fargo required the tax-
    WELLS FARGO   v. US                                      12
    exempt entities to state that at the time of closing they
    had not made any determination whether they would
    exercise their repurchase options. However, based on
    evidence including statements made by the tax-exempt
    entities at the time, the trial court found that “[t]he
    evidence . . . strongly supports a conclusion that the
    [repurchase options] would almost certainly be exercised
    to terminate the transactions.” At oral argument, counsel
    for Wells Fargo acknowledged that affirmance would be
    appropriate if that finding were to stand.
    We have never held that the likelihood of a particular
    outcome in a business transaction must be absolutely
    certain before determining whether the transaction
    constitutes an abuse of the tax system. The appropriate
    inquiry is whether a prudent investor in the taxpayer’s
    position would have reasonably expected that outcome.
    Characterization of a tax transaction based on a highly
    probable outcome may be appropriate, particularly where
    the structure of the transaction is designed to strongly
    discourage alternative outcomes. See Knetsch v. United
    States, 
    364 U.S. 361
    , 366 n.3 (1960) (disregarding “wholly
    unlikely assumption” regarding taxpayer’s future behav-
    ior in the course of concluding that transaction lacked
    substance); Stobie 
    Creek, 608 F.3d at 1378
    (evaluating
    substance of transaction based in part on the “structure of
    the investment,” which eliminated any “reasonable possi-
    bility” of nontax profit); cf. Frank Lyon v. United States,
    
    435 U.S. 561
    , 577 (1978) (finding taxpayer to have in-
    curred risk in transaction where there was “substantial
    risk” of loss, “not just the abstract possibility that some-
    thing will go wrong”).
    Seeking to undermine the trial court’s finding that the
    tax-exempt entities were highly likely to exercise the
    repurchase options, Wells Fargo challenges the testimony
    13                                        WELLS FARGO   v. US
    of Dr. Thomas Lys, the government’s expert on financial
    economics. Dr. Lys concluded that “exercising the [repur-
    chase option] is the most advantageous option for transit
    agencies in virtually all circumstances.” The trial court
    found that Dr. Lys “provided a compelling economic
    analysis of the SILO transaction.” According to the court,
    Dr. Lys “established beyond doubt that no tax-exempt
    entity in its right mind would fail to exercise the purchase
    option.”
    When the tax-exempt entities make the decision
    whether to exercise their repurchase options, they will
    necessarily compare the economic benefits provided by
    the leased assets with the economic benefits of the alter-
    native options. Prior to the transactions at issue in this
    case, Wells Fargo’s appraisers analyzed the expected
    benefits to the tax-exempt entity from the repurchase
    option and the service contract option. The appraisers
    concluded that the respective benefits and costs of the two
    options would be such that the tax-exempt entities would
    not be under any economic compulsion to exercise the
    repurchase option. Dr. Lys conducted the same analysis
    but came to a different result. He determined that the
    value of the leased assets to the tax-exempt entity would
    greatly exceed the exercise price, meaning that the tax-
    exempt entity would have a strong economic incentive to
    repurchase the assets. At trial, Wells Fargo’s expert
    challenged Dr. Lys’s analysis and defended the apprais-
    ers’ analysis.
    The crux of the disagreement between Dr. Lys and the
    appraisers is the discount rate that the tax-exempt entity
    would apply in calculating the net present value of its
    WELLS FARGO   v. US                                       14
    alternatives at the decision point. 2 The appraisers se-
    lected the weighted average cost of capital (“WACC”)
    prevailing in the industry sector in which the tax-exempt
    entity operated (e.g., the rail industry) as the discount
    rate that the entity would use to compare net present
    values of the repurchase and service contract options. 3
    Dr. Lys equated the discount rate with the rate at which
    the tax-exempt entity could borrow funds. According to
    Dr. Lys, that rate “reflect[ed] the general risk” of the tax-
    exempt entity. Wells Fargo’s expert, however, testified
    that Dr. Lys’s choice of discount rate “violate[d] a funda-
    mental tenet of finance,” and that the relevant industry
    WACC was the appropriate discount rate.
    Because Dr. Lys used a lower discount rate than the
    appraisers used, Dr. Lys projected that the leased assets
    would retain their value better than the appraisers ex-
    pected and that the cost of the periodic payments to Wells
    Fargo under the service contract would be higher than the
    2    The present value of a fixed asset incorporates the
    time value of money and investment risk, as well as
    depreciation in the case of a depreciable asset. For a
    stream of future cash payments, the present values of
    each payment are calculated separately and summed to
    yield the net present value. See generally Richard A.
    Brealey et al., eds., Principles of Corporate Finance 14-16
    (9th ed. 2008).
    3    The WACC is the expected return on a portfolio of
    all of an entity’s existing securities. Those securities
    consist of both debt and equity. The WACC is a weighted
    average of the return that security holders expect on their
    investments. For example, if bondholders expect a 3
    percent interest rate and stockholders expect a 6 percent
    return on investment, and a company’s debt-to-equity
    ratio is 2-to-1, the entity’s WACC would be 4 percent. See
    
    Brealey, supra
    , 241-42. The calculation of WACC can be
    very complex, particularly for public entities such as
    municipal transit agencies.
    15                                        WELLS FARGO   v. US
    appraisers expected. Consequently, according to Dr. Lys,
    there would be no reason for the tax-exempt entity to risk
    entering into the service contract because it would receive
    less economic benefit than if it simply repurchased the
    assets.
    Wells Fargo characterizes the trial court’s reliance on
    Dr. Lys’s testimony as reflecting the application of an
    incorrect legal standard. See Lazare Kaplan Int’l, Inc. v.
    Photoscribe Techs., Inc., 
    628 F.3d 1359
    , 1381 (Fed. Cir.
    2010). Wells Fargo argues that, as a matter of law, the
    likelihood of outcomes in SILO transactions must be
    assessed using estimates based on the fair market value
    of the assets in question, and it cites the Frank Lyon case
    to support that argument. In Frank Lyon, however, there
    was no service contract, simply a series of purchase
    options provided to the lessee in a sale-leaseback transac-
    tion at future points in time. The purchase option prices
    were negotiated between the parties and were known at
    the time the parties entered into the transaction. Frank
    Lyon Co. v. United States, 75-2 USTC ¶ 9545, 
    36 A.F.T.R.2d (RIA) 75-5154
    , 75-5156 (E.D. Ark. 1975). The trial
    court in that case concluded that the price of the fixed
    purchase options “represented fair estimates of market
    value [of the leased building] on the applicable dates,” and
    the Supreme Court did not disagree with that 
    conclusion. 435 U.S. at 569
    . The Court, however, did not address the
    appropriate discount rate to use when comparing a fixed
    purchase option to alternative options. Nothing in the
    Frank Lyon case (or any of the other authorities cited by
    Wells Fargo) supports Wells Fargo’s argument that Dr.
    Lys’s testimony was predicated on a legally incorrect
    standard. To the contrary, the question of what discount
    rate is most appropriate to use in estimating the benefits
    and burdens of purchase and service options that will not
    be exercised for a number of years is a distinctly factual
    WELLS FARGO   v. US                                      16
    matter, and Wells Fargo has not shown that the trial
    court’s acceptance of Dr. Lys’s methodology constituted
    clear error.
    In any event, the critical inquiry is whether Wells
    Fargo could have reasonably expected that the tax-
    exempt entities would exercise their repurchase options,
    and the trial court’s resolution of that question does not
    turn on the appropriateness of Dr. Lys’s choice of discount
    rate. While the court found Dr. Lys’s testimony “most
    valuable” on that issue, the court explained that Dr. Lys
    had simply “confirmed” the court’s conclusion based on
    other evidence. The court noted that “[m]any of the
    expert witnesses at trial testified as to the probability
    that the transit agency would elect the [repurchase op-
    tion] instead of becoming subject to the service contract or
    the return of the equipment.”
    For example, an expert on the passenger railcar in-
    dustry testified that the repurchase option was “very
    likely” to be exercised in the NJT, Caltrans, and WMATA
    SILOs, and that there was no foreseeable circumstance in
    which the option would not be exercised. An urban public
    transportation expert similarly testified that the four
    transit agencies were “very likely” to exercise their repur-
    chase options due to the uncertainty and potential diffi-
    culties of complying with the service contract option
    within a short time frame. That expert’s report examined
    each of the transit agency SILOs individually and deter-
    mined that substantial hurdles would hinder the agen-
    cies’ compliance with the service contract option. An
    expert in leasing and asset financing testified that he
    “[could not] think of a set of circumstances under which
    CalTrans would not exercise a purchase option, and the
    requirements [of the alternatives] . . . make it a practical
    certainty that the purchase option will be exercised.” A
    17                                        WELLS FARGO   v. US
    Federal Transit Administration witness testified that
    SILO arrangers consistently told the FTA that it
    “shouldn’t be too concerned about [the possibility of the
    repurchase option not being exercised] because it was
    likely that the transit agencies would exercise [it]. It was
    the easiest way to close the transaction normally.” Fi-
    nally, the court credited the testimony of several wit-
    nesses that service contracts are rare in the domestic
    transit sector, particularly for existing transit services
    and equipment.
    The witness testimony regarding the four transit
    agency SILOs was supported by documentary evidence.
    Credit approval presentations (“CAPs”) prepared by Wells
    Fargo before it entered into each SILO suggest that the
    transactions were designed to encourage the tax-exempt
    entities to exercise their repurchase options. The NJT
    CAP, which is representative, states that the transit
    agency “has all the cash necessary (in the defeasance
    funds) to purchase the Equipment at the [repurchase
    option] point and would incur little or no added expense
    at this point.” In an e-mail exchange, the SILO promoters
    informed representatives of Houston Metro that they
    “fully anticipate[d] that [Houston Metro] will buy the
    buses back with the defeasance proceeds.” The FTA
    approved the NJT and WMATA SILO transactions after
    noting in each case that “[a]t the purchase option date . . .
    [the transit agency] is expected to exercise its option to
    buy out the head lease.” The FTA considered it a “very
    low likelihood” that the transit agencies would not exer-
    cise the repurchase option.
    Wells Fargo points to testimony by employees of the
    transit agencies who testified that at the time their
    agencies entered into the SILO transactions, they did not
    know whether the repurchase options would be exercised.
    WELLS FARGO   v. US                                       18
    In addition, several witnesses testified that the transit
    agencies were under no compulsion to exercise the repur-
    chase options. Those statements do not undermine the
    trial court’s findings. For the most part, the employees’
    testimony did not evaluate the probabilities of the exer-
    cise of the purchase options. To be sure, one Houston
    Metro employee testified that it was a “realistic” possibil-
    ity that the repurchase option would not be exercised. In
    light of the contrary witness testimony and documenta-
    tion described above, however, that isolated statement is
    not enough to call into question the trial court’s conclu-
    sion that the tax-exempt entities were highly unlikely to
    exercise their repurchase options.
    The trial court permissibly found that the expecta-
    tions were no different for the Belgacom SILO. Wells
    Fargo’s CAP for that transaction notes that Belgacom was
    expected to exercise its early buyout option. The presen-
    tation stated that “the equipment return provisions are
    strict and onerous to Belgacom . . . providing significant
    economic incentive for Belgacom to simply purchase the
    equipment.” Similarly, annual internal reviews by Wells
    Fargo in 2001 and 2002 explain that Belgacom was ex-
    pected to exercise its early buyout option. The documents
    are explicit that “[t]he original return provisions of the
    lease were written with the intention of being overly
    onerous to make the lease-end return of any equipment
    an unattractive option.” 4 The trial court considered those
    documents persuasive evidence that Wells Fargo fully
    4   Wells Fargo points to a statement in each annual
    review that “[s]hould the lease run to full maturity,
    Belgacom will most likely purchase the equipment for
    Fair Market Value.” The conditional part of that state-
    ment will be satisfied only if the early buyout option is not
    exercised, which the same document characterizes as a
    very unlikely event.
    19                                        WELLS FARGO   v. US
    anticipated that Belgacom would exercise its option to
    repurchase the assets, and that Belgacom was virtually
    certain to do so. Wells Fargo highlights statements of
    expert witnesses that Belgacom’s exercise of the repur-
    chase option was not completely assured at the outset of
    the transaction. Once again, however, that evidence does
    not undermine the trial court’s finding that the Belgacom
    is virtually certain to exercise the repurchase option.
    In light of the extensive witness testimony and docu-
    mentation relied on by the trial court, the provision in
    each of the five representative SILOs stating that the tax-
    exempt entity has not “taken any official corporate action
    pertaining to the exercise or non-exercise of the Purchase
    Option” rings hollow. The transactions “were part of a
    prepackaged strategy marketed to shelter taxable gain,”
    Stobie 
    Creek, 608 F.3d at 1379
    , promoted to tax-exempt
    participants with the understanding that they would
    exercise their repurchase options. The trial court was
    justified in concluding that “[f]rom the inception of the
    transactions, the economic effects of the alternatives were
    so onerous and detrimental that a rational tax-exempt
    entity would do nothing other than exercise the options.”
    The court’s conclusion is borne out by the fact that Bel-
    gacom did in fact exercise its early buyout options on both
    lots of equipment, and that NJT has exercised its early
    purchase option on every similar transaction in which it
    engaged in the past. Thus, even in the absence of Dr.
    Lys’s economic analysis of the representative SILO trans-
    actions, the trial court had before it compelling evidence
    of the very high likelihood that all of the tax-exempt
    entities would exercise their repurchase options. The
    trial court therefore did not clearly err in concluding that
    the tax-exempt entities were virtually certain to repur-
    chase the assets when the lease periods expired.
    WELLS FARGO   v. US                                      20
    Wells Fargo argues that this case is governed by the
    Supreme Court’s decision in Frank Lyon, in which the
    Court allowed tax deductions for a “sale-leaseback” trans-
    action having some of the same characteristics as the
    SILOs at issue here. However, Frank Lyon involved a
    transaction between two taxable entities, and the Su-
    preme Court noted that the facts in that case “stand in
    contrast to many others in which the form of the transac-
    tion actually created tax advantages that, for one reason
    or another, could not have been enjoyed had the transac-
    tion taken another 
    form.” 435 U.S. at 583
    n.18 (distin-
    guishing Sun Oil Co. v. Comm’r of Internal Revenue, 
    564 F.2d 258
    (3d Cir. 1977), a case involving a sale-and-
    leaseback of land between a taxpayer and a tax-exempt
    entity). 5 Moreover, the trial court in Frank Lyon found
    that the lessee “was highly unlikely” to exercise its pur-
    chase option. 
    Id. at 570.
    Since the decision in Frank
    Lyon, a number of courts have disallowed LILO and SILO
    transactions between taxpayers and tax-exempt entities.
    E.g., BB&T Corp., 
    523 F.3d 461
    (LILO); Altria Grp., Inc.
    5     This court and our predecessor court have not
    looked favorably upon transactions designed to unlock
    previously unavailable tax advantages through manipula-
    tion of taxable income. In Rothschild v. United States,
    
    407 F.2d 404
    (Ct. Cl. 1969), the taxpayer “borrowed funds
    to make an investment which initially would yield him a
    loss. However, by deducting the interest paid from ordi-
    nary income and paying capital gains tax on the profits of
    the investment, the taxpayer’s investment yielded an
    after-tax 
    profit.” 407 F.2d at 405
    . The court denied the
    taxpayer deductions stemming from the difference be-
    tween his marginal income tax rate and the capital gains
    tax rate. More recently, in Stobie Creek, we disallowed
    deductions where a “tax shelter . . . . took advantage of
    the fact that assets and contingent liabilities were treated
    differently for tax purposes when contributed to a part-
    nership, thus enabling the taxpayer to generate an artifi-
    cial 
    loss.” 608 F.3d at 1368-69
    .
    21                                        WELLS FARGO   v. US
    v. United States, 
    694 F. Supp. 2d 259
    (S.D.N.Y. 2010)
    (LILO and SILO); AWG Leasing Trust, 
    592 F. Supp. 2d 953
    (SILO).
    The sole exception is Consolidated Edison Co. v.
    United States, 
    90 Fed. Cl. 228
    (2009), a case that is await-
    ing final judgment. In Consolidated Edison, the Court of
    Federal Claims allowed deductions in a LILO transaction
    after finding that it was uncertain whether the tax-
    exempt entity would exercise its option to repurchase the
    leased assets. Whether that finding was supported by the
    evidence and whether the Consolidated Edison court
    applied the correct legal standard on the issue of probabil-
    ity are not questions that we address today. It is suffi-
    cient to note that the finders of fact in that case and in
    the present case reached different conclusions regarding
    the likelihood of the tax-exempt entity exercising its
    repurchase option.
    IV
    Because we uphold the trial court’s finding that the
    tax-exempt entities are virtually certain to exercise their
    repurchase options, we are left with purely circular trans-
    actions that elevate form over substance. The only flow of
    funds between the parties to the transaction was the
    initial lump sum given to the tax-exempt entity as com-
    pensation for its participation in the transaction. From
    the tax-exempt entity’s point of view, the transaction
    effectively ended as soon as it began. The benefits to
    Wells Fargo continued to flow throughout the term of the
    sublease, however, in the form of deferred tax payments.
    The third-party lender and its affiliate were also compen-
    sated for their participation, as were the creators and
    promoters of the transactions. These transactions were
    win-win situations for all of the parties involved because
    WELLS FARGO   v. US                                   22
    free money—in the form of previously unavailable tax
    benefits utilized by Wells Fargo—was divided among all
    parties. The money was not entirely “free,” of course,
    because it was in effect transferred to Wells Fargo from
    the public fisc.
    We sustain the trial court’s conclusion that the SILO
    transactions at issue in this case run afoul of the sub-
    stance-over-form doctrine and therefore are abusive tax
    shelters. Based on well-grounded factual findings, the
    trial court permissibly found that the claimed tax deduc-
    tions are for depreciation on property Wells Fargo never
    expected to own or operate, interest on debt that existed
    only on a balance sheet, and write-offs for the costs of
    transactions that amounted to nothing more than tax
    deduction arbitrage. We therefore uphold the judgment of
    the Court of Federal Claims.
    AFFIRMED