WIENER, Circuit Judge.
Plaintiffs/Appellants/Cross-Appellees Central States, Southeast and Southwest Areas Pension Fund, a multiemployer pension fund, and its trustees (collectively “Central States”), appeal from the district court’s judgment that dismissed Central States’s pension plan withdrawal liability claims against Defendants/Appel-lees/Cross-Appellants Creative Development Company (“Creative Development”), a Louisiana partnership, and its partners,, Terry Smith, Sandra Theriot Smith, Jack Rome, Jr., and Suzanne McCraine Rome.1 Disagreeing with the district court as a matter of law, we conclude that a written agreement (“the 1986 Agreement”) unambiguously provided for and effectuated the transfer of a capital interest in Creative Development to Sheldon Beychok, the majority owner of a different business organization which had ceased making pension fund contributions to Central States. We therefore reverse the district court’s judgment for Creative and remand for the limited purpose of affording that court the initial opportunity to determine whether, by virtue of such acquisition of a capital interest in Creative Development, Beychok (either alone or in combination with appel-lee Jack Rome, Jr.) owned both a “controlling interest” and “effective control” of each business, within the intendment of the Employment Retirement Income Security Act (“ERISA”)2 as amended by the Multiemployer Pension Plan Amendment Act (“MPPAA”),3 thereby placing Creative Development and Beychok’s other business organization under “common control” 4 and subjecting Creative to responsibility for the other business’s “withdrawal liability” to Central States.
I.
BACKGROUND
A. Statutory Framework
Central States is a multiemployer pension plan within the meaning of §§ 3(37) and 4001(a)(3) of ERISA.5 Central States brought this suit to recover “withdrawal liability” from Creative Development and its individual partners under MPPAA. The term “withdrawal liability” refers to the share of unfunded vested benefits, i.e., the difference between the present value of a pension plan’s assets and the present value of the benefits it will be obligated to pay in the future, that an employer owes to a multiemployer pension plan governed by ERISA when the employer “withdraws” from the plan.6 An [409]*409employer is deemed to have withdrawn from a multiemployer pension plan when the employer “(1) permanently ceases to have an obligation to contribute under the plan, or (2) permanently ceases all covered operations under the plan.”7 ERISA imposes withdrawal liability on an employer in these situations to ensure that “the financial burden of his employees’ vested pension benefits will not be shifted to the other employers in the plan and, ultimately, to the Pension Benefit Guaranty Corporation, which insures such benefits.”8
When an employer officially withdraws from a multiemployer pension plan, the plan sponsor must then (1) determine the amount of the employer’s liability, if any, (2) notify the employer of this amount, and (3) collect the sum from the employer.9 If the withdrawing employer is unable to pay its assessed withdrawal liability in full, the plan may recover the deficiency from other entities that are “trades or businesses” under “common control” with the withdrawing employer.10 Consequently, all such trades or businesses, including the withdrawing employer, that are determined to be under “common control” within the meaning of MPPAA and its regulations, are deemed to belong to a “controlled group” of trades or businesses and are treated as a “single employer.” As such, all are jointly and severally (solidarily) liable for the withdrawal liability incurred by any member of the controlled group.11 This form of liability is commonly referred to as “control group” liability.12
The determination whether particular entities are in fact controlled group members requires resort to several Treasury Department regulations, among which is one that specifies that a trade or business belongs to a “brother-sister” controlled group if:
(i) the same five or fewer persons who are individuals, estates, or trusts own ... a controlling interest in each organization, and (ii) taking into account the ownership of each such person only to the extent such ownership is identical with respect to each such organization, such persons are in effective control of each organization.13
In the case of a trade or business that is a partnership, a “controlling interest” means “ownership of at least 80 percent of the 'profits interest or capital interest of such partnership,14 and ‘effective control’ exists when five or fewer persons” own an aggregate of more than 50 percent of the profits interest or capital interest of such [410]*410partnership.15
In this case, a Baton Rouge, Louisiana bakery business known as Wolf Baking Co. Inc. (“Wolf Baking”) had been a signatory to a collective bargaining agreement (“CBA”) pursuant to which Wolf Baking was required to make contributions to Central States. In December 1986, Wolf Baking filed for bankruptcy and discontinued its operations, thereby permanently terminating its obligation to make contributions to Central States. As a result, Wolf Baking was deemed to have withdrawn from Central States. Accordingly, Central States calculated Wolf Baking’s withdrawal liability and determined it to be $1,352,710.73. Because of its bankruptcy, however, Wolf Baking was able to pay only $289,858 of this obligation to Central States, leaving a deficit in excess of $1 million. Central States now seeks to recoup the Wolf Baking shortfall through a withdrawal liability assessment and recovery against the partnership and the individual partners comprising Creative, asserting that the partnership was, at all pertinent times, a member of a brother-sister controlled group with Wolf Baking. This, Central States posits, resulted from the three-cornered transaction among (1) Wolf Baking and its affiliates, (2) Beychok, individually, and (3) Creative Development, as formalized in the 1986 Agreement.
B. The Brother-Sister Entities: Creative Development and the Bakeries
Creative Development was formed as a Louisiana partnership in 1981 by the above-named individual appellees to develop a residential subdivision near Baton Rouge. The initial capital of the partnership was $5,000, consisting of equal contributions from the founding partners.
In the same year, 1981, W.B.C. Inc. (“WBC”) was formed by Sheldon Beychok, now deceased, appellee Jack Rome, and Harold Salmon, Jr., to acquire the stock of two bakeries that had recently emerged from bankruptcy. One of those bakeries was Wolf Baking; the other was Wm. Wolf Bakery, Inc. (“Wm. Wolf Bakery”). At all times relevant to this case, those two bakeries were wholly owned subsidiaries of the holding company, WBC.16 Furthermore, at all relevant times, Sheldon Bey-chok and appellee Jack Rome collectively owned 85 percent of the issued and outstanding capital stock of the holding company, WBC, with Beychok owning 61.45 percent and Rome owning 23.55 percent.17 Thus, Beychok and Rome, through their controlling interest in the parent corporation, WBC, owned or controlled more than 80 percent of the capital stock of its Wolf Baking and Wm. Wolf Bakery subsidiaries — actually, 100 percent control by virtue of their combined 85 percent control of WBC, which owns 100 percent of the stock of each subsidiary.
During the mid-1980s, the wholly owned subsidiary bakeries of WBC were chronically in need of cash, so Beychok made loans to each from time to time. By June 1, 1986, the outstanding balance of these loans aggregated $324,000.
C. Creative Development’s Initial Involvement with the Bakeries: The Sale and Leaseback of the Bakery Depots
For a better understanding of the 1986 transaction, which is at the vortex of the dispute in this case, we briefly review how Creative Development first became directly involved with the bakeries and Beychok. In the early 1980s, after completion of the [411]*411real estate venture for which it was originally formed, Creative Development decided to invest in two “bakery depots”18 owned by Wm. Wolf Bakery. In March 1982, Creative Development purchased the two bakery depots for a price of $250,000, then immediately leased both depots to Wolf Baking “and/or its affiliates.” Bey-chok — who, with Harold Salmon, had previously purchased two other bakery depots from Wm. Wolf Bakery — confirmed that the purpose of this March 1982 sale to Creative Development was to obtain cash for injection into Wolf Baking and its affiliates so that the bakeries could continue to operate.
The financing for Creative Development’s 1982 purchase of the bakery depots came from two sources: (1) cash, obtained from a $200,000 loan from River City Federal Savings & Loan (“River City”), evidenced by Creative Development’s promissory note, which was secured by a first mortgage on the depot properties, and (2) credit, evidenced by an unsecured $50,000 purchase money promissory note given by Creative Development to the vendor, Wm. Wolf Bakery.
As an additional, inducement for Creative Development to purchase the two depots, Beychok agreed to become a party to the depot leases and personally guarantee the lease payments to Creative Development. In consideration for Beychok’s becoming a party to and personally guaranteeing the leases,19 Rome and Smith executed a counter letter to Beychok, acknowledging that in truth Creative Development owned only an undivided two-thirds (%) interest in the two depots purchased from Wm. Wolf Bakery and that the remaining one-third (%) interest was purchased for the account of Beychok. The counter letter further declared that Rome and Smith would, when called upon to do so, transfer record title to Beychok of his undivided one-third ($) interest in the depots.
Creative Development’s purchase of the two bakery depots in March 1982, coupled with the provisions of the counter letter, produced a joint venture between Creative and Beychok. Subsequent financial statements and tax returns prepared for or filed by the Romes, the Smiths, Creative Development, and the Creative/Beychok joint venture, reflect Creative’s two-thirds (%) and Beychok’s one-third 0&) ownership interests in this depot venture.
D. The 1986 Transaction
By the spring of 1986, Rome and Bey-chok knew that WBC’s subsidiary bakeries were headed for bankruptcy and that as a result Beychok, an insider, would never recover the $824,000 owed to him by the bakers. The parties also knew that the $50,000 promissory note given by Creative Development to Wm. Wolf Bakery as the credit portion of the partnership’s purchase of the bakery depots was still outstanding and would become an asset of the bankruptcy estate of the bakeries.
In an apparent effort to “save” Creative Development’s $50,000, the bakeries’ bankruptcy counsel suggested that these two debts be offset to the extent possible and that the transaction be disclosed to the bankruptcy court. Accordingly, at Rome’s request, Beychok caused the 1986 Agreement to be prepared. It was signed on June 1, 1986 by Terry Smith (on behalf of Creative Development), Jack Rome (as President and CEO of Wm. Wolf Bakery, Inc. and Wolf Baking Company, Inc.) and Sheldon Beychok, individually. In the 1986 Agreement, the parties first acknowledged the existence of (1) Creative Development’s $50,000 promissory note owed to Wm. Wolf Baker, and (2) the bakeries’ cumulative debt of $324,000 owed to Bey-[412]*412chok. The parties then agreed to the following: (1) Beychok authorized the bakeries to reduce the amount of their $324,000 indebtedness to him by $50,000; (2) the bakeries, in turn, agreed to credit the indebtedness owed to them by Creative Development by the same amount ($50,000), “forever extinguishing the obligation [on the promissory note] from Creative to Wolf and/or Baking Company”; and (3) Creative agreed that it
does hereby sell, transfer and assign unto Beychok an interest in that partnership [Creative] equal to said Fifty Thousand ($50,000) Dollar offset as described hereinabove, (emphasis added).
The legal effects of this round robin transaction comprise the crux of the threshold issue of the instant litigation: whether Sheldon Beychok, the now-deceased former majority owner of WBC and its subsidiaries, acquired a capital or equity interest in Creative Development as a result of the 1986 transaction, or merely became its creditor. If Beychok acquired a capital interest and such interest, either alone or in combination with Rome’s, equaled or exceeded the minimum percentages needed to constitute “controlling interest” and “effective control” for purposes of the relevant Treasury Regulation,20 then Creative Development and the bakeries would have been under “common control,” i.e., members of the same brother-sister group of trades or businesses under the common control of Rome and Beychok. As such, Creative would be liable to Central States under MPPAA.21 If, however, Beychok merely became a creditor of the partnership, no such liability would attach.
As shall be seen in the analysis that follows, this case turns on the answers to three questions: (1) Is the subject provision of the 1986 Agreement ambiguous?; (2) regardless of whether that provision is ambiguous, what is the nature of the “interest” acquired by Beychok in Creative Development by virtue of the 1986 Agreement?; and (3) if the interest Beychok acquired was capital and not debt, does Beychok’s percentage of capital interests, or the combined percentage interests of Beychok and Rome, in both Creative Development and the bakeries meet the two-pronged test for “common control” under the applicable Treasury Regulation?22
II.
PROCEEDINGS
Central States filed the instant action in September 1992, alleging that Wolf Baking and Creative Development constitute a controlled group and should be treated as a single employer for, purposes of assessing and recovering withdrawal liability under MPPAA. The district court denied cross notions for summary judgment, finding that the 1986 agreement was ambiguous, and that further inquiry into the intent of the parties was required. After a one day trial, the district court reaffirmed its earlier determination that the 1986 Agreement is ambiguous. The court then held, based on extrinsic evidence, that the 1986 Agreement was entered into solely to make Beychok a $50,000 creditor of Creative Development which, to that extent, simply replaced the bakeries as Beychok’s debtor. The court concluded that Beychok neither became a partner nor acquired a capital interest in Creative Development but merely became its creditor. Accordingly, judgment was entered for Creative, dismissing Central States’s claims at its costs. As it decided the case on that reasoning, the district court never reached the questions of common control or controlled group liability for purposes of assessing withdrawal liability.
Victorious, Creative filed a motion urging the district court to amend its judgment to include an award of costs, expenses, and attorneys’ fees under [413]*413ERISA.23 The district court considered the factors affecting entitlement to such an award and held that it would not be appropriate. Central States timely appealed from the district court’s judgment dismissing the withdrawal liability claims against Creative, and Creative cross-appealed from the district court’s denial of its request for costs, expenses, and attorneys’ fees.
III.
ANALYSIS
A. Standard ofRevietu
The principal thrust of Central States’s contention on appeal is that the district court erred in finding the language of the 1986 Agreement to be ambiguous on the question whether Beychok acquired a capital ownership interest in Creative Development or merely became its creditor. A district court’s interpretation of a written agreement, including its initial determination whether that agreement is ambiguous, presents questions of law and thus is subject to our de novo review.24 Findings of historical or discrete facts are reviewed for clear error.
B. The 1986 Agreement Was Not Ambiguous and Conveyed a Capital Interest in Creative Development to Beychok
Central States insists that the meaning of the 1986 Agreement among Beychok, Creative Development, and the bakeries, is plain: Beychok acquired a capital interest in Creative Development; any other reading simply disregards what the 1986 Agreement actually says. Central States argues that the transaction conveyed to Beychok an equity or “capital” interest in Creative Development equal to $50,000, thus giving Beychok and Rome a combined capital or asset interest in that partnership of more than 80 percent. As together Rome and Beychok also owned more than 80 percent of WBC and thus of its Wolf Baking subsidiary, concludes Central States, a combined ownership interest of 80 percent or more of Creative Development placed Rome and Beychok in “common control” of both businesses. This in turn subjected Creative Development and its partners to solidary liability for the remainder of Wolf Baking’s withdrawal liability under MPPAA.25
Creative, of course, rejects this view of the transaction memorialized in the 1986 Agreement. Creative insists that the district court correctly determined, vis-a-vis Beychok, the transaction amounted to nothing more than an exchange of debtors — Creative Development for the bakeries' — on the $50,000 owed to him. As such, Beychok was substituted for the bakeries as the $50,000 creditor of Creative Development; in essence, Beychok only made a loan to, and became a creditor of, Creative Development. This, asserts Creative, precludes the possibility that Creative Development and Wolf Baking were under common control.
After carefully reading the 1986 Agreement as a whole and giving its words their generally prevailing meaning,26 we agree with Central States’s position to the extent that it characterizes the 1986 Agreement as unambiguously transferring a capital interest in Creative Development to Bey-chok, albeit without admitting him as a partner. We therefore conclude that, as [414]*414this interpretation does not produce any absurd consequences, it must be given effect without resort to extrinsic evidence.27
A contract is not ambiguous merely because the parties disagree upon the correct interpretation.28 In determining the presence of ambiguity vel non, we both parse the provision in question and construe that provision in the context of the entire document. The particular provision of the 1986 Agreement that we examine for ambiguity today states:
Creative does hereby sell, transfer, and assign unto Beychok an interest in that partnership [Creative] equal to said Fifty Thousand ($50,000) Dollar offset as described hereinabove. (Emphasis added).
The functional purpose of this provision is to identify the consideration that Bey-chok received from Creative Development in exchange for the bakeries’ cancellation of the $50,000 debt theretofore owed to them by that partnership. As noted, that cancellation by the bakeries was made possible by Beychok’s $50,000 reduction of the $324,000 then owed to him by the bakeries.29
In endeavoring to determine whether the “interest in that partnership” that Creative expressly sold, transferred, and assigned to Beychok could have more than one sensible meaning and thus be ambiguous, we find it helpful to engage in that venerable deductive exercise known as the process of elimination. In so doing, we first identify all of the possible kinds of interests that the words themselves could conceivably refer to in the context of the entire 1986 Agreement. We then examine each such possibility to see if it withstands legal analysis and remains a sensible reading of the agreement. If two or more of the possibilities remain viable, there is ambiguity; but if only one is left standing, there is no ambiguity.
Like the district court and the parties before us, we discern but three possibilities: (a) membership as a partner in Crea[415]*415tive Development; (b) debt owed by Creative Development to Beychok, or (c) an innominate financial interest (income, capital, or both) in that partnership, which interest is neither debt nor membership in the partnership. We proceed to analyze each possibility in order.
1. Beychok as a Partner in Creative Development
After the district court found the presence of ambiguity, it had no difficulty eliminating membership as a partner in Creative Development as one of the possibilities of the kind of interest that Beychok acquired. And, on appeal, neither the appellants nor the appellees seriously urge that the 1986 Agreement admitted Beychok into Creative Development as a partner. Clearly it did not. As Creative correctly explains, under Louisiana partnership law (1) unanimous action by the parties is required to amend a partnership agreement for the purpose of admitting a new partner unless otherwise agreed,30 (2) neither the number nor the identity of the partners of Creative had changed since it was formed in 1981, and (3) the 1986 Agreement was not signed by or on behalf of all four partners qua partners.31 In short, as the unanimous consent of the partners was not evidenced in the 1986 Agreement, then as a matter of law Beychok could not have been admitted as a partner.32 Moreover, the phrase “interest in that partnership” clearly eschews the contention that the sale, transfer, and assignment of such an interest somehow admitted Beychok as a partner: Memberships in partnerships are not sold, transferred or assigned; rather, persons are “admitted” into partnerships or “made” partners. Obviously, then, the first of the three possibilities — membership in Creative Development — must be eliminated.33
2. Beychok as a Creditor of Creative Development
Differing with the district court, we hold as a matter of law that, whether read “in a vacuum” or in context of the entire 1986 Agreement, the above quoted contractual provision neither transferred to Beychok the old promissory note that Creative Development had given to the bakeries in connection with acquisition of the bread depots nor created a new debt owed by Creative Development to Beychok. First, [416]*416only the bakeries, as payee and holder of the old promissory note, had the legal capacity to transfer it, yet there is no record evidence, much less any provision in the agreement, reflecting such a transfer by the bakeries. Conversely, Creative Development was the maker of the old note, not the payee or the holder, so it had no legal capacity to transfer the note. In fact, the 1986 Agreement states that it “expressly extinguished” the obligation, which under Louisiana law voids the note as well. Thus, neither the bakeries nor Creative ever purported to transfer or assign the old Creative Development note to Beychok.
Second, there is neither record evidence nor any language in the 1986 Agreement to indicate that a new or replacement note was made by Creative when that agreement was executed. There is simply no evidence that a new note was issued and made payable either to Beychok or to “Bearer,” then delivered to Beychok.
Absent express transfer or assignment of the old note or creation and delivery of a new note, only the above quoted language of the 1986 Agreement itself remains as a potential candidate for evidencing the creation or acknowledgment of debt owed by Creative Development to Beychok or the transfer or assignment to Beychok of an old debt owed by Creative to the bakeries. Yet absolutely nothing in that provision sounds in debt. Elsewhere in the 1986 Agreement the parties correctly employed such terms as “indebtedness,” “loan,” “debt,” and “obligation,” and did so with the professional facility we would expect of learned counsel who drafted it, thereby confirming the understanding of these terms and concepts by the parties and their scrivener. Unlike other portions of the 1986 Agreement, the particular provision that we now review for ambiguity employs none of these terms of indebtedness. In fact, none of the traditional objective indicia of a loan or credit relationship are anywhere to be found in the subject provision.34 Notably, there is (1) no reference to a promissory note representing the purported loan or credit obligation, (2) no maturity date for the purported loan, (3) no provision for repayment of the purported loan, (4) no specification of a rate of interest or a way to calculate it, (5) no reference to a due date or “payment on demand,” and (6) no provision concerning default. Perhaps most significantly, the subject provision of the 1986 Agreement contains no stipulation that, in the event of termination or liquidation of the purported debt, assets of that partnership would be paid to Bey-chok as a creditor in preference to monies due to its partners. The complete absence of these objective indicators of a debtor-creditor relationship far outweighs the subjective testimonial evidence proffered by Creative — and relied on by the trial court — to support the contention that the transaction’s purpose was to convey an old creditor’s interest or create a new one.35
[417]*417Furthermore, were we to have found ambiguity and considered extrinsic evidence, we would be compelled to observe the presence of four sworn documents, executed respectively by Jack Rome, Bey-chok and Rome, the bakeries’ bankruptcy counsel, and Wolf Baking’s comptroller, each of which was prepared for admission in various bankruptcy proceedings, and all of which uniformly state that Beychok was either a partner in or an owner of Creative Development. This is far too uniform and consistent to be explained away by press of the lawyers’ business. As a minimum, this independent sworn documentation would cast serious doubt on the subjective, self-serving testimonial evidence relied on by Creative and the district court to support the conclusion of debt, and would further support our conclusion that the 1986 Agreement was not intended to transfer a note to Beychok or to create a debt- or-creditor relationship between Beychok and the partnership.36 It follows inescapably that, like membership in Creative Development, debt too must be eliminated as the type of consideration that Beychok received in the transaction memorialized by the 1986 Agreement.
3. Beychok as the Owner of a “Capital Interest’’ in Creative Development
Having determined that the “interest in that partnership” sold, transferred, and assigned to Beychok was neither membership for him in Creative Development nor debt owed to him by that partnership, all remains to be done is to identify precisely what interest Beychok did acquire from Creative Development in the 1986 transaction, and whether identifying the interest as such would lead to “absurd consequences.” 37
Creative takes the position that the process of elimination supports the district court’s determination that the “interest” Beychok received was that of a creditor. Creative does so first by eliminating the possibility that the Agreement made Bey-chok a partner (with which we and the district court — and, presumably, Central States — all agree). But Creative then asserts that, as a matter of state law, the “in that partnership” that Creative Development transferred to Beychok could not have been a capital interest. Creative insists that Louisiana law does not permit a non-partner to acquire and own a capital or equity interest in a partnership without first being or becoming a partner. In this contention Creative badly misapprehends — or consciously mischaracterizes — ■ Louisiana partnership law.
In a number of circumstances, Louisiana law does in fact permit persons who are not partners to acquire capital or equity interests in the partnership. Perhaps the most commonly encountered example occurs when a partner dies. Although the heirs or legatees of the deceased partner do not themselves become partners, they nevertheless do, in the absence of a con[418]*418trary provision in the partnership agreement, “inherit the interest of the deceased partner, which entitles them to be paid as provided in Civil Code Article 2823 et seq.”38 The same holds true in the instances of (1) a creditor who seizes the interest of a partner, (2) a partner who voluntarily withdraws or is expelled from the partnership, and (3) a partner whose membership in the partnership terminates pursuant to provisions of the partnership agreement.39 In each of these variations, there exists “an interest in the partnership” that has value and must be accounted for, even though the successor to such interest never was or has ceased to be a partner.
The point is even more vividly demonstrated by the situation contemplated in article 2812 which provides that “partner may share his interest in the partnership with a third person without the consent of his partners, but he cannot make [the third person] a member of the partnership ....”40 This code article, which follows the approach of the French Civil Code, recognizes that, “[i]n the absence of an express prohibition in the partnership agreement, a partner may associate a third person in his interest in the partnership [even though] the association would not make the third person a partner.”41 And we are aware of nothing in Creative Development’s partnership agreement that prohibits the total or partial sale, transfer, or assignment of an interest to a nonpartner third person. Obviously, it would be sophistry for Creative to argue that a partnership cannot “transfer, and assign” that which can be alienated by .a partner. Moreover, as such a disposition does not require admission of a new partner or amendment of the partnership agreement, nothing in the Louisiana Civil Code or the partnership agreement mandates unanimous consent of the partners.
In sum, these examples confirm that Louisiana partnership law anticipates and expressly provides for the possibility that a third person may acquire and possess (at least for a time) “an interest in a partnership” — capital or income or both — without being a partner. We are satisfied that our interpretation of the 1986 Agreement as unambiguously transferring to Beyehok a capital interest in Creative Development (and conceivably, though unimportantly, an income interest as well) does not produce any consequences that are impossible under Louisiana partnership law or either nonsensical or absurd. Therefore this reading must be given effect without exiting the four corners of the 1986 Agreement to conduct further inquiry into the intentions of the parties.42 It follows that the district court’s resort to extrinsic evidence of intent was unwarranted and eventually led to reversible error in both methodology and substance.43
C. Withdrawal Liability
1. Membership as a Partner is Not a Prerequisite
As a final observation in the circuitous and arcane route to the determination of withdrawal liability of all members of a controlled group, we underscore the truism that, to recover for withdrawal liability under MPPAA, an ERISA multiemployer pension plan need not prove that [419]*419one who, with others, owns a “controlling interest” in, and exercises “effective control” over, a partnership that is one organization in a purported controlled group of trades or businesses, actually satisfies all of the state law requirements to be a partner in such partnership.44 Rather, all that MPPAA and its implementing regulations require is that such person own the requisite percentage of a “profits interest” or a “capital interest” in that partnership.45 As we have determined that the interest in Creative Development that Beychok acquired by virtue of the 1986 Agreement was a “capital interest” within the meaning of 26 C.F.R. § 414(c)-2, there can be no serious disagreement with the proposition that the interests of Beychok and Rome in both Creative Development and the bakeries are such that they must be tested to determine whether the requisite percentages — over 50 percent for “effective control” and 80 percent or more for the “controlling interest” — are present, irrespective of the fact that Beychok’s capital interest in Creative Development was not owned by him as a partner. If those capital interests are found to be present in such percentages, Creative Development and its partners cannot avoid solidary liability for the deficiency in the bakeries, withdrawal liability to Central States simply because Beychok was not a full-fledged partner in Creative Development.
If there are some who feel that controlled group rules produce unduly harsh results or set traps for the unwary, they should not turn a blind eye to all the facts that Rome and Beychok, as well as able counsel, knew or should have known when they confected the plan to salvage what they could from the impending bankruptcy of the bakeries. They had to have known, for example, that the bakeries (1) employed union labor, (2) were parties to a CBA, (3) were participating employers in a multiemployer pension plan pursuant to the CBA, (4) were approaching imminent bankruptcy, and (5) would, by virtue of bankruptcy, cease to participate in that multiemployer plan, leaving a substantial deficit in funding and thus withdrawal liability. As such, Rome, Beychok, and their counsel also knew or should have known that opting to confect and enter into the 1986 Agreement, which purposefully employed carefully crafted language that clearly eschews partner status for Beychok but just as clearly eschews debtor-creditor relationship between Creative and Bey-chok, was a high-risk endeavor. It amounted to flying perilously close to the flame that always burns brightly when super-majority interests in two separate entities are vested in five or less individuals and one of those entities is a participating employer in a multiemployer pension plan.
Neither should the history of intertwined business dealings among Rome and Beychok and the organizations that they owned and controlled be disregarded. The 1986 Agreement was no chance encounter; these two businessmen had been in business with each other on a number of prior occasions, in both the bakery business and the real estate business. And on at least one occasion — the 1982 bakery depot transaction — both individuals as well as the bakeries, Creative Development, and the Smiths were directly involved. In hindsight, it may well prove to be regrettable for Creative if the tangled web they helped weave by confecting and entering into the 1986 Agreement, and possibly the bakery depot joint venture as well, ultimately traps its weavers. Yet that distinct possibility was — or at least should have been — a known risk.46
[420]*4202. Render or Remand?
Time and again in its briefs and post-argument submittals, Central States expresses or implies that if Beychok’s $50,000 interest in Creative Development is found to be a capital interest and not a creditor’s interest, the conclusion is foregone that Beychok and Rome together owned at least 80 percent of the capital interest in both Creative Development and the bakeries at the time in question, and that those two organizations would be under common control per se. Central States finds this same conclusion implicit in the district court’s opinion as well. Even though at this juncture the presence of the “at least 80 percent” factor is irrefutable as to the bakeries, its presence is less than certain as to Creative Development.
More significant (and curious) is the observation that nowhere does Central States advert to the fact that the 80 percent “controlling interest” factor is but one of two prongs of the conjunctive test for “common control.”47 Although the 80 percent test determines “controlling interest,” that is only one-half of the “common control” calculus.48 The other half — “effective control” — is determined under the second prong of the test for common control in the brother-sister context.49 For partnerships, “effective control” is defined as “an aggregate of more than 50 percent of the ... capital interest of such partnership.”50 Importantly, this second, “effective control” prong takes into account the “ownership of each such person [singular] only to the extent such [person’s] ownership is identical with respect to each such organization.” 51 One need only consider the relevant example set forth in the regulation52 to realize that the “effective control” prong of the common control test is no simple arithmetic exercise; after all, if it were, there would always be a “more than 50 percent” capital interest when the persons [421]*421in question satisfy the controlling interest “at least 80 percent” prong of the test. But in the “effective control” second prong there is a tricky factor lurking just beneath the surface of the facially murky phrase, “only to the extent such ownership is identical with respect to each such organization .53
The turbidity of that phrase, especially the operative word “identical,” clears up considerably, however, when the “effective control” test is applied to actual examples. In this second prong test, the ownership of each person must be examined separately, focusing on one person’s ownership in each organization under consideration to find his or her ownership only to the extent it is identical in each organization. Practical application of this test reveals, every time, that the identity of ownership for each person is the smallest percentage that he or she owns in any of the targeted organizations.
Purely for purposes of illustration, we will employ Central States’s post-argument approach to ascertaining the percentages for Rome and Beychok, i.e., determining the percentages of capital ownership in Creative Development on the basis of the parties’ respective capital contributions. Thus we begin this hypothetical example by assuming that, of the total capital contribution of $55,-000,54 Rome’s -$2,500 represented 4.545 percent and Beychok’s $50,000 represented 90.9 percent. As for the bakeries, it is undisputed that Rome’s percentage of the stock in WBC was 28-55 percent and Beychok’s was 61.45 percent. In this example, then, Rome’s “identical” ownership in the two businesses would be 4.54 percent, i.e., his percentage of ownership interest in the capital of Central States; the difference between that percentage and his larger percentage of ownership in the bakeries drops out as nonidentical. In like manner, Beychok’s “identical” ownership in the two businesses would be 61.45 percent, i.e., his percentage of ownership interest in WBC; the difference between that percentage and his larger hypothetical percentage of ownership in Creative Development drops out as non-identical.
In this illustration, the “identical” ownerships of Rome and Beychok- — 4.54 percent for Rome and 61.45 for Beychok, for a total of 65.99 percent — obviously satisfy the second prong, “effective control” test which only requires an aggregate of more than 50 percent. Indeed, Beychok alone satisfies that test, as he is “five or fewer persons” and his “identical” ownership in each organization is more than 50 percent.
Additionally, in this hypothetical example, the first prong, “controlling interest” test for common control would be satisfied because the first prong examines seriatim the capital owners’ combined percentage in each separate trade or business. Based on his hypothetical 90.9 percent capital interest in Creative Development, Beychok alone would satisfy the “at least 80 percent of the ... capital interest” test for the partnership. And together, Beychok’s 61.45 percent and Rome’s 23.55 percent of the stock of WBC, totaling 85 percent, would satisfy the “at least 80 percent” test of both the voting power and the total value of all shares of all classes in the bakeries.
As the district court concluded that the 1986 Agreement (which it found ambiguous) neither admitted Beychok into Creative Development as a partner nor conveyed “an interest in” Creative Development to him, but instead made him a creditor to the extent of $50,000, the court never reached or addressed the crucial MPPAA question whether separately or jointly Rome and Beychok had “controlling interest” in and “effective control” of both Creative Development and the bakeries. Moreover, the status [422]*422of the record on appeal is such that— without engaging in substantial appellate fact finding regarding matters that at this juncture are not uncontested, stipulated, or otherwise clear beyond cavil— we cannot determine whether Rome and Beychok had controlling interest and effective control.55 And, as we decline to engage in such inappropriate fact finding, we are not able to render a judgment in this case, one way or the other. Instead, we are constrained to remand it to the district court for the limited purpose of adducing the evidence that it needs to make such indispensable factual determinations and calculations.
Inasmuch as the governing regulation on this point expresses “controlling interest” and “effective control” in percentages,56 the fundamental factual determination that the district court must make on remand is the percentage of the capital interest in Creative Development that Beychok owned at the time the bakeries “withdrew” from Central States. This will require the court to adduce sufficient evidence to enable it to convert Beychok’s $50,000 capital interest into a percentage. More specifically, the court must first convert that dollar amount to a percentage as of June 1, 1986, and then find whether, between that date and the effective date of the bakeries’ withdrawal from Central States, Beychok’s percentage changed or remained the same. The district court must also ascertain the percentage of Rome’s capital ownership in Creative Development as of the relevant time or times, presumably one-half of the figure derived by subtracting Beychok’s percentage from 100 percent. Then, with those percentages firmly established, the court must proceed to determine whether Beychok and Rome (or Beychok alone) owned at least 80 percent of the capital interest in the partnership for purposes of “controlling interest.” Thereafter, taking into account Beychok’s and Rome’s respective ownerships “only to the extent such ownership is identical” in both the partnership and the bakeries, the court must ascertain whether those two individuals (or one of them alone) owned more than 50 percent of the capital interest — and thus “effective control” — in both Creative Development and the bakeries.57
Should the district court ultimately conclude that both prongs of the common control test are satisfied, it must then render a judgment assessing Creative’s responsibility (and that of its partners) for the delinquency in the withdrawal liability owed by the bakeries to Central States. If, however, the court concludes that either prong of that test is not satisfied, it must render a judgment dismissing Central States’s action against Creative and its partners.
In the interest of judicial economy and to avoid the need for another panel of this court to “reinvent the wheel” if either or both sides are so disappointed with the district court’s findings and rulings on remand that they appeal, this panel retains appellate jurisdiction pending this limited remand to the district court. Consequent[423]*423ly, regardless of whether or not the district court finds on remand that the required joint or aggregate percentages of “controlling interest” and “effective control” are sufficient to constitute “common control” and thus impose controlled group liability on Creative under MPPAA, any appellate review will be conducted by this panel.
IV.
CONCLUSION
The district court reversibly erred in holding that the 1986 Agreement was ambiguous, and it compounded the error by considering extrinsic evidence of the parties’ intent and basing its judgment on that evidence. We conclude de novo that the 1986 Agreement was not ambiguous and that it conveyed a capital interest in Creative Development to Sheldon Beychok. Consequently, if in combination the interests of Beychok and Rome in both Creative Development and the bakeries are ultimately found to be sufficient to constitute Creative Development a member of the same controlled group of trades and businesses as Wolf Baking, then Creative Development and its partners, the Romes and the Smiths, will be liable in solido to Central States for the outstanding balance of Wolf Baking’s withdrawal liability.
Only one facet of one prong of the two-prong common control test is discernible from the record on appeal: In both combined voting power and total value, Rome’s and Beychok’s shares of stock in WBC were sufficient to vest those two shareholders with a “controlling interest.” Without engaging in inappropriate appellate fact finding, however, we cannot convert Beychok’s dollar interest in Creative Development to a percentage interest. And without that indispensable piece of the puzzle before us, we are unable to determine whether Beychok’s and Rome’s combined capital ownership in Creative Development equaled at least 80 percent and thus constituted a controlling interest in that partnership. For the same reason, we are unable to determine the extent of either Beychok’s or Rome’s ownership interests in those two business organizations to the extent that they are “identical with respect to each,” so we cannot say whether those two individuals had “effective control” of the bakeries and the partnership. It follows that neither we nor the district court can tell whether the two business organizations were under “common control” for purposes of MPPAA, a determination that is critical to either court’s ability to decide whether Creative Development and its partners have soli-dary withdrawal liability to Central States.
Accordingly, we reverse the district court’s judgment that dismissed Central States’s withdrawal liability claims, and we remand the case to that court for the limited purpose of (1) determining the several ownership percentages required to test for the presence of common control; (2) applying the percentages thus determined to both prongs of the common control test; and (3) if common control is found to have been present, assessing the quantum of Creative’s withdrawal liability to Central States and rendering a judgment accordingly. In the interest of judicial economy, this panel retains appellate jurisdiction for the purpose of reviewing the determinations and judgment of the district court on remand, should the parties or any of then elect to appeal.
REVERSED and REMANDED with instructions; appellate jurisdiction retained by this panel.