The Large Business and International Division of the I.R.S. (“LB&I”) periodically develops international practice units (“I.P.U.’s”) that serve as training material for international examiners. I.P.U.’s provide explanations of general tax concepts and information about a specific type of transaction. Because I.P.U.’s are not official pronouncements of law, they cannot be used, cited, or relied upon as authority. Nonetheless, they explain the general approach that will be followed by an LB&I examiner and are helpful when preparing for an I.R.S. examination of a multinational group.
In November 2017, the I.R.S. issued an I.P.U. entitled “Common Ownership or Control Under IRC 482 – Inbound.” It serves as a primer for determining whether sufficient control exists between two parties to bring the arm’s length transfer pricing rules of Code §482 into play.
On the same date, the I.R.S. issued a sister I.P.U. for outbound transactions, “Common Ownership or Control Under IRC 482 – Outbound.” It is based on the same set of principles and is virtually identical to concepts of control for inbound transactions.
This article explains how the I.R.S. looks at the issue of control. How is it defined? In what fact patterns does it exist? In approaching these issues, this article focuses on the context of a non-U.S.-based group with operations in the U.S.
The I.P.U. begins with the acknowledgement that a foreign-based group operating in the U.S. generally do so through a U.S. subsidiary. If the group sources its product outside the U.S. for sale in the U.S., the U.S. subsidiary generally is charged with the task of establishing a marketing plan and implementing that plan through a U.S. sales network.
The I.P.U. identifies the following types of transactions that often exist between the U.S. subsidiary and its parent or affiliates based abroad:
In comparison to business transactions entered into by unrelated parties, where each party is acting solely to increase its own economic goals, the I.P.U. expresses the view that related parties may take steps to price transactions based on other factors. Where that occurs, a U.S. taxpayer may underreport its U.S. taxable income and Federal income taxes.
To prevent tax slippage arising from related-party transactions, Code §482 authorizes the I.R.S. to conduct an examination and to reallocate income among related parties when necessary to reflect arm’s length pricing. The purpose of Code §482 is to ensure that taxpayers clearly reflect income from “controlled transactions” and to prevent U.S. taxpayers from avoiding taxation by artificially shifting income.
However, the I.P.U. acknowledges that the mere fact that two parties are related does not create any presumption that intercompany pricing is other than arm’s length.
A transaction is a controlled transaction if it occurs between two or more organizations, trades, or businesses that are either owned or controlled by the same interests. A controlled group of taxpayers is a group of taxpayers that are owned or controlled directly or indirectly by the same interests.1 Therefore, a controlled transaction is any transaction or transfer between two or more members of the same group of controlled taxpayers.
In contrast, an uncontrolled transaction is any transaction between two or more taxpayers that are not members of the same controlled group.2
Thus, the term “controlled” in the Treasury Regulations is a shorthand that generally refers to the concepts of both common ownership and common control, except where it is necessary to distinguish between those concepts.
The term “controlled” is defined as any kind of control (i) whether direct or indirect, (ii) whether or not legally enforceable, (iii) however exercisable or exercised, and (iv) including arrangements by which two parties act in concert or with a common goal or purpose.3 It is the reality of control that is the decisive factor and not its form or the mode through which control is exercised. Control is presumed to exist if income or deductions have been arbitrarily shifted between related parties.
Common ownership or control is determined at the time the parties agree to perform a transaction, even if the parties perform the transaction later.
CONTROL THROUGH DIRECT OWNERSHIP
The first step is to determine whether the “ownership” test is satisfied. The position of the I.R.S. is that common ownership exists if there is a greater than 50% ownership by the same related-party interests.
Ownership can be direct or indirect. Direct ownership occurs when one party directly owns stock or another ownership interest in its name. This is illustrated in the following diagram:
Indirect ownership occurs when one party owns the stock or other ownership interest of another party indirectly through ownership of one or more other, intervening parties. This is illustrated in the following diagram.
In the diagram, Foreign Corporation owns 100% of the stock of U.S. Corporation A and 35% of the stock of U.S. Corporation B. U.S. Corporation A owns 40% of the stock of U.S. Corporation X, and U.S. Corporation B owns 60% of the stock of U.S. Corporation X. Therefore, Foreign Corporation indirectly owns 61% of the stock of U.S. Corporation X. The facts and circumstances would need further development to determine if common control exists.
If, in the above diagram, U.S. Corporation B were to have only two shareholders, viz., Foreign Corporation and an unrelated U.S. Corporation C (not shown in diagram), Foreign Corporation may not be able to exert actual control over pricing even though it directly owns 40% and indirectly owns another 21% of U.S. Corporation X. On the other hand, if unrelated U.S. Corporation C also purchases goods or services from U.S. Corporation X, both Foreign Corporation and U.S. Corporation C could be acting in concert to keep prices for purchased goods or services below an arm’s length amount. In such case control would exist and the prices charged by U.S. Corporation X to Foreign Corporation and U.S. Corporation C may not be at arm’s length.
The I.P.U. addresses a case in which the facts were almost identical to those in the diagram. In W.L. Gore v. Commr., 4 the U.S. taxpayer entered into a 50/50 joint venture with a Japanese corporation in which it held a 30% ownership interest. The U.S. taxpayer granted to the joint venture the exclusive license to use certain technology in Japan on a royalty-free basis. The joint venture’s rights to license the technology acquired from the petitioner were severely circumscribed. The agreement also provided that, in case either the taxpayer or the joint venture were to improve the technology or develop new technology, each would promptly disclose such technology and grant a royalty-free license to the other. A similar license arrangement existed between the Japanese corporation and the joint venture.
The I.R.S. contended that the joint venture and the U.S. taxpayer were under common control. In part, the I.R.S. argued that the U.S. taxpayer not only owned a direct 50% interest in the joint venture but also an indirect 15% interest through the Japanese corporation. The joint venture agreement provided for royalty-free cross licenses. It also alleged managerial control existed in the U.S. corporation and that the two parties to the joint venture were acting in concert.
The taxpayer filed a motion for summary judgment, in part because it did not directly own sufficient shares in the joint venture to control its activities. The motion was denied. The court accepted several arguments raised by the I.R.S. in support of its position that summary judgment was not appropriate in the circumstances. One of those arguments was the existence of a 15% indirect interest in the joint venture through the taxpayer’s ownership of a 30% interest in its joint venture partner. The court stated that the 30% ownership of the joint venture partner can “properly be considered even if the usual standards for attribution of ownership, such as those found in section 318, are not met.” Whether that 30% ownership provided the U.S. taxpayer with control was a matter of fact that would have to be determined at trial.
The I.P.U. views indirect control through an unrelated party as an important factor, stating:
While W.L. Gore could be viewed as a case that the Tax Court decided based on common ownership (i.e., taxpayer’s 65% overall ownership interest in JV), the Tax Court also addressed common control factors such as managerial control in reaching its decision. Thus, the Tax Court in W.L. Gore addressed both common ownership and control for purposes of IRC 482.
In a situation where Code §482 can apply only if there is common control (due to the absence of common ownership by a majority of the same interests), common control might result in any number of ways depending on the facts of the case.
Where a taxpayer has legal voting control over another entity, the I.P.U. states that the control element will usually be met for purposes of Code §482. This is true even when a taxpayer owns less than 50% of the value of the stock, yet holds a majority of the voting stock, of a corporation.
This fact pattern is illustrated in Diefenthal v. U.S., 5 where one third of the stock in a corporation called Scrapco was owned by a father and the other two thirds were split between two sons. The stock of the two sons was held in trust and the father was named as the trustee. Acting as trustee, the father had the power to vote two thirds of the shares, and acting as shareholder, the father voted the balance of the outstanding shares. The Code §482 issue was whether Scrapco engaged in arm’s length transactions with a corporation owned wholly by the father.
The district court held that common control was present for Code §482 purposes because the father had power to vote 100% of Scrapco’s stock and also owned 100% of the other corporation that participated in a transaction with Scrapco. The court reasoned that, on the basis of voting control, Code §482 was applicable.
The I.P.U. then proceeds to posit the following fact pattern as an illustration of control without ownership of more than 50% of the shares. Unrelated entities Foreign Corporation A, Foreign Corporation B, and Foreign Corporation C all have ownership interests in U.S. Corporation as illustrated in the following diagram:
U.S. Corporation’s governing documents state that all material company decisions will be made by a majority vote of the shareholders. Because voting power is typically controlled by ownership of voting shares, Foreign Corporation B clearly has common control of U.S. Corporation. The I.P.U. then asks whether U.S. Corporation is also controlled by Foreign Corporation A because of its ownership of 80% of the stock. While the I.P.U. does not provide an answer, it would appear that voting control trumps majority ownership, in the absence of other arrangements.
Even if a taxpayer does not have absolute voting control, there are scenarios where the taxpayer, on a practical level, has sufficient control so that common control is met.
One example involves a fact pattern in which 49% of a corporation’s stock is owned by a single entity and the other 51% is widely dispersed among many other shareholders, none of which owns more than 1%. Clearly, the owner of the 49% interest controls the corporation.
A second example involves a non-majority owner of a joint venture entity that provides all the debt financing to the joint venture or supplies the joint venture with essential components under an exclusive supply agreement. The presence or absence of control would depend upon the degree of ownership and the significance and size of the particular transaction relative to the joint venture business.
Common control for purposes of Code §482 may be established based on one party’s management control of another entity. Thus, for example, in Charles Town, Inc. v. Commr., 6 the court held that common control existed where there was only 2% common ownership.
In the facts of the case, two brothers owned all of the stock of Fairmount. The brothers formed a new corporation (“Charles Town”), which acquired a race track. The brothers owned only 2% of the outstanding stock in Charles Town. A cousin owned the other 98%. However, the brothers served as president, secretary-treasurer, and directors of Charles Town. Fairmount advanced funds to Charles Town for operations at the race track. Charles Town operated the track and retained 10% of the net profits, while paying the remaining profits to Fairmount. The facts are illustrated in the following diagram:
The I.R.S. allocated all income received by Fairmount to Charles Town. Charles Town contended that insufficient control existed between Charles Town and Fairmount. However, the court held that of both Charles Town and Fairmount were controlled by the two brothers. While they owned only 2% of Charles Town shares, the court found that the brothers controlled Charles Town because they caused the corporation to be formed, constituted the majority of the board of directors, were principal officers of the corporation active in its management, and made all major decisions with respect to the allocation of income and expenses.
CONTROL IN CONCERT WITH AN UNCONTROLLED PARTY
Another indicator of common control occurs when two or more entities “act in concert.” The I.P.U. acknowledges that non-majority shareholders and owners can have control over another entity if they act in concert as a majority with a common goal to shift income or expenses to or from the entity.
The paradigm case is B. Forman Co. v. Commr., 7 where the Second Circuit Court of Appeals reversed the Tax Court and concluded that common control exists where two unrelated corporations are equal owners in a third corporation and act in concert to direct the actions of the third corporation. In the case, two unrelated corporations made equal interest-free loans to a third corporation “all of whose stock they owned and all of whose directors and officers were their alter egos.” Using a “realistic analysis,” the court found that the two unrelated corporations exerted control even though they had no common shareholders, directors, or officers. The court found that the two unrelated corporations acted with a common goal to shift income. Thus, the court upheld I.R.S. reallocations between the controlled corporation and its two corporate owners.
The I.P.U. points to the importance of the fact that the shareholders’ economic and tax interests were lined up in parallel with each other. This made the income shifting determination more obvious. However, if two taxpayer/owners have clearly adverse interests, a common goal may be absent, which could prevent the application of Code §482.
Another consideration in finding that unrelated taxpayers are acting in concert is the dependence of each company on the other. An example is South Texas Rice Warehouse Co. v Commr.8 The taxpayer owned a rice-drying warehouse, while a related business leased the warehouse from the taxpayer and operated the warehouse. Four families each owned 25% of both entities, although the family members were not always the same.
The I.R.S. proposed an adjustment in the income of the two businesses in order to properly reflect income among controlled parties. The adjustment was contested in the Tax Court on the grounds that the same interests did not control the two businesses within the meaning of Code §482. Acknowledging that the two businesses were controlled by different members of the same two families, the taxpayer argued that Code §482 contains no provision for the imputation of ownership or control because of ownership by related parties.
The court disagreed with the taxpayer, stating that the requisite control may exist even though it is not legally enforceable. It is the reality of the control which is decisive, not its form or the mode of its exercise. Under the facts of the case, common control existed between the two businesses in part because the two business were owned only by members of two families and in part by the interdependence of one business on the other. One owned the facility, the other leased the facility. One leased land and water to rice farmers and the other operated a rice drying and warehousing business. For the business of drying and storing rice, not only was a facility required but also rice throughput for drying. When looked at as a whole, the lands that were leased, the facilities that were operated, and the rice that was grown were parts of an integrated business in which the owners were members of the same two families, who had common interests and always acted in concert.
The I.P.U. goes on to caution that the mere existence of a family unit owning two businesses that conduct business with each other is not always sufficient to justify a conclusion that the companies are under common control. An example is Brittingham v. Commr.9 The case involved two brothers and their immediate families. Each brother owned, directly and through his immediate family, 37% of a U.S. corporation and participated on the board of directors of that corporation. However, only one brother served as an operating officer of that corporation. The U.S. corporation purchased inventory from a Mexican corporation that was wholly owned by the other brother, his mother, and his wife.
The I.R.S. adjusted the intercompany purchase price paid by the U.S. corporation, asserting the price was artificially high in order to shift profits. Code §482 applied in the view of the I.R.S. because the brothers and their families collectively owned or controlled both corporations so that the same interest controlled both corporations.
The Tax Court rejected the I.R.S. position, because no evidence existed that the corporations were operated in concert or that profits were artificially shifted to the Mexican corporation. Only one brother, along with his wife and mother, owned the Mexican corporation. He was not active in operating management of the U.S. corporation, and it was not credible the brother who operated the U.S. corporation would shift profits to a company he did not own. The I.P.U. agreed with the court in concluding that in light of the facts in the case, neither the common ownership test nor the common control test was satisfied and Code §482 was inapplicable.
TIMING OF CONTROL
In many cases, the timing of common ownership or control is not an issue because the relevant events all occurred at a single point in time. But in some cases, where relevant events occur over a period of time, the timing of common ownership or control can be an issue. In these circumstances, it is necessary to determine the appropriate point in time when the parties were commonly owned or controlled and took steps to enter a transaction under a specific pricing arrangement DHL Corp. & Subsidiaries v. Commr.10 involved the global package delivery company DHL, a U.S. corporation. It was part of a global network in which DHL handled U.S. operations exclusively. DHLI held international rights to use the DHL trademark and was based in Hong Kong. Independent local agents conducted DHL’s international operations and paid a network fee to DHLI.
In 1989, two foreign corporations entered into a memorandum of understanding with DHL to purchase (i) a 60% ownership stake in DHLI for $450 million and (ii) the DHL trademark for $50 million, subject to confirmation of the tax effect for the trademark transaction. Subsequently, but before the acquisition was concluded in 1990, DHL and DHLI executed an agreement granting DHLI an option to purchase the DHL trademark for $20 million. Shortly thereafter, a final agreement was reached with the foreign corporations under which the purchasers acquired: (i) a 12.5% stock interest in DHLI, with an option to purchase an additional 45% interest; (ii) a 2.5% interest in DHL; and (3) an option to purchase the DHL trademark for $20 million. On June 7, 1992, the two foreign corporations exercised their stock option, purchasing a majority stake in DHLI. In autumn 1992, the foreign corporations, as majority owners of DHLI, caused DHLI to exercise its option to purchase the DHL trademark rights for $20 million. Clearly, DHLI’s option to acquire the DHL trademark was negotiated when DHL and DHLI were related parties.
Clearly, too, DHLI was not related to DHL at the time the option was exercised. This presented the court with the following question: At the time the purchase of the trademark was completed and DHLI and DHL were not under common control, was Code §482 applicable? The Tax Court answered in the affirmative, and that conclusion was confirmed on appeal.
Code §482 was applicable because DHL and DHLI were commonly controlled entities as of time of their negotiations and the date on which terms were set. The fact that an uncontrolled party was also involved did not remove the terms of the transaction from the ambit of Code §482 because the uncontrolled party was either indifferent to the price or possibly advantaged by the price. The price was determined by a taxpayer intent on shifting income and in position to effect an income shift. Of note was the fact that the net value of the entire transaction was not affected by the reduction in the price of the trademark as other revisions offset that decrease in price.
A similar result was reached in GAC Produce Co. v. Commr., 11 a case in which a U.S. marketing company agreed to help market fresh produce grown and distributed by a group of companies based in Mexico. The agreement with the U.S. marketing company purported to allocate the prices among the U.S. and Mexican members of the Mexican group. When the U.S. member of the Mexican group was examined by the I.R.S., a transfer pricing adjustment was proposed between the U.S. group member and related suppliers in Mexico. The U.S. member argued that Code §482 was not applicable because the internal pricing was controlled by the agreement with the U.S. marketing company. The Tax Court agreed with the I.R.S. The U.S. marketing company was indifferent to the method by which the Mexican based group internally allocated income.
However, the I.P.U. cautions that the result in GAC Produce could be different in other facts. Control would not exist where the unrelated party is not indifferent to the shift of profits – meaning that bears an economic loss from each dollar shifted – and, for that reason, the unrelated party keeps the controlled parties from shifting income.
Arm’s length transfer pricing rules are designed to prevent controlled parties from manipulating transaction values in order to reduce taxes inappropriately. Operating management often believe that control is determined based solely on the existence of common ownership at the time of a transaction. As demonstrated in the I.P.U., this approach to equating control solely with ownership is flawed. For transfer pricing purposes, “control” is given a broader meaning that looks not only to ownership but to control of any kind (i) whether direct or indirect, (ii) whether or not legally enforceable, (iii) however exercisable or exercised, and (iv) including arrangements by which two parties act in concert or with a common goal or purpose.