There is a well travelled path in transfer pricing that in many cases provides both a reasonable result and an adequate response to a government tax examiner. This path involves selecting a method of analysis, picking a tested party, determining its profitability under an appropriate financial measure and then comparing this profitability to a set of comparable companies. This is well and good when it works, but in some instances this path leads to the deep woods, for example, when the tested party's profits can't be readily benchmarked, or when the profit left in a particular jurisdiction doesn't match the existing division of functions and risks. In this article, the authors draw from their experience to describe some practical approaches to these issues.
Interdependent transactions
A number of businesses are faced with difficult transfer pricing questions in cases where there are interrelated transactions, one of which provides all or almost all of the profit. Consider the following fact pattern.
A US-based equipment manufacturer (USCo) sells its equipment into the US market with little or no profit, but earns substantial returns on post-sale maintenance. When USCo is deciding how to set prices, it is trying to maximize profits from the joint sale of product and after-sales services. As a result, the price that is charged for the product sale is related to the price charged for services, and is probably lower than it would be on a stand-alone basis. In this case, the pricing of the product and the services are interdependent because the services profit makes up for the low profits/losses incurred on the initial product sale.
USCo establishes a subsidiary in a European country (EuroSub) and sells its equipment to EuroSub. Let's assume that USCo prices its equipment to EuroSub at the price it sells to third-party customers in the US, and therefore earns little or no profit on such sales. EuroSub resells the equipment to European customers with a 1% markup and earns little or no profit on the equipment sales. Furthermore, assume that maintenance is not formally tied to the equipment sale. However, EuroSub does, in fact, secure a substantial share of the post-sale maintenance work and earns substantial profits from these services.
Due to the interdependent US retail pricing, the sales by USCo to EuroSub may be priced at less than a stand-alone arm's-length price even though: (1) the market for the equipment appears similar between the US and Europe as indicated by the similar third-party pricing (EuroSub sells at the U.S. retail price plus 1%); (2) USCo is using a retail price (USCo's sales to third parties) as a benchmark for a wholesale transaction (USCo's sales to EuroSub); and (3) EuroSub is making little or no profit on the equipment sales. A documentation report that provides the above analysis, however, would carry US transfer pricing risk as the IRS might analyze the transaction and determine that USCo will always earn low returns, or even losses, on its sales to EuroSub. Presumably the European tax authorities would be happy with EuroSub's high profitability on its maintenance work, but they might strongly object to an IRS forced increase in the price of the equipment to provide some minimal profit to USCo when it sells the equipment to EuroSub, where such a price might well establish a negative gross margin on every final sale by EuroSub. This difference in views between the IRS and the European tax authorities could well lead to double tax.
The difference in opinion between the tax authorities probably stems from the problem that the identified intercompany transaction relates solely to the tangible property transfer, when in fact there is little or no system profit-related to the equipment sales due to interdependent pricing with the services. In looking at the business as a whole, USCo should probably share in the profits of EuroSub's post-sale services, as it does with regard to its own sales in the US market. One practical method for sharing these profits is to burden the post-sale maintenance transaction with a royalty or other fee so that EuroSub does not earn more than a routine return for its combined functions and risks. To support this charge by USCo to EuroSub, it will be important to establish a tie between the technology or the brand associated with the equipment and the post-sale maintenance work. It may be possible to establish that EuroSub wins a substantial share of the post-sale maintenance contracts because it markets these services using USCo's trade name. It might also be the case that EuroSub's service technicians use know-how provided by USCo, for example training and service manuals. In taking this approach, there are several issues to consider:
1) What is the appropriate royalty rate? The answer to this question will depend on the level of system profits on equipment sales, the extent of cross subsidization from the services business to the equipment business and the anticipated volume of equipment and service revenues in EuroSub's jurisdiction.
2) What are the non-transfer pricing tax implications of implementing a royalty? For example, while establishing a royalty may result in the parties recognizing the right amount of income, it could change the US tax consequences to the manufacturer because of interplay between the source-of-income rules and foreign tax credit rules. Because additional sales income presumably would be 50% foreign-source income, but the royalty is 100% foreign-source income, the royalty approach will increase USCo's foreign tax credit limitation. Therefore, if the royalty is foreign source and USCo has excess foreign tax credits that it could not otherwise utilize, the royalty may come in essentially tax free. The IRS may, in such a case, want the sales price of the equipment adjusted to reflect an arm's-length price. In other words, it's EuroSub's prerogative if it wants to use the equipment as a loss leader, but USCo should get an arm's-length price on the equipment sale. Royalty payments between some countries may also be subject to withholding tax. If the answers to these questions can be resolved successfully, the taxpayer could implement this practical solution to resolve its problem while avoiding an increase in the product pricing that would probably be unpopular with the European tax authority.
Persistent losses in marginal markets
Sometimes a multinational company has a uniform transfer pricing policy that it believes is arm's length and in general, this transfer pricing policy can be supported with a traditional Comparable Profits Method (CPM) or Transactional Net Margin Method (TNMM). However, the multinational may have a few outposts that consistently lose money. In some cases, for example when the outpost is a simple distributor, it would be straightforward to have the manufacturer lower the transfer price so the distributor earned a routine margin. While this may result in problems, for example, if the tax authority looking at the manufacturer lowering its transfer price takes issue with this "support" payment, the transaction is readily identifiable and the taxpayer should be able to address the issue head on with a solid functional analysis and pragmatic approach with the tax authorities.
There are other cases, however, where related parties in marginal markets also lose money, but there are no or limited direct related-party transactions. Such a fact pattern often occurs in businesses where the value of the business lies in its network. Consider the following situation in the telecommunications industry: Parentco has affiliates in a range of different countries. The success of the business of any one of these affiliates is dependent upon the existence of the other affiliates, as it is important to have a network that covers all of Europe, for example, and not just selected markets where the telecommunications company can earn a profit.
In contrast to the outpost distributor example noted above, in this example there are an insufficient number of intercompany transactions between the network and the loss making affiliate (Country A) to "solve" the problem through traditional transfer pricing. However, while some type of subsidy is needed, such a subsidy is likely to be resisted by the tax authorities of the countries paying the subsidy as not providing a direct benefit to the affiliate operating in their country.
As a first step, it is critical to establish in the functional analysis that the entity in Country A conveys value to the network. Such an analysis probably will be complex, as it will involve demonstrating that the loss incurred in Country A is outweighed by the incremental revenue that the other affiliates earn as a result of being able to provide service to Country A. Once demonstrated, however, several practical solutions can be employed to address the issue.
One such solution involves converting one entity into a centralized key entrepreneurial risk-taking (KERT) entity for the system. The KERT would have overall economic responsibility for the network and would conclude service contracts with the other entities in the network, whereby the KERT would take on any non-routine functions and risks. The KERT would pay the other network entities on a cost-plus basis and would thereby take on the risk of any residual profit or loss in the system, although the local affiliates would remain the entrepreneurs on their local business. Thus, in our earlier example, Country A would earn a profit element above its costs that is consistent with its functions and risks. Similarly, the KERT would earn returns commensurate with its position as the entrepreneur of the network.
Tested party results are outside the CPM range
Many companies with distribution subsidiaries utilize the CPM or the TNMM to document the arm's-length nature of their transfer pricing results. In the most standard case, the distribution subsidiary is selected as the tested party, and its consolidated results are tested against an estimated arm's-length range established by independent, comparable, publicly traded companies. If the multi-year results (for example, the three-year weighted average operating margin) of the tested party falls within the interquartile multi-year range of results of the comparables, the taxpayer concludes that its results are consistent with the arm's-length standard for the tested year. Suppose a taxpayer performs this analysis contemporaneously for its tax return (but post year-end), and discovers its results are not within the arm's-length range? What are the practical approaches to deal with this situation?
Depending on the facts and circumstances, the taxpayer may consider either: (1) adjusting its transfer pricing results on a timely filed tax return to conform with the established comparable company range; or (2) adjusting the transfer pricing method (or its application) to reflect changes in facts that are driving the divergent results. As has been discussed elsewhere in this publication, post-year adjustments can lead to a range of significant issues. Therefore, here we will address the second alternative.
To address this issue, let's consider the following example in the toys/games industry. This industry is characterized by seasonality and cyclicality. Seasonality comes into play as a substantial portion of industry sales are made in connection with the December holidays. Cyclicality comes into play as success in the industry is driven by a company's continuing ability to innovate and create the next hot game. Creating the next hot game typically results in the sale of all existing inventory at strong retail prices with the promise of continued demand for more product. Inability to do this can have disastrous immediate-term and long-term financial consequences. In the immediate-term, the company can be forced to sell its inventory at close-out prices. In the longer term, retail shops may allocate shelf space to other companies.
With this industry backdrop, GameCo, a company headquartered outside the US, has subsidiaries around the world including one in the US, US-SUB. GameCo performs product development with input from US-SUB, conducts all manufacturing and provides broad guidelines to its foreign affiliates regarding the marketing of product. US-SUB is responsible for making a market in the US for GameCo's product, including both marketing and distribution.
In the current fiscal year, US-SUB earns a three-year weighted average operating margin that falls below the interquartile range of the comparable companies. Its single year operating margin falls below the single year interquartile range as well. We will explore three possible reasons for this result, including information on how to spot such situations and ways to resolve these issues.
1. Imperfect comparable companies
Perfect comparables do not exist. Typically, comparable company selection is an exercise in identifying the compromises that can be made that least impair the reliability of the analysis. In the example of GameCo in particular, comparable company selection will be challenging, as there are few publicly-traded, pure play distributors of toys and games in the US. It may be possible to identify distributors of other durable goods that exhibit more closely the cyclicality and seasonality experienced by the tested party (for example, construction and building materials), but it is highly probable that these companies will not exhibit the same swings in profitability from year to year that are specific to the games industry. In addition to broadening the sample of comparable companies, other practical solutions might include examining financial results under a longer time horizon, under the theory that over time US-SUB will earn returns consistent with those of full-fledged distributors, or making adjustments to improve the reliability of the financial results of the comparable companies. In this example, one such adjustment might involve giving effect to US-SUB's role in product development.
2. Changing product mix
In the current year, let's say five products (A-E) constituted 95% of US-SUB's sales. At the beginning of the year, GameCo worked with US-SUB to establish transfer prices that would apply for each product. US-SUB performed substantial market research so that it had a strong idea of the expected retail price and volumes. It used this information to develop a "resale minus" approach, to establish a gross margin that should leave it with a routine distribution return at the operating profit level (that is, taking into consideration its level of fixed costs). At the beginning of the year, US-SUB expected volumes of products A-E each to be 20% and priced each to yield the same gross margin contribution at the expected volume levels.
At the end of the year, products A-D achieved their volume and price targets, but each represented 22% of US-SUB's sales because of a shortfall in product E's sales relative to budget. In turn, that unexpected shortfall in sales of Product E led to the shortfall in profit contribution that led to US-SUB's falling below the interquartile range established by the comparable companies. First and foremost, monitoring product volumes and external pricing on inter-company transactions throughout the year can help identify product mix problems early. The practical solution in that case is to rely on reforecasted volume numbers and pricing data from the market to establish gross margins that yield the right profit contribution. This can be done not only on product E, but also on products A-D.
3. Non-transfer pricing reasons
There are a whole host of reasons completely unrelated to transfer pricing that could be the cause of US-SUB's profit shortfall – from heightened competition that affects the tested party and not the comparable companies to foreign exchange gains and losses, to missing the boat with the "hot" toy. In this example, let's say that one of the major retailers to which US-SUB sells experienced financial distress in the current year. As a result of this distress, the retailer closes stores, slashes its workforce, and sells less of US-SUB's product than anticipated. US-SUB's sales consequently are lower, which provides it with less gross profit to cover its fixed costs and leads to US-SUB's profit shortfall relative to the comparable companies. As a practical matter, the loss of a major customer is a business risk that most full-fledged distributors, including US-SUB, assume. Likewise, the misidentification of this year's "must have" game would be at least in part the responsibility of US-SUB, which participates in product development for its market. In a transfer pricing documentation setting, practical solutions for dealing with such an issue would involve arguing that a gross-profit-oriented transfer pricing method such as the Resale Price Method would provide a more reliable answer than an operating profit-oriented transfer pricing method. Alternatively, one could rely on an operating profit-oriented transfer pricing method and rely on qualitative explanations on why US-SUB's poor results are driven by business factors wholly separate from transfer pricing.
In the "outside the range" examples discussed, the available alternatives and selection of method are based on the facts and circumstances. If the application of a method or the method itself is changed from a prior year, the taxpayer should take care to document the specific changes in facts that lead to the change in conclusion about its choice of best method. If the tested party result appears to be driven more by inaccurate implementation of a transfer pricing policy than by changes in facts and circumstances, a voluntary transfer pricing adjustment may be the most appropriate solution.
Legal rights and economic substance
Legal rights are an important aspect of transfer pricing. Both the OECD transfer pricing guidelines and the US regulations identify contractual rights as a critical factor in establishing the appropriate transfer price. Because taxpayers have control over the content of the inter-company agreements, these agreements can be a powerful tool for establishing and supporting a transfer pricing policy.
In the ideal situation, taxpayers will have legal agreements in place that define the division of rights, risks, and responsibilities. But in many cases, there may be no agreement in place, the legal agreement may be ambiguous, the agreement may not address the specific issue at hand, or the agreement may be inconsistent with the surrounding facts or the parties' behavior.
Ambiguous legal agreements
As in all business, inter-company agreements don't always provide clear answers to transfer pricing questions. There are a wide range of examples. When output is moved from one plant to another, does the manufacturer losing production volume have a right to compensation? Similarly when a distributor is "de-risked," does the distributor have a right to some payment? When a company writes down the fixed assets of a "cost-plus" manufacturer, who should bear this cost?
It is clear that tax authorities take these issues into account. Certainly it would be best if the legal agreements clearly stated answers to these and other questions; however, it is often difficult if not impossible to foresee all contingencies when drafting intercompany agreements. Taxpayers and tax authorities often will look solely to economic reasoning to find answers to these questions. For example, a tax authority might argue that there is an economic loss to the manufacturer or distributor that follows the plant closing or the change in the risk profile. Similarly, the fair market value of the asset, as reflected in the write down, might arguably be the appropriate cost base from an economic perspective.
The answer might not rest, however, solely on economics. An alternative, and in some cases more taxpayer favourable, approach is to analyze the legal rights of the parties under commercial law. For example, in the US, the rights of a manufacturer to a payment in support of a plant closing would be established by state law. In many cases these rights would be very limited, and thus a payment for such a move legally should be de minimis, regardless of the economics. Using commercial law analysis to support transfer pricing is an often overlooked, but potentially powerful tool.
Legal form and economic substance
Sometimes the inter-company contract does not match the economic substance of a transaction. This can happen for a variety of reasons, many related to the nature of related-party relations. Related-party transactions are sometimes not as scrutinized by counsel and other advisers as third-party transactions, so errors can occur. In addition, related-party contracts are not always updated to take into account changing business circumstances. The related parties may adjust their pricing without bothering to amend the contract.
Examples of errors may be an inter-company agreement that calls for a royalty based on net sales, that is, exclusive of returns, but the royalty is paid on gross sales. In another case, an asset, such as a loan with a convertible equity feature, may be sold inter-company, but the related equity interest is inadvertently not transferred at the time of the sale. If the equity interest subsequently becomes very valuable and is then sold to a third party, the wrong company could end up with a gain. One practical approach to the circumstance of an error is to investigate the legal rights of unrelated parties in the circumstances of the related parties. This approach builds on the basic fiction of the arm's-length standard, that is, related companies must act as if they were unrelated. Under local commercial law, unrelated companies will have certain rights that will depend on both legal form and the economic substance. An analysis of these rights may determine, for example, that after a certain period of time, the payor of the excessive royalty may lose the right to receive a refund under the contract. This analysis could then help build the case that the transfer price should not be adjusted downward. In the second example, the legal analysis might demonstrate that the purchaser of the loan has a strong claim to the equity interest that was not transferred. A payment in settlement of this claim could be a viable method of sorting out the proper amount of gain that should accrue to the two related parties. This type of approach, that is, analyzing the economic substance of a transaction as creating a legal claim that creates discounted value for transfer pricing purposes, was the approach ultimately adopted by the US Tax Court in DHL vs Commissioner, 1998 Tax Ct. Memo 461 (1998) and affirmed by the 9th Circuit, 28 F 3d 1210, 1219 (2002). In that case the Tax Court discounted the non-US rights in the DHL trade mark by 50% based on potential problems with DHL's ownership of the foreign trade marks.
In situations where the related parties have adjusted their transfer pricing behaviour, but neglected to update the inter-company agreement, a similar approach of looking to commercial law to support the course of dealing may be of assistance. It may be that where two parties have mutually agreed to amend a contract by clearly following an altered contractual relationship over a period of time, the terms of the contract will be treated as amended.
Obviously, it is best if the inter-company agreements comport with the economic substance and are followed by related parties. The suggestion to look to commercial law for support for transfer pricing that diverges from the inter-company agreement is one that would only be considered in a pinch. However, it is often helpful to have a variety of tools to address difficult transfer pricing problems, and commercial law remedies for unrelated parties in similar circumstances, on the right facts, can be one such practical tool.
Using legal agreements affirmatively
Finally, perhaps the best use of legal agreements is to use them to delineate affirmatively a set of rights and risks that supports unusual profit results. The key issue with this in a documentation context, however, is that these agreements generally must be entered into on an upfront basis to be most effective.
One common transfer pricing documentation problem is one or more distribution entities earning a relatively high return. It may be that the functional analysis has uncovered little information that would tend to support anything other than a routine return for the distributor, for example, an operating margin of the order of 2% to 4%. The actual returns may be double digit. In some cases, management might resist a transfer pricing adjustment to reduce the distribution return to the level of the benchmarks. In some cases this resistance might be due to fundamental business concerns such as the use of the distribution returns as a key performance indicator for compensation decisions.
One response could be to enter into legal agreements with the distribution entities, granting them defined territorial or other market rights in their jurisdiction. These agreements, akin to a franchise agreement, might provide the distributor a specified right to compensation if the franchise is terminated. This type of agreement may provide the necessary underpinning for the observed high distribution returns.
Putting an agreement in place providing the distributor with property rights to its market can raise a transition issue. Should the distributor entering into such an agreement pay a fee to obtain these rights? Conceptually there may be instances at arm's length that such a fee would be paid; however, in other cases a successful distributor may have market power to obtain special consideration. It may be helpful to manage this risk through the development of facts that underlie the distributor's success, for example, experienced management, good customer relationships and favourable locations. These same facts might also be identified as "marketing intangibles"; however, by supplementing these facts with a contractual right to the market, the marketing intangibles should be both enhanced and the high distribution return more clearly supported. The inter-company agreement might also state that it is merely formalizing in writing the parties' undocumented agreement over a number of years and as such does not create a new commercial arrangement.
A further use of market-defining inter-company agreements could be to differentiate one distributor from another. It may be that only one distributor has particularly high returns. To support this result it may be appropriate to grant market rights to this single distributor while providing clear termination clauses without compensation to others. Again, it will be important to provide a business case to support singling out one distributor for preferred treatment, but putting this preferred treatment down in writing should make the story more palatable to the tax authorities
Dependent on facts and circumstances
Not every transfer pricing issue can be solved without an adjustment, amended return, or competent authority intervention. However, there are a number of practical approaches to transfer pricing problems that on the right facts can sometimes avoid the administrative burden of changing the transfer price after the fact. In this article, we have tried to set out some of the approaches that have worked in particular circumstances. It is important, of course, to emphasize that the facts and circumstances of a particular transaction will in the end dictate what is a practical solution, and what is not.
The information contained herein is general in nature and based on authorities that are subject to change. Applicability to specific situations is to be determined through consultation with your tax adviser.
The views and opinions are those of the authors and do not necessarily represent the views and opinions of KPMG in the US.
Biographies |
|||||||||
Sean Foley is the national leader of KPMG's global transfer pricing services practice in the US and is based in Washington, DC. He focuses on advance pricing agreements (APA) and competent authority matters, transfer pricing risk management, and intangible migration strategies. Before joining KPMG, he was the director of the IRS APA Programme. Foley has an LLM in taxation (with distinction) and a JD (summa cum laude) from the Georgetown University Law Centre. He clerked for Justice Ruth Bader Ginsburg when she served on the US Court of Appeals and served as legislative director to Congressman Sander Levin, a member of the House Ways and Means Committee. Foley is an adjunct law professor at the Georgetown University Law Centre. He writes a monthly column for the International Tax Review on US international tax developments and is the current chairman of the American Bar Association's transfer pricing subcommittee on services.
Erik Skarstad is a senior manager in KPMG LLP's (US) economic and valuation services and Washington national tax practices. He has over eight years of transfer pricing experience with particular emphasis in the electronics, telecommunications and financial services industries. He has led various types of engagements including advance pricing agreements, Sarbanes-Oxley 404 transfer pricing reviews, transfer pricing audit defense cases, and global planning and documentation projects. His work has focused primarily on intangible property ownership and rights (that is, cost sharing and licensing), transfers of tangible property, and the inter-company provision of services. Skarstad holds an MBA in finance and economics from Columbia University in New York and a BA in economics from Carleton College in Northfield, Minnesota.
Anne Welsh is a managing director in the economic and valuation services practice of KPMG LLP (US). She is an economist with 10 years experience in transfer pricing engagements, with particular experience in the software, electronics and telecommunications industries. Her work has focused on transfers of rights to intangible property, compensation for high-value services, and transfers of tangible goods. She has led numerous global transfer pricing engagements that have included all aspects of strategic planning, compliance, advance pricing agreements, and audit assistance. She has authored a number of articles on transfer pricing and related issues. |
