Transfer pricing: A challenge for developing countries
A lot of debate about tax and developing countries nowadays tends to focus on how to reduce revenue leakage through offshore tax havens. But there is another hot issue called transfer pricing which developing countries have to be mindful of, particularly if they want to avoid the risk of losing out on tax revenue from cross-border transactions carried out by multinational enterprises. How does it work?
A large proportion of world trade is accounted for by cross-border trade taking place within multinational enterprises, where branches or subsidiaries of the same multinational enterprise exchange goods or services. These transactions within the group are not exposed to the same market forces as transactions between independent enterprises. They are referred to as “controlled transactions”. If the prices of these transactions are artificially lowered or increased they may lead to taxable profits being shifted from one country to another.
Take a beverage company located, say, in Mexico, which has a subsidiary in France. Let’s assume that the French subsidiary pays royalties to the Mexican headquarters for the rights to sell its drinks. Taxes are owed in France based on the French subsidiary’s results. The higher the royalties paid by the French subsidiary, the lower the taxable profits in France. The French tax authorities will be satisfied if they see that the royalties paid by the French company to its headquarters in Mexico are not higher than those that would be paid to an independent enterprise for a similar transaction. But if the royalties are too high, there is a possibility that profits are being shifted out of France to reduce tax liabilities there. The “arm’s length principle” is used to address such issues.
Under the arm’s length principle, one compares the remuneration from crossborder controlled transactions within multinationals with the remuneration from transactions made between independent enterprises in similar circumstances. The arm’s length principle has become the international norm for allocating the tax bases of multinational enterprises among the countries where they operate. All OECD countries use this principle, as do an increasing number of non-OECD countries, such as Argentina, China, India, Russia, Singapore and South Africa.
The question is how to determine an arm’s length price for cross-border transactions within MNEs. The OECD lists as many as five methods for approximating arm’s length outcomes and these are explained in detail in the OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations (“OECD Transfer Pricing Guidelines”).
First, the Comparable Uncontrolled Price (CUP) method compares the price charged for property or services transferred in a controlled transaction within a multinational enterprise to the price charged for property or services in a comparable transaction on the open marketplace, or so-called “uncontrolled” transaction. In theory, this is the most direct way to apply the arm’s length principle and where it can be reliably applied, it is regarded as preferable to all other methods. The trouble is that in many industries, few uncontrolled transactions in the marketplace satisfy the comparability requirements needed for this method to work reliably. In effect, a seemingly minor difference in the property transferred in the controlled and uncontrolled transactions could materially affect the price and the reliability of the comparison between the two. For this reason, the CUP method is mostly used for trading commodities.
The resale price method is most useful for buy-and-sell operations. It starts from the price at which products purchased from an associated enterprise are resold in the market place. A resale price margin is applied to cover sale and other operating expenses, work out an appropriate profit given the functions performed by the reseller, its assets and its risks, and find an arm’s length price for the original property transfer between the associated enterprises. The cost plus method starts from the costs incurred by the supplier of property or services in a controlled transaction. An appropriate mark-up is then added to these costs, for the supplier to make an appropriate profit corresponding to its functions, assets and risks. The method is mostly used for contract work in manufacturing and back-office services.
The same reasoning can also be applied to net profit margins. The net margin taxpayers make from controlled transactions can be compared to the net margin earned by the same or other taxpayers in similar uncontrolled transactions in the marketplace. This is the so-called transactional net margin method or TNMM.
Finally, the division of profits in the controlled transaction can be compared to the one that independent enterprises would have expected to realise from similar transactions, in what is known as the profitsplit method. This profit method is mostly used in cases where both parties to the transaction contribute valuable intangibles.
Historically, OECD countries have tended to treat the profit methods–that is, both TNMM and the profit split method–as last resort for when the other traditional transaction methods do not work. In practice however, profit methods are now widely used in many countries. The transactional net margin method is actually proving easier to use than traditional transaction methods, being less sensitive to minor product differences for instance. The profit split method is also spreading, partly because of the growing importance of intangibles in today’s business transactions, in the e-world or in financial products, for instance, for which comparables can be scarce.
As a result, the OECD is now considering removing the “last resort” status of profit methods, and putting the emphasis on the selection of the most appropriate method for a particular case, and taking account of the respective strengths and weaknesses of the various methods. Public comments have been gathered and new guidance is being prepared (see www.oecd.org/ctp/tp/cpm).
In practice, one key difficulty in applying transfer pricing methods is to find open market transactions between independent enterprises that are comparable to the controlled transactions within a multinational enterprise. This is an issue for developed as well as developing countries, although it is magnified for developing ones due to the smaller size of their economies and smaller number of independent enterprises operating in their markets that can be looked to for comparisons.
One major OECD on-going project that developing countries need to follow is to reach consensus on how the arm’s length principle applies to business restructurings involving the cross-border redeployment of operations by a multinational enterprise. Such restructurings can have dramatic effects on the allocation of the profit (or loss) potential among the members of the multinational enterprise and affect the corporate income tax paid in each of the countries where the group operates. The arm’s length principle and the OECD Transfer Pricing Guidelines can help to lay out the appropriate questions and point to solutions.
For instance, what operations does the restructuring involve? Have valuable intangible assets, such as patents and trademarks, been moved? Who should bear the termination costs that may follow from such restructuring: the restructured entity itself, the parent company that made the decision to restructure, or the entity to which the operation is being relocated? How does the arm’s length principle apply in such cases? The OECD has been working with the business community and stakeholders since 2005 on this issue. A draft report was released for public comment in 2008 (see www.oecd.org/ctp/tp/br) and is expected to be finalised in 2010.
No method is perfect, and the inherent risks for disputes are not hard to spot. It is little wonder that a survey by Ernst and Young in 2007-08 found that 74% of parent and 81% of subsidiary respondents in MNEs believed that transfer pricing will be “very important” or “absolutely critical” for their organisations over the next two years.
Take our example of the Mexican beverage company and its French affiliate. In case of a dispute with the French authorities on the arm’s length price for the royalties, the risk for the firm is that the same amount of profits will be taxed twice–once in Mexico and once in France. The Mexican enterprise will want to avoid such double taxation. But Mexico and France will each rightly want their legitimate share of the firm’s profits. It is therefore critical to reach a consensus on the arm’s length price that the Mexican and French authorities, as well as the taxpayer, can use to arrive at an agreement.
Generally, under the OECD Model Tax Convention, double taxation can be resolved between the states concerned by following the Mutual Agreement Procedure (or MAP for short) which caters for an increasing number of complex international tax disputes, including tricky transfer pricing disputes (see Manual on Effective Mutual Agreement Procedures at www.oecd.org/ctp/memap). The OECD Model Tax Convention now includes compulsory, binding arbitration procedures for cases left unresolved after two years of Mutual Agreement Procedure.
Developing economies are keenly aware of the challenges posed by transfer pricing. Their goal is the same as for OECD countries: protecting their tax base while not hampering foreign direct investment and cross-border trade. The arm’s length principle can help them achieve that goal. The key is to tailor the legislative measures and administrative effort to the strategic needs and resources of each country. Applying the arm’s length principle can become complex and resource-intensive, though policymakers should bear in mind that most OECD countries started modestly and built their transfer pricing legislation and practices gradually over several years. Indeed, they are still in the process of improving them.
Tax authorities in developing countries who wish to implement transfer pricing legislation may focus on the most common types of transactions and sectors in their economy first, for instance the exploitation of natural resources, manufacturing, or service activities. Enforcement objectives should be realistic, given the available capacity, and compliance requirements made reasonable for taxpayers in light of the size of the crossborder trade. So-called “safe harbours” are sometimes used to simplify compliance by small taxpayers, or to deal with small and less complex transactions carried out by multinational enterprises.
Given the global and sometimes controversial nature of transfer pricing, it is important to develop internationally shared principles to help each country fight abusive transfers of profit abroad, while at the same time limiting the risk of double taxation of those profits. This is what the arm’s length principle is for. As more developing countries apply it, new lessons will be learned. This is a key step on the road to building a stronger, cleaner and fairer world economy.
Neighbour, John (2002), “Transfer pricing: Keeping it at arm’s length”, in OECD Observer N° 230, January 2002.
OECD (2009), Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations, Paris.
OECD (2007 update), Dispute Resolution: Country Mutual Agreement Procedure Statistics, available under “transfer pricing” at www.oecd.org/ctp
©OECD Observer No 276-277 December 2009-January 2010
Will the world economy brighten in 2014 compared with 2013?