Tackling Transfer Pricing

In the last few years, the question of what constitutes an equitable incidence of taxation has gained a very high profile within countries and internationally as well. One key aspect is the issue of what level of taxation of investment represents the appropriate balance between recognising an economic footprint or engagement in the taxing country that justifies a contribution to revenue, on the one hand, and acknowledging that too highly taxing such investments might deter the investors to the point that production is not carried out altogether.

Each country makes its own sovereign decisions on such matters, but there is increasing acceptance that governments need to work together more often and more deeply to ensure that such sovereign decisions are respected. This will lend support to the development of a country and its ability to gain fiscal and policy space. How countries cooperate to ensure appropriate taxation behaviour by the multinationals of one country in relation to another country will therefore be a test of how sincere this age of apparently greater international cooperation really is.

The subset of “transfer pricing” cases will be especially important in this context. In referring to transfer pricing, we are talking about the pricing arrangements for transactions between related parties—such as among companies that are members of a corporate or multinational enterprise (MNE) group. These arrangements can relate to prices for goods, services, loans or use of property. Transfer prices are inherently part of the way in which MNEs operate, because MNEs need to assess which parts of the group are profitable and which ones are not. By itself, the term transfer pricing refers simply to this process of allocating values to transactions, without necessarily implying any tax evasion or avoidance.

There is, however, a risk that the prices declared by an MNE will, either deliberately or not, fail to reflect its real economic engagement in a country. In other words, there is “mispricing.” One result of mispricing (whether it constitutes tax avoidance or evasion) is that losses might be reported in a country where, in fact, profits should have been reported.

The reason why transfer pricing represents such a test of international cooperation on tax matters is because, in many respects, developing countries, especially those with weak or small administrations, are especially disadvantaged by factors such as the complexity of transfer pricing concepts and their practical application, the “fuzziness” of some of the concepts involved, the frequent lack of data needed to evaluate the correctness of the profit allocation (and the cost of access to relevant data), the resource-hungry nature and long time frame of transfer pricing cases in terms of person-hours and costs, and the very large sums often involved.

Some General Challenges

A key transfer pricing challenge for both developing and developed countries is how to identify, for tax purposes, a fair price for intra-group transactions. This has to be done bearing in mind the real economic engagement of the multinational in a particular country, and therefore, the extent to which profits are truly being made there or elsewhere. If the transfer pricing does not reflect the true profits earned in that country, the country is unfairly deprived of revenue.

But that is not the only challenge. There is also a need to have internationally accepted “norms” in this area that are fair to countries involved in the transaction generally, and also to taxpayers. There is a risk that if one country adjusts the pricing of a transaction within a group so as to increase profits made and taxable in that country, another country will not decrease its calculation of the profits made in that country. In that case, some part of the profits may be taxed twice, with a possible discouragement or deterrence of investment that can itself hinder development.

There is a further aspect, which in Financing for Development terms, is referred to as a “systemic issue” and involves what could be termed “the rules of the game” for transfer pricing. Whether one agrees with them, or disagrees, current approaches to transfer pricing were elaborated by developed countries and premised on an access to technical resources—including skills, knowledge and data—that are beyond the reach of many developing countries. In the future, indeed, in the present, more attention needs to be paid to the role of developing countries as “norm makers,” rather than as just “norm takers.” Transfer pricing is an area where this issue arises with particular cogency.

Another related issue is whether the home countries of MNEs need to make greater efforts to ensure that the taxation behaviour of their residents—the MNEs—support development in other countries rather than acting as a brake on it. This is indeed a complex issue, but one that deserves greater discussion in an increasingly globalised world.

Controversial Issues

There are particular issues and controversies in the area of transfer pricing that reveal opportunities for international cooperation on these issues and also illuminate the potential role of the United Nations (UN) and other multilateral organisations in that process.

Arm’s-length Approach vs Global Formulary Apportionment: The tax treaties and domestic rules of countries seeking to address transfer pricing issues almost always reflect the approach that pricing of internal transactions within a multinational should be done on basis of the “arm’s-length principle.” This means comparing the transaction between the two related parties to one that occurred on market terms between companies that have no special relationship with each other (that is, that operate at arm’s-length).

The difficulties begin when this is put into practice, especially since many MNE transactions do not have direct market analogies. This can be because they involve intangibles (such as intellectual property rights) that have a unique, but difficult-to-determine value, or because they involve “set-off”¹ arrangements that would not occur between unrelated companies, or because they involve the sharing of costs that would only ever be shared within a related group.

Critics assert that this is a fatal flaw in the arm’s-length approach; that hypothesising an arm’s-length relationship is the antithesis of the synergistic relationship between different parts of an MNE and of the ways in which MNEs use their integrated, non arm’s-length nature to unlock value from their transactions. Those who take that view often argue that a different approach is required, based less on hypothesising an arm’s-length return in a non arm’s-length situation, and more about apportioning the overall profits of an MNE between different economic aspects of its operations on an agreed basis.

Critics of the arm’s-length approach often propose some form of “global formulary apportionment,” which would allocate the profits and losses of an MNE among the different jurisdictions where it operates. A formula would be used, based upon factors showing economic activity such as the proportion of sales (on the demand side) and assets or payrolls/employees (on the supply side) in that jurisdiction. There is room for considerable debate about which of these factors are most relevant, and therefore, about the proportionate weighting such factors should have in determining the areas of real economic activity.

This method is similar to the formulary approach taken by many states of the United States (US), where a Multistate Tax Compact reflects the value to both states (especially smaller ones) and taxpayers of a single approach, more or less uniformly applied by the Multistate Tax Commission.² There are also developments in the European Union (EU), with a proposed (and optional for taxpayers) Common Consolidated Corporate Tax Base that relies on a formulary apportionment method which evenly weights the three factors chosen, and which, therefore, ultimately favours the supply side.

It is worth noting that proponents of formulary apportionment not only emphasise the benefits to jurisdictions of a common approach that addresses profit shifting and helps taxpayers in filing tax returns relevant to several jurisdictions, but also point to the benefits for taxpayers of a consistent approach across several jurisdictions.

Taken in a broader perspective, the debate about the arm’s-length approach versus formulary apportionment reveals a clash of preferences between a more overarching, or “macro”, approach that proposes a generalised fairness, and a more “micro” approach that emphasises the “particularised” examination of transactions offering fairness to individual taxpayers on a case-by-case basis. In view of the resource constraints of many developing countries, there is a particular significance to this debate.

The process of applying the arm’s-length principle in actual practice is very complex. In order to find the arm’s-length price of a transaction, tax authorities and businesses make use of transfer pricing methods that are esoteric and convoluted. These represent ways of calculating the profit margins of transactions or entire enterprises, or of calculating a transfer price (or indeed, a range of prices) that qualifies as being at arm’s length. There are a variety of methods for doing this, each one more appropriate to a particular form of transaction or to cases where relevant data are not available. Access to data and analytical resources is critical to an effective arm’s-length approach.

For example, traditional transfer pricing methods used to find the arm’s-length price are applied, in practice, by establishing comparability between the conditions of a controlled transaction (between parts of an MNE) and of an uncontrolled transaction (between independent parties). They therefore rely, to a great extent, on the availability of data from comparable uncontrolled transactions (that is, sufficiently similar transactions between unrelated parties) in a comparable market.

However, in many developing countries, a market for such transactions does not exist or is not properly documented, and a market price can be difficult or even impossible to determine. There are usually not organised databases where one can find such comparables about activities in the local market. Some developing countries use data extracted from developed country databases, such as EU and US sources. This can be problematic, however, since access to such databases is costly, and market conditions may be so different that for the data to be useful, it must be adjusted for the developing country’s market conditions. Such adjustment is itself a resource-intensive process (and therefore costly, including in terms of other work forgone that might assist revenue collection). It often also leads to contested results.

Fixed Margins: The difficulty in finding suitable and available comparables when determining the arm’s-length price on a case-by-case basis, especially in developing countries, raises the contentious question of whether “fixed margins” should be introduced in transfer pricing regimes. The term fixed margins refers to cases where tax laws indicate what level of profits can be treated as arm’s-length in particular transactions, and consequently, what are treated as being, at least prima facie, mispriced.

Those arguing for a fixed margin approach consider that as long as fixed margins are transparent and “scientific” in their formulation and operation, and genuinely seek to determine prevailing market margins, they provide a more flexible, real world response to what constitutes an arm’s-length price. They take the view that such an approach gives certainty to both taxpayers and authorities, and means that neither is required to resort to resource-intensive case-by-case analyses and expensive, highly specific or tailored outside advice.

Proponents argue that the fixed margin approach is a simplified, but fair regime that may pave the way to a more traditional case-by-case regime. They say that this is especially the case if taxpayers bear the onus of showing that the fixed margins do not reflect arm’s-length pricing in a particular case. This equates to having a “ready-to-wear” approach to what constitutes correct transfer pricing, rather than requiring a “tailored” approach in each individual case, with the time and costs the latter entails.

Those opposing the fixed margins approach say it is intrinsically inconsistent with the arm’s-length principle, and would only work (if at all) in large, competitive markets where profits tend to equalise among different competitors. They argue that it is impossible, in practice, to keep margins so up to date as to reflect the reality of the limitless number of types of transactions in a varied set of market conditions. They consider that it would be especially difficult for smaller countries to keep the fixed margins under review and to update them as necessary, and to cover the many different types of industries and cases.

Once again, this debate illustrates a clash of preferences between, on the one hand, a more overarching approach (this time, in the form of fixed margins) that proposes a generalised fairness and, on the other, an approach favouring a more particularised examination of individual transactions that proposes fairness to taxpayers on a case-by-case basis. In this context, greater recognition is needed on the part of the developed countries that favour the approach that the complexity and individualised attention to each case found in the traditional arm’s-length approach often creates its own unfairness, even though it is designed to give a fair result in each individual case.

In fact, the more complex the exercise to hypothesise an arm’s-length price (in the face of the reality that multinationals are highly integrated in their operations, and even more distant in their internal dealings than in normal market conditions), the more disproportionate will be the impact on those least well equipped to deal with unnecessary complexity. They will be unable to bridge the gap between theory and reality and to fulfil the promise of the arm’s-length principle—the promise of fairness to all stakeholders in the system.

There is no doubt also that the complexity involved in case-by-case analysis has encouraged an industry of “tailored” advice. The role of advisers on tax issues is an important one, but critics have warned that this can produce a bias, and leads to even greater complexity and to unnecessarily individualised attention that is burdensome, and not just upon smaller and less well resourced taxpayers, but also on smaller and less well resourced countries.

It follows that developing countries, particularly those with the fewest resources, will disproportionately bear the cost in practice of what are, in theory, fair solutions tailored to the facts and circumstances of individual cases. Global formulary apportionment is not the solution, at least at this time. Even though it may be a part of it, or may be a framework for other approaches, the global formulary apportionment method is not yet sufficiently worked out in practice at the international level to displace the arm’s length approach.

UN Manual

Assistance in applying the arm’s-length principle can, however, play an important role in equalising the positions of developing countries with those of developed countries and MNEs. For this reason, the UN Tax Committee has prioritised—for its substantive involvement in transfer pricing work—the completion of the UN Practical Manual for Developing Countries (www.un.org/esa/ffd/tax) in 2012. The manual is currently being written to help make the arm’s-length approach as understandable and workable for developing countries as possible.

Efficient capacity building also involves encouraging countries to share experiences, both good and bad, with a view to building enduring networks and limiting the risk of costly policy and administrative dead ends. This South-South sharing of successful tax practices is something we have been working closely on with the Special Unit for South-South Cooperation of the UN Development Programme and others (www.s4tp.org). At the global level, effective capacity building requires that the UN, the Bretton Woods institutions, the OECD, non-governmental organisations and regional institutions, all working closely and effectively together, seek to avoid unnecessary duplication, but also remain true to their varying mandates, memberships and guiding philosophies. This will sometimes mean that developing countries have alternative views to choose from in the context of their situations and priorities, and that is a good thing.

Dispute Settlement

The need for clarity and predictability to promote investment and to avoid long and resource-intensive disputes raises questions regarding the potential role of dispute avoidance and resolution mechanisms such as (1) Mutual Agreement Procedure (MAP), (2) Arbitration, and (3) Advance Pricing Agreements (APAs) in the area of transfer pricing.

The MAP, under bilateral tax treaties, is designed to settle treaty-related questions and to provide a forum where residents of one of the involved states can seek reconsideration of actions they find to be contrary to a tax treaty. It does not compel a result, and MAP proceedings can become bogged down for years with no ultimate resolution.

Arbitration, on the other hand, provides for an alternative dispute settlement outside the courts when the parties have not been able to reach an agreement through the MAP.

Finally, Advance Pricing Agreements are agreements between a taxpayer and a tax authority (or more than one tax authorities) on an appropriate transfer pricing approach for some future set of transactions.

Despite the potential to provide greater clarity and predictability to taxpayers, and possibly assist smaller or weaker countries in dealing with complex transfer pricing issues, arbitration raises some potential concerns, especially for resource-strapped developing countries.

Arbitration is a potentially costly and complex procedure. For example, the costs of the judge, facilities, translators and other support services in arbitration—unlike in a court-based system—must be paid for by the parties. Also, extra-budgetary allocation of amounts for arbitration and the use of foreign currency reserves for payment, for example in dollars or euros, might be required for arbitration. This could tend to increase pressure on countries to settle outstanding cases merely for practical or budgetary reasons, regardless of the merits of the case. This problem becomes especially acute if multiple arbitrations are launched by different taxpayers on essentially the same facts—a situation similar to the many investment protection arbitrations faced by Argentina following its debt restructuring in response to the 2001 economic crisis. There is no apparent provision for consolidating multiple cases into a single arbitration.

Resource constraints can therefore challenge the practical application of such arbitration procedures in developing countries and may unduly weight the scales against them in the way those procedures work. If the cost were borne by the taxpayer, the concern is that this would encourage (or more accurately, be perceived as encouraging) arbitrators to make decisions unfavourable to the tax administration.

Another possible perception is that arbitrators in this area are likely to be predominantly from developed countries for some time and may lack understanding of developing countries’ realities and conditions. Developing countries may have had no experience of arbitration, or may have had a bad experience in an area such as investment treaty arbitrations.

In the long term, arbitration probably has a very important part to play in dealing with transfer pricing disputes, but in the short to medium term, it is an approach that needs to be taken with caution, and a full understanding of all the potential scenarios, including not only the best case scenarios, but also the worst case ones.

APAs also have great long-term potential, but agreeing upon such issues in advance requires a good understanding of the rules of the game and of potential loopholes that need to be closed off. There is also a concern frequently expressed, that in the early years of a transfer pricing regime, APAs can tie up key resources as an administration deals with taxpayers seeking the APAs (taxpayers likely to be broadly in compliance) rather than allocating those resources to address the greatest transfer pricing risks.

United Nations’ Role

The complexities in the area of transfer pricing, and the issues they raise from the perspective of developing countries, lead us to consider what special characteristics the UN, with its current limited tax cooperation resources, can bring to the table in meeting these challenges.

The first such characteristic is its universal membership and legitimacy. The challenges in this area, the growing recognition of the development aspect of tax systems, and even the emergence of multinationals that are “home-grown” to developing countries, all mean that efforts to reduce the complications of transfer pricing, in a way that is fair to all stakeholders, must have broader input and ownership than in the past. An approach of inclusive multilateralism is called for—one that draws upon the experience and perspectives of all stakeholders in international taxation issues, including business and non-governmental organisations.

The United Nations is increasingly looked to in this area as an impartial, responsible, representative and legitimate means to advance international tax cooperation with a view to development, including both fair returns to countries and a climate favourable to investment as part of a long-term partnership for development.

The United Nations can, in short, help safeguard that all with an interest in the “rules of game” will have a full seat at the table, to ensure that those rules meet the challenges and needs of development, and not only the needs of developed countries. In fulfilling this role, both procedures and outcomes must be fair to all those affected, and ultimately remain people-focused, because to be properly people-focused is to be development-focused, and vice versa.

Notes

1 “Set-off” arrangements are where payments may not be made by the purchaser to the seller in the usual direct way. Some other benefit may be given, such as a payment by the purchaser to another entity, who will then give an equivalent benefit to the seller. An example is where amounts are paid to one of the seller’s suppliers, to pay for those supplies.

2 See www.mtc.org. Canada, Germany and Switzerland also have some formulary apportionment experience at the sub-national level.