Tax Transfer Pricing: Under the Arm’s Length and the Sale Country Principles

By Andrea Musselli

The book pays attention to the tax treatment of transfer pricing in a single perspective of analysis since the most important principles (the arm’s length -ALP- i.e. conditions that independent parties would share, and the sale country) are agreed worldwide. They must be applied in the same way regardless of the economic sector or industry. A country survey overlooks the most important issue of the fiscal problem, that is, the ability to project a unitary policy in compliance with the ALP (or with the sale country principle) and that should be audited by one sole (only theoretically) existing tax authority. The practical part and examples disclose how rules should be/have been applied, how legal proceedings can arise/arose regarding their application , how they were decided if litigation truly occurred, and finally the author’s motivated opinion with special focus on which is “the breaking point” of a specific analysis. The term “breaking point” is used to explain which can be the factual and/or the interpretative change that is able to modify such analysis and thus the solution. Extract from the preface of prof. Reuven Avi-Yonah: “this book is a must read for any serious student of the topic and an important contribution to understanding how the ALP is applied today as well as to how it should be applied. It is an invaluable contribution and should be read widely by both tax lawyers and accountants and by tax policy makers”.

• Language: English
• Publisher: IlSole24Ore
• Release date: Sep 23, 2022
• ISBN: 9791254831557

Book preview

Preface

Reuven Avi-Yonah

¹

Andrea Musselli’s outstanding book on Tax Transfer Pricing under the Arm’s Length and the Sale Country Principles is a must read for any serious student of the topic. The book is designed as a general introduction to the subject from a global perspective rather than a country-by-country survey. As the author states in his introduction:

"[The book] pays attention to the tax treatment of transfer pricing in a single perspective of analysis since the most important principle (the arm’s length principle-ALP, i.e. conditions that independent parties would share) is agreed worldwide. It must be applied in the same way even regardless of the economic sector or the specific industry while their regulatory approach should only be taken into account (as it’s logical), for example , in the risk analysis: business rules are applied first than fiscal rules.²

Even the concurrent principle that partly is going to replace the ALP for largest multinationals (the sharing of extra profit in the sale country) is internationally agreed.

In the author’s opinion, a country survey overlooks the most important issue of the fiscal problem, that is, the ability to project a unitary policy in compliance with the ALP (or with the sale country principle) and that should be audited by one sole (only theoretically) existing tax authority. Therefore, the objective of the entire narrative is at setting legal transfer prices while all other aspects are subordinated to that."

In what follows I will briefly describe the evolution of my own views on transfer pricing, because this would be a helpful background to understanding why this book is such an important contribution to the topic.

Transfer pricing was my first love in international taxation. I took the introductory class on international tax in the spring of 1988. That summer, the US Treasury published the White Paper that first introduced the Comparable Profit Method (CPM) and Profit Split methods. The White Paper was in response to a Congressional mandate in the 1986 tax reform act, because of dissatisfaction with the existing transfer pricing methods (comparable uncontrolled price, cost plus and resale price).

What struck me as strange in the White Paper was the insistence that the new methods were consistent with the ALP (CPM was called the Basic Arm’s-Length Return Method) when CPM relied on very loose comparables and profit split in many cases did not rely on comparables at all. It seemed to me that if there are no comparables, then any method including formulary apportionment could be called arm’s-length because then there was no way of proving that it was not what unrelated parties would have agreed to. This was the basis of my first significant tax article, The Rise and Fall of Arm’s Length, which was published in 1995 just after the new US regulations including the CPM and profit split were finalized.³

In that article and in many subsequent articles I advocated a move toward formulary apportionment. But this was rejected by both the US and the OECD. The International tax regime was committed to the ALP since the work of Mitchell Carroll in the 1930s, and the ALP was incorporated into Article 9 of every tax treaty. In addition, the US and OECD (but not the UN) deleted Article 7(4) from their models because it created an opening to formulary methods. 

By the 2000s things seemed to be at an impasse, even though there was evidence that the ALP was still not working well for large multinationals (the IRS kept losing cases; it did not win a transfer pricing case from 1980 until its recent victory in Coca Cola). In that context Kim Clausing (who is now at the US Treasury), Mike Durst (formerly director of the APA program at the US Treasury) and I proposed a compromise: apply the ALP to routine returns where there could be comparables but allocate any residual to the market jurisdiction because that was the least subject to tax competition.⁴

For a long time, the OECD resisted this suggestion, and explicitly rejected it at the beginning of the BEPS project, which did not significantly modify the ALP. But eventually it came around, and in Pillar One of BEPS 2.0 it adopted it for amount A of the largest multinationals (which are the subject of most litigated transfer pricing disputes). 

Nevertheless, as Musselli points out, the ALP still applies to amount B as well as to routine returns, and to all companies that are smaller than the largest ones. In that context this book is an important contribution to understanding how the ALP is applied today as well

as to how it should be applied. It is an invaluable contribution and should be read widely by both tax lawyers and accountants and by tax policy makers. 

---

¹ Irwin I. Cohn Professor of Law, the University of Michigan.

² The financial industry is a clear example of what affirmed in the text; extract from OECD Transfer Pricing Guidelines 2022, note at paragraph (D.1.2.1).

The regulatory approach to risk allocation for regulated entities should be taken into account and reference made as appropriate to the transfer pricing guidance specific to financial services businesses in the Report on the Attribution of Profits to Permanent Establishments (OECD, 2010).

³ Avi-Yonah, The Rise and Fall of Arm’s Length: A Study in the Evolution of U.S. International Taxation, 15 Virginia Tax Rev. 89 (1995).

⁴ Avi-Yonah, Clausing and Durst, Allocating Business Profits for Tax Purposes: A Proposal to Adopt a Formulary Profit Split, 9 Fla. Tax Rev. 497 (2009).

Introduction

Andrea Musselli

¹

A few words on what the book is and what it is not since it is important to explain the author’s method of work in advance.

The focus is on the application of the transfer pricing principles for income tax and not only a survey of current national dispositions, their interpretation, or case law. It pays attention to the tax treatment of transfer pricing in a single perspective of analysis since the most important principle (the arm’s length principle-ALP, i.e. conditions that independent parties would share) is agreed worldwide. The principle must be applied in the same way regardless of the economic sector or the specific industry while the regulatory approach should only be taken into account, for example, in the risk analysis: business rules about a specific economic sector are applied first. ² Even the concurrent principle, that partly is going to replace the ALP for largest multinationals (the sharing of extra profit in the sale country) is internationally agreed.

In the author’s opinion, a country survey overlooks the most important issue of the fiscal problem, that is, the ability to project a unitary policy in compliance with the ALP (or with the sale country principle) and that should be audited by one sole (only theoretically) existing tax authority.³ Therefore, the objective of the entire narrative is at setting arm’s length prices (or prices in compliance with country of sale principle) while all other aspects are subordinated to that.

The book truly (and primarily) outlines economics (Chapters 1 and 7) and in the following also US statutes, regulations, and case law as well as the OECD transfer pricing guidelines- OECD TPG- (Chapters 2,3,4,5,6) but only because both legal systems have the greatest historical experience on transfer pricing. Even a small number of dispositions and case law of other countries (for instance, China, France, Germany, India, Italy, Switzerland et cetera) are considered when they allow analysing some aspects of the ALP.

The practical application of principles, which is the most relevant part of the work (Chapters 8 , 9 and 10), deals with many cases that can be substantial in amount, involving astonishing issues, and regard facts, analyses, and conclusions that even when inspired by real events and true case law, are not country-based since they are paradigmatic; facts are stated to establish the rule in the ALP (or in a concurrent principle as the sale country) compliance.

The examples of the book are based on circumstances that each agent involved in actual transfer pricing (managers of firms, advisors and lawyers, tax officials, et cetera) or in theoretical studies (professors, students, et cetera) can recognize as those that he meets in his practice or studies and regardless of the country of origin.

In that way, the national perspective is no longer relevant: analysis and conclusions are due to specific facts for which the former follow the latter irrespective of the involved country. The discussion on examples may affect any jurisdiction that enforces the ALP and any firm and tax administration that are requested to apply it.

Another aspect tends to give uniformity to the enforceable rules. The architrave of the ALP legal system is the best method (or most appropriate method) rule. The fiscal law of (almost all) countries enforces codified (but also non codified) methods to establish values; however in every concrete case, the legal and appropriate method is not set in advance because it will be the one that most reliably sets the ALP in the true circumstances in which it is going to be actually applied. Methods are only means to an end, i.e. complying with the ALP.

This affords opportunity for a broad use of economics for sourcing rational rules that are behaviors and values modelled on those that would have been agreed upon by independent parties who truly are in conflict of interest. According to that, the book first highlights concepts that underlie or should underlie fiscal rules and that arise from economics just in order to emphasize their main importance.

The practical part and examples disclose how rules should be/have been applied, how legal proceedings can arise/arose regarding their application with special focus on which is the breaking point of a specific analysis, how they were decided if litigation truly occurred, and finally the author’s motivated opinion. The term breaking point of a transfer pricing analysis is used to explain which can be the factual and/or the interpretative change that is able to modify such analysis and thus the solution.

In the logical order, the facts found the analysis and the appropriate solution should follow: what is searched for is a rule that (starting from specific supposed circumstances) is modelled on the behaviors of independent parties who negotiate in conflict of interest.

Furthermore attention is paid to incentives that are provided by new rules to firms: the incentivized behaviors of firms can completely change the forecasted effects of said rules.

In the examples, special focus is on the importance of performed functions (OECD TPG paragraph 6.56) in order to allocate the intangibles’ return and the bulk of profits but it is noticed that such a criterium can create significant uncertainty to what is specifically requested in order to comply with the law. Allowing the gain of residual business claims based on subjective assessments of the importance of functions, their special significance, and their nature in any specific case depending on facts and circumstances could give rise to never-ending discussions. Uncertainty is not only related to marginal cases of subjects to be remunerated with independent comparable returns but especially affects the main driver for attributing the role of residual claimant of the entire group’s profits or losses.

The examples in the book begin from the simple use of comparable prices, but the author makes it clear that comparable independent prices are often an exception for intragroup transactions. Therefore, the attention must shift to the group party performing the less complex functions and in comparison with its activity and similar activity performed by an independent firm. There are two main reasons that often make inappropriate the comparison of prices:

a) A hold-up analysis clearly explains why, in certain circumstances, independent transactions that could be compared to group transactions are not found; they simply do not exist. In specific investment situations when it is difficult to achieve consensus for a complete contract, the investor delays his project, and no transaction occurs between independent parties (see Chapter 1 and the paragraphs on Economic of contract).

b) In capitalist economies, each good or service sold by firms is generally different from those of other firms since they intend to innovate their productions and thereby achieve a quasi monopolistic position and extra profits. For this reason, even intermediate goods are different (see Chapters 7 and 8).

The author emphasizes that the conclusions of examples are based on declared specific facts and not on general statements since only the former can be proved as being consistent or inconsistent with the ALP, by submitting it to a Popperian process of possible empirical falsification. General statements cannot be deemed consistent or inconsistent with the ALP; facts found behaviors of independent parties in mentioned circumstances and, from such behaviors, it is possible to derive general statements that are also ALP compliant.

Under this line of reasoning, the author discusses the important 2015 BEPS⁴ OECD statement that only operational control of activity that sources intangible properties deserves the allocation of related revenues. The change should prevent abuse by groups since many (most industrialized) countries are concerned about cash boxes without substance, which are highly capitalized, low-taxed companies. They can become the owners of intangibles and substantial return claimants through the funding of intangibles development performed by other group companies.

The new statement about control over intangible activity directly derives from example 17 of the OECD TPG on intangibles, however, it is not possible to submit the general statement to an analytical test of consistency or inconsistency with the ALP; instead, it can be done with the analysis included in example 17.

The author uses special concepts of economic of contracts.⁵

A party funding an intangible development who is not able to monitor the operational activity of the development provider would only link the latter’s remuneration to the success or failure of the research project (higher remuneration to the successful result and lower to the failed). However, he (the funding party) would remain the residual claimant of gains or losses of the intangible project after having paid a market remuneration to the development provider.

The author’s conclusion is that example 17 and the derived general statement of control over intangible functions are measures beyond the ALP, despite what the OECD affirm.⁶

Moreover, the practical component of the narrative (Chapters 8 and 9) discusses the most important and sometimes controversial cases that transfer pricing agents (private or public) truly meet during their activity. They are, for instance:

• identification of whether there is/are and which is/are the tested party/ies,

• identification of one sole tested party in lieu of considering both parties as intangible owners/residual claimants,

• possibility of extreme results of a tested party via contractual risk assumption,

• valuation of intangibles at the moment of their transfer, and

• criteria for the division of profits generated from extraordinary cooperation of both parties (parameters of profit split).

Chapter 10 instead focuses on a summary of the new principle of the partial sharing of group extra profit in the sales country (OECD Pillar One, recently agreed for the largest multinationals) coupled with the minimal tax rate of 15% wherever they operate (OECD Pillar Two, agreed for a larger set of firms than Pillar One, but still including great groups). The incentives that such rules provide to firms are analyzed and, last but not least, even the circumstance that the form of group firms (as opposed to independent firms) is incentivized for the distribution activity when a minimal distribution reward is granted for associated companies and such a reward disallows the reallocation of profit in the sale country.

Proposed changes to rules confirm that what moves transfer pricing legislation at national and international levels is the interest to safeguard the tax base of most developed (in many cases high taxation) countries.

The author does not want to ethically judge the interests of highly developed states to protect their tax base: the latter motivates the pressure on the OECD and countries organized in the BEPS framework with the intent to maintain a fair level of taxes from multinationals without exacerbating their public finance sustainability.

OECD always produced a valuable effort in order to include the governments of the majority of countries in the work of projecting new rules (even the less developed or low taxation countries) but it’s self evident which interest comes first and which comes next.

The world tension due to European war in 2022 has already delayed the entry to force of the new Pillars enacted by the OECD and some even doubt that they will ever be applied. Perhaps they will have a different timing.

The 2021 Pillar One agreement about the principle of sale country could be applied by States included in a sort of regional list for that cooperation is still possible, since they adhere to the same military block or have not conflicting political or economic interests.⁷

In June 2022 some commentators argued that OECD is resetting the Pillar One timeline and early intends to release a new consolidated consultation document for comments by stakeholders.

On Pillar Two the 2021 agreement on the 15% minimum tax enforced:

• (i) an Income Inclusion Rule (IIR) that imposes a top-up tax on a parent company in respect of the low taxed income of a group entity; the introduction was planned to be in 2023 but the implementation seems shifted to 2024.

• (ii) an Undertaxed Payment Rule (UTPR) that denies deductions or requires an equivalent adjustment (i.e. withholding tax) to the extent that the low tax income of a group entity is not subject to tax under an IIR; the introduction was planned to be in 2024 but the implementation seems shifted to 2025.⁸

In December 2021 a domestic minimum top-up tax of 15% was additionally planned: in this case Jurisdictions have a significant incentive for the introduction because in the absence of such a tax, profits of multinationals could be taxed outside them.

Pillar Two does not rely on all countries agreeing to a minimum tax, but simply a sufficient number of countries agreeing to do so.⁹

Coming back to present times the author would also like to examine the phenomenon that is currently ¹⁰ existing in the application of the ALP about its worldwide jeopardized (not common) interpretation.

1. The difference between the US legal system, the OECD TPG, and the legal systems of many other countries must be not understood as the result of various enforceable norms as they are primarily the same and are based on the ALP. The difference can be and is due to inconsistent (each other) behaviors of firms and especially of tax authorities of countries that actually and inconsistently apply the same or similar norms. National tax authorities (and not the private parties that are associated with a group) have a conflicting interest in that the increase of the tax base in one country is the decrease of the same in another one.¹¹

2. Since 2015, the OECD TPG has founded the entitlement to extra profits on the value creating activities of the company’s officers in performing the important functions related to the intangibles (development, enhancement, maintenance, protection, exploitation-DEMPE). The renewed standard has the stated objective of preventing abuse and does not recognize intangible ownership that is foreign based, maybe allocated to companies residing in Tax havens with no other activity than the sole funding of the intangible development. However, the disposition will mainly be interpreted through the concept of value creation. This consequently has a side effect of introducing in transfer pricing affairs the powerful, evocative but historically most debated welfare economics concept of value. Philosophers and economists linked value to: i) prices that are freely observable in market economies (the true notion of the ALP), ii) socially necessary labour time that is not observable in market economies (Marxian perspective), and iii) a specific distribution of income between wages and profits (Sraffian perspective).

3. The current reference to the value creating activities and ALP future replacement for greatest groups (OECD Pillar One) lead to delegitimize the economic and market configuration of the principle as a fair profit allocation rule. What was previously stated is valid even if the ALP (in the market configuration) remains the legal principle in force for all past, present and approximately 98% of future transactions. Therefore, uncertainty on the ALP’s application is currently relevant but could still increase.

Therefore the author aims at answering to the question why ALP application is so controversial, in addition to the obvious reason of the contrasting interest of firms but also of (at the least) the two states that audit the group transactions.

Legal rules are primarily founded on a comparison between a group transaction with an independent transaction, setting conditions of the former based on the latter. A comparability analysis can lead to the idea that transfer pricing under the ALP is an objective process, but the book proves that it is a dangerous misunderstanding since it is not able to explain the frequent litigation between companies and tax authorities.

a) First, even when a transfer price is set by comparable independent transactions, a critical assessment typically aims at identifying the party performing the most simple activity of the group companies, known as the tested party. It can be compared to the independent party that constitutes a possible alternative to the group company and whose prices and profits are ALP-compliant by definition since they represent market conditions. However, under said assessment, the most important role in a group of companies will be disclosed, i.e. the counterpart of the tested party who is the residual claimant of group (or transaction) results. This entity retains the possible extra profits or losses of the group (or transaction) given that it achieves a result that is what remains from group profits after tested parties have been remunerated as if they were comparable independent companies.

b) A second critical conclusion is that the concept of comparability is almost ineffective for transfers of highly valuable (or so potentially) intangibles and similar operations. Intangibles are strictly related to innovation: in a perfect competition equilibrium, companies gain no (or low) competitive profit. They are able to achieve extra profits only by innovating in technologies to reduce the costs of existing outcomes or in differentiating them, having new products or services with more valuable features for consumers. However, innovations are unique and not comparable because, if they are imitated by competitors, they no longer allow extra profits. Intangibles are (often or almost always) not comparable with each other and must be valued via the profits projected as arising from their future use but considering the risk of development. Information useful for valuation is information that is available at the moment of sale. It is not ALP compliant to value intangibles via actual profits that are realized and disclosed years after the transfer. Tax authorities are less informed about business information than taxpayers, and they mistrust managers’ projections. Under the said framework, international and national legislators have enacted many specific and detailed legal rules but at the expense of losing the focus on the economic principles underlying them.

The Amazon landmark case was decided by a US Court of Appeal ruling in 2019 and can be used to define the border of the ALP via an intangible transfer that occurred at the end of 2004. Which intangibles allowed the Amazon group to obtain the extra profits achieved after 2004, i.e. those that were current at the end of 2004, the ones developed in the following years, or both of them?

Transfer pricing largely remains a matter of judgment and this conclusion is paradoxical when one thinks that legal rules as OECD TPG¹² can be contained in a book that is much larger than the one being presented here.

The author opines that problems source because it is not completely clear (it’s not fully shared) which economic model (of the ALP) underlies the legal rules, while he discloses his unitary key of interpretation. To limit uncertainty the interpreter must relate the allocation of the global extra profit (or better of global result including losses) to the entity that:

a) Invested in innovating the business through risky projects and activities; innovative expenses may lead to success or failure (they can become irretrievable sunk costs) and are thus risky investments by definition.

b) Invested at the moment that it was not yet in a dominant position in the industry/market.

Under the interpretative key and appropriate contracts set in advance of transactions to furtherly limit uncertainty , many transfer pricing policies can be classified as being in breach of or in compliance with the law; a motivated opinion can even be proposed on what the ALP compliant valuation is among the 11 different valuations ranging from 0.3 to 5.6 USD billion that this book discusses in the Amazon case and the 2004 buy-in paid by the Luxembourg to the US entity.

The key is based on facts that are more objective than a simple discounted cash flow –DCF- valuation which is indeed mainly founded on subjective profit forecasts. These are done by company managers who are more informed about business conditions than tax officers while the author uses more objective facts (that tax authorities, just as any other, can also audit) to reach a conclusion.

The arguments of the author are not alternative to subjective valuations but can additionally support them in validating or rejecting their conclusions and can support the choice of applicational parameters of the financial methodology (discount rates, etc.) that is selected to value an asset.

This is the ontological nature of the ALP in market economies and if something else is desired, it is not the association of the principle with the concept of value creation that can change the legal framework.

A common ALP interpretation can only be achieved when disputes between firms and administrations and between administrations (of different countries) are solved by international third-party judges (or arbitrators).

However almost thirty years have passed since the first enforcement of the renewed 1994 US Regulations and the 1995 OECD TPG, but mentioned judges do not yet exist.¹³ Furthermore, it is difficult to forecast more cooperation between states in a delicate matter that can impact their national budgets when there is currently (June 2022) an ongoing war in Europe and a cold war climate between most powerful countries.

What can interested parties do to improve the current uncertainty in fiscal law application? They can propose examples based on specific facts that anyone can recognize such as the facts that currently exist for setting transfer prices.

National aspects should be forgotten since, when facts are duly extracted from a case, the country of origin is no longer relevant, and the proposed rule (even when reached after a lively and heated debate) can regard whichever jurisdiction is enforcing the ALP.

In every location where the ALP is the current rule (which is practically all over the world), it is always proclaimed as the condition upon which independent parties would agree, but it is often applied in a widely dissimilar manner.¹⁴ However, it is not necessary to have a country survey to imagine how the principle is applied in a specific country. It is sufficient to understand that, without a clear enforceable interpretation, each country will tend (not always but not rarely) to overestimate the factors that allow it to claim a greater part of the profit of the entire activity. Firms will also conversely tend (with the same frequency as that mentioned for countries) to overestimate the factors that allow them to globally pay less in taxes or defend from possible arbitrariness of some income adjustments. The normative answer to uncertainty has historically been the production of an increasing number of rules.¹⁵ Transfer pricing is becoming a matter for specialists, and the truth is that it is becoming very complicated, contradictory and subjective. In the author’s (humble) opinion, this is not an appropriate course of action; there should be a clear reference to a model of ALP application (such as the model that is presented in Chapter 7 of this book). Economic principles should underlie legal rules, and thus there should be consistency between them.

Finally, the work contains all ancillary aspects of the legal matter as an handbook but in a logical sequence even if it is focused on setting transfer prices. It also addresses the primary concepts of governments’ auditing power and companies’ compliance burdens, principles enforceable in different industries (practically in the financial – or other regulated-sector), anti-abuse laws and criminal aspects; the author summarizes rules that are enforced by national laws and focuses on their general principles.

However, the mentioned aspects are only drafted to achieve the objective of a work of about 300 pages (excluding notes). Therefore, it is erroneous to think that the book is just a beginner’s guide or the bare bones of the matter. On the contrary, a concise account of ancillary aspects¹⁶ in respect to the setting of prices allows dealing with the core of the problem in a logical and coherent manner and avoids a complicated and sometimes inconsistent narrative or a mere paste and copy of different countries’ legislation that readers can find by themselves.¹⁷

The OECD has published a very useful section on national legislation on its website with more than sixty transfer pricing country profiles (the section has been recently expanded to rules on hard to value intangibles).¹⁸

The book is rigorous on concepts and develops an interdisciplinary (economic and legal) unitary analysis but with elementary math¹⁹ and always in an applicational perspective for solving true cases or invented examples. This is because theory, even the most sophisticated, must be applied to and is for ruling practical aspects of possible or existing cases.

The author concludes with an issue that is not usually expressed for multinational businesses but that affects their economics in a more general manner than taxation and can even have an indirect relevance on establishing ALP compliant prices.

Extra profits can be achieved even under contestable (competitive) market conditions. However their persistence during long periods may be an indication of a dominant position of some firms.²⁰ It seems that legislators and politicians in many countries are willing to resolve the problem of multinational businesses by only claiming the power to tax their profits. How companies manage their activity and the explanation for the achievement of extra profits is first chronologically, logically, and in terms of importance more than how extra profits are taxed and how tax revenues are distributed among countries.

The author’s hope is that readers will enjoy a short book that gets to the core of tax transfer pricing problems via ALP (or country of sale principle). Problems do not stay in theoretical aspects of principles but in their living application.

He would appreciate receiving comments and motivated critiques in a such way that they will possibly enrich the content and proposed solution of presented cases, when this will be deemed necessary.²¹

Special thanks

The author would like to extend his gratitude to all of the individuals who had the patience to discuss ideas, analyses, and opinions. They are economists, law professors, lawyers, tax advisers and accountants, tax inspectors, OECD officials during public discussions at the Paris Headquarters, and executives of companies. Discussions with all of them allowed to share experiences about the actual application of ALP.

Any mistakes and the opinions expressed in the book are entirely the author’s own.

Above all, thanks to Professor Reuven Avi-Yonah with whom the author exchanged many opinions and who shared his authoritative US and international experience and prepared a short preface of the book.

Secondly, the author would expressly like to thank many economists:

• recently, Professor Oliver Hart who found it interesting to apply economics of contracts to the analysis of the ALP and suggested discussing topics with Professor Massimiliano Vatiero (thanks to the latter, too).

• Professors Luciano Olivotto, former Head of Business Administration Studies at Ca’ Foscari, Venice University, for capital budgeting and financial economics, Giampaolo Arachi, at Econpubblica Bocconi University, Vincenzo Scoppa , Economics of Contracts at Calabria University and Dr. Deloris Wright.

Thanks to lawyer Samuel Maruca, Dr. Rijkele Betten, professors, Maarten Floris de Wilde, Mindy Herzfeld, Pasquale Pistone, Wolfram Richter, and Oddleif Torvik. The author also appreciates feedback from Sunny K. Bilaney, Dr Marta Pankiv, and several others. Finally, thanks to those who discussed issues about the previous book by Andrea Musselli with co-author, Alberto Musselli, Transfer Pricing (eight editions over more than 15 years) but only in Italian, by IlSole24Ore (first edition in 1997).

Thanks to Mr Jean Marie Del Bo, Mr Roberto Esposito and Mrs Maria Carla De Cesari, and Mr Claudio Pagliara, Il Sole 24 Ore group, historical editor of the authoritative financial newspaper IlSole24Ore (in Italian) and of professional books and database.

Last but not least is Mrs Jenny Hill for English language.

---

¹ Andrea Musselli is a master at Economics and Commerce (Venice Cà Foscari University, 1985) and a master at Law (Milan University, 1997) and has been/is dottore commercialista (he thinks it can be indicated as economic, corporate and tax law advisor), chartered accountant, and tribunal advisor for more than 30 years in Milan.

² It’s a wrong perspective to think that ALP application should depend on the involved economic sector: what is (and must be) relevant is how the group segments the entire activity into different functions (production, distribution et cetera) and how segmented functions are performed by associated companies, concerning to assumed risks and used assets.

The financial industry is an important example of that, see OECD TPG 2022, note (1) at paragraph (D.1.2.1).

"The regulatory approach to risk allocation for regulated entities should be taken into account and reference made as appropriate to the transfer pricing guidance specific to financial services businesses in the Report on the Attribution of Profits to Permanent Establishments (OECD, 2010)".

³ In reality, the sole tax authority does not exist because at least two national authorities are always involved in an audit, and they have conflicting interests.

⁴ BEPS means Base Erosion and Profit Shifting.

⁵ The economics of contracts (contract theory) is an important branch of studies focused on well-designed agreements that provide incentives for conflicting parties to exploit the prospective gains sourcing from cooperation.

⁶ Even if the author proved that the OECD TPG has no theoretical foundation on the issue, he accepts that the company funding the investment project (S) must have a level of substance in order to not be considered a pure cash box.

⁷ OECD will detail which jurisdiction will surrender taxing rights in the reallocation of those rights on highly profitable multinationals, see D. Moroses , OECD Prepared To Tackle Taxing Rights, US Official Says, at https://www.law360.com/ , accessed June 27, 2022.

⁸ About the delay for IIR and UTPR see, for instance, the discussions in the EU (April 5, 2022 Ecofin agreement by 27 EU countries, except Poland).

⁹ https://news.bloombergtax.com/daily-tax-report-international/oecds-pillar-one-and-pillar-two-a-question-of-timing

¹⁰ The phenomenon started a long time ago but recently increased a lot.

¹¹ When the same principle is applied in the two States and the case does not end with double taxation at group level.

¹² 2022 OECD TPG have more than 650 pages (including annexes).

¹³ There are few exceptions due to some International Treaties that will be dealt with especially in chapter 11.

¹⁴ The author is quite pessimistic on the future possibility to reach a common interpretation of the principle.

¹⁵ Even some editorial projects on transfer pricing may currently reach a number of thousands of pages and base their narrative on sectorial or country based issues.

¹⁶ For instance, Chapter 11 on administrative aspects outlines only the main principles regarding exchange of information, documentation of firms, governments’ power to adjust prices, and possible remedies that are enforced by each country legislation.

¹⁷ This is just and also the reason why the book has been restricted as much as possible, and is edited by one single author who attempts to provide a sole logical vision of the matter.

¹⁸ https://www.oecd.org/tax/transfer-pricing/transfer-pricing-country-profiles.htm

¹⁹ Requiring basic knowledge of the probability theory but whose concepts will be newly drafted during the narrative.

²⁰ The use of software codes - the prerequisite for generating sufficient revenue to cov "the high costs of development - is constrained by legal means. At the same time, economies of scale and network externalities prevent effective competition; there is a tendency toward natural monopoly. With a market share now exceeding 90 percent, Google’s search engine provides a striking example of that phenomenon". So W Richter, Reforming International Taxation: A Critique of the OECD Plans And a Counterproposal. Tax Notes International, Volume 102, June 28, 2021.

²¹ The book is updated with information available at June 2022.

Chapter 1

Economics of transfer pricing; contract theory and the arm’s length principle (ALP)

1.1 What is Transfer pricing

Transfer pricing is a practice that represents the price at which divisions of a company or different companies controlled by a sole economic interest in a group structure transact with each other. The values of intrafirm transactions have long been studied beginning with Harry Sidgwick’s Principles of Political Economy (1883), but the actual focus on them dates only from World War II.

There are multiple methods for dealing with transfer pricing. The first in neoclassical studies is from a microeconomic viewpoint, however, more recently, especially business administration scholars have delved into other schemes that range beyond the abstract marginalist perspective.

Concepts deriving from studies of new institutional economics and firm theory have also been examined in regard to the reasons why groups of companies arise and why it is more efficient for transactions to occur between divisions of a single firm or companies associated with a group as opposed to independent divisions/companies. The approach of the contract theory is drafted considering the games that result from the negotiation of two economic agents when they must cooperate in order to gain a profit that will be generated from their combined efforts. The analysis of the games is beneficial because, for a long time, the most important fiscal standard of transfer pricing has been the ‘arm’s length principle (ALP), i.e. conditions that would be agreed upon by independent parties in similar circumstances of intra group transactions.

There is necessary discernment in this context because there are many types of political and business economics as well as a myriad of accounting practices, therefore, the substantial amount of economic literature on transfer pricing will not be addressed but only the concepts that are useful for taxation issues.

1.2 Neoclassical microeconomics

In theoretical economics, transfer pricing decisions arise when one division of a firm (or a company associated with a group) sells goods (or services) to another division (or another associated company) within the same firm (or group of companies).

In Hirshleifer’s simple model¹ (1956) that was developed through a marginalist scheme, there are two divisions into a firm and a head office. The latter concerns with the firm and is able to mandate divisional decisions; the seller division produces an intermediate good and supplies it to the buyer division that processes or only sells the intermediate/completed item in the final product market.²

Rational managers of the firm’s divisions act in order to maximize their proportional distribution of profit without considering what occurs to the firm as a whole. The divisions are independent of each other, however, the firm’s best interest is served when the head office selects a transfer price that maximizes the total firm profits that are generated by the aggregate efforts of the firm’s individual units. This is a normative perspective because its objective is to set the right or optimal (maximizing profit) transfer price.

Ultimately, if no market exists for the intermediate product, the appropriate transfer price will equal the marginal cost of the intermediate item itself. Its correct transfer price will equal its market price if there is a competitive market for it.

1.2.1 No competitive market for intermediate product

The head office (able to mandate decisions to the selling and purchasing divisions) will make the sum of marginal costs of both of the divisions equal to the price of the final product³.

When both divisions are provided with the discretion to negotiate, the same solution is obtained. The selling division will request a transfer price (for the intermediate product) equal to its marginal production costs; the buying division will ask for a transfer price in such a way that the intermediate product cost plus the distribution marginal cost equals the price of final product. The agreement and the related transfer price between the divisions is reached when these are achieved.⁴

The transfer price is an optimal price which allows maximum profits for the firm as a whole, and it is also fair since it is reached by agreement between the parties even when both divisions are free to negotiate.

1.2.2 Competitive market for the intermediate product

When a market price for the intermediate product exists, the buying division no longer finds it convenient to purchase the intermediate product at the marginal cost of the selling division since this marginal cost is higher than the market price of the intermediate. Additionally, if the buying division requires a greater quantity of intermediate products to make the marginal production costs⁵ plus marginal distribution costs equal to the price of the final product, it can acquire this on the intermediate market.

This is still an optimal and fair solution. It is optimal because the total profit of the firm is maximum: the marginal production costs (purchases by the buying division from the selling division and possible purchases from the market) plus the marginal distribution costs equal the price of the final product. The solution is also fair because, even when the head office affords opportunities to divisions to negotiate, the buying division will continue to buy from the selling division but only because the marginal cost of the intermediate produced by the selling division is lower than its market price. For transfer prices higher than the market prices of the intermediate product, the buying division will no longer find it convenient to buy from the selling division; therefore, it will do so from the market. The selling division will find it convenient to sell intermediate products to the buying division as the marginal cost of production is lower than its market price.

1.2.3 Heritage and deficiency of the neoclassical pattern

The most important concepts developed under marginalist studies are ‘optimality’ and ‘fairness’ (neutrality).

• Optimal transfer prices allow the firm as a whole to achieve maximum efficiency and profits.

• Fair (or neutral) transfer prices incentivize profit-seeking managers of either the buying or the selling division to take decisions consistent with the target of optimal profit for the entire firm. Fairness (neutrality) is thus the correct measurement of performance of divisions of whole firms or single companies associated with a group.

The concept of neutrality-fairness, which descends from neoclassical economics, must be clearly understood. Neutrality and fairness are equal in a market perspective whereas they are not so under a different distributive principle. For instance, if the intention is to focus on labor as the primary productive factor creating value, that fairness no longer results from the setting of market prices. Market prices also reward capital outlay and risk in capitalist economies, and fairness is thus consequent to such an allocation to productive factors and not to one that is different.

The marginalist view has some important disadvantages.

"The target of transfer pricing at marginal cost is to induce a level of internal transfers that maximizes (only) short-run firm profits. However, short-run profit maximization is its sole desirable characteristic. This is clear when, for example, a situation is considered in which the selling division is responsible for technological innovation that grants the firm significant market power and results in higher profit (i.e. due to a reduction in production costs).

An application of the marginal cost will not credit the selling division with any significant share of profits sourcing from cost reduction. In fact, the transfer price is reduced to the new and lower marginal cost of production. The profits will predominantly accrue to the buying division.

The case resembles the age-old debate about the cost drawback of monopoly power. Monopoly causes a loss of welfare in a way that market prices exceed the cost of production while perfect competition optimally allocates resources. However, only the incentive to earn monopoly profits is able to push firms to make investments for innovation in technology and products."⁶

This is a general critique of any marginalist pattern first developed in political economy by J.A. Schumpeter who identified the concept of innovation as the main profit driver.⁷

"Firms’ pursuit of innovation disrupts perfect competition, which allows no economic (extra) profits. Firms seeking profits must innovate in reducing the cost of existent products or in developing new products (or markets) in order that the temporary absence of competitors can make innovating firms able to gain economic (extra) profits."

Therefore, if the objective of the marginal cost rule is short-run profits for the firm as a whole, that rule raises questions regarding compatibility with long-run (or extra) profit maximization that is achieved only by innovating the business.

Furthermore, the marginal cost rule leads to so-called agency costs "because managers are not worried for rising of costs and could be aimed at rising expenses in their own interest but not in the interest of the company that they are managing. In fact the implementation of marginal cost pricing generates distorted measures of divisional performance, provides little incentive for the selling division to cooperate, and sacrifices divisional autonomy.

The resulting profit measures will be of little use to central management as a basis either for interdivisional allocation of capital or for rewarding divisional management. It is generally acknowledged, however, that performance measurement is a necessary condition for effective allocation of resources within an organization. If divisional profits are based on transfer prices equal to marginal cost, and if those profits are used to evaluate managerial performance, potential agency costs are enormous. If constrained to transfer at marginal cost, management may respond by manipulating those costs to their advantage. For example marginal cost pricing does not provide a built-in incentive to control variable costs or to use the most efficient technology."⁸

These critiques are supported by scholars of business administration whose relevant work will be presented in what follows.⁹

1.3 Business economics

The first contributions of business administration scholars are less abstract than the neoclassical schemes and are associated with more realistic conditions that occur when firm managers project a transfer pricing policy.

When there is a competitive market for the product, the transfer price must be set at market price. If, instead, there is no competitive market but information is available to mimic a market price, the intermediate product must be exchanged at the supposed market price. If information is not available, a transfer price must be established, if possible, at a cost value plus a mark-up (the mark-up is added as profit).¹⁰

MARKET OR HIERARCHY FOR TRANSFER PRICES¹¹

Scholars emphasize the enigma and contradiction of transfer pricing. If the head office’s managers (mother company of a group or headquarters of a divisional firm) mandate transfer prices at marginal costs, they can optimally allocate resources. At the same time, however, headquarter managers demotivate divisional managers (or those of the companies associated with the group) from improving their results.

Indeed, greater efficiency in the selling division or selling company is not recorded as an increase in its result because it is automatically offset by a decrease of transfer prices. This is a paradox since divisions or group companies that perform only a segmented part of the entire production process are usually created just to improve efficiency, and this aim is not incentivized through the aforementioned transfer pricing policy.¹² "A divisional firm can maximize short-run profits, seriously distort divisional profit measurement and incentives, and, consequently, increase the probability that long-run profits will fall. Alternatively, it can sacrifice some short-term profits, obtain superior measures of divisional contributions to firm profits, and improve the chances of maximizing long-run profits (shareholder wealth in the long-term)."¹³

1.3.1 Strategies surrounding transfer pricing policies

Eccles, a business scholar in the mid-1980s maintains that group and firm strategies are more important and are a prerequisite for transfer pricing policies that must incentivize and be consistent with business strategies and not the contrary.¹⁴

Eccles opines that the group of companies or the divisional firm can be classified into one of four pure types of organization under the Manager’s Analytical Plane (MAP).

"As shown in Exhibit I, the first dimension of the plane is vertical integration, that is, the extent to which the company (division) carries on production and distribution activities that other companies (divisions) could perform. Vertical integration results in interdependence between profit centers when each stage of the production and distribution processes is evaluated on the basis of profitability.

The second dimension of the plan is diversification, that is, the extent to which the company (divisions) is engaged in different businesses. Diversification is determined by the extent of product-market segmentation. This results in an emphasis on the independent contributions of each business when they are separated as profit centers."

Competitive, cooperative, and collaborative organizations, for which the main features are illustrated in Exhibit II, must solve the transfer pricing problem (whereas there is no transfer pricing in the collective organization). This marks a Copernican Revolution when compared to the neoclassical (static) vision. The type of organization that managers seek is a prerequisite of transfer pricing since the transfer pricing policy must be consistent with the desired nature of the organization and not the contrary.

The focus is on transfer pricing in competitive and cooperative organizations.¹⁵

1.3.1.1 Competitive organizations and transfer pricing

Highly diversified organizations that have little vertical integration between business units represent the competitive type. Usually, divisions are autonomous and perform all of the necessary functions of a complete business and have little interdivisional dependence.

"A policy of market-based transfer prices with constrained sourcing is appropriate, such as giving inside suppliers a ‘last look’ chance to meet outside quotes. In competitive organizations, transfer prices are used to select suppliers and customers and to value internal exchanges in a way consistent with the measurement of business unit performance."

Setting transfer prices at market values determines whether sourcing will be internal from other business units or companies of the group or external from independent suppliers. When there is no market price, it is appropriate to determine a price set at cost plus a mark-up that is extracted from comparable products, average return on assets (of similar firms) etc.¹⁶

1.3.1.2 Cooperative organizations and transfer pricing

In a cooperative organization, the units (companies) must collaborate with each other and not compete. Usually, these entities are highly vertically integrated and organized on a functional basis where all of them are cost centers except for a single sales force that functions as a revenue center.

The transfer price in cooperative organizations is determined after the decision to source inside has already been taken, and internal transfers are therefore mandated for both buying and selling units at full cost. Standard costs may be used for avoiding problems, hailing from fluctuations of full costs.

"A cost-plus-investment method while retaining mandated internal sourcing may exceptionally be used.

This approach essentially splits up the selling unit into an investment center based on external sales only and a cost center for internal transfers, and makes the buying unit responsible for profits and ROI (Return On Investment) on all internal resources used to manufacture its products."

The objective of cost plus investment is to obtain some of the advantages of the other organizational type but leads to insufficient results.

Exhibit III summarizes the appropriate transfer pricing policy as the strategic decision of independence or vertical integration is taken.

1.3.2 Summary of concepts from business economics

Transfer pricing can be analyzed¹⁷ under:

1) Goals to be reached;

2) Decisional process;

3) Make or buy choice (supply source).

The classical goals of a transfer pricing policy are:

1) Optimality of resource allocation and maximum profit reached at the level of individual associated companies and at the level of the entire group (the consolidated¹⁸ profit of the group);

2) Fairness (neutrality) in the division of the group results between associated companies in a way that correctly measures the contribution of each company to the consolidated profit of the whole group.

The aforementioned goals can both be reached but they may also conflict.

When a market for the intermediate product is existent and through a transfer pricing policy at the market value, the optimality and maximum profits at the level of individual companies of the group and at the level of the group as a whole are both achieved.

Instead, when no market of the intermediate product exists or – which is the same thing – vertical integration has been mandated by the head office (or managers of the mother company), this will create problems from a transfer pricing policy at marginal cost. The group as a whole can achieve maximum profit but the policy will contrast with the individual associated company’s incentive to improve its efficiency.

The following is an example. The producer of the intermediate product is not incentivized to reduce costs of production (through innovation in technology) because each cost saving is immediately absorbed by an equal reduction in the transfer price at which it sells the intermediate to the associated buyer.

Therefore, even if the group’s short-run profit is attained, there is no incentive toward the target of achieving long-run (extra) profits; the latter are enabled only by innovation in technology (cost reduction for existing products) or in products (new products with greater value for consumers).

Business economics scholars have flipped the relationship between transfer pricing and firm strategy. The goals of optimality and fairness (neutrality) are not the group’s final goals but result from a selected decisional process about the policy. Firstly, a group must select its organizational form (vertical integration or autonomy of associated companies) after which the selection of transfer prices and the supply source (between external operators or associated companies) might be consistent in such a way that they increase the effect of the aforesaid choice.

For instance, when vertical integration is decided, a consistent transfer pricing policy must be mandated by the mother company’s managers to those of the associated companies. The policy is based on transfer pricing at cost values (marginal cost) and on the purchase of intermediate products from associated producers (excluding external firms).

Here is the summary of all of the concepts highlighted by business economists¹⁹.

1.4 Valuation of intangibles (a primer)

Three methods have been historically considered appropriate in economics, business administration studies, and professional (also accounting) practice that are explicitly targeted at the valuation of intangible assets: 1) the market-, 2) the cost-, and 3) the income-by use approaches. Intangibles are primarily valuated through financial economics and, e.g. discounted cash flow (DCF) which is derived from the use of properties.

Valuation of intangibles, when such properties are intra-group transferred, are dealt with in a specific paragraph in Chapter 9 (regarding transfer pricing of intangibles in practice).

1.5 New institutional economics and the arm’s length principle

New institutional economics represent a significant variety of economic studies focusing on social and legal rules that represent the playing field of business activity. Such studies are based on concepts resulting from neoclassical microeconomics and early institutional economics²⁰ and enable a survey of the true conditions existing in human societies where economic agents perform their activities.

A branch of the new institutional economics is the transaction cost theory, and Nobel prize-winners Oliver Williamson and Oliver Hart²¹ have been major representative scholars since the late 1970s. Williamson’s theory is focused on the concept of²² asset specificity or idiosyncratic investments which are "durable investments that are aimed at a particular transaction; the opportunity cost of a specific asset is much lower in best alternative uses or by alternative users should be targeted transaction be prematurely terminated."²³

When transactions are:

1) Uncertain about the achievable result, e.g. a profit or a loss;

2) frequent, involving a relevant amount, and (above all);

3) asset specific (idiosyncratic), not having alternative use-value as they are sunk costs,²⁴

the business is not conveniently managed by independent parties as per transaction costs²⁵; independent parties are usually unable to accept uncertainty and to negotiate a complete agreement before beginning a business. A complete contract involves an agreement containing clauses that provide for the conduct (right and duties) of the parties in case of future and unpredictable events.

The costs of negotiating may involve substantial amounts, and it may be more advantageous to conduct business between associated parties since they do not need to forecast the positive and negative aspects of future events. Additionally, if associated parties are considered as a single consolidated unit (as they genuinely are), any loss by one party (due to uncertain future events) is matched by an equal gain by the other associated party.

In this perspective, the higher that the uncertainty, the frequency-amount of transactions, and the asset specificity are, the more probability that there will be a group association (as opposed to independent firms transacting with each other).²⁶

Contract economic theory highlights that, when there is a specific investment, both parties are locked into the relationship. For the seller, the investment has value as he trades with the buyer, and no alternatives exist. For the buyer, the asset specificity of the seller grants a better result than contracting with other potential sellers.

More recently, influential authors clarify that the Williamson’s ‘fundamental transformation’ acts to change a competitive market into a monopoly (on the non-investing side) or to a bilateral monopoly (with bilateral specific investments).²⁷ Sunk costs²⁸ or specific investments play a complicated role in affecting the investor’s incentive and on the nature of a market that is highlighted.

ENVIRONMENT SUPPORTING INTEGRATION OF ECONOMIC OPERATORS