The Arm’s Length Principle – An Adequate Means for Taxing Multinational Corporate Groups?

1 Introduction

Prevailing international tax law is characterized by the idea that (transfer) prices between ‘associated enterprises’^[1] should be determined like prices between ‘independent enterprises’^[2] in comparable transactions in comparable circumstances (OECD 2022a). This so-called ‘arm’s length principle’ aims at ensuring that transfer prices are not influenced by multinational corporate group objectives such as reducing the tax burden. Accordingly, the benchmark for an intra-group transaction price should be a hypothetical price free from distortions such as tax avoidance strategies (Biondi 2017a). However, while one may find comparable transactions for ordinary goods or services, comparable transactions of unique intangibles are hard (or, by definition, virtually impossible) to find. As almost a third of the corporate income in global value chains relates to intangibles (Chen, Los, and Timmer 2018), transfer pricing is, unsurprisingly, seen as a major driver of tax avoidance (Heckemeyer and Overesch 2017; Liu, Schmidt-Eisenlohr, and Guo 2020; Vella 2015; Zucman 2014). Even though studies that estimate the extent of tax revenue losses come to very different results, they all agree that tax avoidance is a serious problem. As Table 1 shows, worldwide tax revenue losses amount to at least 100 billion USD annually.

The consequences of tax revenue losses are obvious: A decrease of tax revenues impedes the reduction of existing inequality since either governments can provide fewer social goods such as social insurance, or other taxpayers must shoulder a heavier tax burden. In fact, since multinational corporate groups that avoid taxes do not pay their fair share of taxes (Scherer and Schmiel 2021), tax avoidance exacerbates inequality. Thus, there have been serious multilateral and unilateral political attempts to decrease base erosion and profit shifting of multinational corporate groups. Considering the OECD’s Two-Pillar Solution (OECD 2021),^[3] previous OECD attempts (OECD 2013, 2015b), EU initiatives (European Commission 2023a, 2023b; European Parliament Legislative Train Schedule 2023a, 2023b, 2023c, 2023d, 2023e), and unilateral initiatives such as digital services taxes (Tax Foundation 2021), international tax law seems to be in a state of change.

The problems with the arm’s length principle are well-known (Avi-Yonah and Benshalom 2011; Sikka and Willmott 2010) and the arm’s length principle has been discussed critically in the literature: A first strand of literature recommends abandoning transfer pricing based on the arm’s length principle completely and switching, e.g., to the formulary apportionment (Quentin 2017) or to a destination-based corporate tax (Avi-Yonah 2017). These recommendations are also supported by the argument that the arm’s length principle ignores the nature of multinational corporate groups by treating associated enterprises as independent enterprises (Biondi 2017a; Eden 2022; Greil et al. 2023; Picciotto 2022). While we may agree that this is a valid argument, it must also be recognized that the arm’s length principle remains important even though international tax law may be on the verge of changing (as mentioned above). The persistent relevance of the arm’s length principle motivates us to deal with the question whether it is possible to improve it. This question has already been addressed in a second strand of literature that suggests improving the arm’s length principle, for instance by reducing its complexity (Cardoso and Petruzzi 2019; Greil et al. 2023; Picciotto 2018; Schreiber et al. 2020). This paper contributes to this second strand of literature by focusing on a different dimension. We argue that the OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations (OECD 2022a; hereinafter ‘OECD Transfer Pricing Guidelines’) do not refer to a theory on the formation of prices but merely state the application of one of the OECD transfer pricing methods (OECD 2022a). Since the arm’s length principle lacks a theoretical foundation, we advocate developing a theoretical substantiation that enables a new interpretation of the arm’s length principle, based mainly on concepts of a general evolutionary theory (Campbell 1965; Dopfer and Potts 2014; Hodgson and Knudsen 2006; Witt 2017) and political-cultural market theory (Fligstein 2001; Fligstein and McAdam 2012). Against this theoretical background, we interpret pricing techniques in actual arm’s length transactions between independent enterprises as rules that are subject to an evolutionary process. We argue that the reference to an adequate market theory leads us to a new interpretation of the arm’s length principle that differs considerably from the current interpretation. This brings us to the important question whether the new interpretation of the arm’s length principle is a better means to reducing tax avoidance than the current interpretation is.^[4]

The present paper is divided into the following four sections. In Section 2, we show that the arm’s length principle remains relevant despite emerging developments and that the current interpretation of the arm’s length principle allows multinational corporate groups to choose transfer pricing methods almost arbitrarily, thus facilitating tax avoidance. In Section 3, we progressively develop a new interpretation of the arm’s length principle. In Section 3.1, we analyze how prices between independent actors are determined from the perspective of a general evolutionary theory and political-cultural market theory. In Section 3.2, we apply our hypotheses to the arm’s length principle and to controlled transactions. Section 4 evaluates whether this new interpretation is an adequate means, particularly regarding its ability to reduce tax avoidance and its feasibility. Section 5 summarizes and discusses the results.

2 The Current Interpretation of the Arm’s Length Principle and Its (Lacking) Theoretical Foundation

OECD member states commonly refer to the arm’s length principle as set out in the OECD Transfer Pricing Guidelines although these are not legally binding (Feinschreiber and Kent 2012; OECD 2022a). This principle means that, for tax accounting purposes, a price for a ‘controlled’^[5] transaction between associated enterprises should be determined like the price independent enterprises with diverging interests would have negotiated ‘at arm’s length’ in a comparable ‘uncontrolled’^[6] transaction in comparable circumstances.

Let us first look at the introduction of the arm’s length principle. The arm’s length principle emerged in a period characterized by the idea that double taxation has negative economic consequences. These damaging effects of double taxation were outlined in the seminal report submitted to the Financial Committee of the League of Nations by four economists (Bruins et al. 1923; Collier and Andrus 2017; Navarro 2023). Subsequently, the idea that double taxation should be avoided took root and the arm’s length principle was seen as the means to achieving this aim. From this perspective, Article 5 of the Draft Convention on the Allocation of Profits presented by the League of Nations in 1933 can already be seen as a forerunner of the arm’s length principle (Collier and Andrus 2017; Navarro 2023). After the dissolution of the League of Nations in 1946, the OECD took over the task of dealing with international double taxation. Predominantly inspired by the U.S. regulations (Avi-Yonah 2007), the OECD incorporated the arm’s length principle in Article 9 of the 1963 Draft Model Double Tax Treaty (OECD 1963). This article was retained in subsequent model treaties (Collier and Andrus 2017). In sum, the arm’s length principle as a means to allocating profits fairly between states and avoiding double taxation is firmly established in international tax law (Navarro 2023) even though international tax law is possibly on the verge of change.

One major driver for this emergent change seems to be a political momentum against tax avoidance in general. Another major driver are the challenges caused by the digitalization of the economy that also facilitates tax avoidance: The current international tax regime was “developed in a ‘bricks-and-mortar’ economic environment” (OECD 2023a). Accordingly, the tax nexus is rather based on physical presence, and cross-border taxation is often built on the notion of permanent establishments by foreign entities. Yet, large multinational corporate groups like Alphabet or Meta generate revenue in countries without being physically present. This has resulted in unilateral countermeasures, with countries such as Austria or Poland having introduced digital services taxes to tax these digitalized business models (Tax Foundation 2021). Moreover, there are proposals for changing the requirements for permanent establishments to encompass ‘virtual permanent establishments’ (European Parliament Legislative Train Schedule 2023f). Despite several political attempts to decrease base erosion and profit shifting of multinational corporate groups, such as the OECD’s Two-Pillar Solution (OECD 2021) or a formulary apportionment approach in the EU (European Commission 2021), the arm’s length principle remains important for the following reasons: Under Pillar One of the OECD’s Two-Pillar Solution (OECD 2021), the main innovation would be that, for multinational corporate groups in scope, a part of the profit would be allocated to market jurisdictions (Amount A). This would result in a new taxing right on these profits for market jurisdictions. Since only a part of the profits, namely, 25 % of the residual profit (which is defined as profit that exceeds 10 % of revenue), would be allocated under Amount A, Amount A would only “operate as an overlay to the existing profit allocation rules” (OECD 2022b, p. 8). The larger part of profits of multinational corporate groups in scope and, generally, profits of multinational corporate groups not in scope (which appears to be the majority^[7]) would still be allocated based on the arm’s length principle (Coleman et al. 2022). As Pillar One is part of a multilateral attempt to attack the challenges from the digitalization of the economy, it was, moreover, agreed that, for the time being, no further unilateral digital services taxes should be introduced and already implemented unilateral digital services taxes should not be imposed (OECD 2021, 2023b) so that digital services taxes would have no influence on the arm’s length principle.^[8] Pillar Two would implement a global minimum taxation for multinational corporate groups (OECD 2021, 2022c).^[9] As Pillar Two focuses on the effective (minimum) tax rate, the role of the arm’s length principle is generally not touched (Coleman et al. 2022). Another proposal concerns the EU only: The EU is currently thinking of reforming corporate income taxation in a project called ‘Business in Europe: Framework for Income Taxation’ or ‘BEFIT’ and reconsidering the introduction of “a single corporate tax rulebook for the EU, based on the key features of a common tax base and the allocation of profits between Member States based on a formula (formulary apportionment)” (European Commission 2021, p. 11f.; see European Commission (2023a) for the proposal for a directive). The formulary apportionment system would completely replace the current profit allocation system based on the arm’s length principle for transactions taking place within the EU. Nevertheless, the arm’s length principle would still be relevant for transactions involving corporations that are not located in the EU. Furthermore, the implementation, if at all successful, will take time, as the previous attempt of implementing a Common Corporate Tax Base shows (European Parliament Legislative Train Schedule 2023a, 2023b, 2023c).

Let us next look at the issues of the arm’s length principle. One important issue in the context of the arm’s length principle is the question of accepted transfer pricing methods. In the late seventies, the OECD followed earlier US regulations by recommending (i) the comparable uncontrolled price method, (ii) the resale price method, and (iii) the cost plus method (Avi-Yonah 2007; Collier and Andrus 2017; OECD 1979). The core idea of the comparable uncontrolled price method is to compare the price for, e.g., a good or service in a controlled transaction to the price in a comparable transaction between independent enterprises in comparable circumstances. The resale price method starts with the selling price of, e.g., a good or service that has been bought from an associated enterprise and is then resold to an independent enterprise. This resale price is then reduced by a gross margin granted to the reselling enterprise. The resulting price (resale price minus gross margin) is considered to be the price for the controlled transaction. Thus, the object of comparison of the resale price method is the gross margin. The core idea of the cost plus method is to start with the costs incurred by an enterprise for, e.g., a good or service in a controlled transaction and to add a mark-up on these costs reflecting the profit for this transaction. The cost mark-up is then compared to mark-ups in comparable uncontrolled transactions.

In 1995, the OECD published the Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations (OECD 1995). Considering changes in US regulations (Avi-Yonah 2007), the set of applicable transfer pricing methods was increased. Henceforward, the OECD Transfer Pricing Guidelines have included five transfer pricing methods (Collier and Andrus 2017; OECD 1995, 2022a); besides the traditional transaction methods described above, there are also (iv) the transactional net margin method and (v) the transactional profit split method (summarized as transactional profit methods). According to the transactional net margin method, the net margin (net profit relative to a base such as costs or sales) in a controlled transaction is compared to the net margin realized in a comparable uncontrolled transaction. The transactional profit split method compares the split of profits or losses in a controlled transaction to the split of profits or losses in a comparable uncontrolled transaction. The change in the canon of allowed transfer pricing methods resulted in particular from frequently missing comparable uncontrolled transactions (Avi-Yonah 2007). The introduction of the transactional net margin method and the transactional profit split method was understood as a paradigm shift (Collier and Andrus 2017). While the traditional version of the arm’s length principle was restricted to comparable-based methods, the enhanced set of methods reflects an arm’s length version that accepts “any methods reaching arm’s length results” (Avi-Yonah 2007, p. 24). In consequence, the originally clear distinction between transfer pricing and formulary apportionment blurred (Navarro 2023), yet without the benefits that a consistent formulary apportionment concept would have provided (Avi-Yonah 2007). The following Table 2 summarizes the transfer pricing methods (also called “OECD recognised” (OECD 2022a, p. 95) methods) that are included in the OECD Transfer Pricing Guidelines since the paradigm shift in 1995:

Additionally, multinational corporate groups can apply methods not described in the OECD Transfer Pricing Guidelines (so called “other methods” (OECD 2022a, p. 95)) if these other methods lead to prices that are in line with the arm’s length principle (OECD 2022a). However, other methods should not be used when the OECD recognized methods are more appropriate for that specific case. Moreover, if other methods are used, the OECD Transfer Pricing Guidelines suggest that the multinational corporate group should present additional information on why no OECD recognized method was applied and why the other method is more appropriate (OECD 2022a).^[10] The following Figure 1 summarizes the current interpretation of the arm’s length principle:

Figure 1:

Current interpretation of the arm’s length principle.

According to the current interpretation of the arm’s length principle, multinational corporate groups should apply one of the described transfer pricing methods, reason why the respective method is chosen, and justify why the transfer price is adequate according to the method used. Regarding the question which of the described methods should be used, the OECD Transfer Pricing Guidelines merely state the following:

2.2. The selection of a transfer pricing method always aims at finding the most appropriate method for a particular case. For this purpose, the selection process should take account of the respective strengths and weaknesses of the OECD recognised methods; the appropriateness of the method considered in view of the nature of the controlled transactions, determined in particular through a functional analysis; the availability of reliable information (in particular on uncontrolled comparables) needed to apply the selected method and/or other methods; and the degree of comparability between controlled and uncontrolled transactions, including the reliability of comparability adjustments that may be needed to eliminate material differences between them. (…)

2.3. Traditional transaction methods are regarded as the most direct means of establishing whether conditions in the commercial and financial relations between associated enterprises are arm’s length. This is because any difference in the price of a controlled transaction from the price in a comparable uncontrolled transaction can normally be traced directly to the commercial and financial relations made or imposed between the enterprises, and the arm’s length conditions can be established by directly substituting the price in the comparable uncontrolled transaction for the price of the controlled transaction. As a result, where, taking account of the criteria described in paragraph 2.2, a traditional transaction method and a transactional profit method can be applied in an equally reliable manner, the traditional transaction method is preferable to the transactional profit method. Moreover, where, taking account of the criteria described in paragraph 2.2, the comparable uncontrolled price method (CUP) and another transfer pricing method can be applied in an equally reliable manner, the CUP method is to be preferred. (…). (OECD 2022a, pp. 93–94)

Considering this, especially the comparability of the transaction between independent enterprises and the transaction between associated enterprises is of major importance (OECD 2022a). According to the OECD Transfer Pricing Guidelines,

transactions are comparable if none of the differences between the transactions could materially affect the factor being examined in the methodology (e.g. price or margin), or if reasonably accurate adjustments can be made to eliminate the material effects of any such difference. (OECD 2022a, p. 20)

Such material effects can result especially from differences in functions performed and risks taken by the enterprises or from differences in the products or services that are transacted as this often leads to differences in their value (OECD 2022a).

It is often argued that the comparability criterion can be fulfilled for transactions of usual assets in well-frequented markets with observable market prices (Schreiber et al. 2020). Nevertheless, assessing the comparability of a controlled transaction to an uncontrolled transaction gives multinational corporate groups room to maneuver as the comparability of transactions depends on the understanding of ‘comparability’ and can only be assessed on a subjective basis. This room to maneuver is exacerbated in the case of unique intangibles that have special characteristics which may make it impossible to find comparable transactions. In this regard, the OECD Transfer Pricing Guidelines refer to unique and valuable intangibles and define them as intangibles that, inter alia, “are not comparable to intangibles used by or available to parties to potentially comparable transactions” (OECD 2022a, p. 250). When there are transactions concerning intangibles where no comparable transactions can be found, the OECD Transfer Pricing Guidelines advise using such valuation techniques that discount expected future income generated by the intangible and taking the result as an arm’s length price (OECD 2022a). The problem of unique intangibles gives multinational corporate groups even more room to maneuver to choose the transfer pricing method and the transfer price almost arbitrarily. When there are only barely comparable transactions, if any, multinational corporate groups cannot only choose their favorite transfer pricing method but can also conduct comparability adjustments rather arbitrarily. This gives multinational corporate groups an immense room to maneuver, effectively enabling tax avoidance (Büttner and Thiemann 2017). This is even more problematic as transactions regarding intangibles are becoming more and more important (Chen, Los, and Timmer 2018). All in all, the more unique the good or service is, the more arbitrarily the method and the price can be chosen.

Although, according to the arm’s length principle, prices between associated enterprises should be determined like prices between independent enterprises, the OECD Transfer Pricing Guidelines do not explicitly refer to theories which deal with the question how prices on markets are determined. If at all, the OECD Transfer Pricing Guidelines may seem to refer at least implicitly to neoclassical market theory when, for example, they advise using valuation techniques that discount expected future income generated by an intangible (OECD 2022a). The original theoretical foundation of defining a present value by discounting future cash flows is the Fisher separation theorem (Fisher 1930), which was later enhanced to the theory of optimal investment decision (Hirshleifer 1958). Its core proposition is as follows: If an individual has access to a perfect and complete capital market without risk or uncertainty and, thus, can save as well as borrow any amount of money at a uniform, risk-free interest rate, the individual’s optimal investment decision can be separated from the optimal consumption decision. In contrast, if such perfect capital market access is missing, these two decisions depend on each other and, thus, are not separable anymore. When discounting an intangible’s expected future cash flow, the OECD Transfer Pricing Guidelines also advise considering risk. They explicitly mention the example of taking weighted average costs of capital (WACC), whose equity return is usually determined by the capital asset pricing model (Lintner 1965; Sharpe 1964), as a discounting rate (OECD 2022a). The capital asset pricing model’s original theoretical foundation by portfolio selection theory (Markowitz 1952, 1959) as well as its alternative foundation by arbitrage pricing theory (Ross 1976) also require access to a perfect and complete capital market (Follert 2023).^[11] Since neoclassical assumptions are neither fulfilled nor feasible (Buchanan and Vanberg 1991; Shubik 2019), neoclassical market theory is, from our point of view, not adequate for substantiating transfer pricing. Furthermore, referring to neoclassical market theory is self-contradictory since transfer pricing guidelines are superfluous on perfect markets where prices are objective and observable. Finally, perfectly competitive markets assume that actors are price-takers (Mas-Colell, Winston, and Green 1995), therefore, neoclassical theory does not deal with the question how actors determine prices.

In summary, even though, according to the arm’s length principle, prices between associated enterprises should be determined like prices between independent enterprises, the OECD Transfer Pricing Guidelines do not refer to adequate theories about how prices between independent enterprises are determined. Because of this lacking theoretical foundation, multinational corporate groups can choose transfer pricing methods and, thus, prices, almost arbitrarily. Due to the ambiguity of the current system, they can use transfer pricing for tax avoidance strategies.

3 New Interpretation of the Arm’s Length Principle from the Perspective of a General Evolutionary Theory and Political-Cultural Market Theory

3.1 Hypotheses on the Formation of Prices

Section 2 shows that the current interpretation of the arm’s length principle lacks a theoretical foundation and allows multinational corporate groups to choose transfer pricing methods almost arbitrarily. For that reason, we need an alternative interpretation of the arm’s length principle. According to the arm’s length principle, prices between associated enterprises should be determined like prices between independent enterprises in comparable transactions (OECD 2022a). Consequently, we need to know how prices between independent actors, such as independent individuals, independent enterprises, or independent multinational corporate groups, are determined.

We start our discussion with the idea that the question how actors determine prices can be traced back to the more general question how actors behave. Firstly, we draw on the hypothesis that actors have imperfect foresight and behave only boundedly rationally (Furubotn and Richter 2005; Simon 1955, 1972). Furthermore, individuals can act only under genuine uncertainty, and they cannot maximize outcomes but can merely increase their utility in a subjective sense. Bounded rationality results from actors’ crucial limits to their predictive abilities since actors can only be incompletely informed about complex future alternatives. Another point worth considering is that actors learn and, thus, may ‘create’ the future so that the future is unknowable and cannot be foreseen at the time of the decision (Buchanan and Vanberg 1991; Schmiel and Sander 2022; Shackle 1983). Because of their imperfect knowledge, actors can only evaluate ends and means from their subjective perspective. This implies that they choose actions they individually assume to be an appropriate means to achieving their ends (Buchanan and Vanberg 1991; Vanberg 2002).

Secondly, we apply the evolutionary hypothesis that actors are rule-makers and rule-users (Dopfer and Potts 2014) and we substantiate this hypothesis by referring to the political-cultural market theory. The political-cultural market theory views markets as social and political arenas or fields that enable structured exchange (Bourdieu 2005; Fligstein 2001; Fligstein and McAdam 2012). Actors in a field have a shared understanding of the rules of the field that are part of the culture of the field. They also have shared understandings of the actors in a field, of their respective status, and of the subject of the field (Fligstein 2016; Fligstein and McAdam 2012). Actors use shared understandings to interpret the actions of other actors and they behave according to the rules and the culture of the field (Fligstein 2016; Fligstein and McAdam 2012). Another key characteristic of markets is power (Fligstein 2001; Fligstein and McAdam 2012). Actors are usually differently powerful: There are less powerful actors, so-called challengers, and there are more powerful actors, so-called incumbents. While powerful actors try to maintain their position, the challengers attempt to defy the current system of domination. According to the political-cultural market theory, power means being able to influence the rules of the field and the behavior of other actors. Thus, incumbents can influence the rules and thereby the culture of the field by maintaining or changing rules. Generally, the culture reflects the interests of powerful actors.

In this light, independent actors such as independent individuals, independent enterprises, or independent multinational corporate groups use rules which are part of the field to find a transaction price. We call those rules ‘pricing techniques’ and define them as units of conjectural knowledge about how to calculate a price that include all cognitive patterns that emerge for calculating a minimum or maximum price.^[12] To give an example, one type of pricing technique could be the discounted cash flow method (Berk and DeMarzo 2017; Brealey, Myers, and Allen 2017; Ross, Westerfield, and Jaffe 2002). Pricing techniques also include habits regarding what data sources to use. Self-gathered accounting data of an enterprise may be a data source just as well as publicly available databases. When a supplying actor aims to gain profits by exchanging, e.g., a good, a service, or an intangible against a transaction price, this end will be missed if the transaction price is lower than the minimum price that this actor may calculate by applying pricing techniques. In contrast, from the demanding actor’s point of view, the end will be missed if the price is higher than the maximum price calculated by applying pricing techniques. Since we understand pricing techniques as part of boundedly rational means-end beliefs, they can be diverse. In consequence, even when facing the same transaction within the same situation, actors may estimate the transactions completely differently when their boundedly rational pricing techniques are different.

From an evolutionary perspective, pricing techniques in actual arm’s length transactions between independent enterprises are subject to an evolutionary process of variation, selection, and retention.^[13] They are expected to evolve over time because they are based on evolving knowledge.^[14] To be clear, Section 2 shows that the evolutionary process is not limited to the evolution of pricing techniques as rules of the field that deal with the determination of prices. It also covers the arm’s length principle or the OECD Transfer Pricing Guidelines^[15] as rules regulating the taxation of multinational corporate groups. Since answering our research question whether it is possible to improve the current arm’s length principle requires examining how prices between independent actors are determined and since pricing techniques on markets play a crucial role in this context, our priority is to analyze the evolution of pricing techniques (and, thereby, the evolution of the price-determining rules). Even though it is not central to this study, we pick up on the evolution of the arm’s length principle itself (and, thus, the tax regulation rules) in Section 4 when we ask whether the new interpretation of the arm’s length principle is feasible. The evolution of rules for determining prices and rules in tax regulation have in common that the evolution of both is influenced by culture and power. However, there may be differences regarding the relevant actors and the way they exercise their power. As mentioned in Section 2, the evolution of the arm’s length principle is predominantly driven by the OECD or dominant states like the US, which use their regulatory power.

Let us now deal with the evolution of pricing techniques. The first step of an evolutionary process is variation, as illustrated in Figure 2.

Figure 2:

First step of the evolution of pricing techniques.

We apply theoretical approaches to variation processes of cultural units (Boyd and Richerson 1985; Campbell 1960, 1965; Durham 1991; Hodgson 2002, 2003; Hodgson and Knudsen 2006, 2010; Knudsen 2002, 2004; Witt 2017) to the evolution of pricing techniques. Accordingly, evolution starts with single incidents of variations of pricing techniques. Variations of pricing techniques means that actors invent new pricing techniques that have not existed before. As theories of cultural evolution (Boyd and Richerson 1985), evolutionary economics (Vromen 1995), and organization research (Levitt and March 1988; Rerup and Feldman 2011) highlight, creative processes of trial and error play a vital role in the emergence of variations. This also includes the creative construction of knowledge to fill gaps in actors’ prior knowledge. We understand variation in a very broad sense and classify not only inventions of completely new pricing techniques but also, among others, modifications, combinations, extensions, simplifications, or any other changes to already existing pricing techniques as variations. The common characteristic of variations is that knowledge, in our case a novel pricing technique, emerges that was not available before. A good example is the pricing technique which determines a maximum price or a minimum price based on the discounted cash flow method (Fisher 1930; Hirshleifer 1958). This invention was a variation in the sense described above. Adding techniques of estimating future cash flows for different scenarios and integrating probability-weighted cash flows for the first time, as, e.g., in the form of option price models (Merton 1973; Scholes and Black 1973), belongs to variation, too. Combining initially different pricing techniques, e.g., the invention of applying the capital asset pricing model to present value calculation by risk-adjusting discount rates (Rappaport 1986), is also a variation within pricing techniques.

The second step of the evolutionary process, as illustrated in Figure 3, is the evolutionary selection of pricing techniques (Boyd and Richerson 1985; Campbell 1960, 1965; Durham 1991; Hodgson 2002, 2003; Hodgson and Knudsen 2006, 2010; Knudsen 2002, 2004; Witt 2017). When a new pricing technique has been invented, other actors can adopt the new knowledge and select the novel pricing technique, or they can ignore it (Dopfer and Potts 2014). Adoption means that actors acquire knowledge and routines by learning from other actors, and it also implies learning by socialization and education (Levitt and March 1988). Learning by adoption is not limited to learning through personal conversation. Adoption can also be mediated by information sources such as books or digital media (Bass 1980; Witt 2017) or by social media (Hutcheon 2012) as knowledge can be recorded in information sources, though not exhaustively (Levitt and March 1988). Learning by adoption presupposes that other actors have the chance to know about the new knowledge. When new pricing techniques have emerged in the form of an actor’s novel knowledge and other actors become aware of it, the next step is either to apply or at least principally accept this knowledge (positive selection) or to reject this pricing technique and to not apply it (negative selection). Say, e.g., that one day an independent actor had ‘invented’ the discounted cash flow method as a pricing technique to calculate the price of an intangible. This actor then applied the new pricing technique regularly and, thus, selected it positively. When this actor communicated this pricing technique to other actors, the other actors had the choice to select it positively or negatively. Processes of selecting pricing techniques, whether positive or negative, may take differently long. Generally, this raises the question of what may influence the positive or negative selection of a pricing technique.

Figure 3:

Second step of the evolution of pricing techniques.

Since actors consider the rules and the culture of a field, we may assume that actors positively select pricing techniques that are part of the rules and the culture of a field to justify their behavior and their calculated price. If there is a change in the culture of the field regarding pricing techniques, actors may select different pricing techniques positively (negatively) that (do not) fit the new culture of the field. This may also apply to a change in the information sources used by actors. Additionally, according to political-cultural market theory, the selection of pricing techniques will be influenced by power since powerful actors will try to (at least) maintain their position by influencing other actors and the rules of the field. Consequently, powerful actors may try to influence the positive or negative selection of pricing techniques in a way that best fits their interest.^[16] Less powerful actors may select pricing techniques positively (negatively) that are (not) accepted by powerful actors.

When pricing techniques are widely selected in a field, we reach the third step of the evolutionary process called retention, as illustrated in Figure 4 (Campbell 1965; Dopfer and Potts 2014). When a whole industry has widely adopted the knowledge of a pricing technique such that every new member of the sector is already socialized accordingly and the pricing technique is widely selected, we have retention of pricing techniques. Because of these socialization effects, a pricing technique that was developed or first selected by former members of the sector can outlast these members because new members will select it as well. Evolved pricing techniques may be incorporated in textbooks or standards. One example may be the International Valuation Standards (IVSC 2022), which are, in turn, incorporated in valuation textbooks (Fazzini 2018; Parker 2016). When textbooks or standards recommend the use of certain pricing techniques, we may interpret these recommendations as indicators for evolved pricing techniques in this context. It is worth noting that a retention of pricing techniques does not necessarily imply that all actors are in favor of these rules (Fligstein 2001; Fligstein and McAdam 2012). A retention of pricing techniques only shows that actors have a shared understanding of these pricing techniques as techniques to calculate prices.

Figure 4:

Third step of the evolution of pricing techniques.

In sum, since, according to the arm’s length principle, prices between associated enterprises should be determined like prices between independent enterprises in comparable transactions (OECD 2022a), we need to know how prices between independent actors are determined. From the perspective of a general evolutionary theory and political-cultural market theory, actors determine their prices by using pricing techniques that are widely adopted and selected in a field, with the evolutionary process being influenced by culture and power. Unlike the theoretical basis of the current interpretation, the underlying theories are neither based on unrealistic assumptions nor is referring to these theories self-contradictory.^[17] Moreover, one major advantage of using the political-cultural market theory here is that it highlights the context, the relevant actors, their interests and their power, and the role of rules (Fligstein 2016). The following Table 3 compares the theoretical basis developed here with the theoretical basis of the current interpretation of the arm’s length principle described in Section 2.

Table 3:

The theoretical foundation of the current OECD interpretation in comparison to the theoretical foundation of the new interpretation of the arm’s length principle.

	Current interpretation of the arm’s length principle	New interpretation of the arm’s length principle
Reference to theory	No explicit reference to adequate theories about the formation of prices Implicit reference to neoclassical market theory, which is an inadequate theoretical basis	Explicit reference to a general evolutionary theory and political-cultural market theory
Determination of prices in uncontrolled transactions between independent actors	On perfect markets, Prices are objective and observable Actors are price-takers The question how actors determine prices is irrelevant	Independent actors use rules (pricing techniques) to find a transaction price Pricing techniques are expected to evolve over time Actors determine their prices by using pricing techniques that are widely adopted and selected in a field, with the evolutionary process being influenced by culture and power

3.2 The Arm’s Length Principle in the Light of Hypotheses on Pricing Techniques

Our theoretical analysis in Section 3.1 provides insights into how prices in uncontrolled transactions are determined. From the perspective of a general evolutionary theory and political-cultural market theory, actors determine their prices by using pricing techniques that are widely adopted and selected in a field. The evolutionary process is influenced by culture and power. Let us now apply our hypotheses on pricing techniques to the arm’s length principle and to controlled transactions. According to the arm’s length principle, prices between associated enterprises should be determined like prices between independent enterprises.^[18] Thus, the arm’s length principle will be met best if evolved pricing techniques are also used in controlled transactions.^[19] Hence, the first step should be to identify evolved pricing techniques established within the temporal and situational context in cases of comparable uncontrolled transactions between independent enterprises. The temporal context is relevant because the evolutionary process will continue. Hence, to determine an arm’s length price requires identifying the evolved pricing techniques that are at the level of retention for comparable uncontrolled transactions at the time of the controlled transaction. This also includes identifying the evolved procedures to determine input data like, e.g., prediction techniques for future cash flows. Moreover, the situational context is relevant. We characterize the situational context especially by the industry and the type of assets or services involved. This is relevant because different situational contexts may have different evolved pricing techniques. In a second step, these evolved pricing techniques should be applied to the controlled transaction. The following Figure 5 illustrates the new interpretation of the arm’s length principle.

Figure 5:

New interpretation of the arm’s length principle in the light of hypotheses on pricing techniques.

Accordingly, multinational corporate groups should not only present information on why a transfer price is adequate according to the pricing technique used (as currently required for the choice of one of the OECD transfer pricing methods). They should also provide evidence that the chosen pricing technique itself fits the evolved pricing techniques for comparable uncontrolled transactions within the temporal and situational context, such as, e.g., the industry. In line with that, tax administrations should not only ask for evidence that a transfer price is adequate according to the pricing technique used by the multinational corporate group, but they should also additionally ask for evidence that the pricing technique chosen fits evolved pricing techniques actually used in comparable uncontrolled transactions within the temporal and situational context.

A key question is how to identify these evolved pricing techniques within the temporal and situational context since this is a central aspect in checking the arm’s length nature of the controlled transaction. One way to identify evolved pricing techniques may be to acquire up-to-date knowledge about currently applied pricing techniques, including techniques of generating input data. This requires empirical research that relates characteristics of actually used pricing techniques to characteristics of the contexts of the transactions. However, a first complication arises from the fact that such empirical data is usually scarce and confidential (Luckhaupt, Overesch, and Schreiber 2012). The problem is exacerbated by the fact that the evolutionary process may continue so that pricing techniques will change. Despite these difficulties, as discussed in Section 3.1, from the perspective of the political-cultural market theory, the selection of pricing techniques is influenced by culture since pricing techniques are part of the shared understanding on how to calculate a price. Due to the relevance of culture, a change in the culture of the field regarding pricing techniques may imply a change in the pricing techniques used, which may also lead to a change in the information sources. Remember that actors behave according to the rules and the culture of a field and that actors may positively select pricing techniques that are part of the culture of the field to justify their calculated prices. From this follows that general information on the culture and on the pricing techniques used within a specific context may exist. As outlined in Section 3.1, general information on evolved pricing techniques may be incorporated in textbooks or valuation standards.

In contrast to the current interpretation of the arm’s length principle, the interpretation of the arm’s length principle provided here refers to evolutionary theory and the political-cultural market theory and deals with the question how prices emerge in uncontrolled transactions. The following Table 4 compares the new interpretation of the arm’s length principle with the current interpretation of the arm’s length principle described in Section 2.

Table 4:

The current OECD interpretation in comparison to the new interpretation of the arm’s length principle.

	Current interpretation of the arm’s length principle	New interpretation of the arm’s length principle
General rule	Multinational corporate groups should: Apply one of the described OECD methods, Reason why the respective method is chosen, Justify why the transfer price is adequate according to the respective method used.	Multinational corporate groups should: Identify evolved pricing techniques that are established for comparable uncontrolled transactions within the temporal and situational context, Apply these evolved pricing techniques to determine an arm’s length price within a controlled transaction Present information on why the transfer price is adequate according to the pricing technique used and also provide evidence that the chosen pricing technique itself fits the evolved pricing techniques for comparable uncontrolled transactions within the temporal and situational context

The differences between the new interpretation presented here and the current interpretation of the arm’s length principle are not limited to the theoretical basis and the general rule how transfer pricing methods should be selected mentioned above. Rather, there are obvious differences in their practical application when we consider concrete cases that deal with the selection of the most appropriate transfer pricing method. Let us refer to example 9 of the OECD Transfer Pricing Guidelines, where the application of the profit split method is recommended in a case of granting the right to use an intangible for producing a drug (OECD 2022a). In this example, there are two companies, ACo and BCo, which are subsidiaries of AB Inc.^[20] ACo, located in country A, has developed the unique Compound A and owns worldwide patents on it. BCo, located in country B, has developed the unique Enzyme B. Working together, ACo and BCo discover that they can produce a highly effective and valuable drug by combining Compound A and Enzyme B. Thus, ACo grants the right to use Compound A to BCo against an annual royalty.^[21] BCo combines Compound A with its self-developed Enzyme B and develops the valuable drug. Figure 6 visualizes the legal-economic structure of this example.

Figure 6:

Legal-economic structure of the mentioned example.

Under the current interpretation of the arm’s length principle, one of the methods described in the OECD Transfer Pricing Guidelines is selected to determine the royalty. The OECD Transfer Pricing Guidelines argue that, in this case, the profit split method is the most appropriate method (OECD 2022a) as both parties make unique and valuable contributions. In contrast, from the perspective of the new interpretation of the arm’s length principle presented here, applying the profit split method requires that this method is an evolved pricing technique at the level of retention for comparable uncontrolled transactions as described above. Although it seems not completely impossible that the profit split method could be an evolved pricing technique, it is important not to ignore that other pricing techniques are more likely to have evolved. We may, for instance, consider that, as mentioned in Section 3.1, evolved pricing techniques may be incorporated in standards like the International Valuation Standards. According to the International Valuation Standards, which represent the global standards for the valuation profession (Deloitte 2024), “most royalties are paid as a percentage of revenue” (IVSC 2022, p. 75). If calculating royalties as a percentage of revenue is an evolved pricing technique at the level of retention in the given (here: pharmaceutical) industry, this pricing technique conforms with the new interpretation of the arm’s length principle.^[22]

Let us take another example. A recent decision of the German Federal Tax Court (2021)^[23] deals with the question how to determine adequate intra-group interest rates. In this case, the taxpayer used the comparable uncontrolled price method: They determined the interest rate of the controlled loan by referring to interest rates for loans between independent enterprises where borrowers had an equal credit rating (German Federal Tax Court 2021; OECD 2022a, p. 415). These interest rates were derived from market reports and from a credit contract with an independent bank. Yet, the tax authorities and the lower court deemed the comparable uncontrolled price method not applicable. The rejection of the comparable uncontrolled price method was reasoned, among others, with the argument that the relevant intercompany loan was not sufficiently comparable to a loan from a third party. Instead, the tax authorities and the lower court applied the cost plus method: They primarily referred to refinancing costs of the lender, added payroll costs as well as (low) costs of materials. They finally added a mark-up reflecting the lender’s profit and took the result as transfer price (German Federal Tax Court 2021). However, the German Federal Tax Court referred the case back to the lower court by arguing that, for intra-group interest rates, the comparable uncontrolled price method is, generally, applicable since granting a loan has a homogeneous character and is objectively comparable. The German Federal Tax Court argued further that, as in this case several transfer pricing methods can be applied in an equally reliable manner, the comparable uncontrolled price method was also preferable according to the OECD Transfer Pricing Guidelines (Bärsch and Engelen 2022; German Federal Tax Court 2021). In contrast to this argumentation, from the perspective of the new interpretation of the arm’s length principle presented here, a transfer pricing method can only be applied if it is an evolved pricing technique at the level of retention for comparable uncontrolled transactions. According to textbook literature concerning loan pricing methods in certain temporal and situational contexts (Dermine 2015; Ghosh 2012), the comparable uncontrolled price method appears to be less relevant in comparable uncontrolled transactions than a cost-oriented approach like the cost plus method.

In sum, the new interpretation of the arm’s length principle presented here leads at least to a different reasoning why a transfer pricing method is applicable, and it may lead to different results regarding the applicable transfer pricing method.

4 Evaluation of the New Interpretation of the Arm’s Length Principle

The interpretation of the arm’s length principle provided here differs considerably from the current interpretation. Let us now analyze whether the new interpretation of the arm’s length principle may be an adequate means to reducing tax avoidance and whether it is feasible. Both points call for further discussion.

Under the current interpretation of the arm’s length principle, multinational corporate groups should apply one of the described OECD methods, reason why the respective method is chosen, and justify why the transfer price is adequate according to the respective method used. Although prices between associated enterprises should be determined like prices between independent enterprises,^[24] the OECD Transfer Pricing Guidelines do not refer to adequate theories about how prices between independent enterprises are determined. In consequence, multinational corporate groups can, in fact, choose transfer pricing methods arbitrarily. This is clearly expressed in the decision of the German Federal Tax Court (2021) described above which indicates that the final choice of a transfer pricing method can only be assessed subjectively, which also gives room to maneuver. Additionally, if there is no comparable transaction to show the adequateness of the chosen transfer price, multinational corporate groups can choose transfer prices almost arbitrarily. As a result, the current interpretation of the arm’s length principle facilitates tax avoidance (Avi-Yonah 2007; Heckemeyer and Overesch 2017; Liu, Schmidt-Eisenlohr, and Guo 2020; Sikka and Willmott 2010). In contrast, under the new interpretation of the arm’s length principle, multinational corporate groups would have to identify evolved pricing techniques that are established for comparable uncontrolled transactions within the temporal and situational context and apply these evolved pricing techniques to determine an arm’s length price within a controlled transaction. Moreover, they would have to present information on why the transfer price is adequate according to the pricing technique used and also provide evidence that the chosen pricing technique itself fits the evolved pricing techniques for comparable uncontrolled transactions within the temporal and situational context. Since multinational corporate groups would have to demonstrate not only the adequateness of the transfer price but also of the pricing technique in the sense described above, it is a reasonable assumption that this additional criterion may help hinder tax avoidance.

Nevertheless, the new interpretation of the arm’s length principle also faces challenges in reducing tax avoidance: It is important not to ignore that finding a comparable transaction is inherent in the arm’s length principle and, therefore, also necessary under the new interpretation. Thus, under the new interpretation of the arm’s length principle, the comparability problem remains. Further, next to the requirement that information on comparable transactions should be provided, the new interpretation also presupposes that both, multinational corporate groups and tax administrations, know which pricing technique is the currently evolved one in a specific temporal and situational context. Although valuation textbooks, valuation standards, or pricing techniques actually used in previous cases of uncontrolled transactions in the same temporal and situational context may point to the existence of evolved pricing techniques, there are, once more, options for tax avoidance as multinational corporate groups may argue that a pricing technique resulting in favorable transfer prices has evolved when, in fact, it has not.^[25] We should also keep in mind that the evolutionary process continues. Thus, the availability of information is also an issue that would have to be considered under the new interpretation of the arm’s length principle. Another point worth considering is that there may be more than one evolved pricing technique, possibly resulting in different prices. Applying different pricing techniques may result in a range of potential arm’s length prices which would give multinational corporate groups room to maneuver. This issue is similar to the current interpretation of the arm’s length principle (Avi-Yonah 2017). Since the new interpretation of the arm’s length principle introduces a new criterion for checking the adequateness of transfer prices, this may help to reduce the opportunity for choosing transfer prices arbitrarily and, thus, for tax avoidance. At the same time, comparability remains a problem, as does the availability of information.

However, we come to a more optimistic result when we evaluate the new interpretation of the arm’s length principle not only on a stand-alone basis but in connection with other (multilateral and unilateral) means to reducing tax avoidance. Pillar One, for instance, grants market jurisdictions a new taxing right on profits allocated to them under Amount A. Thus, Pillar One, at least partially, addresses problems resulting from the digitalization of the economy.^[26] Pillar Two mainly introduces an effective global minimum tax rate of 15 % which, although still far lower than the average statutory corporate income tax rate (OECD 2023c; Tax Foundation 2022), tackles tax avoidance practices of multinational corporate groups: If they have a lower effective tax rate than the effective minimum tax rate, multinational corporate groups will at least be adjusted to the effective minimum tax rate of 15 %.^[27] In sum, combined with other current initiatives, the new interpretation of the arm’s length principle might have a substantial effect on reducing tax avoidance of multinational corporate groups.^[28]

Let us now turn to the feasibility of the new interpretation of the arm’s length principle. The arm’s length principle and its core idea, treating multinational corporate group members as independent enterprises, are well-known (OECD 2022a). Against this background, the new interpretation of the arm’s length principle may be generally feasible but there may be obstacles regarding its feasibility on closer examination. A first obstacle may appear when we consider the role of power. Powerful actors may try to prevent the introduction of the new interpretation of the arm’s length principle if it is not in line with their interests. Our insights above show that the new interpretation of the arm’s length principle might help to limit tax avoidance. Thus, powerful multinational corporate groups that fear losing options for tax avoidance may try to use their power to prevent the implementation of the new interpretation of the arm’s length principle. In addition to multinational corporate groups, we should also consider states to be powerful actors. There may be states in favor of the new interpretation while others may be against it. On the one hand, reducing tax avoidance should generally be in the interest of states as corporate taxes are an important financial source. Thus, they may support the implementation of the new interpretation of the arm’s length principle. On the other hand, states may also have reasons for trying to maintain the current interpretation of the arm’s length principle: Due to the ambiguity of the current system, multinational corporate groups can easily shift profits. This results in some (low taxing) states profiting and other (higher taxing) states potentially being disadvantaged by the current situation. Consequently, the new interpretation of the arm’s length principle may lead to a reduction of tax payments in one state while it may increase tax payments in another state. Thus, some actors may try to prevent change while others may try to foster it (Büttner and Thiemann 2017). Thus, the feasibility of the new interpretation of the arm’s length principle depends on the power of its supporters.^[29]

Another potential obstacle to the feasibility of the new interpretation of the arm’s length principle may be its complexity. The current interpretation of the arm’s length principle is already quite complex as it requires a lot of information for the comparability and for the functional analysis (Couzin 2013; Schreiber et al. 2020). Transfer pricing rules are actually often seen as the most complex regulations of tax law (Harst et al. 2021; Schreiber et al. 2020). There are already calls to reduce the complexity of the current system, and the current rules are perceived as difficult to implement (Cardoso and Petruzzi 2019; Greil et al. 2023; Picciotto 2018; Schreiber et al. 2020). Due to the additional requirement to reason why the pricing technique applied fits evolved pricing techniques in the specific temporal and situational context, the application of the new interpretation may be seen as even more complex and increase compliance costs.^[30] This new requirement leads to more effort for taxpayers as well as for tax administrations as, firstly, they must find reliable data and, secondly, tax administrations need to check the information provided. Due to a possible lack of information on evolved pricing techniques, checking the information provided may be difficult and take time. Thus, increasing complexity may limit the feasibility of the new interpretation of the arm’s length principle. Increasing complexity may also foster tax avoidance. Accordingly, if tax avoidance is in the interest of powerful actors, they may actually advocate the introduction of even more complex rules to facilitate tax avoidance.

Moreover, the political-cultural view provided here highlights the fundamental role of power in the transfer pricing process. As we said, power may have an influence on the introduction of the new interpretation of the arm’s length principle, but it does not end there. In the light of the analysis in Section 3.2, we must expect that under the new interpretation of the arm’s length principle, powerful actors may try to influence the evolutionary process in a way that best fits their interests. Thus, if it is in their interest to reduce or to avoid taxes, they will support the selection of pricing techniques that help them to reduce or avoid taxes. Additionally, powerful actors may establish a culture in which pricing techniques in their favor are considered evolved pricing techniques. Thus, we should keep in mind that the evolution of an evolved pricing technique may be influenced by powerful actors. A second point is that the insights we gain from the political-cultural market theory shed light on the currently underestimated role of power: The current interpretation of the arm’s length principle refers to prices between independent actors. However, according to political-cultural market theory, such prices may be distorted by powerful actors who use their power to influence them in their favor. The radical implication, if taken to its logical conclusion, is that the tax burden resulting from transfer prices depends on the power relations on markets. If we pursue this line of argument, we reach the conclusion that power may not only influence the introduction of the new interpretation of the arm’s length principle but may also have influenced the evolution of the arm’s length principle up to now. In this view, the evolution of the arm’s length principle is probably also influenced by actors who use their power to maintain or change the rules in their favor.^[31]

All in all, the new interpretation of the arm’s length principle may be an improvement compared to the current interpretation as it is adequately theoretically founded and may reduce tax avoidance. Nevertheless, the new interpretation also faces challenges considering its ability to reduce tax avoidance and its feasibility. Furthermore, the theoretical insights into power relations discussed here question the arm’s length concept as a whole. The following Table 5 summarizes the strengths and weaknesses of the current interpretation of the arm’s length principle in comparison to the new interpretation of the arm’s length principle:

5 Conclusion and Discussion

This paper deals with the question whether it is possible to improve the arm’s length principle. This is motivated on the one hand by the deficiencies of the arm’s length principle and the significant societal effects resulting from tax avoidance. On the other hand, it is motivated by the fact that the arm’s length principle will remain relevant even though international tax law is in a state of change, which the OECD’s Two-Pillar Solution (OECD 2021), previous OECD attempts (OECD 2013, 2015b), EU initiatives (European Commission 2023a, 2023b; European Parliament Legislative Train Schedule 2023a, 2023b, 2023c, 2023d, 2023e), and unilateral initiatives such as digital services taxes (Tax Foundation 2021) show. We start our discussion with the criticism that the OECD Transfer Pricing Guidelines lack a theoretical foundation. Therefore, we advocate developing a theoretical substantiation by referring to concepts of a general evolutionary theory (Campbell 1965; Dopfer and Potts 2014; Hodgson and Knudsen 2006; Witt 2017) and political-cultural market theory (Fligstein 2001; Fligstein and McAdam 2012).

The overall results can be summarized as follows: First, even though, according to the arm’s length principle, prices between associated enterprises shall be determined like prices between independent enterprises, the OECD Transfer Pricing Guidelines do not refer to adequate theories about how prices between independent enterprises are actually determined. In consequence, multinational corporate groups can choose transfer pricing methods and, thus, prices almost arbitrarily.

Second, from the perspective of a general evolutionary theory and political-cultural market theory, boundedly rational actors use pricing techniques to determine prices that are expected to evolve over time because they are based on evolving knowledge. Moreover, the evolution of pricing techniques is influenced by culture and power.

Third, the new interpretation of the arm’s length principle implies that multinational corporate groups should not only present information on why a transfer price is adequate according to the pricing technique used. They should also provide evidence that the chosen pricing technique itself fits the evolved pricing technique for comparable uncontrolled transactions within the temporal and situational context. In other words, the core idea is to identify the evolved pricing techniques established in the temporal and situational context of comparable uncontrolled transactions between independent enterprises and to apply these identified pricing techniques to determine an arm’s length price within a controlled transaction between associated enterprises.

Fourth, in contrast to the current interpretation of the arm’s length principle, the new interpretation of the arm’s length principle is based on an adequate theoretical foundation. Moreover, due to an additional criterion in checking the arm’s length nature of a transfer price, the new interpretation of the arm’s length principle may reduce the opportunity of multinational corporate groups to arbitrarily choose transfer prices. In consequence, the new interpretation may be a means to reducing tax avoidance. Nevertheless, comparability is also a problem under the new interpretation. Additionally, finding reliable information on evolved pricing techniques is difficult. Thus, tax avoidance is still a problem under the new interpretation of the arm’s length principle. Another aspect is the feasibility of the new interpretation: Since the arm’s length principle and its core idea are well known, the new interpretation of the arm’s length principle may be feasible. However, powerful actors that benefit from the current system may try to prevent change. Moreover, the new interpretation of the arm’s length principle is complex and requires more effort, which lessens its feasibility. As a consequence, we doubt that the new interpretation of the arm’s length principle alone is an adequate means to reducing tax avoidance significantly. However, because of the wide range of current initiatives to reduce tax avoidance, we argue for evaluating the new interpretation of the arm’s length principle as one instrument in a mix of instruments. Obviously, the character of the other instruments affects the scope of the arm’s length principle. If, for example, a formulary apportionment, which is addressed in the so-called BEFIT project of the EU (European Commission 2021, 2023a), were introduced in the EU,^[32] the scope of the arm’s length principle would be reduced (in comparison to the OECD’s Two-Pillar Solution (OECD 2021)), but the arm’s length principle would not be completely replaced. All in all, regardless of whether the other instruments tend to be more similar to the Two-Pillar Solution or to a formulary apportionment, understood as one instrument in a mix of instruments, we evaluate the new interpretation of the arm’s length principle as a more adequate means to fight tax avoidance than its current interpretation.