Sources of Law and Sources of Obligation in International Tax Law-making: An Autumn Note on the Legal Status of OECD Commentaries under International Law

By Diego Mejía-Lemos

This note is part of a series of invited contributions which will appear each season of the year. Their aim is to explore various issues of international tax law, from the perspective of public international law. Given the paucity of space, the overview of issues explored is brief, and the potential avenues for further research into them are merely suggested tentatively.

The legal status of the Commentaries of the Organisation for Economic Cooperation and Development (OECD) to the OECD Model Tax Convention have raised issues as to their nature and legal consequences, including under general international law. The literature has recently considered the potential roles of the OECD Commentaries in the interpretation of double taxation treaties (DTTs) (Broekhuijsen (2013) 9; Machado da Silva (2015) 573), and the application of DTTs which make reference (renvoi) to them (eg Protocol to Netherlands/Barbados DTT, discussed in Engelen (2008) 52 n 74; Colombia/France and Colombia/United Kingdom DTTs, discussed in Girón Uribe (2018) 118). While these issues seem merely academic, they often arise before courts, including, most recently, in C Danmark I v Skatteministeriet (cf, AG Opinion, C-119/16 (2018)).

The scholarly analyses of the nature of the OECD Commentaries often rely on conceptual frameworks distinguishing between ‘binding’ and ‘non-binding’, and the ‘no man’s land’ in between (Blokker (2008) 20), political and legal commitments (Ward (2008) 79), and ‘hard’ and ‘soft’ law (eg Girón Uribe (2018)). The prevailing position among scholars and in judicial practice is that the OECD Model Tax Convention, and, a fortiori, the OECD Commentaries, are not legally binding per se (eg AG Opinion, C-119/16 (2018), para 50). Nevertheless, there are vexed aspects of the debate which remain unresolved (see contributions to Douma & Engelen (2008) and other literature surveyed), and yet the conceptual frameworks deployed to address them are still the same. The fact that no solution has been achieved calls into question the heuristic usefulness of those conceptual frameworks.

This Note proposes that the distinction between sources of law and sources of obligation, among others, may provide a more fruitful heuristic tool, as far as general international law aspects are concerned. This distinction has been tangentially mentioned in scholarship, but not fully explored (eg Blokker (2008) 27, referring to the notion of absence of ‘a legal obligation’ in this area).

Whether the OECD Commentaries may take the form of one of the recognized sources of law has been discussed, particularly with regard to custom (Blokker (2008) 25); and, in this particular respect, the conclusion tends to be that neither practice nor acceptance as law thereof have been established in relation to the use of OECD Commentaries by states (Blokker (2008) 25).

The debate over whether other ‘sources of international law’ may account for the legal consequences, if any, of the OECD Commentaries, best illustrates the greater explanatory power of the distinction between sources of law and sources of obligation over the hard/soft law dichotomy. In fact, the hard/soft law distinction has often proven to be a false dichotomy, since, ‘[w]here treaties are hard in terms of bindingness and soft […] in terms of meaning, the Commentaries are inversely proportional to these treaties in matters of interpretation.’ (Pijl (2008) 102). Unsurprisingly, it has been noted that whether the OECD Commentaries are ‘binding’ has proven ‘too broad a formulation of the problem to be meaningful’ (Thirlway (2008) 32).

In particular, the debate over the place of acquiescence and estoppel is of relevance, both to scholarship and judicial practice (cf Judgment C-460 of 2010 – discussed in Girón Uribe (2018) 107-, concluding that the OECD Commentaries may not become binding by virtue of estoppel or acquiescence where the requirements for the operation of either are not met). In this vein, it has been claimed that the OECD Commentaries may become binding for OECD Member states whose DTT incorporate OECD Model Convention provisions: those legal consequences would extend to any such DTT by virtue of acquiescence or estoppel (Engelen (2008) 59, 71-72). This claim has been opposed with greater or lesser vigour. When plainly rebuked, it is argued that, given the clear non-binding character of the OECD Commentaries, acquiescence or estoppel are not relevant (Ward (2008) 74).

While the positions adopted in this particular debate might reflect divergent understandings of acquiescence and estoppel, it seems that progress is further hindered by a conflation of levels of analysis, namely of the general law-making level of sources of law and the specific obligation-making level of sources of obligation. For instance, it has been argued that ‘[a]lthough tax treaties give rise to international obligations, the OECD Commentaries do not’, and, thus, a state’s departure from the latter ‘would not in and by itself be an internationally wrongful act’ (Ward (2008) 84). And yet, an internationally wrongful act exists whenever an international obligation is breached, ‘regardless of its origin or character’ (Article 12, ARS (2001)). Indeed, the law of international responsibility is primarily concerned with international obligations, and not confined to those obligations flowing from sources of law (cf Commentary (3) to Article 12, ARS (2001) 55, noting that ‘States may assume international obligations by a unilateral act’, in addition to custom or treaty, and that ‘[t]he formula “regardless of its origin” refers to all possible sources of international obligations’).

The use of the distinction between sources of law and sources of obligation may provide a more solid basis for the analysis and legal consequences of OECD Commentaries. It provides a criterion to identify the correct level of analysis, where sources of obligations are the unit of analysis. At this concrete level of analysis, the only legal significance of soft law instruments, what is often called their ‘normative effect’ (Pijl (2008) 136-137), finds a more proper explanation: that normative effect results from the process of determination of the content of an obligation, where the non-binding instrument is connected to a source of obligation (cf Commentary (9) to Article 2, ARS (2001) 36, noting that certain issues may concern ‘content and interpretation of the primary obligation and cannot be determined in the abstract’). More importantly, where the nature and legal consequences of the OECD Commentaries is analysed, placing them at the level of sources of obligation allows for a more nuanced, and accurate, analysis: one in which, for instance, where acquiescence or estoppel are resorted to, emphasis is placed on the ‘specific conduct’ (Thirlway (2008) 39) in connection with the OECD Commentaries, instead of the ultimately empty notion of the OECD Commentaries as a non-binding instrument which may nevertheless become binding. The distinction may pave the way for progress in the enquiry into the nature of the OECD Commentaries.