Gaspar Lopes Dias V.S.
This article thoroughly examines the dividend and interest provisions in Model Conventions as they apply to characterize income from cross-border instruments, considering cases of overlap and thin capitalization.
This article scrutinizes the dividend and interest articles of the OECD Model Convention (OECD Model), UN Model Convention (UN Model) and US Model Convention (US Model) for the purpose of treaty characterization of income from cross-border financial instruments. In particular, articles 10 and 11 correspond directly to the dichotomy of hybrid financial instruments’ characterization, between the tax treatment given to the return on equity and debt investments, respectively. In tax treaties, both the dividend and interest articles grant limited taxing rights to the state of source, typically – though not necessarily – exercised by means of a withholding tax, while providing for an unrestricted right to tax to the residence state.
In order to mitigate double taxation, tax treaties require the residence state to provide relief for the withholding tax levied in the source state, through the exemption or credit method. In this context, portfolio dividends (i.e. those not arising in the context of a substantial shareholding) are commonly subject to higher withholding tax rates than interest payments. A substantial shareholding is considered to be 10% of the capital under the UN Model, 25% of the capital under the OECD Model and 10% of the voting rights under the US Model.
In addition, both the OECD Model and the US Model establish that the withholding tax rate in the framework of a substantial shareholding should not surpass 5% of the gross amount of the dividends, and provide for a maximum withholding tax rate of 15% for portfolio dividend payments, while the UN Model does not include numeric values for these rates.
Interest payments are subject to a withholding tax rate at source not exceeding 10% of the gross amount of the interest under the OECD Model, while the US Model generally provides for no withholding tax at source for interest payments.
The US Model defines dividends as “income from shares or other rights, not being debt-claims, participating in profits, as well as income that is subject to the same taxation treatment as income from shares under the law of the State of which the payer is a resident”. On the other hand, the OECD and the UN Model Conventions define the term “dividends” as follows:
the term “dividends” as used in this Article means income from shares, “jouissance” shares or “jouissance” rights, mining shares, founders’ shares or other rights not being debt-claims, participating in profits, as well as income from other corporate rights, which is subjected to the same taxation treatment as income from shares by the laws of the State of which the company making the distribution is a resident.
The term “dividend” is not defined exhaustively, and both definitions refer to the domestic law of the source state. In particular, under article 3(2) of the OECD Model, the expression “income from other corporate rights” is not defined in the Convention, and its meaning should, in this case, be derived from the domestic law of the source state, unless the context requires otherwise. ,  However, it is essential to clarify the extent of this reference.
In this context, the dividend article covers:
First, the addition of the term “other” to “corporate rights” confirms that all the items specified in article 10(3) are corporate rights. Second, from the wording of article 10 it can be clearly understood that the term “corporate rights” must be interpreted independently from the domestic law of the source state (“the term ‘dividends’ as used in this Article means income from […] or other rights not being debt-claims, participating in profits”). Finally, the term “other corporate rights” cannot be interpreted as imposing whatever definition of dividends the source state applies, since to qualify as a dividend under article 10 such definitions must be in accordance with the concept of “corporate rights” as autonomously interpreted in the treaty. ,  Thus, the reference to the domestic law of the source state relates only to the “taxation treatment as income from shares”, and not to the meaning of the term “corporate rights”.
However, the US Model’s definition of dividends does not require an autonomous interpretation, instead accepting the source state’s definition, unrestrained by the concept of “corporate rights” under the treaty. To the domestic definition, the US Model adds only that, specifically, “income from shares or other rights, not being debt-claims, participating in profits”, falls under article 10.
Interest is defined in a similar way by the Model Conventions as:
income from debt-claims of every kind, whether or not secured by mortgage and whether or not carrying a right to participate in the debtor’s profits, and in particular, income from government securities and income from bonds or debentures, including premiums and prizes attaching to such securities, bonds or debentures.
This is an exhaustive (closed) definition, i.e. no reference is made to the domestic law of the contracting states. As a consequence, the definition is to be autonomously interpreted. ,  The Commentary on the OECD Model states that the exhaustive and autonomous nature of the definition of interest in article 11 is warranted by the fact that (i) it encompasses the essence of most domestic law definitions of interest, (ii) a closed definition enhances both legal certainty and the longevity of the rule and (iii) references to the domestic law of contracting states should, to the extent possible, be avoided.
On the other hand, in the US Model, the following is added to the above-mentioned definition, providing for a reference to the domestic law of the source state: “and all other income that is subjected to the same taxation treatment as income from money lent by the taxation law of the Contracting State in which the income arises”. , 
Again, article 11(3) of the OECD Model states that “the term ‘interest’ as used in this Article means income from debtclaim of every kind”. In the same direction, the Commentary on the OECD Model provides that “the definition of interest […] does not normally apply to payments made under certain kind of nontraditional financial instruments where there is no underlying debt (for example interest rate swaps)”, provided that there is no abuse.
From the examination of article 11 of the OECD Model and the respective Commentary, as well as doctrinal contributions in the literature that indicate the logical consequences of the concepts, the characterization as interest income requires (i) the existence of a legally enforceable “debt-claim” that may be “of every kind”, (ii) providing remuneration (iii) for making capital available, rather than paid as a price, (iv) implying the actual exchange of the face value. Nevertheless, the term “debt-claim” is not defined in the OECD Model or the Commentary thereon. Instead, some examples are provided.
In accordance with the provided criteria, credit default swaps or total return swaps, for example, do not generally yield interest income, as there is no actual exchange of the face value of the contract and capital is not made available. In the case of payments made under a guarantee, the same conclusion can be reached, given that there is no underlying debt-claim unless the credit event for which the guarantee is provided does take place, in which case the payments under the guarantee cease. However, as under a credit-linked note capital is actually exchanged, payments made under such an instrument may be characterized as interest for treaty purposes.
Moreover, the Commentary on the OECD Model refers to participating bonds and convertible bonds as giving rise to income falling under the interest article until the moment of actual conversion occurs, unless the instrument shares the business risk or in the case of reclassification under thin capitalization rules. The 1994 OECD report on the taxation of new financial instruments states that “even payments on interest rate swaps, which may be a stream of payments calculated on an interest basis, are not income from a debt-claim. They are income from an interest rate swap agreement”.
Furthermore, the interest article should, in most situations, cover an original issue discount (OID). However, if not duly considered, this may entail perils for taxpayers and tax authorities alike, namely where the accrued interest not yet paid is subject to withholding at source, as well as to tax arbitrage opportunities if the unpaid interest may be deducted in the source state while being subject to a different treatment abroad (e.g. as capital gains), or where rights in the underlying instrument are transferred.
Overall, the OECD and the UN Model Conventions provide autonomous definitions of the concepts of dividends and interest, with the exception of the term “other corporate rights” (as discussed). Diverging from this approach, the US Model establishes that both treaty concepts of dividend and interest cover the respective definitions under the domestic law of the source state.
The following is a discussion on the criteria to resolve the overlap between the tax treaty definitions of dividends and interest. For the sake of simplicity, the focus is specifically on the OECD Model. First, the classical formulation of the corporate-rights test is examined, followed by the analysis of the debt-claim test. Finally, overall conclusions are drawn regarding the interpretation of tax treaty law in order to (i) identify the common legal practice, (ii) provide arguments for the criteria which logically dictate the characterization of income and, from a tax policy perspective, (iii) elaborate on the adequate basis for the adoption of a consistent rule.
As a starting point, the characterization of an item of income as dividends or interest under articles 10(3) and 11(3) of the OECD Model has in common a reliance on the features of the financial arrangement under which the payments are made.
The meaning of the treaty concept of “corporate rights” yielding dividend income is far from consensual among the doctrine. According to the widely cited interpretation advanced by Vogel, a corporate right implies a right to benefit from the potential increase in value of the enterprise as remuneration for sharing the business risk, which also comprises the potential loss of the invested capital in the image of a regular shareholder.
Therefore, the shareholder test or corporate-rights test establishes the scope of the dividend article to depend on the existence of remuneration from an investment catering both a right to (i) participate in the profits and (ii) in the liquidation proceeds. Each of these requirements is a condition to qualify as a dividend, though not per se sufficient. Hence, it must be understood that an item of income comprising only one of these features may be characterized under any other applicable article of the treaty. Moreover, entitlement to the liquidation proceeds of an entity is identified with a share in the hidden reserves, such that the reimbursement of the invested capital is necessarily subordinated to the claims of other stakeholders.
The twofold criteria are to be ascertained through a subsequent test in view of all the facts and circumstances.
However, such an analogy vis-à-vis the position of the shareholder refers exclusively to aspects impacting the assumption of business risks and not, for instance, to voting and control rights. Wholly considered, the pivotal fault of this test (as acknowledged by its supporters) is that, in light of all the possible features of a given instrument under which payments are made, it is far from clear when such instrument can be said to be sufficiently participating in the profits and in the liquidation proceeds in order to render a dividend characterization under the treaty.
In general, interest is defined as remuneration for making capital available on a temporary basis – thusly requiring repayment.
The OECD Model’s characterization of income as interest finds the core of its meaning by reference to the expression “debt-claims of every kind”, itself not defined in the Model Convention. Ergo, the treaty notion of interest relies on a debt-claim test which, autonomously interpreted, requires the existence of a legally enforceable right of the provider of capital to its redemption, i.e. to reclaim the repayment of the face value of the advanced sum. This interpretation, as asserted by Rotondaro, constitutes, in his view, the only reliable criterion to distinguish between income falling under the dividend or interest articles of the treaty. ,  It could be sustained in support to this view that, since the interest article – as any other exhaustive treaty definition – favours an autonomous and thus homogeneous treaty interpretation, it should take precedence over other income articles.
Nevertheless, regardless of whether an autonomous interpretation should – merely by that fact – take precedence, treaty definitions which themselves rely on terms not defined in the treaty ultimately find their meaning in the internal law of the contracting states, thus compromising the desired hermeneutical homogeneity. This is the case of the concept of debtclaim, on which relies the entire sense of the interest article.
Now, this does not mean that (a) any income taxed as interest or (b) any definition of debt-claim, under the law of the source state gives rise to interest under the treaty. Rather, characterization as interest for treaty purposes unequivocally requires income to be yielded in connection with an underlying debt-claim, (autonomously) defined as a legally enforceable right to the repayment of the advanced capital. As such, the reference to the source state’s law is restricted to ascertaining the existence of such a legally enforceable right to be repaid. Therefore, the purportedly autonomous nature of the interest definition cannot, in itself, constitute sufficient basis to establish its preponderance over the other income articles.
In addition, it is well established that interest must be paid as such and not, for example, as a price. This concept has been further clarified by several commentators, inferring that the interest character of the income manifestly does not require fixed rates, collateral or a guarantee, which do not impact the existence or otherwise of an underlying debt-claim. Furthermore, the uncertainty as to whether remuneration payments will actually be made and at which amount, does not preclude interest characterization, given that article 11(3) encompasses debt-claims of every kind “whether or not carrying a right to participate in the debtors’ profits”, as interpreted by the Commentary on the OECD Model.
Furthermore, it is imperative to distinguish business risk (as a concept identified with the possibility of loss of the invested amount, and thus opposed to the notion of debt) from the solvency risk of the debtor (referred to in literature as the “creditor’s general hazard”, which is intrinsic to any obligation and does not prevent a characterization as interest for treaty purposes). Even the concept of “business profits” under article 7 is defined in opposition to the concept of debtclaim in article 11, which is characterized by the absence of business risk.
In this context, it has been asserted that a legal debt-claim ought to be supported by an economic claim, understood as the need to meet minimum economic conditions of actual repayment (i.e. a minimum standard of solvency), in the absence of which the investor necessarily shares the business risk of the borrower. Accordingly, as long as there is a substantive legal claim to the invested capital, an unsubordinated nature of the underlying instrument is not essential to characterize income under article 11. Yet, the value of the interest should reflect the associated risk, and not entail a significant risk of loss akin to equity investments.
Under this view, for example, a perpetual or super-maturity debt instrument is likely to have a term sufficiently protracted in time so as to shrink the net present value of the debt claim to a negligible figure. Instead, convertible debt necessarily requires the existence of a debt-claim until the moment of actual conversion (if it ever takes place), after which such right to repayment perishes.Notably, it is immaterial whether the potential conversion operates by the potestative right of any party (for instance through a put or call option) or automatically by the outset of a certain date or event (for example as with reverse convertibles).
Lastly, it is relevant to clarify that, according to the Commentary on the OECD Model, income from derivatives is, in principle, to be treated as “other income” under article 21, and the contracting states may opt to enforce its provisions only to the extent that the income derived from non-traditional instruments is at arm’s length.
This section aims to constitute a detailed and critical examination of how these tests are applied and are to be applied in accordance with the current legal regime of tax treaty law. In addition, a comment is offered on the existing paradigm for the treaty characterization of income, distinguishing between the dividend and interest articles, and suggestions are made regarding the basis for viable alternatives. Within the framework of the previous two subsections, each of the two tests endeavours to determine alone whether income should be characterized as dividends or interest under the treaty, thus constituting the key tie-breaking test.
As there are several income articles, the fact that an item of income is excluded from the scope of one article, does not contribute to actually classify such item under another specific article. In addition, it is ironclad that income must be classified in its entirety under a particular article of the treaty. In this sense, the corporate-rights test is understood to impart dividend characterization if its two limbs are present, indicative of the existence of business risk, regardless of whether the underlying financial arrangement comprises a legally enforceable debt-claim. Furthermore, it is uncontroverted that under any test, the profit share is a conditio sine qua non of dividend characterization, yet it is irrelevant for the distinction between the scope of articles 10 and 11, since having a share in the profits does not exclude interest characterization.
Henceforth, according to this test, the only distinction is determined by the existence or otherwise of business risk. In this case, it is necessary to establish the legal basis for the relevance of business risk as revealed by a participation in the liquidation proceeds, which in turn assures that the repayment is subordinated to that of all other classes of stakeholders.
Effectively, the requirement of the second limb of the corporate-rights test appears to be solely based on its inherent logic, which deduces from the requirement to share the profits a share in the global appreciation of the enterprise as a consideration for the business risk assumed in a way akin to that of a shareholder, i.e. the risk of loss of the invested capital as expressed by the right to share in the liquidation proceeds.
The predicament here resides in that the risk of not obtaining profits is not the same as the risk of losing the invested capital, as it is perfectly possible to hold a right to profits in a company and yet not have a share in the liquidation proceeds; one thing does not result from the other. Moreover, if the access to profits were necessarily a consideration for the total risk assumed, then it would be antithetical to accept that payments could be characterized as interest for treaty purposes if they were profit sharing. Therefore, the author fails to see a legally binding link that allows such requirement (to share in the liquidation proceeds indicating a business risk of total loss) to be attained based on the wording of article 10. Consequently, the corporate rights test cannot be adopted as a key tie-breaking factor between the scope of the dividend and interest articles of the treaty.
However, apparently the Commentary on the OECD Model, along with numerous commentators and jurisprudence from multiple countries, have accepted this requirement, as it is conceptually sound. This fact may allow it to become enforceable in some jurisdictions and under some treaties. For all intents, in order to be applied the corporate-rights test needs further clarifications, considering that (i) the criteria regarding entrepreneurial risk are altogether unclear and (ii) it is not possible to attribute characterization as interest for treaty purposes to an item of income by failing to meet the conditions of the test. This, since the absence of either of its limbs does not imply the existence of a debt-claim required by the wording of article 11, without referring and precisely accessing such concept in the test. Furthermore, as article 10 includes the expression “rights not being debt-claims”, doctrine agrees that income from corporate rights and from debt-claims cannot coincide under treaty definitions.
Against this background, the Commentary on the OECD Model, in an attempt to resolve the overlap and recognizing that the distinction may be problematic, makes clear that “the term ‘interest’ as used in Article 11 does not include items of income which are dealt with under Article 10”. Naturally, in general, this statement is true to the extent that the opposite is also true (article 10 does not deal with items of income that are dealt with under article 11); this is a corollary of articles 10 and 11 being mutually exclusive.
Nevertheless, this statement is made here so that, in the context of thin capitalization situations, article 10 is applied to income from corporate rights, even though – outside such context – the income would be regarded as interest, due to also being remuneration from debt claims. This is so when one considers that the wording of article 10(3) restricts dividend characterization to “other [corporate] rights, not being debt-claims”. In other words, dividends are income from corporate rights (i.e. payments from corporate profits) to the extent these are not also remuneration from debt-claims.
Thus, it can be understood from the referred statement among others in the same way that, in the context of thin capitalization situations, the Commentary on the OECD Model asserts that either (a) as an exception, the debt-claim should be ignored and the interest recharacterized as dividends or, in fact, (b) the debt-claim no longer exists, so that one is left with dividends. In this case, only the last option may be compatible with the wording of article 10(3) (“other rights, not being debt-claims”), provided that article 11 is understood to require, in order to grant interest characterization, that a debt-claim be not only legally, but also economically enforceable, as a thin capitalization situation does not affect the existence of a legally enforceable debt-claim, (although it may condition its actual economic enforceability).
In this framework, the debate on the binding nature of the Commentary on the OECD Model does not influence these conclusions. Even if it were binding, it may not contradict the textual wording of the treaty itself.
Notwithstanding this analysis of paragraph 19 of the Commentary on Article 11 of the OECD Model, the explicit exclusion of debt-claims in the definition of dividends in article 10(3) implies that it is necessary to first ascertain whether the underlying instrument comprises a debt-claim. In this setting, the debt-claim test is the key tie-breaking factor for the distinction between dividends and interest under the treaty, as – in addition to attributing interest characterization to income from a legally enforceable debt-claim – it is possible to characterize income as a dividend by failing the debtclaim test, given that it is compatible and incorporates the twofold definition of dividends under article 10(3) as (i) profit participating and (ii) not a debt-claim.
However, indeed, regardless of the wording of the Commentary on the OECD Model and its applicability, if a loan has no possibility at all of ever being repaid (either at inception or at some point of its duration), then in such case the lender effectively shares the business risk, as the formal legally enforceable claim can only regain substance (i.e. the only chance of repayment) depending on the enterprise’s success. Therefore, at that point, the investor has, in fact, a claim to the remaining capital of the company akin to a share in the liquidation proceeds, because the fact that this claim is unsubordinated and limited to the lent amount is irrelevant, since that quota will not be met in any case.
The Commentary on the OECD Model addresses this and further clarifies, in several instances, the mentioned process by which interest should be (re)characterized as dividends in accordance with the thin capitalization legislation of the borrower’s country. The Commentary adds that this takes place whenever the creditor assumes the preponderant extent of the entrepreneurial risk, “interest […] should be considered as a dividend if the loan effectively shares the risks run by the debtor.” In the same way, “Article 10 deals not only with dividends as such but also with interest on loans insofar as the lender effectively shares the risks run by the company, i.e. when repayment depends largely on the success or otherwise of the enterprise’s business”. ,  In fact, in this last sentence, the Commentary on the OECD Model refers to the uncertainty about the repayment as a key factor differentiating interest from dividends, thus requiring a risk analysis, distinguishing (i) solvency risk as risk-free income from (ii) business risk conveying no certainty of repayment.
However, by listing factors that may indicate the existence of business risk, the Commentary on the OECD Model – just like the corporate-rights test – gives rise to uncertainty. Thus, it leaves the risk distinction to depend on the thin capitalization legislation of the source states and the ultimate burden of shedding light on the matter to the national courts’ free assessment. This may have a far-reaching impact, allowing the recharacterization of interest into dividends even in the presence of a substantial debt-claim, i.e. in situations where there is exposure to solvency risk only.
Again, the reading of article 11 establishes that it grants interest characterization to “debt-claims of every kind”, where the word “debt-claim” means a right to be repaid – not a natural right, but rather a legal right. That is to say, the right to repayment must be legally enforceable. In addition, the concept of (substantial) debt-claim is incompatible with the notion of business risk, as there is no such thing as an “absolute and unconditional” right to repayment that accepts the possibility of total loss. Therefore, article 11 can be interpreted as requiring more than a mere legally enforceable right to repayment, as such a restrictive interpretation would have to be explicitly stated.
Nonetheless, it could be argued that the expression “debt-claims of every kind” entails a broad notion that encompasses both strictly legally enforceable and substantial debt-claims. In any case, the interpretation of article 11 requiring a substantial debt-claim reconciles the debt-claim test with the situation addressed by the Commentary on the OECD Model and taken into account under the corporate rights test (although through an entirely different justification), instead of considering that in situations where there is business risk the existence of a debt-claim should be ignored, rather it should be recognized that in the presence of business risk there can no longer be a debt-claim.
This interpretation establishes a presumption that in the presence of a debt-claim, the income is interest unless it can be proven that the risk assumed clearly supersedes the solvency risk and can therefore be considered business risk. That is, it can be logically ascertained that a debt-claim has no substance only when, in spite of the existence of a legally enforceable right to repayment, it is blatant that there is no actual economic possibility of repayment, given that, cumulatively, (i) the borrower does not have the capital, (ii) the profits are not sufficient and (iii) there is no reliable prospect of obtaining either in the foreseeable future. The other factors listed by the Commentary on the OECD Model are not relevant, as they are not determinative.
As a result, under this view, the application of article 11 requires a minimum level of solvency for a claim to be economically enforceable, although there are other aspects that may condition such actual enforceability, thereby preventing a characterization as interest for treaty purposes. Accordingly, a debt-claim that has a residual (almost zero) net present value in the context of a perpetual or super-maturity debt instrument, arguably would not fall under article 11, even though the borrower may be solvent, as the right to repayment is purely formal.
Another question that has been raised in this context concerns whether the two tests considered here may be concurrently applied. This may initially appear viable if the exclusion of debt-claims from the scope of article 10 is made relative, even though these tests were not conceived to be applied in such a manner. In such case, if payments do not fall under a particular test (for example the debt-claim test), the corporate rights test becomes relevant, or vice versa. If the hypothetical payment does not fulfil this test either, the income would have to be characterized as something else, business income, capital gains or other income.
Naturally, in this frame, the question as to which test should prevail gains relevance only in relation to income that passes both tests, for example, an instrument comprising a legal debt-claim and a subsequent share in the liquidation proceeds. This, considering that there is the legal need to characterize the income for treaty purposes as either dividends or interest, given that, as previously stated, an item of income cannot be characterized under more than one income article. Hence, if an item of income passes only one of the two tests, it is immaterial which is applied first. Additionally, because if the income fulfils the requirements of both tests, only one will prevail; the one which prevails may be applied first, in view of the fact that if the income passes such test there is no need to apply the other.
However, the combined application of the two reformulated tests inevitably leads to the dilemma regarding which is to prevail in situations of overlap which, in turn, is resolvable only by applying the tests as they were conceived. Consequently, by its inherent logic, a combined application of the tests is not feasible.
To date, there has been no global, consistent application of a uniform key tie-breaking test, as shown by case law, since the interpretation of articles 10 and 11 is not altogether clear or consensual. Nevertheless, the analysis in this article has presented a framework from where arguments may be retrieved to support a position in a particular case. However, a secure course of action, in addition to obtaining a ruling when available, would be to draw conclusions from the history of justifications favoured by the national courts of a particular country.
From the undertaken hermeneutic of the treaty articles, in light of significant national court decisions and leading literature, ultimately the debt-claim test is sustained as the key tie-breaking test, as the wording of article 10 expressly excludes debt-claims from its scope, while entailing a debt-claim is the single requirement under article 11. Therefore, the interpretation of the treaty concept of “debt-claim” has been the focus of the ensuing analysis, from which the most critical conclusion with far reaching effects is that – depending on the accepted understanding of article 11 – a debt-claim must be not only legally, but also economically enforceable in order to render interest characterization for treaty purposes.
This convoluted dilemma emerges as a consequence of the absence of a consistent rule to distinguish between the concepts of dividends and interest, both of which represent remuneration for capital made available. Overall, from a
tax policy perspective, this discussion cannot avoid the fact that the criteria for classification of an item of income as a dividend or interest could be clearer. Instead of an all-encompassing definition, the criteria should rely on a single, decisive factor that can be objectively analysed.
Ultimately, the pattern of risk could be said to be the key between dividend and interest characterization. However, identifying the risk pattern is sufficient only to distinguish interest, but not to specifically characterize income as dividends. , 
The author would like to thank Prof. Frans Vanistendael for his supervision of the Adv. LL.M thesis that is the basis for this article. All views expressed are personal.
 OECD Model Tax Convention on Income and on Capital art. 12 (15 July 2005), Models IBFD; United Nations Model Double Taxation Convention between Developed and Developing Countries (2011), Models IBFD; United States Model Income Tax Convention of November 15, 2006; United States Model Technical Explanation Accompanying the United States Model Income Tax Convention of November 15, 2006.
 OECD Model (2010), art. 23A and B; UN Model (2011), art. 23A & B; US Model (2006), art. 23. See also OECD Model Tax Convention on Income and on Capital: Commentary on Article 23A and B para. 47 (2010) (naturally stipulates that the credit method be applied instead of the exemption method if the residence state chooses to tax the income).
 Articles 10(2) and 11(2) of the OECD and UN Models and articles 10(2) and 11(1) of the US Model, require that the recipient of the income be its beneficial owner in order to enjoy treaty benefits.
 E. Eberhartinger & M. Six, Taxation of Cross-Border Hybrid Finance: A Legal Analysis, 37 Intertax 1 (2009), at 7-8 (“these limits are an important subject of treaty negotiations […] it is also possible for a specific DTC to deny the source state the right to levy withholding tax, more frequently in the case of interest payments”.).
 Art. 10(5) US Model (2006).
 Art. 10(3) OECD Model (2010); art. 10(3) UN Model (2011).
 J.F. Avery Jones et al., The Definitions of Dividends and Interest in the OECD Model: Something Lost in Translation?, Brit. Tax Rev. 4 (2009), at 406 (“the use of the ambiguous expression “income from other corporate rights” should be avoided, as was recommended in the OECD Thin Capitalization Report. […] So far as the potential overlap between the dividend and interest definitions is concerned we support the Thin Capitalization Report’s recommendation that it should be made clear that Art. 11 does not include anything dealt with in Art. 10. […] In addition, the Commentary should state that there can be no conflict between the Model’s interest definition and limbs 1 and 2 of the dividend definition, so that the only case where the priority rule would be necessary is in the case of limb 3 of the dividend definition amended as we have suggested”.). Along the same lines, see M. Tenore, Taxation of Dividends: A Comparison of Selected Issues under Article 10 OECD Model and the ParentSubsidiary Directive, 38 Intertax 4 (2010), at 226 (“The definition of Dividends under Article 10.3 OECD Model raises complex issues, mainly due […] [to the fact that] this provision endorses the qualification under domestic law for treaty purposes”.).
 See section 3, which further clarifies when the definitions under the domestic law of the source state prevail for treaty purposes. Furthermore, paragraph 11 of the Commentary on Article 3 of the OECD Model refers to the domestic law “in force when the Convention is being applied”. Also, article 31 of the Vienna Convention on the Law of Treaties, by obliging an interpretation in good faith, precludes, within article 3(2), the possibility that domestic law might violate a treaty provision.
 M. Helminen, Classification of Cross-Border Payments on Hybrid Instruments, 58 Bull. Intl. Taxn. 2 (2004), at 58, Journals IBFD (“[Some states] seem to be of the opinion that as long as they, as the source state, tax an item of income as a dividend, the income also qualifies as a dividend under the tax treaties following the OECD Model and the residence state should accept the dividend classification. This way of applying a tax treaty, however, cannot be correct if the payment was not made on a corporate right”.). In addition, “the OECD Commentary clearly means not only rights in the form of equity, but also rights in the form of debt-claims which, in their true nature, are actually equity”. Helminen, at 59. See also OECD Model: Commentary on Article 10(2) (2010), para. 15(d).
 This position is widely acknowledged in literature. See e.g. M. Six, Hybrid Finance and Double Taxation Treaties, 63 Bull. Intl. Taxn. 1 (2009), at 23, Journals IBFD (“this means that the source state’s classification is relevant only if, from the perspective of an autonomous interpretation, it is qualified as a corporate right”.). See also K. Vogel, Klaus Vogel on Double Taxation Conventions (Kluwer 1997), at 649.
 The OECD view is that the term “other corporate rights” should be strictly interpreted as referring to “disguised profit distributions”. See e.g. H. Pijl, Interest from Hybrid Debts in Tax Treaties, 65 Bull. Intl. Taxn. 9 (2011), at 492, Journals IBFD (“‘other corporate rights’ through a limited interpretation of paragraph 25 of the Commentary on Article 10 of the OECD Model, which, in their view, would only apply if the holder of the debt claim is at the same time the shareholder (so that the interest flows through the shares, instead of through the debt claim)”.). See also N. Bammens, Articles 24(4
 and 24(5) of the OECD Model Applied to Domestic Thin Capitalization Rules, 5 World Tax J. 2 (2013), at 152, para. 5, Journals IBFD.
US Model: Tech. Explanation (2006), para. 165 (“The definition [of dividends] is intended to cover all arrangements that yield a return on an equity investment in a corporation as determined under the tax law of the state of source, as well as arrangements that might be developed in the future”.).
 J. Bundgaard, Perpetual and Super-Maturity Debt Instruments in International Tax Law, 10 Derivs. & Fin. Instrums. 4 (2008), at 139, Journals IBFD; J. Bundgaard & K.J. Dyppel, Profit-Participating Loans in International Tax Law, 38 Intertax 12 (2010), at 658.
 OECD Model: Commentary on Article 11 (2010), para. 21.
 US Model (2006), Art. 11(3). In addition, paragraph 186 of the US Model Technical Explanation (2006) further clarifies the reference to the source state’s domestic law by including the US definition of interest.
 OECD Model (2010), Reservations on Article 10(3), para. 81.1, (added in 1994) (“Portugal reserves the right to amplify the definition of dividends in paragraph 3 so as to cover certain payments, made under profit participation arrangements, which are treated as distributions under its domestic law.”). Nevertheless, Law 67-a/2007 of 31 December provides that income from interest rate swaps, currency swaps and exchange forwards is interest income domestically and for tax treaty purposes. Even though article 11 includes a closed definition of interest, the new law will impact the tax treaties negotiated by Portugal in which the definition of interest included the wording “other income assimilated to income from money lent by the taxation law of the State in which the income arises”, as was stated in the 1963 Draft Income Tax Treaty. In these cases, income from derivatives will be subject to withholding tax as if it were interest. In the author’s opinion, this does not constitute treaty override, as treaties following the 1963 Draft allocated taxing rights providing for this possibility that Portugal would choose not to exercise, until 2008. Recent treaties remain unaltered by this change in the domestic law.
 OECD Model: Commentary on Article 11(3) (2010), para. 21.1. The reference to non-traditional financial instruments mentions only interest rate swaps, financial futures, options to buy shares and deep discount bonds.
 Vogel, supra n. 10, at 646 et seq.
 Eberhartinger & Six, supra n. 4, at 9.
 V. Thuronyi, United Nations Ad Hoc Group of Experts on International Cooperation in Tax Matters, Taxation of New Financial Instruments (Sept. 2001), at 9, para. 28.
 OECD Model: Commentary on Article 11 (2010), para. 19.
 OECD, Taxation of New Financial Instruments (1994). See Thuronyi, United Nations Ad Hoc Group of Experts on International Cooperation in Tax Matters ST/SG/AC.8/2001/CRP.10, Sept. 2001, supra n. 21, at 9, paras. 26-28.
 OECD Model: Commentary on Article 11 (2010), para. 20 (“…what constitutes interest yielded by a loan security, and may properly be taxed as such in the State of source, is all that the institution issuing the loan pays over and above the amount paid by the subscriber, that is to say, the interest accruing plus any premium paid at redemption or at issue”).
 OECD, Taxation of New Financial Instruments, supra n. 24, para. 108.
 Vogel, supra n. 10, at 651.
 Art. 10 OECD Model (2010) (“participating in profits” referring to “other rights”, i.e. to the meaning of corporate rights).
 E.g. a share category without participation in the profits, yet entitled to the liquidation proceeds.
 Vogel, supra n. 10, at 646.
 M. Helminen, The Dividend Concept in International Tax Law (Kluwer 1999), at 272 (“Examples: (1) the loan very heavily outweighs any other contribution to the enterprise’s capital (or was taken out to replace a substantial portion of capital which has been lost) and is substantially unmatched by redeemable assets; (2) the creditor will share in any profits of the company; (3) repayment of the loan is subordinated to claims of other creditors or to the payment of dividends; (4) the level of payment of interest would depend on the profits of the company; and (5) the loan contract contains no fixed provisions for repayment by a definite date”.). In another example, income from perpetual debt cannot per se be regarded as dividends. Helminen, at 283. See also Bundgaard, supra n. 13, at 140.
 Six, supra n. 10, at 24 (“the extent of the control rights connected with a financial instrument, however, should have no influence on the nature of the instrument as equity or debt because limitations on a shareholder’s control rights do not directly increase or decrease the risk component”. See also Bundgaard, supra n. 13, at 140.).
 Eberhartinger & Six, supra n. 4, at 9.
 Vogel, supra n. 10, at 731.
 OECD Model (2010), art. 11(3). For further clarification of the concept of “debt-claim” through several examples, see OECD Model: Commentary on Article 11 (2010), para. 18.
 Rotondaro used the expression “absolute and unconditional”, while Fehér employed the term “legally enforceable”. T. Fehér, Conflicts of Qualification and Hybrid Financial Instruments, in Conflicts of Qualification in Tax Treaty Law (E. Burgstaller, K. Haslinger & M. Lang eds., Linde Verlag 2007), at 242. See C. Rotondaro, The Right to Redemption as a Key Characterization Factor in the OECD Model Convention Passive Income Taxation System: The Case of Reverse Convertibles, 2 Derivs. & Fin. Instrums. 5 (2000), at 264, Journals IBFD.
 This concept controversially relies on paragraph 19 of the Commentary on Article 11 of the OECD Model (2010).
 The face value may not coincide with the principal, as debt may be issued at a premium, which would constitute interest under the OECD Model (2010) and paragraph 20 of the Commentary on Article 11 (2010) (“when a bond or debenture has been issued at a premium, the excess of the amount paid by the subscriber over that repaid to him may constitute negative interest which should be deducted from the interest that is taxable” (emphasis added)).
 In a landmark case, interest under the Canada-Germany income tax treaty was seen as remuneration received by the lender for making the principal available to be used by the borrower on a temporary basis. Thus, such interest compensation necessarily entails (i) an actual exchange of a principal (ii) with the corresponding right to the repayment thereof (in other words, an underlying debt-claim). CA: FCA, 15 Jan. 1981, Melford v. Her Majesty the Queen, Case A-147-80, Tax Treaty Case Law IBFD.
 The next section scrutinizes the grounds for the acceptance of the debt-claim test as the only reliable criterion to determine the treaty characterization of the income.
 Rotondaro, supra n. 36, at 265 (“the existence/non-existence of an unconditional and certain right to redemption is the only effective and correct criterion available in order to distinguish between interest and dividend under the treaty”.). This interpretation also finds its roots on the presence of the word “repayment” in paragraph 25 of the Commentary on Article 10 and in the text of paragraph 19 of the Commentary on Article 11 of the OECD Model (2010).
In another case in the same direction as the Melford case, the sentence read, “the boundary [between dividends and interest under the treaty] does not accommodate a distinction between interest in form only and interest in substance (but not in form); the only relevant criterion for the application of Art. 11 […], rather than Art. 10, is whether the income is subjected to the same taxation treatment as income from money lent by the law of Australia”. (emphasis added). AU: FC, 22 Oct. 2008, Deutsche Asia Pacific Finance Inc v. Federal Commissioner of Taxation, (2008) FCA 1570, para. 82, Tax Treaty Case Law IBFD.
 For agreement with Rotondaro on the scope of the interest article, see P. Brown, Debt-equity conundrum, General Report, in IFA Cahiers de droit fiscal international, Vol. XCVIIb (Kluwer International 2012), at 40 (“[the] touchstone for whether an instrument will be treated as debt is whether the holder of the instrument has the right to be paid back its investment regardless of the profitability of the issuer”.).
 F.A.G. Prats, Qualification of Hybrid Financial Instruments in Tax Treaties, 3 Diritto e Pratica Tributaria Internazionale 3 (2011), at 986.
 Prats, supra n. 43, at 987.
 Vogel, supra n. 10, at 731. For instance, in paragraph 17 of the referred Melford case, the Court found that guarantee fees must not be characterized as interest, as a lender that receives interest from the money it lends, has, until the money is repaid, lost control over its money. For that loss of control and the risk inherent therein, the lender is paid interest. On the other hand, a guarantor retains complete control of its money until, if ever, the guarantor is called upon to honour its guarantee. The fee that the guarantor receives for providing the guarantee cannot, therefore, be characterized as interest for provision of money on loan, over which money the guarantor has lost control. It is strictly a fee which the guarantor receives for assuming a risk for which it may never be called upon to indemnify the lender”.
 As listed by Fehér, supra n. 36, at 242.
 Rotondaro, supra n. 36, at 264.
 OECD Model: Commentary on Article 11 (2010), para. 19 (“interest on participating bonds should not normally be considered as a dividend… [unless] the loan effectively shares the risks run by the debtor company”). For example a German decision assimilated income from jouissance shares to profit participating loans yielding interest under the treaty. Considering that the treaty does not include a definition of “participation in profits” (Gewinnbeteiligung), the term does refer to any claim somehow dependent on the amount or existence of profits. Furthermore, unlike standard interest, the profit share was expressed as a percentage of the nominal value of the shares which do not render payments in a loss situation. The fact that, in a loss situation, the taxpayer received a lower profit share than agreed was, in the Court’s opinion, not relevant, as this could be recouped by a higher profit share in a following year. Thus, “there was no loss participation prior to the profit participation”. DE: Bundesfinanzhof (Federal Tax Court, BFH), 26 Aug. 2010, Case I R 53/09, BFHE 231, 63, Tax Treaty Case Law IBFD.
 Business risk concept. The total risk of an investment, also designated as business risk, is characterized by a positive or negative departure from the expected return of the investment. However, the downside risk (a concept frequently used in insurance) refers to such deviation only when it is negative (the risk of loss). D. Shaviro, Risk-Based Rules and the Taxation of Capital Income, 50 Tax L.Rev. (1995), at 643; O.G. Wood, Evolution of the Concept of Risk, 31 J. Risk & Insur. 1 (1964), at 83; J.L. Athearn, What Is Risk?, 38 J. Risk & Insur. 4 (1971), at 639; J. Brooks Taxation, Risk, and Portfolio Choice: The Treatment of Returns to Risk Under a Normative Income Tax, Georgetown Law Publ. 1248 (2013).
 Vogel, supra n. 10, at 733. In the same way, see Helminen, Classification of Cross-Border Payments, supra n. 9, at 59.
 E.g. F.M. Giuliani, Article 10(3) of the OECD Model and Borderline Cases of Corporate Distributions, 56 Bull. Intl. Taxn. 1 (2002), at 11-14, Journals IBFD.
 Fehér, supra n. 36, at 242. This idea can also be found in the wording employed by Rotondaro, who defines a debt-claim as an “absolute and unconditional” right to repayment.
 This economic claim must not be mistaken for the psychological will to have the invested capital returned, which is not reflected in tax treaty law and would require subjective tests on the intentions of the economic agents.
 For further examination of this concept, see section 3.
 That is, the specific elements of solvability of the case in question.
 For example in these two cases, income received under an agreement denominated as “jouissance right contract” was deemed to be interest under the tax treaty in place, given that the taxpayer did not participate in the losses (entrepreneurial risk) and had the right to terminate the contract without losing his claim. In addition, profits only influenced the rate of interest applied to the capital lent. Lastly, in Case 2 K 2536/94 the need for the taxpayer’s consent to the further issue of jouissance rights was regarded as not decisive. DE: Finanzgericht Cologne (Tax Court of First Instance), 23 May 1996, Case 2 K 2536/94, published in EFG 1996, 836, Tax Treaty Case Law IBFD. Similarly, see DE: FG Cologne, 29 Apr. 1999, Case 2 K 3998/95, Haufe-Index 981245, Tax Treaty Case Law IBFD.
 Fehér, supra n. 36, at 242. See also Bundgaard, supra n. 13, at 141-142. By way of comparison, under the corporate-rights test, if such instrument is not subordinated enough to entitle the investor to a share in the liquidation proceeds, in addition to being profit participating, it cannot be seen as a dividend for treaty purposes.
 IN: AAR, 10 Oct. 2008, LMN India Limited v. Commissioner of Income, Tax-4, Tax Treaty Case Law IBFD (“Until the date of conversion, payments from convertible debt are interest payments, because a bond does not cease to be ‘debt’ merely because it is eligible for conversion into equity on a future date”.). In the same way, see OECD Model: Commentary on Article 11 (2010), para. 19 (“interest on convertible bonds [should not normally be considered as a dividend] until such time as the bonds are actually converted into shares”.). In yet another case, an agreement of jouissance rights convertible into shares was regarded by the Court as interest under the treaty, until actual conversion. To that end, the inclusion of jouissance rights in the treaty definition of dividends was not seen as decisive, as such concept is not defined in the treaty, thereby not allowing a distinction from income under article 11. The Court additionally clarified that the accounting classification as equity was not relevant for tax treaty purposes. DE: FG Cologne, 11 Dec. 2003, Case 2 K 7273/00, EFG 2004, 659-663; StE 2004, 139, Tax Treaty Case Law IBFD.
 OECD Model: Commentary on Article 10 (2010), para 21.1.
 OECD Model: Commentary on Article 21 (2010), para. 7.
 In general, income is attributed to the lex specialis.
 See section 2.2. (participate in the profits and share in the liquidation proceeds).
 Eberhartinger & Six, supra n. 4, at 9 (e.g. “A right to participate in profits is undoubtedly a necessary condition for classification as equity, but by itself it clearly does not change the debt to equity”.).
 That is, under a teleological interpretation of the purpose of article 10, which can take place only when a literal interpretation is insufficient to clarify the whole range of meaning of the provision and yet only to the extent permitted by such a literal interpretation of the wording in question.
 E.g. Eberhartinger & Six, supra n. 4, at 9. (“The wording of Article 10 OECD Model neither explicitly lists such criteria nor goes into detail on how these two criteria must be formulated so that a particular hybrid instrument qualifies as equity”.).
 In several instances in the Commentary on the OECD Model (2010), namely paragraph 25 of the Commentary on Article 10 and paragraph 19 of the Commentary on Article 11.
 As previously discussed.
 Art. 11(3) OECD Model (2010) (“interest […] means income from debt-claims of every kind”).
 OECD Model: Commentary on Article 11 (2010), para. 19.
 The actual wording employed is “presumed thin capitalization”, as assessed under the source state’s domestic law. The word “presumed” should be regarded as allowing a rebuttal of such presumption. See OECD Model: Commentary on Article 11 (2010), para. 19.
 OECD Model: Commentary on Article 11 (2010), para. 19; OECD Model: Commentary on Article 10, para. 25.
 In cases where the corporate-rights test is applicable, the justification of point (i) is valid, given that these operate in the exact same way.
 The Legal Status of the OECD Commentaries (S. Douma & F. Engelen eds., International Tax Law Series, IBFD 2008).
 Pijl, supra n. 11, at 487 (“The Commentary alone may give important guidance on the interpretation of tax treaties, but, first, the Commentary does not have the status of a tax treaty, second, it may be wrong and, third, it is questionable as to whether or not new Commentaries […] have an effect on previous tax treaties”.).
 This interpretation admits the view according to which “the term ‘corporate rights’ refers to instruments where a company is the ‘counterparty’, whether contractually or based in company law, of the holder. In the author’s view, “corporate” was intended to exclude rights in individuals or partnerships. Rights that are granted to employees or suppliers can also be corporate rights”. Pijl, supra n. 11, at 502. Regarding “the notion of dividends basically concerns distributions by companies”, see OECD Model: Commentary on Article 10 (2010), para. 24. The term “company” is understood as “any body corporate or any entity that is treated as a body corporate for tax purposes”. Art. 3(1)(b) OECD Model (2010).
 OECD Model: Commentary on Article 11 (2010), para. 19.
 Emphasis added. OECD Model: Commentary on Article 10 (2010), para. 25 (“Articles 10 and 11 do not therefore prevent the treatment of this type of interest as dividends under the national rules on thin capitalisation applied in the borrower’s country. The question whether the contributor of the loan shares the risks run by the enterprise must be determined in each individual case in the light of all the circumstances, as for example the following: – the loan very heavily outweighs any other contribution to the enterprise’s capital (or was taken out to replace a substantial proportion of capital which has been lost) and is substantially unmatched by redeemable assets; – the creditor will share in any profits of the company; – repayment of the loan is subordinated to claims of other creditors or to the payment of dividends; – the level or payment of interest would depend on the profits of the company; – the loan contract contains no fixed provisions for repayment by a definite date”.).
 CZ: SAC, 10 Feb. 2005, AAA v. Financial Directoratej, Case 2Afs 108/2004-106. For a translation, see Int. Tax L. Rep. 8 (2005), at 178-205, IBFD. In this case, the Czech Supreme Administrative Court did not consider the excessive interest paid in a context of presumed thin capitalization to be income from ‘other corporate rights’ and thus subject to Art. 10 of the respective treaty, despite the fact that the payment of interest was subject to the profitability of the debtor. See also Pijl, supra n. 11, at 502 (“the Commentaries assert that income from debts with preponderant equity characteristics should fall under article 10 instead of article 11, but article 10’s definition must be able to support this reclassification”.).
 The expected return taxation theory could be used to make the distinction. See G. Lopes Dias, Tax Arbitrage through Financial Engineering (Kluwer Law International No. 50 Series on International Taxation 2015), at 9, 13-15, 30, 32, 38, 40-41, 53, 153, 154.
 Rotondaro, supra n. 36, at 264.
 In any case, a substantial claim that is not also legally enforceable (be that what it may) cannot constitute a debt-claim for purposes of the application of article 11.
 OECD Model: Commentary on Article 10 (2010), para. 25; OECD Model: Commentary on Article 11 (2010), para. 19.
 These factors can be shown, for instance, by the leverage ratio; by the debt to revenue ratio (e.g. Debt/EBITDA); by the ability of the profits to cover the interest expense (e.g. EBITDA/Interest Expense); or by an economically based expectation of either these indicators to improve in a way that allows the debt to be repaid at maturity.
 Such income may or not be a dividend, depending on whether it is profit sharing. This can be better understood by means of an analogy with the mechanism for US purposes to attribute debt treatment to financial lease operations, where the central question is “did the parties reasonably expect the funds to be repaid in full”. However, it is known that there are sound business reasons for the issuance of super-maturity or perpetual debt, which are regarded as interest under the treaty if the interpretation of article 11 as requiring a debt-claim to be only legally and not actually enforceable, is regarded as correct.
 Helminen, Classification of Cross-Border Payments, supra n. 9, at 57 (“From an economic point of view, the only actual difference between debt and equity may be the variability of the return, and the distinction should therefore be made according to the risk involved. A shareholder’s intention is to bear the risks of the enterprise, but a creditor does not intend to assume such a risk. Creditors in principle bear a smaller risk than equity holders that the return on the investment will not be paid or that the investment will not be paid back”.).
 Avery Jones et al., supra n. 7, at 41 (“…the Commentary is not satisfactory because it attempts to draw an artificial dividing line between debt and equity. […] The real difficulty is that it is not possible to draw a line with any certainty on the basis of effective sharing of risk… […] In any event, these factors [risk-based rules] can be relevant only to the determination whether the return is interest, not whether it is a dividend”.). See also supra n. 84.
 In this context, it has been advanced that the dividend-bearing criteria may be amended to be granted, e.g. where (i) remuneration is uncertain, (ii) the repayment is subject to total risk and thus subordinated or (iii) there is no fixed term of repayment or this is unusually long. Alternatively, the yield from hybrids could be 50% treated as a dividend and 50% as debt. The first option seems even less precise than the presently outstanding rule, while the second may ignore the instrument’s underlying substance.
The article appeared in Derivatives & Financial Instruments, 2015 (Volume 17), No. 2, published by IBFD.