I read Dan Neidle’s comments (Tax Journal, 3 February 2017) on my analysis of the hybrid rules with interest. I think we can both agree that the drafting of the new legislation leaves a lot to be desired. As it has departed from the original OECD proposal and now lacks any underpinning in principle, it is all the more difficult to interpret – which is why we have arrived at such divergent interpretations.
My reading of the legislation is that the ‘relevant assumptions’ in s 259EB(3)(b) are the successors to the ‘permitted reasons’ in the initial draft of the legislation – as are the equivalent assumptions found in ss 259CB and 259DC. As such, they operate in the taxpayer’s favour.
This can be seen if we take the example that Mr Neidle posits in his piece: that of a US company receiving interest from a disregarded US subsidiary, which is also exempt from US tax in respect of it. For the hybrid rules to apply, the excess must arise ‘by reason of’ the hybridity. Is this test satisfied?
We would both answer ‘yes’ to this question – but for slightly different reasons. Leaving aside the deeming provisions of the hybrid legislation for one moment, there are two reasons why the interest income would not be taxed in the US: first, because the US tax code does not recognise any income as arising on a payment from a disregarded entity; and second, because even if it did, that income would be exempt. The first of these reasons is linked to the payee’s hybridity; the second is not.
We then need to overlay s 259EB(3) (a), which states that in applying the causation test, if one of the reasons for the mismatch is the existence of a hybrid entity, it does not matter if there is also another reason (which, in our example, is the exemption). This is the case even if that exemption would have caused the mismatch in and of itself without there being any hybridity. As a result, the arrangement is within the hybrid rules without any need to refer to s 259EB(3) (b) and the relevant assumptions (which only apply in determining whether the mismatch is linked to hybridity ‘so far as would not otherwise be the case’).
The effect of the relevant assumptions, when re-running the test of causality, is to eliminate the mismatch in circumstances where they apply. Thus, where the recipient is (for example) a charity that is exempted from tax on the interest, it will be treated as being taxable on it. This means that in the counterfactual example, there is no mismatch, from which it follows that the actual mismatch is not treated as being caused by the payer’s hybridity.
In practice, we both seem to agree that HMRC is determined to counteract hybrid structures, even where the hybridity is not crucial to the tax benefit. Where we differ is that I read the relevant assumptions as providing some escapes from the application of the rules, albeit very limited in scope. Paragraph 553080 of the draft guidance seems (at least to me) to support this analysis, but is almost as Delphic as the legislation. It can only be hoped that the final draft will provide some greater clarity.
I read Dan Neidle’s comments (Tax Journal, 3 February 2017) on my analysis of the hybrid rules with interest. I think we can both agree that the drafting of the new legislation leaves a lot to be desired. As it has departed from the original OECD proposal and now lacks any underpinning in principle, it is all the more difficult to interpret – which is why we have arrived at such divergent interpretations.
My reading of the legislation is that the ‘relevant assumptions’ in s 259EB(3)(b) are the successors to the ‘permitted reasons’ in the initial draft of the legislation – as are the equivalent assumptions found in ss 259CB and 259DC. As such, they operate in the taxpayer’s favour.
This can be seen if we take the example that Mr Neidle posits in his piece: that of a US company receiving interest from a disregarded US subsidiary, which is also exempt from US tax in respect of it. For the hybrid rules to apply, the excess must arise ‘by reason of’ the hybridity. Is this test satisfied?
We would both answer ‘yes’ to this question – but for slightly different reasons. Leaving aside the deeming provisions of the hybrid legislation for one moment, there are two reasons why the interest income would not be taxed in the US: first, because the US tax code does not recognise any income as arising on a payment from a disregarded entity; and second, because even if it did, that income would be exempt. The first of these reasons is linked to the payee’s hybridity; the second is not.
We then need to overlay s 259EB(3) (a), which states that in applying the causation test, if one of the reasons for the mismatch is the existence of a hybrid entity, it does not matter if there is also another reason (which, in our example, is the exemption). This is the case even if that exemption would have caused the mismatch in and of itself without there being any hybridity. As a result, the arrangement is within the hybrid rules without any need to refer to s 259EB(3) (b) and the relevant assumptions (which only apply in determining whether the mismatch is linked to hybridity ‘so far as would not otherwise be the case’).
The effect of the relevant assumptions, when re-running the test of causality, is to eliminate the mismatch in circumstances where they apply. Thus, where the recipient is (for example) a charity that is exempted from tax on the interest, it will be treated as being taxable on it. This means that in the counterfactual example, there is no mismatch, from which it follows that the actual mismatch is not treated as being caused by the payer’s hybridity.
In practice, we both seem to agree that HMRC is determined to counteract hybrid structures, even where the hybridity is not crucial to the tax benefit. Where we differ is that I read the relevant assumptions as providing some escapes from the application of the rules, albeit very limited in scope. Paragraph 553080 of the draft guidance seems (at least to me) to support this analysis, but is almost as Delphic as the legislation. It can only be hoped that the final draft will provide some greater clarity.
90a1d87b-c1b5-43c3-a720-cd324830adee.tmb-editorspic.jpg?sfvrsn=1b2d0a7e_1)
