In brief
Written by Senior Regulatory Counsel Michael Mathieson
The exposure draft Bill for the taxation of 'attribution managed investment trusts' has been received with much hand-wringing and plaintive cries of 'what does it all mean?'. A threshold question is whether your MIT will fall within the regime. The text of the Bill makes that question difficult to answer.
A MIT can only be an AMIT if the beneficiaries have 'clearly defined interests' in the trust. They can only have clearly defined interests if two conditions are satisfied. Neither of them is very clear – in fact one of them is very obscure.
Condition 1: no streaming
The first condition appears to be a 'no-streaming' rule. The EM says there must be 'an objective benchmark for the attribution of tax consequences of the activities of the trust to its members'. It also says the trustee must not have any discretion 'to determine the character of income distributed to each member'.
So far so good. However, the proposed terms of the condition itself will require that, 'assuming that the trust is an AMIT … , the amount of each member component for the income year of each member of the trust can be worked out on a fair and reasonable basis ...'.
The link between what the EM says and the draft law is obscure. First, the Bill requires you to assume that your MIT is an AMIT as part of determining whether your MIT is an AMIT – which is not easy to get your head around. Secondly, if you are game enough to work through the relevant provisions, you will encounter the notion of a 'member's member component of an AMIT character' – which does not exactly illuminate matters. Thirdly, the Bill contains a clear 'no streaming' rule in another subdivision, which begs the question – why wasn’t that provision used here? The answer is not obvious.
Condition 2: fixed trust
The second condition is a 'fixed trust' rule. The EM says 'only trusts which are sufficiently non-discretionary receive the benefits available to attribution MITs – such as deemed fixed trust treatment'. It also says the trustee must not have any discretion 'to determine the entitlement to the income and capital of the trust of each member of the trust'.
Again, so far so good but, again, the proposed terms of the condition itself are unclear: they require that 'the right of each member of the trust to the income and capital of the trust cannot be materially diminished through the exercise of a power or right'.
The main difficulty with this drafting is that the rights of a beneficiary of a trust are, to a very large extent, defined, negatively, by the rights of the trustee. If the trustee has the right to do something with respect to the trust's income or capital, it follows that the beneficiary does not have the right to prevent the trustee doing that thing.
In my view, the drafting is misconceived because it assumes that a beneficiary's right can be materially diminished by the trustee doing something when the outer limit of the beneficiary's right is, ordinarily, defined by the trustee's very ability to do that thing. If that is right, then most trusts, including a conventional discretionary trust, would satisfy the 'fixed trust' rule. Plainly, that is not the intention – but if it is even possible that that is the result of applying the drafting, its appropriateness must be questioned.
An objection to my criticism might be that the proposed terms of the condition are very similar to the restriction imposed by the Corporations Act on the power of a responsible entity of a registered scheme to make unilateral amendments to the scheme's constitution. The problem with that objection, as every funds lawyer knows, is that the Corporations Act restriction on unilateral amendments has been the subject of intense judicial disagreement as to its meaning. That controversy was described in 360 Capital Re Ltd v Watts [2012] VSCA 234. Why use a concept which has been shown to be so unclear?
The other problem with the drafting is that it does not confine the powers and rights that may be exercised so as to cause material diminishment, to the powers and rights of the trustee. Specifically, the drafting does not exclude the possibility of an exercise of a power or right by the beneficiaries causing material diminishment – for example, by the beneficiaries passing a special resolution at a meeting. The EM and the surrounding provisions in the Bill suggest that it is only the trustee's actions that are meant to be relevant – but the drafting does not say that. If the beneficiaries' actions could be relevant to the question of material diminishment then, subject to my earlier point, very few trusts would satisfy the 'fixed trust' rule.
There has to be a better way.
A better test for a fixed trust?
Unsurprisingly, this is not the first time the income tax law has had to grapple with the concept of a fixed trust.
The concept appears in the rules governing CGT rollover relief, when there is a transfer between two trusts. In that context, a trust is considered to be fixed if 'the manner or extent to which each beneficiary … can benefit from the trust is not capable of being significantly affected by the exercise, or non-exercise, of a power'.
In this provision, we already have drafting which clearly conveys what most people would understand to be a fixed trust.
A plea to Treasury: Please don't try to fix something that wasn't broken – why not use this test instead?