Schrodinger’s Tax – Quantum mechanics and principles of tax

accapr —  3 December 2013 — 4 Comments

JP  03

By Jason Piper, technical manager, tax and business law, ACCA

Certainty is considered by many businessmen to be a useful characteristic when they are trying to make decisions. The ability to predict outcomes is key to the process. Equally, uncertainty introduces unknowable risk, and for the majority of decision makers this will naturally reduce their incentive to follow that particular course. And ultimately, if every possible course of action results in uncertain, unacceptable risk then the outcome will be inactivity – in economic terms, stagnation.

For business decision makers, the uncertainty which surrounds the outcome for tax purposes when the bogeyman of ‘tax avoidance’ is invoked has often been compared to the famous quantum mechanics thought experiment of Schrodinger’s Cat. For those who are not familiar, the cat is sealed (alive) into a box containing a mechanism controlled by radioactive delay which has exactly a 50% chance of triggering, and killing the cat, before the box is opened again to check on the cat’s welfare. Until the box is opened, we cannot know whether the cat is alive or dead.

But that’s exactly the situation taxpayers find themselves in with principles-based anti-avoidance approaches. The application of such methodologies is notoriously difficult, and will ultimately amount to ‘I know it when I see it – but I have to see it to know it’. That is to say, we cannot be certain as to whether a particular scenario falls foul of the principles rule until it is tested against the rule by the courts. Until then, the tax liability is in the same conceptual limbo as the cat – and with it the business decision maker.

Until the case is tested, the business will not know how much cash it will have left at the end of the fiscal year to reinvest, or pay to its employees as salary. Which is why the Schrodinger’s Cat situation would be an anathema to the cautious business decision maker. The outcome of the transaction could not be known in advance – and once the outcome is known, it is potentially irreversible, with catastrophic consequences for the cat.

In order to try to resolve this difficulty, many jurisdictions have introduced general anti-avoidance rules in one form or another. The argument is that in practice, taxpayers might actually see a net increase in certainty in the system by introducing the meta-law of purposive interpretation as a matter of statute. The truly compliant taxpayer remains in the same state of certainty as without the rule; under neither the strict letter of the tax provisions nor the principles-based rule will their dealings be condemned as ‘illegal’. And with the principles-based system to back up the strict letter, uncertainty is also resolved for the most egregious of schemes – they know with certainty that the proposed structures will fail, because they look wrong. There is no need to go through the complex observation process of interpreting every element of the structures, documentation and returns to see if they comply with the strict wording of the legislation in the view of the particular judge hearing the case.

From the perspective of the compliant this approach might seem to have much to recommend it. But we’re no longer here just looking at Schrodinger’s Cat, the ‘single particle’ uncertainty analysis. There are two boxes, one labelled “underlying tax law” and the other labelled “overarching anti-avoidance rule”. We’ve got two particles to predict, and we’re into the territory of quantum entanglement. That’s the bit about how you can establish the state of one half of a pair of related subatomic particles simply by observing the other – until you’ve observed either though, the system (like Schrodinger’s Cat) is effectively unresolved.

Resolution of the wave form is dependent upon both the strict law and the purposive rule. But if we can predict with absolute certainty the outcome of one ‘observation’ then we know both, and there is no need to stick the cat in the box. For example, a ‘clearly egregious’ case will fail on principles, so there is no need to analyse the strict law position. The net level of uncertainty in the system has been reduced. The limits of uncertainty are now restricted to those few cases where we can predict neither the legal, nor the purposive, outcome.

But the interpretation of intent, of the spirit of the law, is not fixed, and as a result neither are the boundaries of the uncertainty. The defence of principles-based GAARs is that we have ‘certain uncertainty’ – but the risk of populist information of the system is that we have ‘uncertain uncertainty’ in respect of application of the rule, a position no better than that we already inhabit. Business should know whether it’s going to get the cream, or pay the price of being too curious, before ever entering into a transaction. But if the range of uncertainty is itself uncertain, then more and more businesses are going to find themselves in the box, anxiously awaiting their fate. And that is simply not a place they should be.

Post script: There is of course one further wrinkle to all this, which is that the box may never get opened even once you’ve put the cat into it* (David Quentin discusses the issues well here). But the uncertainty about that is more a function of revenue authority enquiry resource than it is underlying tax policy, and while it may be every bit as in need of resolution, the question of whether governments should fund their tax collection authorities properly is hopefully not quite so difficult to answer.

*We’re assuming an otherwise immortal cat. Or perhaps a cat flap round the back that we can’t see. No cats were harmed in the performance of this thought experiment.


4 responses to Schrodinger’s Tax – Quantum mechanics and principles of tax


    I think there may be a bit of inadvertent confusion here between purposive interpretation and principles-based anti-avoidance. Although, personally, I thought the Furniss v Dawson rule as it stood before BMBF was fine, what the House of Lords said in that case was that purposive interpretation is a general canon of statutory interpretation – it is, if you like, part of the underlying law.

    What a GAAR does is to put on top of that a further test – even if the tax advantage was the purpose of the legislation, in the strictly limited sense that purposive is used in BMBF, if there is tax avoidance (or tax abuse depending on what the second A stands for), then you are still caught.

    Principles-based legislation seems to have two potential elements. The first is the one used for disguised interest after the 2007 consultation, where the principle that “a return designed to be economically equivalent to interest is to be taxed in the same way as interest” is put into the legislation. That gives the judges a starting point for purposive interpretation. it also means that you know you have a problem if you start by thinking “I want something that is equivalent to interest but won’t actually be taxed as interest”.

    The other sense in which principle is invoked is Judith Freedman’s general anti avoidance principle (GANTIP*), which would let the court consider what would have been within the legislation had Parliament considered the particular scheme that was before it. That would be an over-arching back-up to any principle or purpose clauses in the legislation.

    A GAAR or GANTIP does admittedly add a second – radioactive isotope vial, surely, rather than box? Unless the cat itself is quantum…? But purposive interpretation does not; rather, it completely replaces the idea of literal interpretation, mostly because there is no such thing: “every decoding is another encoding”. Principles-based legislation is simply legislation drafted to be fit for purposive interpretation.

    Whether or not the cat has survived is still a known unknown, but the Duke of Westminster is definitely deceased…

    *(the original, not Richard Murphy’s very different idea to which he gave the same acronym)


    Thanks Mike, you’ve highlighted an important distinction which I glossed over rather in the constraints of the blog. My take is that purposive interpretation is the realm of the judiciary, and can of course be applied to any kind of underlying law, while principles based legislation is the legislature’s proactive mechanism to drive a ‘purposive interpretation’ style result. You generally get to the same result but by different means, but of course the mechanisms are every bit as important as the ends in an administratively burdensome process like tax.

    IIRC, the MkI Australian GAAR was derailed as a result of dogmatically literal interpretation of the underlying tax law by senior judiciary, hence a MKII which made the purposive style outcome a result of both literal and purposive interpretation of the GAAR itself, though that may be a bit too much of an oversimplification.
    There’s also the view that there are varying levels of purposive interpretation, and it’s linked to the level of extrinsic evidence you adduce. A pure literalist will only look at the words of the statute whereas at the opposite extreme, you may even disregard the precise words of the statute if they seem to contradict wider policy considerations. Of course, the only way you can work out what those wider considerations are is to look beyond the statute and the tax return to things like economic return or eg Hansard in the UK. And therein lies the potential for more uncertainty if you don’t know which factors will carry what weight.

    So ultimately you still end up at the mercy of the judiciary when trying to work out if a particular transaction is, or is not, caught by a GAAR (where GAAR is a separate general provision) as you’ll be having to look to extrinsic factors to decide whether a transaction is ‘reasonable’ or not.

    The targeted principles based rules, like the ‘equivalent to interest’ provision, are probably less susceptible to changes in external factors, and as you say give the judiciary a very clear starting point on the particular issue at hand. But of course they lack the breadth and scope of a GAAR, so add to length and complexity in the written law, especially if you managed to write two which conflict with each other…


    Howdy! I’m at work browsing your blog from my new
    apple iphone! Just wanted to say I love reading through your blog and look forward to all your posts!
    Carry on the excellent work!

Trackbacks and Pingbacks:

  1. A general anti-abuse rule will not fix the corporate tax system | Andrew Goodall - December 4, 2013

    […] I don’t know whether that is true (how much lower is “lower”?), but have a look at Jason Piper’s blog in which he refers to the “boundaries of […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s