Market leading insight for tax experts
View online issue

GAAR: a practitioner’s view

printer Mail

In its first Budget following the election, the Government published a discussion document setting out a new approach to tax policy making. A fundamental part of that approach was a commitment to tackle tax avoidance in order to protect the Exchequer and to maintain fairness in the tax system and the Government announced that it would be considering whether to enact a General Anti-Avoidance Rule (GAAR) as one element of a package of defences against avoidance transactions.


It is perhaps natural to assume that a GAAR would be based on a subjective purpose test along the lines of the various Targeted Anti-Avoidance Rules (TAARs) which have proliferated in recent years. However, if it were to be introduced, the GAAR might just as easily be based on certain objective principles or hallmarks.

For instance, the provision could apply to a transaction if the transaction failed one or more of a specified number of hallmarks set out in the legislation or could be expressed in a more nebulous manner so that one was required to weigh up various stated factors in much the same way as the badges of trade have historically been used to distinguish trading transactions from investment transactions.

It is also unclear whether the GAAR would relate just to corporation tax or would extend across the whole range of taxes, with the possible exclusion of VAT because of the direct application of European legislation in that area and the principles laid down in the Halifax case ([2006] STC 919). Logically, a GAAR would not be confined to corporation tax.

If it is intended to be an expression of a policy of tackling tax avoidance, there is no reason why the provision’s application should be confined to one tax. Moreover, even if the primary target of the provision is the corporate sector, income tax and national insurance avoidance is a feature of that sector as well as corporation tax avoidance. Of course, if the GAAR were to apply to more than one tax, consideration would need to be given to whether the GAAR should take a slightly different form in relation to each tax.

Pros and cons

Leaving aside questions relating to the form which a GAAR might take, it is worth reflecting on whether or not a GAAR is necessary.

The main argument in favour of a GAAR is that it could be a valuable tool in shortening and simplifying the tax legislation. Specific anti-avoidance provisions such as the TAARs could be repealed and it would reduce the need for the current ‘cat and mouse’ practice of closing off loopholes which come to light (by virtue of the disclosure legislation or during the compliance process) with specific legislation of ever-increasing complexity.

Moreover, proponents of a GAAR would say that GAARs are in place in most European jurisdictions (as well as jurisdictions like Canada, Australia and South Africa) and that therefore the UK would not be placing itself at a competitive disadvantage by introducing a GAAR. On the contrary, because some of those GAARs are of relative long standing, it ought to be possible for the UK to benefit from the experiences of those other jurisdictions in producing a more effective GAAR, in much the same way that the UK disclosure legislation benefited from the operation of the US disclosure rules.

In contrast, opponents of the GAAR would say that it might well promote uncertainty in the tax system, particularly if it were to be based on a subjective purpose test rather than objective principles and that it might well involve conferring too much power on the executive branch of government. Whether or not these concerns are valid is likely to depend on two features of how a GAAR would operate in practice.

The first is whether, once a GAAR is enacted, it would be the subject of fairly regular consideration by the judiciary, so that a substantial body of case law built up in relation to the application of the provision. One can easily foresee the possibility that HMRC might use a GAAR in a similar way to its past practice in relation to the TAAR in CTA 2009 s 441, which is to raise the possible application of the provision in any ongoing dispute but not take the matter to litigation, with the result that the exact parameters of the provision are never circumscribed by a court.

The second is whether HMRC would be willing to operate a clearance process in relation to the GAAR and whether those clearances would be published. While a clearance regime, unlike judicial precedent, would not be the subject of external arbitration and would therefore enable HMRC to apply its own interpretation of the legislation unilaterally, the publication of clearances would at least reduce the scope for inconsistencies to develop.

Clearance process

Of course, whether or not a GAAR needs to be accompanied by a clearance process at all is worthy of debate in its own right. Readers may recall that the previous attempt to introduce a GAAR foundered on the Inland Revenue’s reluctance to have a clearance process. There are some grounds for sympathy in this context. One can see how there might not be a rush for the position of clearance provider within HMRC!

Whether or not a clearance process is thought to be necessary may ultimately depend on:

  • the form which the GAAR takes – a subjective GAAR would seem to require a clearance process somewhat more than an objective GAAR would do; and
  • whether or not the GAAR is to be the subject of self-assessment (in the same way as, for example, the TAAR in CTA 2009 s 441) or should simply be available to HMRC as a means of challenging the tax affairs of a taxpayer (in the same way as, for example, CTA 2010 s 731) – if the GAAR is part of a self-assessment regime, it would seem more necessary to have a clearance process than if the GAAR is simply a means of challenging returns.

One potential issue with a published clearance process would be the potential damage that it could cause to taxpayers by virtue of the loss of confidentiality. Even if clearances were to be published in anonymised form, it is easy to think of circumstances in which the anonymised facts would nevertheless reveal the identity of the taxpayer in question.

Disclosure regime

Another question which would need to be addressed if a GAAR were to be introduced is how its introduction would affect the existing disclosure legislation. Arguably, the existence of a GAAR would mean that there was no need for disclosure rules, consistent with the proposition that the GAAR should be bringing to an end the need to close off loopholes in the legislation with specific anti-avoidance provisions.

On the other hand, the disclosure rules do provide a convenient early warning system for HMRC in relation to tax-avoidance and would enable HMRC to promote a legislative solution in relation to a loophole where there was some uncertainty as to whether or not the GAAR applied.

It is perhaps unrealistic to expect HMRC to give up this particular weapon that is currently at its disposal until it has seen how the GAAR actually operates over a number of years.

Some final thoughts

It is the thinking underlying this last point which raises doubts as to whether a GAAR would actually achieve the objective of simplifying and shortening the tax code.

A more likely outcome is that the existing TAARs and other specific anti-avoidance legislation would be kept in place and that the process of closing off loopholes with specific provisions would continue.

In those circumstances, the GAAR would just be an additional weapon for HMRC with which to threaten taxpayers in cases where the technical position was unclear.

If that concern proves to be justified, then introducing a GAAR would not be a meaningful step forward so far as making the UK a more attractive place in which to do business.

That suggests that, if the GAAR is to achieve its purpose, the Government will need to have the courage to repeal much of the specific anti-avoidance legislation (or at least to desist from enacting new specific anti-avoidance provisions as and when loopholes come to light) and instead for HMRC to litigate in those circumstances where it believes that the GAAR applies.



Tony Beare is Head of Tax at Slaughter and May. His main area of expertise is structured finance but he also has considerable experience in private and public mergers and acquisitions and debt and equity capital market transactions. Email:; tel: 020 7090 5023.