|
Assistant Treasurer David Bradbury says the GAAR alterations focus on the definition of tax benefit |
On March 1 2012, Assistant Treasurer Senator Mark Arbib announced that the federal government would act to protect the integrity of Australia's revenue by amending Part IVA of the Income Tax Assessment Act 1936. A press release in May 2012 by the new Assistant Treasurer, David Bradbury, confirmed that the amendments would be "targeted changes, focussed on the definition of 'tax benefit'". To successfully apply Part IVA, the Commissioner of Taxation must prove:
That there is a scheme in the requisite sense;
That the taxpayer has derived a tax benefit; and
That, objectively viewed, the scheme was entered into with the dominant purpose of obtaining the tax benefit.
While details of the proposed amendments have not been announced, they will undoubtedly result in an expansion of the Commissioner's ability to apply Part IVA. The changes are expected to take effect from the date of the announcement, notwithstanding a lack of any real detail as to the nature and extent of the changes. The government's response appears to be a reaction to several high-profile Part IVA decisions recently handed down by the Federal Court, which limited the extent to which the Commissioner can deploy a counterfactual analysis to prove the critical tax benefit element in the hands of the taxpayer.
To prove a tax benefit in the hands of the taxpayer, the Commissioner is able to point to an alternative, hypothetical set of transactional facts as a reasonable alternative to the actual transaction. If accepted by the Court as reasonable, a tax benefit is determined by reference to the actual transaction. To neutralise the Commissioner's counterfactual, taxpayers often claim that in place of the actual transaction, any alternative would have been so unreasonable that the taxpayer would have done nothing at all. By successfully arguing that there was no tax benefit, taxpayers are able to avoid an inquiry into whether the actions actually taken were for the dominant purpose of obtaining a tax benefit.
The proposals are directed at limiting taxpayers' ability to claim that they would have done nothing, and therefore derived no tax benefit referable to the scheme. Nevertheless, the proposed changes raise several issues. At a general level, the retrospectivity of the proposals has caused a significant degree of disquiet, with key stakeholders arguing that this has increased transactional uncertainty. At a technical level, it is possible that the proposals overstep the intended scope of the legislation.
Technical deficiency?
In his May 2012 address to the 5th Annual Tax Forum, the Assistant Treasurer conceded that the changes to Part IVA are being driven by several recent court decisions that
suggest ... there is a technical deficiency in the current rules regarding whether or not a taxpayer had obtained a 'tax benefit' in connection with a scheme.
Two recent Full Federal Court decisions stand out as likely drivers behind the scramble by Treasury and the ATO to address the alleged technical deficiency in Part IVA. Both decisions, read in context, make it harder for the Commissioner to prove the tax benefit element critical to applying Part IVA.
In FCT v AXA Pacific Holdings Ltd, the Full Federal Court dismissed the Commissioner's appeal from the Federal Court, holding that AXA qualified for a scrip-for-scrip rollover on the sale of a subsidiary under a leveraged buyout. The majority found that it was unreasonable to expect the taxpayer to have acted in line with the Commissioner's counterfactual – that is, if AXA had not entered into the scheme, it would have disposed of the subsidiary to another entity, in circumstances that would not have permitted rollover relief.
Not long after, the Full Federal Court came to a similar view in RCI Pty Ltd v FCT. RCI was a member of the James Hardie group of companies. In 1998, the group underwent a restructure, including the transfer of RCI shares in a subsidiary to another member of the Group, RCI Malta. In the run-up to the restructure, the subsidiary paid a dividend of $318 million. As a result, the capital gain made by RCI on the transfer of the shares to the Maltese subsidiary was substantially reduced. The Commissioner determined under Part IVA to include about A$480 million in RCI's assessable income for the 1999 income year. In essence, the Commissioner argued this would have been the capital gain on the transfer of shares to the Maltese subsidiary in 1998 if a smaller dividend had been paid.
Once again, the Court held that the Commissioner's counterfactual – that, but for the scheme, RCI would have executed the share transfer using different steps – was manifestly unreasonable. Had RCI conducted the transfer without first announcing the dividend, the transfer would have resulted in over A$170 million in capital gains tax.
According to their Honours, it would be unreasonable to expect RCI to have executed the transfer in any other manner – not least in the way suggested by the Commissioner. The Court held:
... if the scheme ... had not been ... carried out, the reasonable expectation is that the relevant parties would have either abandoned the proposal, indefinitely deferred it, altered it so that it did not involve the transfer by RCI of its shares ... to RCI Malta. But they would not have proceeded to have RCI transfer its shares in JHH(O) to RCI Malta at a tax cost of $172 million. On this view, RCI did not obtain the tax benefit it was alleged ... to have obtained in connection with the scheme.
Both taxpayers therefore succeeded in attacking the Commissioner's counterfactual, by arguing that in the absence of their executed schemes, both would have done nothing. The argument effectively breaks the critical nexus between the identified scheme, and the all important tax benefit. With no tax benefit referable to a scheme, there is no need to consider the bona fides of the transaction – that is, whether it was undertaken with the dominant purpose of achieving the tax benefit.
The High Court of Australia refused the Commissioner special leave to appeal both decisions, leaving the Full Federal Court's approach to tax benefit undisturbed as the highest appellate authority on the matter. The Commissioner's frustration with the Full Court's position was demonstrated at the AXA special leave hearing. Counsel for the Commissioner submitted that
...what has occurred in this case, and ... other cases now is that it has become a major forensic exercise to determine whether or not there is a tax benefit. In our submission ... the main in Part IVA is the question of dominant purpose ...
Does the shift in focus to "tax benefit" mean the application of Part IVA will be unduly extended? Does it go against the spirit of the legislation, as enacted?
It seems clear from the Explanatory Memorandum (EM) to Part IVA that the intended lynchpin is "dominant purpose", with a view to counter tax avoidance schemes that are "blatant, artificial and contrived". The EM explicitly provides that arrangements of a "normal business or family kind, including those of a tax planning nature, will be beyond the scope of Part IVA".
Part IVA is a reconstructive provision. It allows the Commissioner to restore the fiscal position of the taxpayer to that which would have existed in the absence of the avoidance steps. This begs the question as to what would have happened in the absence of the posited avoidance. This was one of the features of Part IVA which was intended to overcome inadequacies in the previous GAAR, section 260. The proposed changes appear directed to eroding this fundamental aspect of the GAAR.
For the most part, the "do nothing" defence has been applied in circumstances where a controlled group has undertaken, voluntarily, some sort of restructuring transaction. The tax "avoided" is generally referable to an amount of income which has not been returned by the taxpayer. Nor would it have been returned, given the commercial imperatives governing the transaction. Can it really be said to be contrary to public policy if a group enters into a discretionary transaction in a tax effective manner? By the same token, is it fair to levy tax on phantom income which would, in no conceivable circumstance, have been derived by the taxpayer?
The proposed reforms will effectively silence any claim that the impugned transaction would not have been done in any other way, and did not therefore cause a tax benefit.
Issues with Retrospectivity
Another major concern with the proposed changes is their retrospective application. Despite receiving no detailed explanation, taxpayers will be required to comply with the reforms from the date of their announcement, March 1 2012. This raises enormous uncertainty for transactions commenced before that date, and yet to complete. Taxpayers need to be mindful of each aspect of their proposed scheme, to ensure that steps already taken do not give rise to a tax benefit.
In response to widespread criticism, the Assistant Treasurer noted at his address to the Tax Institute that retrospective law is generally only legislated where the law is operating in a manner inconsistent with the Parliament's intention, and there is a risk of significant revenue loss. However, the tax benefit test has essentially remained in its present form since 1981, and is supported by clear and consistent pronouncements by the highest appellate courts.
Details of the amendments have not, at the time of writing, been revealed. Only minor amendments are needed if it is intended that additional income tax cost be removed as a relevant consideration in coming to a view on the counterfactual. However, if the amendments are intended to introduce more significant changes, there will likely be some disquiet in the business community.
Moving forward, we note that the government has appointed an "Expert Roundtable" to consult and advise on the Part IVA reforms, with the intention to avoid inadvertently affecting genuine commercial and business activities.
Mark Friezer (mfriezer@claytonutz.com) is a partner, and John Boyagi (JBoyagi@claytonutz.com) is a lawyer, at Clayton Utz in Sydney.
