Treasurer Joe Hockey’s budget promise last week to “stop multinationals using complex schemes to escape paying tax” suggested a comprehensive regime to address tax avoidance by companies such as Apple, Google and Microsoft, as well as BHP and Rio Tinto.
All these five companies appeared before the recent Senate Committee hearings on corporate tax avoidance. However, a close look at the draft legislation suggests that the proposal is far from comprehensive.
Tax avoidance image via Shutterstock.
It is a step in the right direction - but possibly not a big enough step. Out of the five multi-nationals mentioned above, only two – Google and Microsoft – are likely to be covered by the proposal. Even for these two companies, it is doubtful if the proposal is powerful enough to be an effective weapon for the Australian Taxation Office to challenge their tax avoidance structures.
In broad terms, the proposed law will apply if a number of conditions are satisfied. It is important to summarise the key conditions here to highlight the limitations of the proposal. First, a foreign multinational – with an annual global revenue of more than A$1 billion – derives income from sales made to Australian customers. Second, it avoids booking the sales income in Australia. Third, the profits generated from the sales are subject to low corporate tax rate overseas. Fourth, the structure is designed with a “principal purpose” of avoiding Australian income tax.
The conditions imposed by the proposal dictates that many multinationals that are engaging in tax avoidance structures are not covered. For a start, the proposal will apply only to foreign multinationals. Local multinationals are immune from the anti-avoidance rule. This is despite the fact that recent news reports as well as the Senate Committee enquiry on corporate tax avoidance have revealed that major domestic multinationals – such as BHP and Rio Tinto – have been shifting profits to their marketing hubs in Singapore.
Many foreign multinationals will also be immune from the proposal. For example, it is unlikely that Apple will be subject to the proposed law. Its tax structure does not rely on avoiding booking sales income in Australia, which is one of the key conditions before the proposed anti-avoidance law will apply. In fact, Apple does book its sales income in Australia, but shifts over 90% of the sales income to Ireland through intra-group sales.
The limited scope of the proposed law also means that it is unlikely to be effective to deal with new tax avoidance structures that inevitably will be invented by multinationals and their highly adept tax advisors.
Will The Proposal Be Effective To Its Intended Targets?
Tax avoidance image via Shutterstock.
It is also doubtful whether it will be a very effective weapon to allow the ATO to challenge tax avoidance structures of companies such as Google and Microsoft.
A key test to satisfy before the proposal law can apply is that in general a “principal purpose” of the structure is to avoid income tax. This is an untested concept in Australian income tax law. It may eventually be subject to interpretation of the courts. However, some lessons may be learnt from the experience of the existing general anti-avoidance rule in the tax law, known as Part IVA.
In an internal ATO document disclosed under Freedom of Information Act and revealed in the Senate Committee enquiry on corporate tax avoidance, the ATO admitted that the Part IVA is unlikely to be effective to deal with most tax avoidance structures of multinationals. The main reason for this shortcoming is that Part IVA will not apply unless in general the ATO can prove that the “sole or dominant purpose” of a tax structure is for a tax benefit.
In practice, it is very difficult for the ATO to substantiate that the “dominant purpose” of a complex tax avoidance corporate structure is for tax avoidance. Multinationals often have substantial resources to engage industry experts and tax lawyers to argue that the structure is driven primarily by commercial reasons, and tax benefit is merely an incidental consequence.
Information asymmetry often dictates that the ATO lacks full information about multinationals and their tax avoidance structures. This issue further tilts the odds against the ATO in applying Part IVA to these structures.
It is likely that the “principal purpose” test in the proposed law may suffer similar problems as the “dominant purpose” test.
For anti-avoidance purposes, a more effective purpose test may be found in an existing provision in Part IVA. While Part IVA in general requires a “dominant purpose”, it has a specific section that in broad terms is applicable if “a” purpose of the structure is for tax avoidance.
In other words, as long as one of the purposes of a corporate structure is for tax benefit, that section may apply. It is unclear why the government does not adopt the “a purpose” test in the proposed law.
Given the limitations and doubts on the effectiveness of the proposed law, it may not be surprising that the Treasurer refused to provide an estimate for the tax revenue that may be raised from the proposal.