31 August 2010

Mr Haydn Daw
International Tax Projects Unit
International Tax and Treaties Division
The Treasury
Langton Crescent
PARKES ACT 2600

Dear Sir

Controlled Foreign Companies - June 2010 Consultation Paper

The Group of 100 (G100) is an organization of chief financial officers from Australia's largest business enterprises with the purpose of advancing Australia's financial competitiveness.

We appreciate the opportunity to comment on the development of the revised controlled foreign company ('CFC') measures.

The G100 supports the direction of the changes, which will undoubtedly assist in ensuring that Australian income tax considerations do not unduly restrict the competitiveness of Australian multinational companies, as well as further develop Australia as a location for regional holding companies.

We set out below some matters the G100 believes should be considered in the on-going re-design of the CFC measures.

Commencement date of revised measures
The delays to the implementation of the changes to the CFC measures have meant that the current impediments to Australian multinational's competitiveness persist. Further delays should be minimised.

The G100 believes, at the latest, the changes to the CFC measures should take effect from 1 July 2011 and that an elective start-up date for the 2010-11 financial year should also be available.

Concept of control
The proposed changes contain a concept of control that aligns with the concept of control under accounting standard AASB 127 'Consolidated and Separate Financial Statements'. Broadly, the Australian resident single controller of a foreign company will be an attributable taxpayer of that foreign company.

The interpretation and application of accounting standards can be contentious. AASB 127 includes within the concept of control the capacity to control an entity by reason of the exercise of potential voting rights. Paragraphs 14 and 15 of AASB 127 indicate that the existence and effect of potential voting rights that are currently exercisable or convertible need to be considered in determining whether an entity has the power to govern the financial and operating policies of a company. The uncertainty associated with control by reason of potential voting rights is evidenced by the existence of Implementation Guidance to AASB 127, AASB 128 'Investments in Associates' and AASB 131 'Interests in Joint Ventures' published by the Australian Accounting Standards Board. That Guidance (although not a part of the standards to which it relates) provides guidance as to the circumstances in which there is a power to control via potential voting rights and contains illustrative examples.

Some further issues in relation to the direct linkage with the accounting standards are as follows:

• If the accounting standards change on the meaning of control, so too will the tax law. This would require the revenue authorities (Treasury and the ATO) to monitor the accounting changes on an ongoing basis to see whether the proposed accounting changes continue to be appropriate in a CFC context and whether any ATO interpretive guidance needs to be revised in light of the changes. We understand that AASB 127 is expected to change (new ED 10: Consolidated Financial Statements is being proposed).
   
• As noted above, the accounting rule is proposed for determining a CFC and this will revolve around the concept of the accounting 'control' notion, but the tax rules will be used to determine a 'participation interest' in a CFC. Once control and a CFC is identified, attribution of income of a CFC for tax purposes is then determined by 'participation interests', which is not dependent upon accounting concepts, but another rule altogether. The notion of participation interest draws on the debt/equity concepts for tax law purposes e.g. page 20 of the Treasury Consultation Paper refers to 'total rights to returns on equity interests in the other entity to its members 'that are 'distribution of profits'. It seems strange (and potentially increases taxpayer compliance cost) that there are two different rules to measure an interest in a CFC and query whether it would be simpler to have one set of rules to determine 'control' and 'a participation interest' in a CFC.
   
• The accounting control definition was developed in a different context to the relevant tax law. The accounting control concept is to ensure the consolidation of entities within the group for accounting purposes & to ensure relevant information is presented to users of the financial statements. For tax purposes, the CFC rules are to ensure residents cannot accumulate income offshore and thereby, defer or avoid Australian tax (the attribution rules would apply to tax residents on an accruals basis on their share of income accumulating offshore, ensuring offshore investments are not favoured over domestic investments for taxation reasons).
   
• The Treasury consultation paper only refers to AASB 127 and AASB 131 (joint ventures) in determining control for CFC purposes. There is no mention of related accounting principles such as those governing special purpose vehicles in UIG 112 'Service Concession Arrangements'.
   
• Page 16 of the Treasury consultation paper sets out the definition of control from the AASB 127 but does not make reference to 'so as to obtain benefits from its activities' as in paragraph 4 of AASB 127. It is not clear whether Treasury has inadvertently left out this reference? This raises interesting issues where the CFC is held via a trust structure i.e. the person (beneficiary) that benefits from the activities of the CFC may be different from the person (trustee) that controls the entity itself.
   
• Not all entities would be required to prepare financial accounts in accordance with AASB 127 and thus hypothetically requiring these taxpayers to apply AASB 127 (and or related accounting principles) would not offer much compliance benefit.
   
• If the meaning of 'control' in a CFC context (and therefore an accounting context) was ever litigated, the question is whether the Courts would seek to adopt their own interpretation of the accounting standard, and the practical ramification this might have from a regulatory perspective (e.g. would this mean ASIC would require entities (that would not ordinarily be consolidated) to consolidate for accounting because of how the Courts interpret the accounting standards)?

The G100 suggests that the concept of control should be expressed in the legislation rather than being cross-referenced to AASB 127. It will be a matter for the legislative drafting, as instructed by Treasury, to consider how broad or narrow the concept should be drawn in the legislation. For example, it will probably not be sufficient to merely define control as 'the power to govern the financial and operating policies of an entity so as to obtain benefits from its activities'. It may be necessary to define 'power' including deciding whether a capacity to control in the sense described in the Implementation Guidance mentioned above is necessary. In this regard, we are not convinced that it is necessary given the likely disputes and costs associated with determining whether a capacity to control exists in the circumstances.

Active income test
The proposed rules put forward an exclusion from the CFC regime where passive income arises in the 'ordinary course of the active conduct of a trade or business'. The expression 'active conduct of a trade or business' is briefly defined in very general words.

The G100 is concerned that the meaning to be attributed to the active income test set out in the draft legislative rules is discussed in 12 pages in the Consultation Paper (including 21 examples and various principles). In our view some of the conclusions arrived at in the examples do not follow from the words of the proposed test. Moreover, whilst the Consultation Paper derives 'principles' from the examples it sets out, it is debatable whether all of these principles can be derived from the words of the test. Our main concern is that a court in interpreting the words of the active income test will not feel constrained or influenced by matters contained in an explanatory memorandum. Where a Court believes that the meanings of legislative words in their ordinary sense are clear and unambiguous it is less concerned to discover what the Parliament's intention was as expressed in an explanatory memorandum.

The G100 proposes that, in drafting the active income test, the outcomes desired to be achieved under the active income test (and in this regard a good starting point would be the principles gleaned from the examples set out at page 13 of the Consultation Paper) should also follow from the words of the legislation. This will ensure that a desired outcome will not miscarry because the words of the legislation do not result in that outcome.

A further issue (relating to this exclusion) which is not contemplated in the examples provided in the Consultation Paper, is whether passive income that arises on funds retained in foreign subsidiaries of Australian parents to satisfy foreign regulatory requirements (e.g. local capital requirements of a regulated business) may be viewed as being appropriately related to the active conduct of a trade or business. The G100 recommends that the amended CFC rules expand on the concept of the 'active conduct of a trade or business' currently used in Rule 4-11 to ensure that activities to satisfy regulatory requirements in the country of operation form part of the definition.

Proposed integrity rule is misconceived
Any integrity rule should only apply to protect the Australian tax base where it would otherwise be open to taxpayers to organise their affairs to inappropriately reduce their Australian tax base. In our view the proposed integrity rule goes further than this.

The G100 believes that the proposed integrity rule denies deductions for amounts which, in the hands of a recipient CFC, are characterised under the proposed rules as forming part of its passive income and is not included in the attributable income for the CFC.

The rule applies notwithstanding that, in order to be deductible (leaving aside the proposed integrity rule) the amount would:

1. need to have been incurred in a manner which qualifies for deduction under one or more of the provisions allowing deductions. For instance the amount would need to have been necessarily incurred in gaining or producing the assessable income (or notional assessable income) of the relevant taxpayer (or CFC) (per S.8(1) of the Income Tax Assessment Act, 1997);
    
2. need to satisfy relevant transfer pricing principles set out in Division 13 of Part III of the Income Tax Assessment Act, 1936; and thereby need to be capable of being defended as consideration which would be paid according to the arm's length conventions set out in those provisions;
    
3. in the context of interest (and interest-like) payments by Australian resident taxpayers, need to satisfy Australian thin capitalisation rules contained within Division 820 of the Income Tax Assessment Act, 1997.

Further, the rule does not seek to consider the level of taxation which might apply to the recipient of the income; and thereby in many instances will lead to double taxation.

The clearest example of when the integrity rule may apply is where a CFC of an Australian multinational group lends money to its Australian parent. There are many rational and reasonable commercial reasons why such indebtedness may arise; including:

• to fund the short term working capital requirements of the Australian group. In this instance, it may be prudent for the CFC to temporarily, or even for an extended term, to lend surplus monies to its Australian parent rather than repatriating such amounts as dividends. Dividends represent a permanent return of funds to Australia, which may not be in the best interests of the CFC, and can be subject to foreign dividend withholding tax;
   
• however, the interest expense would not be an allowable deduction to the Australian parent. In addition, withholding tax is normally payable on the interest at 10%. The overall effect of the integrity rule is a tax impost of 10%. This is so notwithstanding that the interest income may be subject to income tax in the country in which the relevant CFC is resident, or is carrying out its business activities. In fact, it is not unusual for the interest income to be subjected to income tax in a foreign country at rates which exceed 30%. In these instances particularly, it is difficult to understand the policy rationale for having the integrity rule operate as currently designed. Conversely, a similar transaction under the existing CFC measures would be tax neutral in most cases; and
   
• many groups choose to access foreign capital markets when raising debt capital which is ultimately destined to be deployed in Australia. The exemption from Australian interest withholding tax in S.128F(8) of the Income Tax Assessment Act, 1936 specifically contemplates that Australian companies can access the exemption where a foreign subsidiary which is wholly-owned by the Australian parent issues the debt interests in a manner which benefits from the public offer test set out in that section. We believe that in this situation it would be absurd for the Australian parent to be denied an interest deduction.

It is unclear whether the integrity rule is proposed to apply where interest is paid to a CFC resident or conducting business in a listed country. The G100 considers that there is no basis for the application of the rule in this instance since currently such interest may be both deductible in Australia under ordinary principles, and not attributable from the relevant CFC.

Moreover, Treasury has indicated that it is not intended that the integrity rule would apply to payments by a resident company to a CFC resident in a listed country. Another example which illustrates unintended consequences of the operation of the integrity rule is its application to captive insurance companies.

In particular, Australian companies using a captive insurance company to address insurance and risk management requirements will be adversely impacted and placed at a competitive and economic disadvantage (i.e. as a result of the denial of Australian tax deductions for insurance premiums paid to offshore captives). We believe that this outcome is contrary to the intention of the CFC reforms.

The proposed CFC rules contain a fundamental shift in what income is taxed and what expenditure is deductible in Australia. Example 18 in the Paper illustrates the operation of a captive and the interaction with the integrity provision ('Integrity rule').

The broad implications arising from Example 18 under the proposed rules can be summarised as follows:

• the income derived by a captive will be considered to be from an 'active business';
• even if the captive was not carrying on an 'active business', the 'Grouping relief' provisions would appear to give rise to the same result (i.e. the captive's income is not taxable in Australia); and
• the Integrity rule will operate to deny the Australian company a tax deduction for the insurance premiums it pays to the captive.

The G100 believes that to the extent that integrity concerns arise, the correct approach would be to attribute the income of the CFC arising from the payment, as is the case under the existing system, so that there is no Australian tax incentive for a taxpayer to divert income to a CFC. Consistently with the goal of simplifying the rules, an exception to the integrity rule should arise where the relevant CFC is a resident of a listed country and the income is not eligible designated concession income; or the income is otherwise subject to tax in a listed country, for instance because of a permanent establishment of the CFC in a listed country.

Treatment of financial intermediaries
Under the existing CFC measures, an exemption (the AFI subsidiary exemption) applies to certain CFCs that carry on a banking or money lending business. However, the exemption is highly prescriptive and does not take account of current practices in the finance industry.

The exemption under the proposed CFC measures has been modernised and will apply to a wider range of banking businesses in accordance with the wide range of services provided by banks. However, the exemption no longer applies to the non-bank sector. It appears that the non-bank sector must rely solely on the active business exemption in order for their income not to be included as passive income. The G100 considers that there is no basis for discriminating against the non-bank financial sector in this exemption and it should be extended to a business of money-lending.

Extension of passive income
The types of passive income and gains to which the proposed CFC measures might apply are generally consistent with the existing CFC measures. That is, it applies to interest and payments in the nature of interest, dividends, rent, royalties annuities and other income from the disposal of passive assets (such as financial arrangements, shares and intellectual property).

However, the scope of the measures has been significantly expanded by the inclusion of profits from 'financial arrangements'. It is not yet clear what type of financial arrangements will be covered. However, it appears that the expression may take its meaning from the provisions of the Tax Act dealing with the taxation of financial arrangements ('TOFA'). If this is the case, like TOFA, the items that would fall within the definition of 'financial arrangement' would be extensive. Many arrangements that are excluded from TOFA do not appear to be excluded from passive income as envisaged under the proposed CFC measures.

The G100 believes that where the TOFA rules exclude an arrangement from being a financial arrangement for the purpose of the application of the TOFA rules, similarly, those arrangements should be excluded from the CFC rules.

Subtractive approach
We understand that the subtractive approach to the attribution of the attributable income of a CFC, canvassed in Attachment A of the Consultation Paper, would, if adopted, be the only basis for the calculation of attributable income.

In our view, the disadvantages of the Subtractive Approach, as identified in Attachment A of the Consultative Document, outweigh the perceived advantages of the approach. The G100 believes that the subtractive approach should not be pursued, or alternatively that it be available only by election.

CFC grouping exemption
Proposed rule 4-30 indicates that a CFC cannot obtain the benefit of the grouping exemption unless that CFC has been a member of the CFC Group throughout the statutory accounting period of the CFC. The restriction also applies where the CFC providing the benefit is not a member of the CFC group for the whole of the statutory accounting period. There is no obvious policy basis for this distinction.

The G100 considers that the grouping exemption should apply at all times where a financial benefit is provided between entities that at the time of provision are members of a CFC group.

Treatment of losses is unclear
Many existing CFCs may have losses from previous transactions that were in respect of the derivation of passive income under the existing CFC measures. It is not yet clear how these losses will be treated under the proposed CFC measures.

We consider that the transitional rules should reasonably and fairly treat existing CFC losses to ensure that no disadvantage is suffered in adopting the revised CFC regime.

Yours sincerely
Group of 100 Inc

Peter Lewis
National President