Public advice and guidance compendium

The submitter notes that stakeholders have expressed views that significant aspects of three recent draft rulings are inconsistent with the existing GSTR 2006/3:

•

Draft Goods and Services Tax Determination GSTD 2018/D1 Goods and services tax: determining the creditable purpose of acquisitions in a credit card issuing business

•

Draft Goods and Services Tax Ruling update GSTR 2004/4DC1 Goods and services tax: assignment of payment streams including under a typical securitisation arrangement, and

•

Draft Goods and Services Tax Ruling GSTR 2019/D1 Goods and services tax: determining the creditable purpose of acquisitions in relation to transaction accounts.

The proposed amendments do not answer specific points of contradiction raised previously by the submitter in relation to the three draft rulings.

Summary of points raised in previous submissions

In previous submissions, the points raised by the submitter include that:

•

Each of the Commissioner's draft views are at odds with the guidance in GSTR 2006/3 concerning the use of direct estimation methods (which at paragraph 35 of GSTR 2006/3 are endorsed as best practice according with the basic principles in paragraph 33 of the Ruling).

•

Practically, it is through a cost allocation approach (by way of a management costing or financial accounting system), which the Commissioner accepts is a 'direct estimation method', that the relevant costs are identified as relating to the various supplies made in the relevant area (that is, the credit card issuing business, the retail lending enterprise or the transaction account business). Furthermore, such systems would not generally differentiate 'fully creditable' acquisitions from others.

•

Paragraphs 92 to 101 of GSTR 2006/3 can be relied upon to determine the application of paragraph 11-15(2)(a) where the cost allocation approach adopted provides an accurate reflection of 'intended use' (in an aggregate sense) of all acquisitions in the relevant business area.

•

Taxpayers' (externally audited) accounting systems have the attributes of being accurate and objective and preclude the capacity for manipulation for GST purposes. Indeed GSTR 2006/3 is clear that the direct estimation method being a cost allocation approach provides an accurate reflection of 'intended use' and its expected that taxpayers would use such a method if available to them.

In particular, previous submissions suggest that stakeholders may be incorrectly interpreting GSTR 2006/3 as providing support for the following propositions:

The use of a direct estimation system (for example, an activity-based costing system) that allocates acquisitions to the level of a business unit means that it is acceptable to determine the connection between acquisitions and supplies at the business unit level for the purpose of paragraph 11-15(2)(a) of the A New Tax System (Goods and Services Tax) Act 1999.^[1]

Once a direct estimation system allocates acquisitions to a business unit that makes both input taxed and taxable or GST-free supplies (for instance, the credit card issuing business), GSTR 2006/3 automatically allows the use of an indirect apportionment method that applies on the assumption that all acquisitions relate to all supplies made in that business unit.

In our view, this type of approach skips to the second step under Division 11 of determining an apportionment method under section 11-30, without undertaking the first step under paragraph 11-15(2)(a) of identifying the relationship between acquisitions and supplies that are input taxed.

It appears that such arguments arise from reading certain paragraphs of GSTR 2006/3 in isolation and disregard the overarching requirement that the method used reflects a fair and reasonable estimate of the intended use of acquisitions.

Therefore paragraphs 35, 38A, 92A, 92B and 101A in the Addendum to GSTR 2006/3 are broadly intended to clarify that:

Where a direct estimation system allocates acquisitions to a business unit that undertakes both input taxed and taxable or GST-free supplies, the use of the direct estimation system will only provide the foundation upon which further apportionment methods must be used to determine the extent of creditable purpose of those acquisitions.

It is a fundamental prerequisite that such an apportionment method is fair and reasonable, in that the method gives a fair reflection of the extent of the relationships between acquisitions and supplies.

The allocation of costs to a business unit that makes input taxed and taxable or GST-free supplies does not give rise to an assumption that once the cost allocation has occurred, all acquisitions relate to all supplies made in that business unit. Instead, the method used must have regard to whether some of the acquisitions only relate to making financial supplies (or equally, to making taxable or GST-free supplies).

Where direct methods are available for use in combination with your direct estimation system, this will best reflect the intended use of the acquisitions.

The proposed amendments do not in the submitter's view make any substantial alterations to the view currently expressed in GSTR 2006/3 (and in many cases they involve reiterations of general principles expressed elsewhere in the Ruling).

If the ATO believes the three draft Rulings do not contradict GSTR 2006/3, what is the rationale for amending GSTR 2006/3?

Given that GSTR 2006/3 is the foundation GST ruling used by all financial supply providers, there needs to be a very clear reason for amending it.

The ATO has not clearly articulated the rationale for the issue of the three draft rulings (GSTD 2018/D1, GSTR 2004/4DC1 and GSTR 2019/D1) - if the three draft rulings do not contradict GSTR 2006/3, what is the rationale for releasing them?

If instead the three draft rulings do (as the submitter firmly believes) contradict GSTR 2006/3, why do the proposed amendments contain little if any substantive change? On the assumption the amendments to GSTR 2006/3 do have the effect of changing the Commissioner's interpretation as currently expressed in GSTR 2006/3, it should be made clear those changes will only apply prospectively.

The general principles in relation to the application of paragraph 11-15(2)(a) are set out in Goods and Services Tax Ruling GSTR 2008/1 Goods and services tax: when do you acquire anything or import goods solely or partly for a creditable purpose? and the general principles in relation to apportionment are set out in GSTR 2006/3.

The reason the changes to GSTR 2006/3 are not substantial is that they are refinements made for the avoidance of any doubt about the ATO's view, in response to comments raised by stakeholders (further explained at Issue 1 of this Compendium).

Prior to the release of the three draft rulings, the ATO had not issued public guidance on the application of paragraph 11-15(2)(a) to common acquisitions in a credit card issuing business, home loans business or transaction accounts business.

These rulings provide additional guidance that complements the general principles in GSTR 2008/1 and GSTR 2006/3. They are intended to provide a clear statement of the ATO's views on the application of paragraph 11-15(2)(a) in these contexts, which provides the foundation that is necessary to determine whether apportionment methods used are fair and reasonable.

Our broader intention is to provide greater certainty and consistency across the industry in terms of the approach for determining the creditable purpose of acquisitions in these specific retail banking contexts.

We agree with the submitter that the amendments are consistent with the views already expressed in GSTR 2006/3. Therefore the Addendum will apply retrospectively.

It is hard to understand what the sentence added to paragraph 36 of the Ruling is intended to achieve, given the 'fair and reasonable' requirement is already expressed in the existing second sentence of that paragraph, and is also developed more fully in the existing paragraph 36 reference to paragraph 34 of the Ruling.

This change removes any potential for doubt that it is a prerequisite for the use of any method discussed in the Explanation part of the Ruling that it is fair and reasonable in your circumstances.

The proposed paragraph 38A appears to be a direct reiteration of paragraphs 26 to 30 of the Ruling.

The proposed paragraph 88A of the Ruling includes the 'expectation' that certain types of data, used as the basis for direct or indirect methods, be updated annually.

Apart from the apparent contradiction between this new requirement and the overarching statement in paragraph 150 of the Ruling, the submitter notes that when GSTR 2006/3 was first introduced, one of the few specific changes from the original Goods and Services Tax Ruling GSTR 2000/22 Goods and services tax: determining the extent of creditable purpose for providers of financial supplies was the removal of a similar annual review process, which was considered at the time to be too prescriptive. The ATO should specifically address the reason for now reversing this long-standing position, and also provide more guidance on situations where annual reviews are required, as well as the nature of those reviews.

The submitter's preference would be to remove the proposed paragraph 88A of the Ruling entirely.

In its current form, the new requirement is unclear, overly prescriptive and will lead to additional uncertainty.

For example, when using an entity wide revenue method it is recognised that Forex and Derivative revenue is more difficult to calculate than other revenue components, even so it must be 'recalculated periodically having regard to the frequency with which source data is refreshed or recalculated within the enterprise' (at paragraph 164 of GSTR 2006/3, see also paragraph 173). It is implicit, that if the most difficult revenue component is updated periodically, the other revenue items would also be updated at least as often.

The purpose of paragraph 88A of the Ruling is to clearly state this requirement. The reason for adding this statement is that we have observed that in some cases entities will adopt a method and use the inputs from the first year, and then not undertake any further updates of the relevant inputs to ensure the method reflects the current circumstances of the enterprise.

The example of updating revenue and transaction count methods annually is consistent with paragraph 164 of the Ruling.

Paragraph 88A of the Ruling does not conflict with paragraph 150 of the Ruling. Paragraph 150 is not referring to the periodic updating of inputs in a method, as it instead refers to monitoring whether the actual method selected continues to be fair and reasonable (for example, does a change in the enterprise mean that transaction count is no longer a suitable method).

To ensure consistency, we have changed paragraph 88A of the Ruling to reflect the wording in paragraph 164 of the Ruling, by adding 'the frequency of these updates will have regard to the frequency with which that source data is refreshed or recalculated within the enterprise'.

On the proposed paragraphs 92A to 92B of the Ruling, we note that the paragraphs add little to the existing statements in paragraphs 114 to 116 of the Ruling, and may in fact be contradictory with the existing paragraphs 92 to 95 of the Ruling which discuss the use of existing costing systems as the basis for direct estimation methods. The proposed additional wording adds to uncertainty around the use of these methods. We note also the use of the highly imprecise term 'observation' in proposed paragraph 92A of the Ruling which creates further uncertainty on the intended status of the factors (bullet points) in that paragraph.

They do not contradict paragraphs 92 to 95 of the Ruling. As the submitter points out, paragraphs 114 to 116 of the Ruling already indicate that direct estimation may need to be combined with other apportionment methods. The inclusion of 92A to 92B of the Ruling makes this point clear at the start of the discussion on direct estimation.

We have removed the term 'observation' to make the meaning clearer: 'Paragraphs 92 to 101A of the Ruling are to be read as being subject to the following propositions:.'