CHEVRON AUSTRALIA HOLDINGS PTY LTD v FC of T

Judges:
Robertson J

Court:
Federal Court, Sydney

MEDIA NEUTRAL CITATION: [2015] FCA 1092

Judgment date: 23 October 2015

Robertson J

Introduction

1. The applicant is Chevron Australia Holdings Pty Ltd ( CAHPL ). These proceedings concern the financial years 2004-2008, inclusive.

2. Central to the proceedings is a Credit Facility Agreement dated 6 June 2003 between CAHPL and ChevronTexaco Funding Corporation ( CFC ) under which CFC agreed to make advances from time to time to CAHPL "in the aggregate the equivalent in Australian Dollars ... of Two Billion Five Hundred


ATC 17790

Million United States Dollars". Interest was payable monthly at a rate equal to "1-month AUD-LIBOR-BBA as determined with respect to each Interest Period +4.14% per annum" and the final maturity date was 30 June 2008 (the Credit Facility Agreement ). The loan was repayable in full after five years but with provision for early repayment at CAHPL's option. CAHPL provided no guarantee to CFC and did not provide to CFC any security over its other assets. CFC was entitled to terminate the Credit Facility Agreement at any time without cause. CAHPL had the right to prepay any advance made to it. CAHPL and CFC are related, each having a common parent, Chevron Corporation ( CVX ), and CFC being a subsidiary of CAHPL. CAHPL and CFC were not dealing with each other at arm's length.

3. The proceedings do not involve the general anti-avoidance provisions in Part IVA of the Income Tax Assessment Act 1936 (Cth) ( ITAA 1936 ). Neither do they involve any allegation that the Credit Facility Agreement was a sham. The proceedings do involve: under the ITAA 1936, the issue of arm's length consideration where a taxpayer, here CAHPL, has acquired property under an international agreement; under the Income Tax Assessment Act 1997 (Cth) ( ITAA 1997 ), the cross-border transfer pricing rules; and the transfer pricing rules in Australia's double tax agreements, particularly the Convention between the Government of Australia and the Government of the United States of America for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with respect to Taxes on Income (6 August 1982,
[1983] ATS 16 (entered into force 31 October 1983)) (the United States convention) .

4. A general procedural history is as follows.

NSD 569 to 578 of 2012

5. These ten tax appeals concern determinations dated 30 April 2010 under s 136AD(3) of the ITAA 1936 for each of the years ended 31 December 2003 to 2007 (2004 to 2008 tax years) inclusive. The determinations were made by Mr Gavin Roberts, an officer in the Large Business and International line of the Australian Taxation Office, acting in the name of Ms Cheryl-Lea Field, an Acting Deputy Commissioner of Taxation. On 20 May 2010, the Commissioner issued notices of amended assessment to CAHPL for each of the 2004 to 2008 tax years inclusive (the 2010 amended assessments ). On 21 May 2010, the Commissioner issued notices of assessment of scheme shortfall penalty (the 2010 penalty assessments ). CAHPL objected against the 2010 amended assessments and the 2010 penalty assessments which were deemed to be disallowed as the Commissioner did not make an objection decision within 60 days of receiving notices under s 14ZYA of the Taxation Administration Act 1953 (Cth). On 20 April 2012, CAHPL filed notices of appeal in this Court.

6. The determinations dated 30 April 2010 took the following form, referring to CAHPL:

DETERMINATIONS MADE PURSUANT TO SUB-SECTION 136AD(3) OF THE INCOME TAX ASSESSMENT ACT 1936 ("THE ACT")

...

I, Cheryl-Lea Field, Acting Deputy Commissioner of Taxation, Large Business and International, in the exercise of the powers and functions delegated to me by the Commissioner of Taxation:

  • 1. find that, for the purposes of paragraph 136AD(3)(a) of the Act, Chevron Holdings Pty Ltd has acquired property under an international agreement;
  • 2. am satisfied for the purposes of paragraph 136AD(3)(b) of the Act, that having regard to
    • (a) the connection between any 2 or more parties to the international agreement; and
    • (b) to (sic) the other relevant circumstances,

      that the parties to the international agreement or any 2 or more of those parties were not dealing at arm's length with each other in relation to the acquisition;

  • 3. find that, for the purposes of paragraph 136AD(3)(c) of the Act, Chevron Holdings Pty Ltd gave or agreed to give consideration in respect of the acquisition and the amount of that consideration (that is, $162,854,342)

    ATC 17791

    exceeded the arms (sic) length consideration in respect of the acquisition (that is, $91,048,496); and
  • 4. determine, for the purposes of paragraph 136AD(3)(d) of the Act, that sub-section 136AD(3) should apply in relation to Chevron Holdings Pty Ltd in relation to the acquisition.

It follows from the above that, for all purposes of the application of the Act in relation to Chevron Holdings Pty Ltd, consideration equal to the arm's length consideration in respect of the acquisition shall be deemed to be the consideration given or agreed to be given by Chevron Holdings Pty Ltd in respect of the acquisition.

Dated the 30th day of April 2010:

Cheryl-Lea Field [Signature of Gavin Roberts] p.p Gavin Roberts

Cheryl-Lea Field

Acting Deputy Commissioner of Taxation

Large Business and International

7. The determinations dated 30 April 2010 were accompanied by a document of the same date entitled "Reasons for Decision to Apply s 136AD of the Income Tax Assessment Act 1936 (the Act)".

NSD 151 to 156 of 2013

8. These six tax appeals concern only the years ended 31 December 2005 to 2007 (2006 to 2008 tax years) inclusive.

9. On 24 October 2012, the Commissioner made determinations under s 815-30 of the ITAA 1997 for each of these years. On 26 October 2012, the Commissioner issued notices of amended assessment to CAHPL for each of the 2006 to 2008 tax years, inclusive (the 2012 amended assessments ). On 26 October 2012, the Commissioner issued notices of assessment of scheme shortfall penalty for these years (the 2012 penalty assessments ). On 31 January 2013, CAHPL filed notices of appeal in this Court. CAHPL also filed notices of a constitutional matter under s 78B of the Judiciary Act 1903 (Cth) with respect to NSD 151 to 156 of 2013.

10. The determinations made on 24 October 2012 took the following form, using the year ended 31 December 2005 as an example:

Determination made pursuant to section 815-30 of Division 815 of the Income Tax Assessment Act 1997

I, Annette Chooi, Deputy Commissioner, Large Business and International, in the exercise of the powers and functions delegated to me by the Commissioner of Taxation, determine under paragraph 815-30(1)(a) of the Income Tax Assessment Act 1997 (ITAA 1997) that the taxable income of Chevron Australia Holdings Pty Ltd ... ("the taxpayer") be increased by the amount of $149,639,013 for the year ended 31 December 2005 (in lieu of the year of income ended 30 June 2006).

I further determine under paragraph 815-30(2)(b) of the ITAA 1997, that the amount of the increase is attributable to a decrease of $149,639,013 in interest deductions of the taxpayer in the year ended 31 December 2005 (in lieu of the year of income ended 30 June 2006).

NSD 440 of 2013

11. On 14 March 2013, CAHPL filed an application for relief under s 39B of the Judiciary Act. This application challenged the validity of the notices of amended assessment dated 20 May 2010 in respect of the years ended 31 December 2005, 31 December 2006 and 31 December 2007. Alternatively, the application sought a declaration that Subdiv 815-A of the ITAA 1997, as introduced by the Tax Laws Amendment (Cross-Border Transfer Pricing) Act (No. 1) 2012 (Cth), and when read with s 815-1 of the Income Tax (Transitional Provisions) Act 1997 (Cth), were not valid laws of the Commonwealth within s 51 of the Constitution and/or were laws with respect to the acquisition of property which contravened s 51(xxxi) of the Constitution and the notices called notices of amended assessment dated 26 October 2012 in respect of the years ended 31 December 2005, 31 December 2006 and 31 December 2007 were not valid notices of assessment under the ITAA 1936 and/or the ITAA 1997.


ATC 17792

The legislation

12. Section 136AD of the ITAA 1936, so far as relevant, was in the following terms.

136AD Arm's length consideration deemed to be received or given

...

  • (3) Where:
    • (a) a taxpayer has acquired property under an international agreement;
    • (b) the Commissioner, having regard to any connection between any 2 or more of the parties to the agreement or to any other relevant circumstances, is satisfied that the parties to the agreement, or any 2 or more of those parties, were not dealing at arm's length with each other in relation to the acquisition;
    • (c) the taxpayer gave or agreed to give consideration in respect of the acquisition and the amount of that consideration exceeded the arm's length consideration in respect of the acquisition; and
    • (d) the Commissioner determines that this subsection should apply in relation to the taxpayer in relation to the acquisition;

    then, for all purposes of the application of this Act in relation to the taxpayer, consideration equal to the arm's length consideration in respect of the acquisition shall be deemed to be the consideration given or agreed to be given by the taxpayer in respect of the acquisition.

  • (4) For the purposes of this section, where, for any reason (including an insufficiency of information available to the Commissioner), it is not possible or not practicable for the Commissioner to ascertain the arm's length consideration in respect of the supply or acquisition of property, the arm's length consideration in respect of the supply or acquisition shall be deemed to be such amount as the Commissioner determines.

Section 136AD(4) is reproduced to provide statutory context for the interpretation of the balance of the provision. No determination under that provision relevant to these proceedings was made by the Commissioner.

13. Relevant definitions were set out in s 136AA of the ITAA 1936, as follows.

(1) In this Division, unless the contrary intention appears:

acquire includes:

  • (a) acquire by way of purchase, exchange, lease, hire or hire-purchase; and
  • (b) obtain, gain or receive.

agreement means any agreement, arrangement, transaction, understanding or scheme, whether formal or informal, whether express or implied and whether or not enforceable, or intended to be enforceable, by legal proceedings.

...

property includes:

  • (a) a chose in action;
  • (b) any estate, interest, right or power, whether at law or in equity, in or over property;
  • (c) any right to receive income; and
  • (d) services.

...

services includes any rights, benefits, privileges or facilities and, without limiting the generality of the foregoing, includes the rights, benefits, privileges or facilities that are, or are to be, provided, granted or conferred under:

  • (a) an agreement for or in relation to:
    • (i) the performance of work (including work of a professional nature);
    • (ii) the provision of, or the use or enjoyment of facilities for, amusement, entertainment, recreation or instruction;
    • (iii) the conferring of rights, benefits or privileges for which consideration is payable in the form of a royalty, tribute, levy or similar exaction; or
    • (iv) the carriage, storage or packaging of any property or the doing of any other act in relation to property;

      ATC 17793

  • (b) an agreement of insurance;
  • (c) an agreement between a banker and a customer of the banker entered into in the course of the carrying on by the banker of the business of banking; or
  • (d) an agreement for or in relation to the lending of moneys.

14. Section 136AC of the ITAA 1936 was in the following terms.

136AC International agreements

For the purposes of this Division, an agreement is an international agreement if:

  • (a) a non-resident supplied or acquired property under the agreement otherwise than in connection with a business carried on in Australia by the non-resident at or through a permanent establishment of the non-resident in Australia; or
  • (b) a resident carrying on a business outside Australia supplied or acquired property under the agreement, being property supplied or acquired in connection with that business; or
  • (c) a taxpayer:
    • (i) supplied or acquired property under the agreement in connection with a business; and
    • (ii) carries on that business in an area covered by an international tax sharing treaty.

15. Section 170 of the ITAA 1936 provided, so far as relevant:

...

(9B) Subject to subsection (9C), nothing in this section prevents the amendment, at any time, of an assessment for the purpose of giving effect to a prescribed provision, a relevant provision, or Subdivision 815-A of the Income Tax Assessment Act 1997.

Note: Subdivision 815-A of the Income Tax Assessment Act 1997 is about cross-border transfer pricing.

(9C) Subsection (9B) does not authorize the Commissioner, for the purpose of giving effect to a prescribed provision or a relevant provision, to amend an assessment made in relation to a taxpayer in relation to a year of income where:

  • (a) in a case where the purpose of the amendment is to give effect to the prescribed provision in relation to the supply or acquisition of property-the prescribed provision has been previously applied, in relation to that supply or acquisition, in making or amending an assessment in relation to the taxpayer in relation to the year of income; or
  • (b) in any other case-the prescribed provision, the relevant provision, or Subdivision 815-A of the Income Tax Assessment Act 1997, as the case may be, has been previously applied, in relation to the same subject matter, in making or amending an assessment in relation to the taxpayer in relation to the year of income.

...

Definitions

(14) In this section, unless the contrary intention appears:

double taxation agreement means an agreement within the meaning of the International Tax Agreements Act 1953.

limited amendment period , for an assessment, means the period within which the Commissioner may amend the assessment:

  • (a) under item 1, 2, 3 or 4 of the table in subsection (1); or
  • (b) under paragraph (3)(a) or (b).

prescribed provision means section 136AD or 136AE.

relevant provision means:

  • (a) a provision of a double taxation agreement that attributes to a permanent establishment or to an enterprise the profits it might be expected to derive if it were independent and dealing at arm's length; or
  • (b) paragraph 7, 8 or 9 of Article 5, or Article 7, of the Taxation Code in Annex G to the Timor Sea Treaty or a provision of any other international tax sharing treaty that corresponds with any of those paragraphs or that Article.


ATC 17794

scheme has the meaning given by subsection 995-1(1) of the Income Tax Assessment Act 1997.

scheme benefit has the meaning given by section 284-150 in Schedule 1 to the Taxation Administration Act 1953.

16. Sections 175 and 177(1) of the ITAA 1936 provided:

175 Validity of assessment

The validity of any assessment shall not be affected by reason that any of the provisions of this Act have not been complied with.

177 Evidence

(1) The production of a notice of assessment, or of a document under the hand of the Commissioner, a Second Commissioner, or a Deputy Commissioner, purporting to be a copy of a notice of assessment, shall be conclusive evidence of the due making of the assessment and, except in proceedings under Part IVC of the Taxation Administration Act 1953 on a review or appeal relating to the assessment, that the amount and all the particulars of the assessment are correct.

17. Section 815-1 of the Income Tax (Transitional Provisions) Act provided as follows:

815-1 Application of Subdivision 815-A of the Income Tax Assessment Act 1997

  • (1) Subdivision 815-A of the Income Tax Assessment Act 1997 applies to income years starting on or after 1 July 2004.
  • (2) However, Subdivision 815-A does not apply to an income year to which Subdivisions 815-B and 815-C of that Act apply.

    Note: For the income years to which Subdivisions 815-B and 815-C apply, see section 815-15 of this Act.

815-5 Cross-border transfer pricing guidance

Despite section 815-20 of the Income Tax Assessment Act 1997, the documents covered by that section for an income year that starts before 1 July 2012 are taken to be as follows:

  • (a) the Model Tax Convention on Income and on Capital, and its Commentaries, as adopted by the Council of the Organisation for Economic Cooperation and Development and last amended before the start of the income year;
  • (b) the Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations, as approved by that Council and last amended before the start of the income year.

18. Section 815-15 of that Act referred to the start date for Subdivs 815-B, 815-C and 815-D of the ITAA 1997 in respect of tax other than withholding tax as the earlier of 1 July 2013 and the day the Tax Laws Amendment (Countering Tax Avoidance and Multinational Profit Shifting) Act 2013 (Cth) received the Royal Assent. That date was 29 June 2013.

19. Subdivision 815-A of the ITAA 1997 was in the following terms.

815-1 What this Subdivision is about

The cross-border transfer pricing rules in this Subdivision are equivalent to, but independent of, the transfer pricing rules in Australia's double tax agreements.

...

Operative provisions

815-5 Object

The object of this Subdivision is to ensure the following amounts are appropriately brought to tax in Australia, consistent with the arm's length principle:

  • (a) profits which would have accrued to an Australian entity if it had been dealing at *arm's length, but, by reason of non-arm's length conditions operating between the entity and its foreign associated entities, have not so accrued;
  • (b) profits which an Australian permanent establishment (within the meaning of the relevant *international tax agreement) of a foreign entity might have been expected to make if it were a distinct and separate entity engaged in the same or similar activities under the same or similar conditions, but dealing wholly independently.

    ATC 17795

815-10 Transfer pricing benefit may be negated

(1) The Commissioner may make a determination mentioned in subsection 815-30(1), in writing, for the purpose of negating a *transfer pricing benefit an entity gets.

Treaty requirement

(2) However, this section only applies to an entity if:

  • (a) the entity gets the *transfer pricing benefit under subsection 815-15(1) at a time when an *international tax agreement containing an *associated enterprises article applies to the entity; or
  • (b) the entity gets the transfer pricing benefit under subsection 815-15(2) at a time when an international tax agreement containing a *business profits article applies to the entity.

815-15 When an entity gets a transfer pricing benefit

Transfer pricing benefit-associated enterprises

(1) An entity gets a transfer pricing benefit if:

  • (a) the entity is an Australian resident; and
  • (b) the requirements in the *associated enterprises article for the application of that article to the entity are met; and
  • (c) an amount of profits which, but for the conditions mentioned in the article, might have been expected to accrue to the entity, has, by reason of those conditions, not so accrued; and
  • (d) had that amount of profits so accrued to the entity:
    • (i) the amount of the taxable income of the entity for an income year would be greater than its actual amount; or
    • (ii) the amount of a tax loss of the entity for an income year would be less than its actual amount; or
    • (iii) the amount of a *net capital loss of the entity for an income year would be less than its actual amount.

    The amount of the transfer pricing benefit is the difference between the amounts mentioned in subparagraph (d)(i), (ii) or (iii) (as the case requires).

Transfer pricing benefit-business profits

(2) A foreign resident entity gets a transfer pricing benefit if:

  • (a) the entity has a permanent establishment (within the meaning of the *international tax agreement) in Australia; and
  • (b) the amount of profits attributed to the permanent establishment falls short of the amount of profits the permanent establishment might be expected to make if it were a distinct and separate entity engaged, and dealing, in the manner mentioned in the *business profits article; and
  • (c) had the profits attributed to the permanent establishment included that shortfall:
    • (i) the amount of the taxable income of the entity for an income year would be greater than its actual amount; or
    • (ii) the amount of a tax loss of the entity for an income year would be less than its actual amount; or
    • (iii) the amount of a *net capital loss of the entity for an income year would be less than its actual amount.

The amount of the transfer pricing benefit is the difference between the amounts mentioned in subparagraph (c)(i), (ii) or (iii) (as the case requires).

Nil amounts

(3) For the purposes of working out whether an entity gets a *transfer pricing benefit, and of negating that benefit under subsection 815-30(1):

  • (a) treat an entity that has no taxable income for an income year as having a taxable income for the year of a nil amount; and
  • (b) treat an entity that has no tax loss for an income year as having a tax loss for the year of a nil amount; and
  • (c) treat an entity that has no *net capital loss for an income year as having a net capital loss for the year of a nil amount.

    ATC 17796

Multiple transfer pricing benefits

(4) To avoid doubt, an entity may get 2 or more *transfer pricing benefits, in one or more income years, in relation to one amount of profits, or one shortfall of profits.

Meaning of associated enterprises article

(5) An associated enterprises article is:

  • (a) Article 9 of the United Kingdom convention (within the meaning of the International Tax Agreements Act 1953); or
  • (b) a corresponding provision of another *international tax agreement.

Meaning of business profits article

(6) A business profits article is:

  • (a) Article 7 of the United Kingdom convention (within the meaning of the International Tax Agreements Act 1953); or
  • (b) a corresponding provision of another *international tax agreement.

815-20 Cross-border transfer pricing guidance

(1) For the purpose of determining the effect this Subdivision has in relation to an entity:

  • (a) work out whether an entity gets a *transfer pricing benefit consistently with the documents covered by this section, to the extent the documents are relevant; and
  • (b) interpret a provision of an *international tax agreement consistently with those documents, to the extent they are relevant.

(2) The documents covered by this section are as follows:

  • (a) the Model Tax Convention on Income and on Capital, and its Commentaries, as adopted by the Council of the Organisation for Economic Cooperation and Development and last amended on 22 July 2010;
  • (b) the Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations, as approved by that Council and last amended on 22 July 2010;
  • (c) a document, or part of a document, prescribed by the regulations for the purposes of this paragraph.

(3) However, a document, or a part of a document, mentioned in paragraph (2)(a) or (b) is not covered by this section if the regulations so prescribe.

(4) Regulations made for the purposes of paragraph (2)(c) or subsection (3) may prescribe different documents or parts of documents for different circumstances.

...

815-30 Determinations negating transfer pricing benefit

(1) The determinations the Commissioner may make are as follows:

  • (a) a determination of an amount by which the taxable income of the entity for an income year is increased;
  • (b) a determination of an amount by which the tax loss of the entity for an income year is decreased;
  • (c) a determination of an amount by which the *net capital loss of the entity for an income year is decreased.
(2) If the Commissioner makes a determination under subsection (1), the determination is taken to be attributable, to the relevant extent, to such of the following as the Commissioner may determine:
  • (a) an increase of a particular amount in assessable income of the entity for an income year under a particular provision of this Act;
  • (b) a decrease of a particular amount in particular deductions of the entity for an income year;
  • (c) an increase of a particular amount in particular capital gains of the entity for an income year;
  • (d) a decrease of a particular amount in particular capital losses of the entity for an income year.
(3) If the Commissioner makes a determination under subsection (1), the Commissioner must make a determination under subsection (2), unless it is not possible or practicable for the Commissioner to do so.

Example: If section 815-25 is relevant in working out the transfer pricing benefit an entity gets, this subsection requires the Commissioner to make a determination relating to the debt deductions of the entity.

(4) Nothing done under subsection (2) affects the validity of a determination made under subsection (1).

(5) The Commissioner may take such action as the Commissioner considers necessary to give effect to a determination under this section.


ATC 17797

(6) The Commissioner must give a copy of a determination under this section to the entity.

(7) A failure to comply with subsection (6) does not affect the validity of the determination.

(8) To avoid doubt, the Commissioner may include all or any determinations under this section in relation to a particular entity, including determinations of different kinds, in the same document.

815-35 Consequential adjustments

Consequential adjustment-associated enterprises

(1) The Commissioner may make a determination under subsection (4) in relation to an entity (the disadvantaged entity ) if:

  • (a) the Commissioner makes a determination under subsection 815-30(1) in relation to a *transfer pricing benefit an entity gets under subsection 815-15(1); and
  • (b) the Commissioner considers that, but for the conditions mentioned in the *associated enterprises article:
    • (i) the amount of the taxable income of the disadvantaged entity for an income year might have been expected to be less than its actual amount; or
    • (ii) the amount of a tax loss of the disadvantaged entity for an income year might have been expected to be greater than its actual amount; or
    • (iii) the amount of a *net capital loss of the disadvantaged entity for an income year might have been expected to be greater than its actual amount; or
    • (iv) an amount of *withholding tax payable in respect of interest or royalties by the disadvantaged entity might have been expected to be less than its actual amount; and
  • (c) the Commissioner considers that it is fair and reasonable that the actual amount mentioned in subparagraph (b)(i), (ii), (iii) or (iv) (as the case requires) be adjusted accordingly.

Consequential adjustment-business profits

(1) The Commissioner may make a determination under subsection (4) in relation to an entity (the disadvantaged entity ) if:

  • (a) the Commissioner makes a determination under subsection 815-30(1) in relation to a *transfer pricing benefit an entity gets under subsection 815-15(2); and
  • (b) the Commissioner considers that, if the permanent establishment were a distinct and separate entity engaged, and dealing, in the manner mentioned in the *business profits article:
    • (i) the amount of the taxable income of the disadvantaged entity for an income year might have been expected to be less than its actual amount; or
    • (ii) the amount of a tax loss of the disadvantaged entity for an income year might have been expected to be greater than its actual amount; or
    • (iii) the amount of a *net capital loss of the disadvantaged entity for an income year might have been expected to be greater than its actual amount; or
    • (iv) an amount of *withholding tax payable in respect of interest or royalties by the disadvantaged entity might have been expected to be less than its actual amount; and
  • (c) the Commissioner considers that it is fair and reasonable that the actual amount mentioned in subparagraph (b)(i), (ii), (iii) or (iv) (as the case requires) be adjusted accordingly.

...

Nil amounts

(3) For the purposes of this section:

  • (a) treat an entity that has no taxable income for an income year as having a taxable income for the year of a nil amount; and
  • (b) treat an entity that has no tax loss for an income year as having a tax loss for the year of a nil amount; and
  • (c) treat an entity that has no *net capital loss for an income year as having a net capital loss for the

    ATC 17798

    year of a nil amount.

Consequential adjustment-determinations

(4) The Commissioner may make one or more of the following determinations, in writing, for the purpose of adjusting an amount as mentioned in paragraph (1)(c) or (2)(c):

  • (a) a determination of an amount by which the taxable income of the disadvantaged entity for an income year is decreased;
  • (b) a determination of an amount by which the tax loss of the disadvantaged entity for an income year is increased;
  • (c) a determination of an amount by which the *net capital loss of the disadvantaged entity for an income year is increased;
  • (d) a determination of an amount by which the *withholding tax payable by the disadvantaged entity in respect of interest or royalties is decreased.

(5) The Commissioner may take such action as the Commissioner considers necessary to give effect to a determination under this section.

(6) The Commissioner must give a copy of a determination under this section to the disadvantaged entity.

(7) A failure to comply with subsection (6) does not affect the validity of the determination.

(8) To avoid doubt, the Commissioner may include all or any determinations under this section in relation to a particular entity, including determinations of different kinds, in the same document.

(9) An entity may give the Commissioner a written request to make a determination under this section relating to the entity. The Commissioner must decide whether or not to grant the request, and give the entity notice of the Commissioner's decision.

(10) If an entity is dissatisfied with the Commissioner's decision, the entity may object, in the manner set out in Part IVC of the Taxation Administration Act 1953, against that decision.

815-40 No double taxation

(1) The amount of a *transfer pricing benefit that is negated under this Subdivision for an entity is not to be taken into account again under another provision of this Act to increase the entity's assessable income, reduce the entity's deductions or reduce a *net capital loss of the entity.

(2) Subsection (1) has effect despite section 136AB of the Income Tax Assessment Act 1936.

20. Section 6-25 of the ITAA 1997, relied on by the respondent Commissioner, was in the following terms:

6-25 Relationships among various rules about ordinary income

(1) Sometimes more than one rule includes an amount in your assessable income:

  • • the same amount may be *ordinary income and may also be included in your assessable income by one or more provisions about assessable income; or
  • • the same amount may be included in your assessable income by more than one provision about assessable income.

For a summary list of the provisions about assessable income, see section 10-5.

However, the amount is included only once in your assessable income for an income year, and is then not included in your assessable income for any other income year.

(2) Unless the contrary intention appears, the provisions of this Act (outside this Part)


ATC 17799

prevail over the rules about *ordinary income.

Note: This Act contains some specific provisions about how far the rules about ordinary income prevail over the other provisions of this Act.

21. Relevant definitions were in s 995-1 of the ITAA 1997, as follows.

...

arm's length : in determining whether parties deal at arm's length , consider any connection between them and any other relevant circumstance.

...

associated enterprises article has the meaning given by subsection 815-15(5).

...

business profits article has the meaning given by subsection 815-15(6).

...

international tax agreement means an agreement (within the meaning of the International Tax Agreements Act 1953) to which that Act gives the force of law.

...

transfer pricing benefit has the meaning given by section 815-15.

22. The International Tax Agreements Act 1953 (Cth) provided in ss 3(1), 3(2) and 3AAA:

3 Interpretation

(1) In this Act:

agreement means a treaty or other agreement described in section 3AAA (about current agreements) or 3AAB (about agreements for earlier periods).

Note: Most of the conventions, protocols and other agreements described in these sections are set out in the Australian Treaty Series. In 2011, the text of an agreement in the Australian Treaty Series was accessible through the Australian Treaties Library on the AustLII website (www.austlii.edu.au)

...

(2) For the purposes of this Act and the Assessment Act, a reference in an agreement to profits of an activity or business shall, in relation to Australian tax, be read, where the context so permits, as a reference to taxable income derived from that activity or business.

3AAA Definitions - current agreements

(1) In this Act:

...

United Kingdom convention means:

  • (a) the Convention between the Government of Australia and the Government of the United Kingdom of Great Britain and Northern Ireland for the avoidance of double taxation and the prevention of fiscal evasion with respect to taxes on income and on capital gains; and
  • (b) the exchange of notes relating to that convention;

each done at Canberra on 21 August 2003.

Note: The text of this convention and notes is set out in Australian Treaty Series 2003 No. 22 (
[2003] ATS 22).

United States convention means the Convention between the Government of Australia and the Government of the United States of America for the avoidance of double taxation and the prevention of fiscal evasion with respect to taxes on income, done at Sydney on 6 August 1982.

Note: The text of this convention is set out in Australian Treaty Series 1983 No. 16 ([1983] ATS 16).

United States protocol (No. 1) means the protocol, done at Canberra on 27 September 2001, amending the United States convention.

Note: The text of this protocol is set out in Australian Treaty Series 2003 No. 14 ([2003] ATS 14).

...

23. Section 4 of the International Tax Agreements Act provided:


ATC 17800

4 Incorporation of Assessment Act

(1) Subject to subsection (2), the Assessment Act is incorporated and shall be read as one with this Act.

Note: An effect of this provision is that people who acquire information under this Act are subject to the confidentiality obligations and exceptions in Division 355 in Schedule 1 to the Taxation Administration Act 1953.

(2) The provisions of this Act have effect notwithstanding anything inconsistent with those provisions contained in the Assessment Act (other than Part IVA of the Income Tax Assessment Act 1936) or in an Act imposing Australian tax.

24. Section 5 of the International Tax Agreements Act provided:

5 Current agreements have the force of law

(1) Subject to this Act, on and after the date of entry into force of a provision of an agreement mentioned below, the provision has the force of law according to its tenor.

The United States convention was such an agreement.

25. Section 6 of the International Tax Agreements Act provided:

6 Convention with United States of America

The United States convention (as amended by the United States protocol (No. 1)) does not subject to Australian tax any interest paid by a resident of Australia to a resident of the United States of America that, apart from that convention, would not be subject to Australian tax.

26. Articles 7 and 9 of the Convention between the Government of Australia and the Government of the United Kingdom of Great Britain and Northern Ireland for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with respect to Taxes on Income and on Capital Gains (21 August 2003,
[2003] ATS 22 (entered into force 17 December 2003)) (the United Kingdom convention ) (see ss 815-15(5) and (6) at [19] above) provided:

ARTICLE 7

Business profits

1 The profits of an enterprise of a Contracting State shall be taxable only in that State unless the enterprise carries on business in the other Contracting State through a permanent establishment situated in that other State. If the enterprise carries on business in that manner, the profits of the enterprise may be taxed in the other State but only so much of them as is attributable to that permanent establishment.

2 Subject to the provisions of paragraph 3 of this Article, where an enterprise of a Contracting State carries on business in the other Contracting State through a permanent establishment situated in that other State, there shall in each Contracting State be attributed to that permanent establishment the profits which it might be expected to make if it were a distinct and separate enterprise engaged in the same or similar activities under the same or similar conditions and dealing wholly independently with the enterprise of which it is a permanent establishment or with other enterprises.

3 In determining the profits of a permanent establishment, there shall be allowed as deductions expenses of the enterprise, being expenses which are incurred for the purposes of the permanent establishment, including executive and general administrative expenses so incurred, whether in the Contracting State in which the permanent establishment is situated or elsewhere.

4 Nothing in this Article shall affect the application of any law of a Contracting State relating to the determination of the tax liability of a person in cases where the information available to the competent authority of that State is inadequate to determine the profits to be attributed to a permanent establishment. In such cases that law shall be applied, having regard to the information that is available, consistently with the principles of this Article.

5 No profits shall be attributed to a permanent establishment by reason of the mere purchase by that permanent establishment of goods or merchandise for the enterprise.

6 Where profits include items of income or gains which are dealt with separately in other Articles of this Convention, then the provisions of those Articles shall not be affected by the provisions of this Article.

7 Nothing in this Article shall affect the operation of any law of a Contracting State relating to tax imposed on profits from insurance with non-residents provided that if the relevant law in force in either Contracting State at the date of signature of this Convention is varied (otherwise than in minor respects so as not to affect its general character) the Contracting States shall consult with each other with a view to agreeing to any amendment of this paragraph that may be appropriate.

...

ARTICLE 9

Associated enterprises

  • 1 Where:
    • (a) an enterprise of a Contracting State participates directly or indirectly in the management, control or capital of an enterprise of the other Contracting State; or
    • (b) the same persons participate directly or indirectly in the management, control or capital of an enterprise of a Contracting State and an enterprise of the other Contracting State;

    and in either case conditions operate between the two enterprises in their commercial or financial relations which differ from those which might be expected to operate between independent enterprises dealing wholly independently with one another, then any profits which might, but for those conditions, have been expected to accrue to one of the enterprises, but, by reason of those conditions, have not so accrued, may be included in the profits of that enterprise and taxed accordingly.

  • 2 Nothing in this Article shall affect the application of any law of a Contracting State relating to the determination of the tax liability of a person in cases where the information available to the competent authority of that State is inadequate to determine the profits accruing to an enterprise. In such cases that law shall be applied, having regard to the information that is available, consistently with the principles of this Article.
  • 3 Where profits on which an enterprise of a Contracting State has been charged to tax in that State are also included, by virtue of the provisions of paragraphs 1 or 2, in the profits of an enterprise of the other Contracting State and charged to tax in that other State, and the profits so included are profits which might have been expected to have accrued to that enterprise of the other State if the conditions operative between the enterprises had been those which might have been expected to have operated between independent enterprises dealing wholly independently with one another, then the first-mentioned State shall make an appropriate adjustment to the amount of tax it has charged on those profits. In determining such adjustment, due regard shall be had to the other provisions of this Convention and the competent authorities of the Contracting States shall if necessary consult each other.

27. Of primary relevance is the United States convention. Articles 7 and 9 were in the following terms:

Article 7

Business profits

  • (1) The business profits of an enterprise of one of the Contracting States shall be taxable only in that State unless the enterprise carries on business in the other Contracting State through a permanent establishment situated therein. If the enterprise carries on business as aforesaid, the business profits of the enterprise may be taxed in the other State but only so much of them as is attributable to that permanent establishment.
  • (2) Subject to the provisions of paragraph (3), where an enterprise of one of the Contracting States carries on business in the other Contracting State through a permanent establishment situated therein, there shall in each Contracting State be attributed to that permanent establishment the business profits which it might be expected to make if it were a distinct and independent enterprise engaged in the same or similar activities under the same or similar conditions and dealing wholly independently with the enterprise of which it is a permanent establishment or with other enterprises with which it deals.
  • (3) In the determination of the business profits of a permanent establishment, there shall be allowed as deductions expenses which are reasonably connected with the profits (including executive and general administrative expenses) and which would be deductible if the permanent establishment were an independent entity which paid those expenses, whether incurred in the Contracting State in which the permanent establishment is situated or elsewhere.
  • (4) No business profits shall be attributed to a permanent establishment by reason of the mere purchase by that permanent establishment of goods or merchandise for the enterprise.
  • (5) For the purposes of the preceding paragraphs of this Article, the business profits to be attributed to the permanent establishment shall be determined by the same method year by year unless there is good and sufficient reason to the contrary.
  • (6) Where business profits include items of income which are dealt with separately in other Articles of this Convention, then the provisions of those Articles shall not be affected by the provisions of this Article.
  • (7) Nothing in this Article shall affect the application of any law of a Contracting State relating to the determination of the tax liability of a person in cases where the information available to the competent authority of that State is inadequate to determine the profits to be attributed to a permanent establishment, provided that, on the basis of the available information, the determination of the profits of the permanent establishment is consistent with the principles stated in this Article.
  • (8) Nothing in this Article shall in a Contracting State prevent the operation in that State of its law relating specifically to the taxation of any person who carries on the business of any form of insurance (as long as that law as in effect on the date of signature of this Convention is not varied otherwise than in minor respects so as not to affect its general character).

...

Article 9

Associated enterprises

1 Where:

  • (a) an enterprise of a Contracting State participates directly or indirectly in the management, control or capital of an enterprise of the other Contracting State; or
  • (b) the same persons participate directly or indirectly in the management, control or capital of an enterprise of a Contracting State and an enterprise of the other Contracting State;

and in either case conditions operate between the two enterprises in their commercial or financial relations which differ from those which might be expected to operate between independent enterprises dealing wholly independently with one another, then any profits which might, but for those conditions, have been expected to accrue to one of the enterprises, but, by reason of those conditions, have not so accrued, may be included in the profits of that enterprise and taxed accordingly.

(2) Where profits on which an enterprise of one of the Contracting States has been charged to tax in that State are also included, by virtue of paragraph (1), in the profits of an enterprise of the other Contracting State and taxed accordingly, and the profits so included are profits which might have been expected to have accrued to that enterprise of the other State if the conditions operative between the enterprises had been those which might have been expected to have operated between independent enterprises dealing wholly independently with one another, then the first-mentioned State shall make an appropriate adjustment to the amount of tax charged on those profits in the first-mentioned State. In determining such an adjustment, due regard shall be had to the other provisions of this Convention and the competent authorities of the Contracting States shall if necessary consult each other.

(3) Nothing in this Article shall affect the application of any law of a Contracting State relating to the determination of the tax liability of a person, including determinations in cases where the information available to the competent authority of that State is inadequate to determine the income to be attributed to an enterprise, provided that, on the basis of the available information, the determination of that tax liability is consistent with the principles stated in this Article.

Structure of applicant's case

28. The applicant, CAHPL, outlined its primary and alternative cases as follows.

29. CAHPL's primary case was that the determinations made under Div 13 of Pt III of the ITAA 1936 were invalid or inoperative and could not be relied upon by the Commissioner in support of the Div 13 2010 amended assessments because:

  • (a) they were made by a person who was not authorised to make them, or;
  • (b) in relation to the 2006 to 2008 years, because they ceased to be operative once the 2012 amended assessments were made under Subdiv 815-A of the ITAA 1997 for those years.

30. If the determinations made under Div 13 of Pt III of the ITAA 1936 were invalid, the Div 13 2010 amended assessments were excessive with the consequence that the only issue in relation to the 2004 and 2005 years was whether a liability arose directly under Art 9 of the United States convention. In that respect CAHPL contended:

  • (i) Art 9 did not confer a separate and independent power to tax; and even if it did,
  • (ii) there were no criteria for liability under Art 9; or
  • (iii) if there were criteria for liability under Art 9, they were not made out in the present case.

CAHPL needed to succeed on only one of those arguments to demonstrate that the amended assessments in respect of the 2004 and 2005 years were excessive.

31. With respect to the


ATC 17801

2006, 2007 and 2008 years , in addition to the issues set out in relation to Art 9 there was a further issue which was whether the determinations under Subdiv 815-A of the ITAA 1997 supported the Subdiv 815-A 2012 amended assessments for those years. The alternative contentions raised by CAHPL in relation to that issue were:
  • (i) Subdiv 815-A was constitutionally invalid; alternatively,
  • (ii) the statutory preconditions for the making of a Subdiv 815-A determination were not satisfied.

CAHPL needed to succeed on only one of those arguments to demonstrate that the amended assessments in the 2006, 2007 and


ATC 17803


ATC 17802

2008 years were excessive.

32. CAHPL's alternative case proceeded from the assumption that the determinations made under Div 13 of Pt III of the ITAA 1936 were valid and that they supported the Div 13 2010 amended assessments. In that case, the first consequence was that the Subdiv 815-A determinations and 2012 amended assessments necessarily fell away because an essential precondition for the making of the Subdiv 815-A determinations did not exist; there could be no "transfer pricing benefit" and the Commissioner could not make a determination under s 815-10 "for the purpose of" negating a transfer pricing benefit that did not exist.

33. Further, in any case, CAHPL submitted that the Div 13 2010 amended assessments, if valid, were still excessive because:

  • (a) the interest paid by CAHPL did not exceed "the arm's length consideration" for the purposes of Div 13 of Pt III of the ITAA 1936; and
  • (b) Art 9 did not confer a separate and independent power to tax; and even if it did, there were no criteria for liability under Art 9, or, if there were criteria for liability under Art 9, they were not made out here. On any view, the Div 13 2010 amended assessments could not be supported by reliance on Art 9.

34. Finally, if, contrary to CAHPL's submissions, both the determinations made under Div 13 of Pt III of the ITAA 1936 and the Subdiv 815-A determinations were valid and simultaneously operative as alternatives, then the Subdiv 815-A 2012 amended assessments of necessity supplanted the Div 13 2010 amended assessments (which thereby ceased to be operative). In that case, the contentions in [31] above were relied upon.

Structure of respondent's case

35. The respondent submitted there was no foundation for the suggestion that "alternative" assessments had been issued for the 2004 to 2008 income years. First, further amended assessments were issued in October 2012 in relation only to the 2006 to 2008 years, so that questions concerning the relationship between the two sets of determinations and assessments applied only to those years. As to those years, 2006 to 2008, there was nothing on the face of the notices of assessment issued in October 2012 to indicate that they were intended to operate otherwise than as amended assessments in the ordinary way so as to amend the existing assessments to operate as altered or added to. The true nature of what was intended was a change to the process of calculation of liability to tax which, in this instance, did not lead to a change in the amount of tax payable.

36. For each of the 2006 to 2008 income years, the 2012 amended assessments became the definitive statement of CAHPL's income tax liability. However, the amendments made in 2010 continued to have effect. They remained extant and incorporated into the 2012 assessments. Further, the Div 13 determinations did not cease to be operative upon the making of the later amended assessments. The Div 13 determinations had not been revoked and therefore, to the extent that they had operation when made, they continued to have that operation. The Div 13 determinations were relied on to support the 2010 amended assessments and they also supported the 2012 amended assessments. The amendment effected in 2012 was the inclusion of determinations under Subdiv 815-A as an additional basis for the existing ascertainment of taxable income.

37. For the income years to which the 2012 amended assessments applied, the Div 13 determinations and the Subdiv 815-A determinations operated in the alternative to each other as support for the 2012 amended assessments. That was the consequence of s 815-40 and s 6-25. Alternatively, if the applicant was correct in submitting that the deeming effect of the Div 13 determinations correspondingly reduced or eliminated the "transfer pricing benefits" upon which the Subdiv 815-A determinations would operate, to that extent the Subdiv 815-A determinations had no work to do and fell away. It might be that, if the Div 13 determinations were given their full effect, the amendments effected in October 2012 would be seen to be unnecessary, but it would not follow that the assessments were excessive. This was an example of the common case where the Commissioner defended an assessment on an alternative basis.

38. As to the Div 13 determinations, the Commissioner submitted that the relevant determinations and the statements of reasons which accompanied them recorded the state of satisfaction referred to in s 136AD(3)(b) and it was not necessary to record the existence of that state of satisfaction in the Appeal Statement.

39. The Commissioner did not submit that the state of satisfaction thus recorded was evidence of the relevant state of satisfaction


ATC 17804

having been personally held by Ms Field. It was accepted that the state of mind which the documents recorded was that of Mr Roberts.

40. It was not in issue that at relevant times Mr Roberts was an Executive Level 2 ( EL2 ) officer and the Commissioner conceded that the instrument of authorisation (described at [41] below) did not confer authority on Mr Roberts to make a determination under s 136AD(3)(d). Otherwise, Mr Roberts had the general authority to form a state of satisfaction as to whether circumstances met a statutory description, such as that required by s 136AD(3)(b). Thus, if the state of satisfaction was an independent and separately examinable determinant of liability to tax, Mr Roberts had the necessary authority to form that state of satisfaction on behalf of the Commissioner or a delegate. On the other hand, if the existence of the relevant state of satisfaction was a factor going only to the validity of a determination under s 136AD(3)(d), it was an aspect of the "due making" of the assessment which was not open to challenge on judicial review grounds in Part IVC proceedings. It was accepted that the relevant instrument of authorisation did not confer authority on Mr Roberts to make determinations under s 136AD(3)(d) and the effect of ss 175 and 177(1) of the ITAA 1936 was therefore critical. Authorities binding on the Court established the position that defects which would render a determination under s 136AD liable to be set aside in judicial review proceedings did not establish the excessiveness of the relevant assessment and were thus irrelevant in proceedings under Part IVC.

41. The instrument of authorisation dated 11 August 2009 authorised all officers from time to time holding or occupying positions or assigned to duties in Large Business and International and/or who exercised powers and functions in relation to any matters arising in Large Business and International to exercise in the name of the person from time to time holding or occupying the position or assigned to the duties of Deputy Commissioner of Taxation, Large Business and International, all the powers and functions delegated to the office of the Deputy Commissioner of Taxation, Large Business and International, and the powers and functions which the Deputy Commissioner of Taxation, Large Business and International, exercised in his or her own right, including those under the Acts listed in Sch 1 and the regulations made under those Acts, subject to the limitations listed in Schedules 2 to 9. Schedule 2 dealt with authorisations for EL2 officers, and subtracted from the authority of those officers, relevantly, the authority to: "make determinations under Division 13 of Part III of the Income Tax Assessment Act 1936 and make decisions on the business profits and associated enterprises articles of international tax agreements and associated treaties relating to profit shifting".

Consideration of the parties' administrative submissions

42. In my opinion, a consideration of the instrument of authorisation shows that Mr Roberts was authorised to form a view as to whether or not a taxpayer had acquired property under an international agreement (s 136AD(3)(a)); to be satisfied (or not) that the parties were not dealing at arm's length with each other in relation to the acquisition (s 136AD(3)(b)); and to form a view on whether the amount of the consideration given or agreed to be given by the taxpayer in respect of the acquisition exceeded the arm's length consideration (s 136AD(3)(c)). Mr Roberts was not authorised to make the determination that the subsection should apply in relation to the taxpayer in relation to the acquisition (s 136AD(3)(d)).

43. Next to be considered is the function of s 136AD(3) and its relationship with ss 175 and 177(1). The judgment in
WR Carpenter Holdings Pty Ltd v Commissioner of Taxation (Cth) [2007] FCAFC 103; 161 FCR 1 at [43] and [48] shows that in a tax appeal the determination under s 136AD(3)(d) is not subject to examination on judicial review grounds, including the ground that the person who made the determination was not authorised to make it. The Full Court said as follows:

[43] [W]here Parliament has exhaustively set out the criteria for liability by reference to objective matters, but has made the application of those criteria dependent upon a step being taken by the Commissioner, the step is procedural in the sense that it is not a step which forms part of the criteria for liability. The due making of such a determination is not subject to examination on judicial review grounds.

...

[48] [T]he Commissioner's determination under s 177F(1) is posited not on the Commissioner's opinion about the tax benefit or that such a benefit results in connection with the scheme, for these are matters of objective fact. They are elements or criteria for liability to tax, but the Commissioner's opinion about them is not. In this sense, the determination is procedural and the due making of it is beyond examination.

44. To the same effect are the earlier observations at [27]-[29] as follows:

[27] Division 13 sets up a number of objectively ascertainable criteria, the satisfaction of which will create liability. Relevantly for present purposes, those are:

  • • an international agreement
  • • between parties not dealing with each other at arm's length
  • • under which property
  • • is supplied
  • • for less than the arm's length consideration in respect of the supply or for no consideration.

[28] In Pt IVC proceedings a taxpayer may challenge, by evidence and argument, the existence of all or any of those criteria. The taxpayer bears the burden of doing so. However, the matters in respect of which the applicants in the present case seek particulars do not concern the existence or otherwise of any of these criteria.

[29] In making the para (d) determination that s 136AD(1) or s 136AD(2) should apply, the Commissioner is not making any finding as to an element or criterion of tax liability.

45. The reasoning of the Full Court in
WR Carpenter Holdings 161 FCR 1, particularly at [48], was expressly approved in
Commissioner of Taxation (Cth) v Trail Bros Steel & Plastics Pty Ltd [2010] FCAFC 94; 186 FCR 410 at [57].

46. The applicant CAHPL relied on what Lindgren J had said at first instance in
WR Carpenter Holdings Pty Ltd v Commissioner of Taxation (Cth) [2006] FCA 1252; 234 ALR 451 at [144]:


ATC 17805

Neither section [177F nor 136AD] makes the existence of any particular state of mind of the Commissioner in relation to the making of the determination, a condition of the power to make it. Sleight [
Commissioner of Taxation v Sleight (2004) 136 FCR 211] should be regarded as establishing that the legislature has revealed an intention that even in an appeal under Pt IVC of the TAA [Taxation Administration Act], the Commissioner's reasoning that led him to make the determination is shielded by s 177(1) of the ITAA 1936 from attack on judicial review grounds as part of the "due making" of the assessment. Of course, the fact itself of the making of the determination goes to the substantive liability to tax: if a determination was not even purportedly made, or if a determination purportedly made was not authorised by the ITAA 1936 because the statutorily prescribed conditions of the enlivening of the power were not satisfied, or, I suggest, failed to satisfy the Hickman principle, the assessment will be shown to be excessive.

47. In the present case, however, there was a determination which was purportedly made, by Mr Roberts. It follows that I reject the applicant's submission that because Mr Roberts was not authorised to make determinations under Div 13, the determinations are a nullity and cannot be relied upon to defend the amended assessments and that as a consequence the Div 13 2010 amended assessments are excessive. I do not accept the submission that what the High Court said in
WR Carpenter Holdings Pty Ltd v Commissioner of Taxation (Cth) [2008] HCA 33; 237 CLR 198 at [40], in referring to the vitiation of a determination by extraneous purposes amounting to jurisdictional error, applies to the present case, as their Honours' reference to
Commissioner of Taxation (Cth) v Futuris Corp Ltd [2008] HCA 32; 237 CLR 146 makes clear. (In Futuris at [25] the High Court referred to tentative or provisional assessments which for that reason do not answer the statutory description in s 175 and which may attract a remedy for jurisdictional error, and to conscious maladministration of the assessment process.) It follows that I also reject the applicant's submission that s 175 is not engaged as the applicant does not seek to impugn the validity of the amended assessment, but rather relies on the invalidity of the determination to demonstrate excessiveness of the assessment based upon it. In my opinion, since s 177(1) establishes, on the production of a notice of assessment, or of a copy of a notice of assessment under the hand of an officer there specified, the "due making" of the assessments, a defect of the kind presently under consideration in a determination under s 136AD(3)(d) which forms part of the making of the assessments does not demonstrate excessiveness of the assessment. Neither, in my opinion, does it lead to the assessments being liable to be set aside in circumstances outside those with which Futuris deals.

48. I regard the cases on s 170 of the ITAA 1936, on which the applicant relied, as distinguishable. In particular, the applicant relied on what was said in
McAndrew v Commissioner of Taxation (Cth) [1956] HCA 62; 98 CLR 263 at 271 concerning s 170(2) which conferred authority on the Commissioner to amend an assessment where the taxpayer had not made to the Commissioner a full and true disclosure of all the material facts necessary for his assessment and there had been an avoidance of tax. Dixon CJ, McTiernan and Webb JJ said, at 271:

But bearing in mind that the word "excessive" relates to the amount of the substantive liability it is not difficult to see that it will extend over the area in which the conditions mentioned in s. 170(2) find a place. For the fulfilment of those conditions goes to the power of the commissioner to impose the liability by amendment. If he cannot amend consistently with s. 170(2) and so increase the amount of the assessment then it must be excessive.

49. In the present case, however, what is excluded is whether the decision to issue a determination was made in accordance with the statutory requirements. It remains to consider whether the assessment is or is not excessive by reference to what may be called the objective facts.

Article 9 of the United States convention

50.


ATC 17806

In relation to Art 9 of the United States convention and the ITAA 1936, the respondent Commissioner submitted that Art 9 operated by itself without s 815. I understood this submission to mean that Art 9 could be relied on also in relation to the amended assessments under the ITAA 1936. Reference was made by the respondent to the decision of the primary judge in
SNF (Australia) Pty Ltd v Commissioner of Taxation (Cth) [2010] FCA 635; 79 ATR 193 and to the decision of the Full Court in
Commissioner of Taxation (Cth) v SNF (Australia) Pty Limited [2011] FCAFC 74; 193 FCR 149.

51. The applicant submitted that Art 9 did not, and could not, confer a separate and independent imposition of taxation on its income (or deemed income). This contention went, in part, to whether Art 9, independently of the transfer pricing provisions in the domestic legislation, could be relied on to support the amended assessments. The applicant submitted that existing authority supported the contention that Art 9 could not, by itself, be so relied on. The applicant referred to the Full Court in
Commissioner of Taxation (Cth) v Lamesa Holdings BV [1997] FCA 785; 77 FCR 597; to
Chong v Commissioner of Taxation (Cth) [2000] FCA 635; 101 FCR 134; to
GE Capital Finance Pty Ltd v Commissioner of Taxation (Cth) [2007] FCA 558; 159 FCR 473; to
Re Roche Products Pty Ltd v Commissioner of Taxation (Cth) [2008] AATA 639; 70 ATR 703; to
Undershaft (No 1) Ltd v Commissioner of Taxation (Cth) [2009] FCA 41; 175 FCR 150; and to the Full Court in
SNF 193 FCR 149.

52. In my opinion, the decisions to which the respondent refers in this respect do not establish a freestanding substantive operation for Art 9. Further, in my opinion, the authorities on which the applicant relies tend strongly against that conclusion. Different considerations arise in relation to the role of Art 9 when considered in the context of Subdiv 815-A of the ITAA 1997. I turn to consider the authorities relied on by the parties.

53. Lamesa concerned the Netherlands-Australia Double Taxation Agreement. The Full Court said that the Agreement substantially concerned allocation of taxing powers. Their Honours said, at 600-601:

The Agreement is an agreement for the avoidance of double taxation and the prevention of fiscal evasion with respect to taxes on income. Although, therefore, the Agreement has this dual object, the Agreement substantially concerns allocation of taxing power.

Thus, as will be seen, the agreement allocates to the State, where business is carried on or through a permanent establishment, the right to tax business profits of that State (Art 7). It allocates to the country of residence the power to tax aircraft and ship profits (Art 8). Sometimes, as with Arts 7 and 8, the power allocated to the jurisdiction named is exclusive. Sometimes, as is the case with interest, both jurisdictions may tax but with a nominated limit of 10% in one (Art 11). The allocation is of the right to tax. There is nothing in the Agreement which compels a jurisdiction to exercise that right. Australia, for example, does not tax "exempt income", although such income could fall within the business profits Article.

So far as the treaty then under consideration was concerned, these observations tend against the submission that a double taxation agreement is a grant of a stand-alone taxing power.

54. In GE Capital Finance, Middleton J said, at [27] and [36]:

  • [27] The USA Double Tax Treaty is one of the many double tax treaties entered into by Australia, and has been entered into for the avoidance of double taxation with respect to taxes on income. To achieve its aim of avoiding double taxation, the USA Double Tax Treaty allocates taxing "rights" between the treaty partners. As with all international treaties to which Australia is a party, it forms part of domestic law only because there is legislation which provides for the treaty to be incorporated into Australian law. The Agreements Act gives the force of law to the various international double taxation agreements scheduled to it.
  • ...
  • [36] It is important to recall that s 3(11) was introduced to amend the Agreements Act and to impact upon the operation of the USA Double Tax Treaty. The Agreements Act and the USA Double Tax Treaty, and in particular Art 7, establish the scope within which the Australian legislature may impose tax. Article 7 provides that in certain circumstances the Contracting State may tax the business profits (which is permissive), but only so much of the business profits as is attributable to the permanent establishment (which involves a prohibition or limitation). Section 3(11) is similarly directed to the ability to impose a tax or the allocation of the power to tax. It is a provision which is to be read and used "for the purpose of determining whether the beneficiary's share of the income may be taxed in Australia" (emphasis added).
  • (Original emphasis.)

This analysis also tends against the respondent's present submission.

55. In Chong, Goldberg J considered the Malaysian tax treaty, in particular Art 18(2). The respondent in that case, the Commissioner, appears to have put a different argument to the one presently advanced. The Commissioner's argument was dealt with by Goldberg J, at [24]-[26]:

  • [24] The respondent submitted that double tax agreements do not allocate the right to tax as such a right already exists by domestic law. Rather, it was said that double tax agreements qualify or limit that right by imposing limitations on the right to tax. The respondent said that a more accurate way to describe the purpose and effect of a double tax agreement was the language used by the Full Court [in Lamesa] at 607 where it said (at 607): "If Art 13 applies, then profit from the alienation is authorised to be taxed in the place where the realty referred to in the Article is."
  • [25] I do not consider that there is a significant difference between the concept of allocating taxing power and authorising the subject to be taxed in this context. When the Full Court in Commissioner of Taxation v Lamesa Holdings BV referred to the allocation of taxing power it was referring to the fact that as between two sovereign States with power to impose taxation on particular persons, receipts and events, the agreement was concerned to identify those areas where such power would not be applied. The Full Court saw the concept of the allocation of taxing power as involving the acceptance, if so agreed, of a limitation on an existing taxing power. This view is demonstrated by Art 23 of the Malaysian Agreement which recognises that the taxation laws of each Contracting State continue to govern the taxation of income in that State except where the Agreement provides otherwise.
  • [26] As a matter of principle it is appropriate to describe the purpose and effect of a double tax agreement, where there are two existing tax systems in two contracting states, as one where areas of taxation are allocated between the two contracting states. The allocation of taxing power in a double tax agreement is predicated on the existence of a sovereign right by a contracting state to impose taxation and the existence of taxation legislation. When one refers to an allocation of taxing power one is doing no more than saying that in an area where both contracting states have the right to impose taxation, and may have already imposed taxation, they have agreed that one contracting state, rather than the other or, as the case may be, both contracting states, shall have the right to impose taxation in that area. Whether one uses the language of allocation of power or the language of limitation of power, the result is the same; there is designated or agreed who shall have the right under the agreement to impose taxation in the particular area.
  • (Emphasis added. Citation omitted.)

56. This approach, with respect, seems to me to be correct and it avoids any difficulties otherwise arising from the terms of s 55 of the Constitution. Section 55 provides that laws imposing taxation shall deal only with the imposition of taxation, and any provision therein dealing with any other matter shall be of no effect and laws imposing taxation shall deal with one subject of taxation only.

57. Undershaft (No 1), concerned whether a capital gain on the sale of shares was not subject to income tax in Australia by reason of Art 5 of the United Kingdom Agreement (a predecessor to the 2003 United Kingdom convention) and whether a capital gain was not subject to income tax in Australia by reason of Art 7 of the Netherlands agreement. Articles 5


ATC 17808

and 7 were referred to as "Business Profits" Articles. If the respective capital gains fell within the Business Profits Articles, Australia was prevented from taxing those capital gains. At [45]-[46], Lindgren J said:
  • [45] A purpose of a DTA [double taxation agreement] is to avoid the potential for the imposition of tax by both of the Contracting States on the same income. It is appropriate to say that the Contracting States achieve their objective by "allocating" as between themselves the right to bring to tax a particular item to one Contracting State while the other State agrees to abstain from doing so (Lamesa 77 FCR at 600;
    Chong v Federal Commissioner of Taxation (2000) 101 FCR 134 at [24]-[27]).
  • [46] A DTA does not give a Contracting State power to tax, or oblige it to tax an amount over which it is allocated the right to tax by the DTA. Rather, a DTA avoids the potential for double taxation by restricting one Contracting State's taxing power.

This latter dictum stands squarely against the respondent's present contention.

58. In
SNF 79 ATR 193 at first instance, Middleton J noted at [18] the Commissioner's submission that the associated enterprises articles of the double taxation agreements there under consideration themselves empowered the Commissioner to assess the taxpayer on profits so included, characterising each double taxation agreement as being a stand-alone taxing power. However, at [19] Middleton J noted that the Commissioner was content to "follow" the taxpayer in relying solely on Div 13, and, his Honour said at [20], in view of that approach he would proceed on the basis that Div 13 contained the statutory provisions the Court needed to interpret and apply and the Court needed to go no further. His Honour observed that a similar approach was taken by Downes J in Re Roche Products. In light of the approach of the primary judge in
SNF 79 ATR 193, one would not expect to find, and one does not find, any consideration of that issue by the Full Court in
SNF 193 FCR 149.

59. In Re Roche Products, Downes J, sitting as the Administrative Appeals Tribunal, dealt with this issue at [189]-[191] as follows:

  • [189] Submissions were put to me on 2 particular matters I have not dealt with so far. The first was whether the double tax treaties as incorporated into Australian law conferred a power to assess. ...
  • [190] So far as the first is concerned I note that the submissions were limited (particularly those of the Commissioner) and both parties accepted that the result in this case would not be affected if the treaties conferred no power to assess. This is because the issues in this case concerned pricing and, to the extent that the double tax treaties relate to profits, the only ultimate relevance of profit was that it reflected prices. Notwithstanding the different tests of independent pricing and arm's length dealing it was accepted that these are essentially the same tests, a proposition which is supported by the OECD Guidelines.
  • [191] In the result I do not need to decide the issue although I note that there is a lot to be said for the proposition that the treaties, even as enacted as part of the law of Australia, do not go past authorising legislation and do not confer power on the Commissioner to assess. They allocate taxing power between the treaty parties rather than conferring any power to assess on the assessing body. On this basis Div 13 of Pt III of the ITAA 1936 should be seen as the relevant legislative enactment pursuant to the power allocated.

60. As to the terms of the Treaty in question, I see nothing in the terms of the United States convention which counts against this conclusion. For example,


ATC 17809

Art 9 speaks of profits which "may be" included in the profits of the enterprise and taxed accordingly. Thus the Article may, as Art 9(3) contemplates, affect the application of any law of a Contracting State relating to the determination of the tax liability of a person.

61. I reject the respondent's submission that Art 9 of the United States convention, independently of the transfer pricing provisions in the domestic legislation, may be relied on to support the 2010 or the 2012 amended assessments.

62. I note that the applicant relied on an affidavit by Professor Harry Rosenbloom who was asked to address the question whether under the law of the United States in the period from 2003 to 2009 Art 9 of the United States convention imposed tax. His conclusion was that under the law of the United States in those years Art 9 did not impose tax. The applicant submitted that it was important to take into account the practice of each Treaty partner in relation to the Treaty by reference to uniformity of practice. The applicant submitted that the fact that the United States regarded the United States convention as not capable of imposing tax was an important fact in concluding whether it could do so from an Australian point of view. I note that Professor Rosenbloom was not required for cross-examination. Nevertheless, I do not find it necessary to rely on Professor Rosenbloom's opinion in reaching my conclusion that Art 9 may not independently of the transfer pricing provisions in the domestic legislation be relied on to support the amended assessments.

Division 13 of the ITAA 1936

63. I proceed to consider s 136AD of the ITAA 1936 in accordance with its terms.

Arm's length consideration

64. I turn to the question of arm's length consideration for the purposes of s 136AD of the ITAA 1936. At [121]-[122] in
SNF 193 FCR 149, the Full Court described as a correct statement what Middleton J had said at first instance at [44]:

The essential task is to determine the arm's length consideration in respect of the acquisition. One way to do this is to find truly comparable transactions involving the acquisition of the same or sufficiently similar products in the same or similar circumstances, where those transactions are undertaken at arm's length, or if not taken at arm's length, where suitable adjustment can be made to determine the arm's length consideration that would have taken place if the acquisition was at arm's length.

65. The Full Court in SNF rejected the Commissioner's submission (set out at [95] of the Full Court's reasons) that the statutory hypothesis was to consist of all of the facts in the real world with one fact changed by the deeming provision, namely, the fact that the taxpayer and its parent were not independent of each other and that this required one to assume a set of facts in which the purchaser was an entity with all of the qualities of the taxpayer except its relationship to the parent manufacturers. At [96]-[99] the Full Court said:

  • [96] The Commissioner's argument hinges, as he submitted in reply, on the proposition that one of the independent parties referred to in the definition of "arm's length consideration" in s 136AA(3)(d) was the taxpayer. This was the case, so he submitted, because the definition provision had to "be read in the context of s 136AD, the operative provision" and that provision commenced "its inquiry with the 'taxpayer' and it is the 'taxpayer' which is the subject of the hypothetical in s 136AA". The three propositions for which the Commissioner contends are, therefore:
    • (a) the definition provision must be read in the context of the operative provision; and
    • (b) the operative provision - s 136AD(3) - begins its inquiry with the taxpayer;
    • (c) therefore the hypothesis required by the definitive provision - s 136AA(3)(d) - must relate to the taxpayer so that "arm's length" in that provision means "arm's length from the taxpayer".
  • [97] The critical words of s 136AA(3)(d) in question are "between independent parties dealing at arm's length". It is the burden of the Commissioner's argument that one of the independent parties to whom the provision refers must necessarily be the taxpayer. That

    ATC 17810

    observation lays bare an ambiguity in the expression "independent parties at arm's length". In the context of ss 136AD(3) and 136AA(3)(d) those words could equally cover any of the following three situations:
    • (a) a purchase by the taxpayer from a hypothetical arm's length supplier;
    • (b) a purchase by a hypothetical purchaser from the taxpayer's actual supplier;
    • (c) a purchase by a hypothetical purchaser from a hypothetical arm's length supplier.
  • [98] That ambiguity underscores the fact that the description of a transaction as being at arm's length is a statement about the independence of two parties from each other. The connexion thus disclosed is a relative one. Generally speaking a statement that two parties have a relative connexion of a particular kind does not carry with it any information about their absolute status. A requirement, for example, that two businesses be more than 20 km apart says nothing about where either business is situated. If one were to look at the definition provision in s 136AA(3)(d) in isolation it would be unsound to read it as requiring any more than that the two parties in question should be independent of each other; that is, the ordinary meaning is not as the Commissioner contends.
  • [99] The question then is whether the ordinary meaning is somehow displaced or modified by the fact that the definition provision feeds into an operative provision - s 136AD(3) - which in turn utilises it to assess the position of the taxpayer. There is no doubt that s 136AD(3) is, as the Commissioner submits, about the taxpayer; that it requires a comparison between that which was actually paid by the taxpayer and an arm's length consideration; and, that, in appropriate circumstances, it then substitutes one for the other. However, it does not follow from acceptance of all those features that arm's length consideration - which does not, in general, refer to the actual position of either party - must be treated as overlaid by a further requirement that the consideration not only be at arm's length but that the arm in question be attached to the taxpayer.

The statutory integers - ITAA 1936

66. I next address in turn each of the statutory integers, so as to provide a framework within which to consider the evidence.

67. Section 136AD(3), set out at [12] above, proceeds by reference, first, to whether a taxpayer has acquired property under an international agreement; second, to whether the parties to the agreement were not dealing at arm's length with each other in relation to the acquisition (of property); and third, to whether the taxpayer gave or agreed to give consideration in respect of the acquisition and the amount of that consideration exceeded the arm's length consideration in respect of the acquisition. It is then that, relevantly, on a determination by the Commissioner that the subsection should apply, consideration equal to the arm's length consideration in respect of the acquisition is to be deemed to be the consideration given or agreed to be given by the taxpayer in respect of the acquisition.

68. Section 136AA(1), set out at [13] above, provides that in Div 13, unless the contrary intention appears, acquire includes "obtain, gain or receive"; agreement means any agreement, arrangement, transaction, understanding or scheme, whether formal or informal, whether express or implied and whether or not enforceable, or intended to be enforceable, by legal proceedings; property includes a chose in action; any estate, interest, right or power, whether at law or in equity, in or over property; any right to receive income; and services; and services includes any rights, benefits, privileges or facilities and, without limiting the generality of the foregoing, includes the rights, benefits, privileges or facilities that are, or are to be provided, granted or conferred under, relevantly, an agreement for or in relation to the lending of moneys.

69. By s 136AA(3), unless the contrary intention appears, "a reference to the arm's length consideration in respect of the acquisition of property is a reference to the consideration that might reasonably be expected to have been given or agreed to be given in respect of the acquisition if the property had been acquired under an agreement between independent parties dealing at arm's length with each other in relation to the acquisition".

70. The parties accepted that the Credit Facility Agreement was an international agreement for the purposes of s 136AC.

71. I do not accept the submission of the applicant that "property" is to be read down or that "services" is to be read down so as to refer, relevantly, only to an agreement for or in relation to the lending of


ATC 17811

moneys, given the non-limiting characteristic of the examples in light of the words "without limiting the generality of the foregoing" in the definition of services in s 136AA(1).

72. There remain, however, a number of questions. Beginning with the operative provisions in s 136AD(3), the taxpayer has acquired property under an international agreement and the question is whether the consideration it gave or agreed to give in respect of the acquisition exceeded the arm's length consideration, that is, the consideration that might reasonably be expected to have been given or agreed to be given in respect of the acquisition if the property had been acquired under an agreement between independent parties dealing at arm's length with each other in relation to the acquisition.

73. Before making the statutory comparison, therefore, it is necessary to identify the property, that is, the rights, benefits, privileges or facilities that are, or are to be provided, granted or conferred by the agreement and acquired by the taxpayer, and, if there has been an agreement between independent parties dealing at arm's length to acquire that property, to proceed by reference to what is the consideration that might reasonably have been expected to have been given or agreed to be given in respect of the acquisition. It is that latter consideration, the consideration equal to the arm's length consideration in respect of the acquisition, that is deemed to be the consideration given or agreed to be


ATC 17812

given by the taxpayer in respect of the acquisition "for all purposes of the application of this Act in relation to the taxpayer". In my opinion, the property CAHPL acquired under the international agreement, the Credit Facility Agreement, was the rights or benefits granted or conferred under that Credit Facility, including the sums lent.

74. The first question, therefore, is whether CAHPL gave or agreed to give consideration and the amount of the consideration exceeded the arm's length consideration in respect of the acquisition. It is for the applicant to show that the amended assessments were excessive: Taxation Administration Act s 14ZZO(b).

75. I do not accept the applicant's submission that the Commissioner has declined to plead as a fact that the state of satisfaction referred to in s 136AD(3)(b) was in fact reached and that the Court should therefore conclude that the state of satisfaction was never duly reached by an authorised officer of the Commissioner. The state of satisfaction is set out in the evidence reproduced at [6] above.

76. What is required, in my opinion, is to depersonalise the agreement to acquire so as to make it, hypothetically, between independent parties dealing at arm's length, but not so as to alter the property acquired. Division 13 of the ITAA 1936 does not, in my opinion, require or authorise the creation of an agreement with terms different from those of the actual agreement, other than the consideration. I reject the Commissioner's submission that the use of the word "an" in s 136AA(3)(d) showed that terms different from those of the actual agreement could be taken into account. In my opinion, the indefinite article is used because the agreement there referred to is the hypothetical agreement. By way of contrast, references earlier in the same paragraph are to "the acquisition of property", "the consideration", "of the acquisition" and "the property".

77. The questions posed by the concept of arm's length consideration "cannot include the requirement of any investigation or consideration ... of motive and purpose":
WR Carpenter Holdings 237 CLR 198 at [38]:
SNF 193 FCR 149 at [7].

78. But the hypothesis must be made to work: it is clear, for example, from the terms of the definition of "property" in s 136AA that property a taxpayer acquires under an international agreement may include the rights, benefits, privileges or facilities that are to be provided under an agreement for or in relation to the lending of moneys. It is not enough, in my opinion, to proceed on the basis that hypothetical independent parties would not have made an agreement on the same terms.

79. One issue of statutory construction between the parties was whether s 136AA(3)(d) required that the borrower as an independent party be considered as a stand-alone company. In my opinion, the answer is "no" as this would be to use the word "independent" as if it meant entirely independent rather than, in a case such as the present, independent of the lender.

80. For present purposes, it is useful to adopt the tool of analysis that, in the hypothetical, the hypothetical independent parties have the characteristics relevant to the pricing of the loan so as to enable the hypothesis to work. Thus, for example, the borrower will be an oil and gas exploration and production ( E&P ) subsidiary.

81. This has implications for the interest rate, which would change depending on the borrower's credit rating (and on the rating of the loan agreement), and on whether or not the interest rate should reflect the absence of security, the absence of guarantee or the absence of covenants given by a borrower for the protection of a lender.

82. Implicit in this approach is that the word "consideration" in s 136AA(3)(d), which is not exhaustively defined but is extended by s 136AA(3)(b), should not be taken to mean more than it otherwise would, in context.


ATC 17813

This means that I do not construe "consideration" in s 136AA(3)(d) to mean the terms of the contract which do no more than affect the interest rate. For example, although the interest rate will be affected by the financial viability of the borrower, I do not construe the word "consideration" as extending to those characteristics.

83. The applicant submitted that the issue was to price the loan, which I understood to mean that the only permissible variable was the interest rate. The primary evidence called by the applicant was directed to the interest rate. The respondent Commissioner submitted that the term "consideration" in the context of legislation dealing with the acquisition of property looked to what was received by the acquiring party "so as to move" the transaction, and could include both money and money's worth:
Archibald Howie Pty Ltd v Commissioner of Stamp Duties (NSW) [1948] HCA 28; 77 CLR 143 at 152 per Dixon J;
Chief Commissioner of State Revenue v Dick Smith Electronics Holdings Pty Ltd [2005] HCA 3; 221 CLR 496 at [71]-[72] per Gummow, Kirby and Hayne JJ. In the context of a loan, it was the promise to repay principal with interest, and any promises to provide security and covenants, which moved the advance of the lender's funds. Each of those promises was to be examined when determining the arm's length "consideration" for the transaction under Div 13. There was nothing in Div 13 which restricted the "consideration" in respect of a loan to the interest rate only. Earlier in his submissions, the respondent contended that the promise to repay with interest was the consideration which flowed from the borrower in exchange for the sum paid to him and cited
McGain v Commissioner of Taxation (Cth) [1965] HCA 41; 112 CLR 523 per Taylor J (upheld in
McGain v Commissioner of Taxation (Cth) [1966] HCA 34; 116 CLR 172 at 173);
ASIC v Great Northern Developments Pty Ltd [2010] NSWSC 1087; 242 FLR 444 at [62] per White J. The respondent referred to the consideration which flowed from the borrower in a loan transaction (that is, the promise to repay with interest and any other covenants) with the consideration which flowed from the lender (that is, the advance of funds). In its very terms s 136AD(3) drew an important distinction between the property acquired and the consideration given. The proper analysis was that CFC paid over USD2.45 billion in return for CAHPL's promise to pay the principal in AUD on maturity and to pay interest:
McGain 112 CLR 523 at 529.

84. In my opinion, neither the context provided by s 136AD(3) nor the non-inclusive definition of "consideration" in s 136AA(3) provides a basis for concluding that the word "consideration" is limited, in the case of a loan, to the interest rate. Insofar as the applicant contended otherwise, I reject that submission.

85. In
McGain 112 CLR 523, Taylor J was considering the definition of "gift" in s 4 of the Gift Duty Assessment Act 1941 (Cth), that word being defined to mean "any disposition of property ... without consideration in money or money's worth passing from the disponee to the disponor, or with such consideration so passing if the consideration is not, or, in the opinion of the Commissioner, is not, fully adequate". The expression "disposition of property" was also defined. Justice Taylor held that a loan of money was a "disposition of property" within the meaning of s 4. In that context, Taylor J said, at 529: "From the point of view of the lender, he has paid over or alienated his money in return for a promise to repay the whole or part of it in money or money's worth." Later in his reasons, at 530-531, Taylor J asked what was the consideration which passed from the company, the borrower. His Honour held that the consideration was not money but a promise by the company to make payments and repayments extending over a period of many years. What the lender got in return for each disposition was a contractual promise enforceable from time to time in accordance with the agreed terms. As such, it was money's worth. On appeal, Barwick CJ, Menzies and Owen JJ rejected the appellant's contention that the loan transactions involved no disposition of property within the meaning of s 4. Their Honours said, at 174, that the real question was whether the payment of money by the lender to the borrower constituted a disposition of property, not whether a loan was a disposition of property. Having concluded that the payments were dispositions of property, the question arose whether such a payment constituted a gift and it then became necessary to look to the consideration for the payment. At 174-175, their Honours said the consideration which passed to the lender from the borrower was a cash payment and a promise to pay the balance of the purchase price over 50 years without interest. What passed at the time of the disposition was, in part, a promise to pay. It could not be said that the promise, together with the money paid, was the equivalent of the property transferred and the inadequacy of the consideration arose from the terms of the promise itself, that is, that the present value of the property, less the cash payment, should be paid over 50 years without interest.

86. In Archibald Howie the High Court was concerned with the Stamp Duties Act 1920 (NSW) and in Dick Smith Electronics with the Duties Act 1997 (NSW). In Archibald Howie, Dixon J held, at 152, that the word "consideration" in s 66 should receive the wider meaning or operation that belonged to it in conveyancing rather than the more precise meaning of the law of simple contracts. Under s 66, his Honour said, the consideration was not involved with offer and acceptance but rather with the money or value passing which moved the conveyance or transfer. In Dick Smith Electronics, the parties accepted that "consideration" in s 21 was not to be read as requiring identification of the consideration sufficient to support a contract. At [71], Gummow, Kirby and Hayne JJ said that so much followed inevitably from the recognition of the fact that s 21(1)(a) (and the expression "the consideration ... for the dutiable transaction") would find application in cases in which a transfer of dutiable property was not made pursuant to contract.

87. Applying ss 136AD(3) and 136AA(3) of the ITAA 1936 in the present case, I accept that the promises by the borrower, CAHPL, were the consideration given or agreed to be given by it. I also accept that the wide definitions of "agreement" and of "acquire" in s 136AA(1) mean that the acquisition of property by a taxpayer under an international agreement is not limited to acquisitions made pursuant to contract. In the present case I find that CAHPL did not give security or operational and financial covenants, which would have affected that part of the consideration which was the interest rate: the interest rate was higher in the absence of those promises or covenants. If the property had been acquired under an agreement between independent parties dealing at arm's length with each other, I find that the borrower would have given such security and operational and financial covenants and the interest rate, as a consequence, would have been lower. The limited scope of the consideration given or agreed to be given by CAHPL resulted in the consideration which CAHPL did give, the promise to pay the interest rate, exceeding the arm's length consideration in respect of the acquisition. It follows, on that basis, that the applicant has not shown that the arm's length consideration assessed by the respondent Commissioner was excessive.

88. However, in my opinion, s 136AD does not require or allow a different agreement to be substituted, on the basis that two independent parties would not have entered into a loan agreement on the same terms as the Credit Facility Agreement. It is not, therefore, an appropriate approach under Div 13 of the ITAA 1936 to ask what the terms, other than the consideration, would have been if CAHPL had been negotiating with a third party lender unrelated to it or if two unrelated parties neither of whom was CAHPL or CFC had been negotiating.

89. I turn now to the


ATC 17814

evidence.

The evidence

90. It is important, in my opinion, to consider the evidence of each witness as it was given and in its context. Because of the variety of complex assumptions, and factual permutations, to take one or two sentences out of their context may be to misstate the evidence. Where the evidence is of limited or no relevance to my ultimate factual conclusions it is referred to more shortly in these reasons.

Applicant's overview of its witnesses

91. The applicant relied on the evidence of two experienced commercial lenders, Mr Eugene Martin and Mr Richard Gross. They prepared reports as to how lenders would go about pricing the loan. Their evidence was supported by Mr Timothy Long formerly of the United States Office of the Comptroller of the Currency ( OCC ). The applicant submitted that those expert reports were evidence in support of its primary case that the Court could and should stop in pricing a loan in accordance with the tests in the legislation. The balance of the applicant's expert evidence was directed to responding to the case or cases put by the respondent Commissioner. First, in a series of further reports, both Mr Gross and Mr Martin responded to criticisms made of them by one of the Commissioner's witnesses and economists, Mr John Hollas, who had had some banking experience. Mr Gross was also criticised by, and responded to, two further witnesses for the Commissioner, Ms Tanya Azarchs and Mr Matthew Taylor. Mr Martin responded to the respondent's other expert with banking experience, Mr John McCormick. (I note that ultimately the Commissioner did not call Mr McCormick to give evidence). Mr Long was criticised by another of the respondent's witnesses, Mr Neil Gaskell, and Mr Long responded to this criticism.

92. Secondly, the applicant responded to the case which the Commissioner put forward on pricing the loan, which asserted that it would be done by reference to how a credit rating agency would rate the actual borrower in this case. In the alternative therefore, the applicant relied upon the evidence of Mr John Thieroff who assigned a BB rating to the applicant on a stand-alone basis. For that purpose, Mr Thieroff was instructed not to take into account parentage. In his second report he was asked to take into account the effect of parentage in the process of a rating's agency rating and that caused him to increase the rating by one notch, resulting in a BB+ rating, which was still non-investment grade. Mr Thieroff also responded to the Commissioner's ratings witnesses, Mr Edward Emmer, who was in management at Standard & Poor's ( S&P ); Mr Taylor, who was a ratings advisor at a merchant bank; and Ms Azarchs who was a financial institutions analyst at S&P. An aspect of Mr Thieroff's evidence was supported by a further ratings witness, Ms Andrea Esposito. Ms Esposito confirmed Mr Thieroff's evidence that caution must be exercised in moving from a non-investment grade rating to an investment grade rating. She also responded to the report of Ms Azarchs.

93. Thirdly, the applicant responded to the respondent's re-characterisation case with four strands of further evidence. In the first strand the applicant relied on the


ATC 17815

evidence of Dr Brian Becker, an economist, who had put on a series of reports responding to the Commissioner's economist, Dr Thomas Horst. Dr Becker also responded to Mr Hollas in another series of reports. Dr Becker was not put forward as a primary or pricing witness, but to demonstrate the deficiencies in the work undertaken by Dr Horst and Mr Hollas to the extent that that was necessary. The second strand in the evidence was that of Ms Caroline Silberztein, the former head of the OECD transfer pricing unit whose report concerned the interpretation of the OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations (the OECD Guidelines ), in response to Dr Horst's reliance on his own construction of those Guidelines. The third strand comprised the evidence of Mr Peter Wasow and Mr Marcus Rowland who gave evidence as to the borrowing practices of oil and gas companies. Also Dr Anthony Webber, an economist, gave evidence about the natural hedging benefit that arose for oil and gas companies borrowing in Australian dollars at the time this loan was made and criticised the evidence of two of the Commissioner's witnesses, Mr Gaskell and Mr McCormick. The fourth strand of evidence was expert accounting evidence. The applicant relied on the evidence of Professor Robert Walker, and he responded to the report of Professor David Boymal, the Commissioner's expert accountant, who opined that the optimal currency of the loan was US dollars.

94. In response to the Commissioner's argument that Art 9 of the United States convention contained an independent power to tax, there was one further leg to the applicant's expert evidence. The evidence of Professor Rosenbloom, formerly international tax counsel at the US Department of Treasury, and the evidence of Mr Leslie Schreyer, who was directly involved in the 14-year negotiation of the United States convention. Neither Professor Rosenbloom nor Mr Schreyer was required for cross-examination. I held that the report of Mr Schreyer was inadmissible (see [318] below.)

95. As to the lay witnesses, the applicant relied primarily on the evidence of Australian Chevron employees, Mr William Dalzell and Mr David Lewis, who gave the background to the loan the subject of the pricing dispute. Mr Dalzell was the person who prepared the report that went to the Board of the applicant for the meeting at which the loan was approved. Further evidence in relation to the background and actual pricing of the loan was given by Ms Sezaneh Taherian who was responsible for communicating with Goldman Sachs and Deutsche Bank engaged to advise on the quantum and price of the loan. The applicant also relied on the evidence of Mr Steven Callaghan, an accountant employed by Chevron Australia. He gave evidence about how the functional currency of the applicant's accounts, Australian dollars, was determined; the structure of the ownership of the assets of the applicant's subsidiaries; and the nature of the revenue of the applicant's subsidiaries. The applicant also relied on the evidence of Mr Paul Oen who was a director of the applicant from its incorporation until November 2001. From 2002 to August 2005 he was the general manager of the Greater Gorgon area. Mr Oen explained the status of that project at the time the loan was entered into. Mr Oen was not required for cross-examination.

Mr Lewis

96. Mr Lewis was called by the applicant. He was a chartered accountant. In March 2000 he joined Chevron Australia as its Australian taxation manager. In June 2004 he relocated to California to work for CVX as a senior tax advisor. He left the Chevron group in late 2005.

97. In his first affidavit, sworn on 17 April 2013, Mr Lewis gave evidence of the execution on 26 June 2002 of an International Restructuring Agreement for the reorganisation of the ownership structure of the Australian subsidiaries of the ChevronTexaco group. As part of that International Restructuring Agreement a new subsidiary of CAHPL, to be known as CFC, was incorporated. It was to become the issuer of the commercial paper in the United States.

98. Mr Lewis participated in an application for a private binding ruling from the Commissioner that the interest payable by CAHPL to CFC would not be subject to interest withholding tax under s 128F of the ITAA 1936. That private binding ruling issued on 2 June 2003.

99. Mr Lewis also gave evidence that at the same time as the reorganisation of the ownership structure of the Australian group was occurring, CVX Treasury was developing a recommendation on the appropriate level of debt for CAHPL. He referred to a recommendation from Treasury in December 2002 that CAHPL incur debt of USD2.5 billion and that such a sum should be raised by CFC and on-lent to CAHPL, leaving that company with a debt to equity ratio of approximately 47%. CFC began raising commercial paper on 2 June 2003. Mr Lewis said that he received an email on or about 3 June 2003 confirming that CVX's Internal Finance Committee had approved an intercompany loan facility between CFC and CAHPL for the AUD equivalent of USD2.5 billion, as well as an intercompany loan facility between CAHPL and Chevron Australia to enable Chevron Australia to repay its existing loans.

100. Mr Lewis then referred to the 6 June 2003 Credit Facility between CFC and CAHPL under which CFC agreed to make advances to CAHPL of the AUD equivalent of USD2.5 billion at an interest rate of 1 month AUD LIBOR BBA +4.14% per annum.

101. Mr Lewis deposed that CAHPL immediately drew down AUD2,180,451,128 under the Credit Facility, which was the AUD equivalent of USD1.45 billion. CAHPL agreed to make advances to Chevron Australia of up to AUD3,034,000,000 at an interest rate of 1 month AUD LIBOR BBA +4.14% per annum. CAHPL used the balance of the funds drawn down from CFC, AUD37,451,128, to make a partial repayment of an outstanding loan to it from Getty Mining Internal Inc ( GMII ). On 26 August 2003, after the completion of the issue by CFC of its second tranche of commercial paper, CAHPL drew down a further AUD1,526,717,557, (the AUD equivalent of USD1 billion). Chevron Australia drew down a further AUD1,091,000,000 under its credit facility with CAHPL. CAHPL used the balance of the funds drawn down, being AUD435,717,557.25, to make a partial repayment of its loan from GMII.

102. Mr Lewis' second affidavit, sworn on 1 August 2013, concerned the currency of the loan made by CFC to CAHPL pursuant to the Credit Facility dated 6 June 2003 being denominated in Australian dollars. Mr Lewis gave evidence that his view had been that, having regard to the size of the debt, no foreign exchange risk for Australian tax purposes should be taken. Given that the borrower was a company resident in Australia whose functional currency was Australian dollars, and given that the debt being repaid was also denominated in Australian dollars, it was his recommendation that the funds be borrowed in Australian dollars. The recommendation on the currency of the proposed loan was made by CVX Treasury.

103. Mr Lewis' third affidavit, sworn on 14 March 2014, also concerned whether the Credit Facility between CAHPL and CFC should be denominated in Australian dollars or in US dollars. He also deposed that the difference between AUD LIBOR and USD LIBOR, the former exceeding the latter in 2002, was not discussed or otherwise raised with him or


ATC 17817

by him as being a reason for preferring one currency over another at the time.

104. Mr Lewis gave oral evidence over an extended period. Very little, in my opinion, was relevant to the present issue but some of it was relevant to penalty. I note that the evidence was given under the umbrella of the objection of the applicant that the issues of purpose and motivation were irrelevant because "at the end of the day we're here to price a loan". As I have said, motive and purpose are not relevant. Mr Lewis gave the following evidence:

The reason why the loan - you say the reason why the loan from CFC was denominated in Australian dollars - they were tax - Australian tax issues that you were concerned with, were they not?---Largely. Yes.

Because you didn't want to - you didn't want any unrealised exchange gains or losses to be taxable in Australia. I'm not critical of that. But that was the reasoning why you wanted - the reason you gave, in fact, for wanting the loan in A dollars?---Sorry. That was one of the reasons. Yes.

And there were other reasons, were there not?---Yes.

Now, a little while ago when I was asking you about the arbitrage and you said, "Well, the - whether the CFC arrangement was cash positive, depended on the interest rate", I think you said, did you not?---Yes.

Yes. And the position was, was it not, that the higher the interest rate on the CFC - CAHPL loan, the greater the cash benefit would be as a - resulting from the CFC structure?---Yes.

So the CVX group - sorry, I will withdraw that - it follows, I suggest to you, that it was in the interests of the CVX group to maximise the rate of interest that was payable under the loan by CAHPL from CFC?---It was in the interests of the group to have a maximum interest deduction. Yes.

And the maximum interest deduction gave the maximum cash positive effects of the CFC arrangement, did it not?---That is true. But that's irrespective of whether that's a CFC arrangement or whether it's any financing into Australia in Australian dollars.

I'm not going to debate that with you. I'm talking about the maximum cash benefit for the group would be obtained by achieving the highest rate reasonably possible - the highest rate of interest reasonably possible?---The maximum benefit would be the highest rate and the highest amount of debt.

Yes. And, of course, that's what you and the tax team in Australia and Chevron treasury were seeking to achieve in the period 2000 to 2003, were you not?--- Sorry. What was it that we were seeking to achieve? Could you ---

What you just told us: the highest interest rate and the highest amount of debt?---No. Not at all, Mr de Wijn.

But you do agree with me that it was in the interest, the financial interests of the Chevron group to have the highest amount of debt possible and the highest interest rate possible?---That would have given the biggest Australian tax deduction.

And the biggest Australian tax deduction would give the biggest net cash benefit, positive benefit, to the group?---That's correct.

...

Mr Lewis, a borrowing at Australian LIBOR or BBSW plus 4.14 per cent would not have been consistent with the group policy of borrowing at the lowest cost of funding, would it?---I believe it would have been consistent with the strategy of optimising the capital structure of the Australian business unit.

And that was because, by borrowing from CFC at LIBOR plus 4.14 per cent, the group got a tax deduction for the interest, and didn't have to pay tax at the other end, and made a net cash benefit out of borrowing at LIBOR plus 4.14 per cent?---I think, very importantly, the group avoided having any exposure to foreign exchange implications for tax purposes.

I'm sorry. I was asking you about the interest rate; I wasn't asking you about foreign exchange?---I would - I would - - -

We will come back to foreign exchange. Mr Lewis, you will just have to listen to the question and answer the question. Borrowing at Australian LIBOR plus 4.14 per cent was not consistent with the group policy of borrowing at the lowest cost of funding?---It was not the lowest interest rate, but I don't believe that's a fair question, to look at that in isolation.

What made borrowing at LIBOR plus 4.14 per cent attractive for the Chevron group was the fact that it got an interest deduction for the nine or 10 per cent that was paid, and didn't have to pay tax on that at the other end. That's right, isn't it?---The group had, for many years, been financed in Australian dollars, and was continuing to be financed in Australian dollars.

I'm not asking you about Australian dollars. I'm asking you about the interest rate?---Well, yes. Factually, the interest rate on an Australian dollar loan was higher at that stage than a US dollar loan.

Well, not only was it higher, there was a margin of 4.14 per cent and the group was able to borrow at close to LIBOR as a result of its AA rating. You understand that, don't you?---Yes.

Yes. So borrowing at a margin of 4.14 per cent above any LIBOR was not consistent with the group policy of borrowing at the lowest cost to funding, was it?---I'm not sure if we're comparing like instruments at this stage.

...

Now, I assume that you, as tax manager of the Australian business unit, would have been concerned to make sure that for transfer pricing purposes the loan between CAHPL and CFC was appropriately priced?---Yes. All cross-border transactions needed to be appropriately priced.

But you were concerned, in particular, that that loan be appropriately priced?---That's correct.

But you were also concerned to try and get the interest rate as high as reasonably possible?---I don't think that was a concern. The concern was to get the appropriate interest rate.

Okay. Now, what steps did you take to ensure that the debt was appropriately priced?---Well, working with corporation treasury, outside advisers were engaged, and Goldman Sachs' report flowed through from that, which recommended a debt level and an interest rate, and, in addition, I sought advice from PricewaterhouseCoopers to ensure that the interest rate would satisfy the ATO's transfer pricing rules under the four step process.

You don't refer to that advice anywhere in your affidavit?---Noted.

The reason for that?---No, there's no reason for that.

The Goldman Sachs report dated 16 January 2003 described itself as discussion materials. The conclusions in that report were not relied on by the applicant but, as I understood it, the fact that it was obtained was pointed to as a matter of history.

105. I accept Mr Lewis' evidence as to the circumstances in which the Credit Facility was made, the amounts drawn down under it and his reasons for recommending that the funds be borrowed in Australian dollars.

Mr Dalzell

106. Mr Dalzell was a director and the finance and compliance manager of the applicant, retiring from his role as finance and compliance manager on 31 July 2013. He was a chartered accountant. In April 1999 he became finance manager of Chevron Australia Pty Ltd ( CAPL ) (a subsidiary of CAHPL) and in that role he was and continued to be responsible for the finance department which encompassed Treasury, finance, taxation and compliance for the Australian operations of the Chevron group. He swore three affidavits in the proceedings.

107. In his first affidavit, sworn on 16 April 2013, Mr Dalzell gave evidence about the incorporation of a new subsidiary of CAHPL to be known as CFC. CFC was to become the issuer of the commercial paper in the United States. Mr Dalzell said he was aware that the issuer of commercial paper in the United States had to be a US resident entity to attract US investors and that the commercial paper had to be denominated in US dollars. He recalled a conversation in which Mr Lewis told him that if the commercial paper were issued by a United State subsidiary of the Australian group rather than by a direct subsidiary of CVX, and the proceeds on-lent to the Australian group, Australian withholding tax would not be payable on the interest paid by the Australian group to the issuer of the commercial paper. Mr Dalzell recalled that he, together with his fellow directors of CAHPL, accepted this advice.

108. Mr Dalzell also gave evidence about discussing with Mr Lewis the currency of the loan and, in particular, that the loan to CAHPL would be in Australian dollars to prevent foreign exchange gains and losses arising in CAHPL. He also said he was aware that the obligations of the issuer of the commercial paper had to be guaranteed by CVX to ensure that the paper was rated by the credit rating agencies as highly as possible. At that time, he said, he understood the Chevron group raised funds from the commercial paper market in the United States because it was, for the group, the cheapest means of borrowing money. Also, in order to ensure that funds could be raised at the cheapest rate, it was his understanding that the issuer of the commercial paper required a guarantee from CVX.

109. Mr Dalzell gave evidence about Goldman Sachs and Deutsche Bank being asked to assist


ATC 17819


ATC 17818

in determining the sustainable debt capacity of CAHPL for the purpose of determining the appropriate interest rate should that amount be lent to CAHPL by CFC.

110. Mr Dalzell also gave evidence that in about December 2002 he had been told by Ms Taherian or Mr Howard Sheppard from CVX Treasury that Goldman Sachs had advised CVX that the appropriate interest rate for a loan of USD2.5 billion to CAHPL was AUD LIBOR +4.14%.

111. Mr Dalzell gave evidence that because CFC had issued commercial paper denominated in US dollars but had provided an Australian dollar loan facility to CAHPL, CFC was subject to the risk of movements in the value of the AUD/USD exchange rate. He also recalled that CFC did not enter into any form of hedging arrangement, as result of the United States Securities and Exchange Commission ( SEC ) rules, the details of which he did not recall.

112. In his second affidavit, sworn on 31 July 2013, Mr Dalzell said that to the best of his knowledge none of the wholly owned Australian subsidiaries of the Chevron group had been rated by a credit rating agency such as S&P or Moody's: an occasion for seeking such a credit rating had never arisen. Where it was necessary to determine interest rates on intercompany loans, Treasury personnel in the United States would suggest the interest rate. He adopted the interest rates that were recommended by Treasury on intercompany loans.

113. Mr Dalzell said that the purpose of the proposed loan from CFC to CAHPL was to refinance existing Australasian Business Unit debt denominated in Australian dollars. That existing debt had been previously incurred to fund a return of capital denominated in Australian dollars and to fund the acquisition of Texaco Australia Pty Ltd ( TAPL ). The proposed loan, refinancing existing Australian dollar loans, was denominated in Australian dollars. Mr Dalzell said he would not have been in favour of a US dollar denominated loan because exchange rate movements between the Australian dollar and the US dollar could have given rise to taxable foreign exchange gains and/or losses in Australia. The quantum of a foreign exchange gain or loss on the USD2.5 billion loan could be a very significant sum over five years.

114. In his third affidavit, sworn on 13 March 2014, Mr Dalzell stated that before the Credit Facility between CAHPL and CFC was entered into, all of the debt funding in Australia was denominated in Australian dollars. CAHPL's annual financial reports were required to be presented in Australian currency by accounting standard AASB 1034 "Financial Report Presentation and Disclosures". Because CAHPL's accounts had to be prepared in Australian dollars, if CAHPL had


ATC 17820

liabilities in US dollars, the Australian dollar value of those US dollar liabilities had to be calculated at the end of each financial year. Any movement in the Australian dollar value of those liabilities over the financial year was required to be recognised in CAHPL's profit and loss statement. Mr Dalzell's state of mind in 2002-2003 was that an Australian dollar denominated credit facility seemed to be an appropriate choice for an Australian company with Australian dollar denominated accounts. It also made sense because it was refinancing existing Australian dollar loans and the overwhelming majority of the CAHPL consolidated accounting group's operating expenses were in Australian dollars. He was concerned that if the Credit Facility between CAHPL and CFC were denominated in US dollars, the currency fluctuations on a US dollar liability equivalent to AUD3,707,168,685.25 could have a significant impact on the annual accounting profits of CAHPL for the duration of the Credit Facility. Up to 2002-2003, the movements in the value of the Australian dollar against the US dollar had been volatile.

115. Mr Dalzell said that in considering the proposed Credit Facility he was concerned that if the exchange rate of the Australian dollar depreciated against the US dollar, unrealised foreign exchange losses on a US dollar denominated borrowing would have reduced the profits out of which dividends could have been paid by CAHPL to its shareholder. It was his intention that CAHPL should pay significant dividends during the term of the Credit Facility. As such, he thought it was important to avoid potential foreign exchange losses that could prejudice CAHPL's earnings. Mr Dalzell said that CAHPL did pay significant dividends in the years 2004 through to 2010.

116. Mr Dalzell also referred to discussions with Mr Lewis about the uncertainty of the future Australian taxation treatment of unrealised foreign exchange gains and losses and the then proposed Australian tax reforms whereby CAHPL would have been required annually to include unrealised foreign exchange movements on a US dollar denominated borrowing in the determination of its taxable income. Mr Dalzell said that he was concerned that in a given year the liability could have been hundreds of millions of dollars. He said that Mr Lewis recommended, and he, Mr Dalzell, agreed, that the Credit Facility between CAHPL and CFC should be denominated in Australian dollars. Although he was generally familiar with the fact that the AUD LIBOR was higher at that time than the USD LIBOR, this played no part in his decision to accept Mr Lewis' recommendation that the Credit Facility should be denominated in Australian dollars. Mr Dalzell said that at no stage did Mr Lewis or anyone else propose that the Credit Facility should be denominated in Australian dollars because the AUD LIBOR exceeded the USD LIBOR.

117. Mr Dalzell said that after the directors of CAHPL approved the Credit Facility, bank accounts were opened with Citibank in the United States for CAHPL and for CAPL in accordance with Chevron group policy and practice. Consequently, on 6 June 2003 CAHPL drew down on the Credit Facility with CFC and a credit of USD1.45 billion was made to CAHPL's Citibank specified account number, and on 26 August 2003 CAHPL drew down on the Credit Facility and a credit of USD1 billion was made to the same CAHPL Citibank account number. Mr Dalzell said that a purchase of AUD3,707 million in Australian currency would have incurred significant bank fees. Interest payments under the Credit Facility were similarly effected through debits to the US dollar bank account. The amounts of the interest payments made to CFC were calculated by reference to the AUD principal amount borrowed, being the AUD equivalent of the aggregate amount of the drawdowns of USD1.45 billion and USD1 billion. The interest payments were initially funded by short-term interest-free advances from Chevron Overseas Petroleum Inc, a division of Chevron USA Inc and the umbrella entity for CVX's overseas operations, including Australia.

118. Mr Dalzell recalled that the maturity of the Credit Facility between CAHPL and CFC was extended beyond the initial term of five years and the principal amount that was ultimately repaid on 29 December 2010 was the USD equivalent of AUD3,707,168,685.25. At that time the USD equivalent of AUD3.7 billion was approximately USD3.75 billion. The repayment was effected by a debit to the US dollar bank account of CAHPL. The amount debited to CAHPL's bank account was USD3.84 billion, comprised of the principal repayment equivalent to a debit of USD3,746,464,673.31 (being the USD equivalent of the AUD liability to pay principal of AUD3,707,168,685.25 translated into USD using the Reserve Bank of Australia exchange rate of the AUD against the USD on that day of USD1.0106), an interest payment equivalent to a debit of USD22,499,081.39 and an overpayment of USD71,036,245.30. The repayment of the principal was funded by borrowing from CAHPL's then immediate parent company Chevron Australia Petroleum Company under the terms of a credit agreement dated 17 November 2009.

119. In cross-examination, Mr Dalzell agreed that CFC did not have any CAHPL directors on its Board; that CFC did not have any staff of its own; that CVX Treasury had the lion's share of the management of CFC; and that all of the activities that resulted in CFC doing anything were originated from CVX Treasury in the United States and the decisions that were made by CFC were made by CVX Treasury staff. CAHPL was not involved in the commercial paper raising in the United States.

120. As to the currency of the loan, Mr Dalzell agreed that it made no difference to the CVX group whether the loan from CFC to CAHPL was in US dollars or in Australian dollars. He also agreed there was never any debate about the currency in which the third party borrowing should take place and that was always to be US dollars. CFC raised funds on the commercial market at close to LIBOR, whatever the AA rating would grant, which was significantly less than the LIBOR +4.14% which it was resolved would be payable on the CFC loan to CAHPL.

121. The following exchanges occurred in cross-examination:


ATC 17821

You were aware, were you not, that the gearing of the balance sheet and the payment of interest on the increased borrowings, resulted in net cash advantages to the CVX group and to CAHPL?---Yes.

Yes. So that the higher the interest rate on the CAHPL CFC loan the higher the cash benefit that CAHPL made and the CVX group made?---Yes.

And that was because the interest on the loan from CFC was, as you understood it, tax deductable in Australia?---Yes.

And not taxable in the US?---I can't recall whether it was taxable in the US.

...

... My question was that at the time the CFC arrangement was put in place, it was your - I'm suggesting to you that it was your understanding that the interest that was payable on the CFC CAHPL loan would not be taxable in the US, either because it was just not taxable or because there was a deduction for an amount and an equivalent amount to set it off?---I can't say that.

You can't say that? Do you recall using the word "arbitrage" in relation to this arrangement?---Yes.

And the arbitrage was to give rise to significant cash benefits for the CAPL group, was it not?---Potentially. Yes.

And that was because the interest would be deductable in Australia and any profit made by CFC ..... was returned by way of dividend to CAHPL?---Yes.

And that dividend was tax-free in Australia?---Yes.

And what I'm suggesting to you was that the - you were fully aware of the fact that there was a positive cash benefit to CAHPL as a result of the CFC arrangement?---Yes.

And that you were fully aware that the higher the interest rate, the higher the cash benefit to CAHPL?---Yes.

And what I'm suggesting to you that it was in the interests - I'm not asking you about purpose at the moment - but what I'm suggesting to you was that it was in the interests of CAHPL, in its commercial interests, to ensure that the debt was maximised and the interest rate on the debt was also maximised, is that right?---You can say that. But that was not the consideration. The consideration we had was what was the reasonable amount of debt and what was an arm's length interest rate that would be applicable.

...

I'm suggesting to you that you were aware in 2002 and 2003 that the effect of maximising the amount of debt and maximising the interest rate payable on that debt was to maximise the commercial benefit accruing to CAHPL?---When you talk of maximising, I - whatever the debt was and whatever the interest rate was, the interest would be deductible.

Okay?---I appreciate that.

I will do it - - -?---But I don't recall any situation where we were maximising either the value of the debt or the interest rate.

I will do it a different way. I - the position is that CAHPL's commercial interests were advanced by having a higher, rather than a lower, level of debt. Do you agree with that?---Yes.

And do you agree with me that CAHPL's commercial interests were advanced by having a higher, rather than a lower, interest rate?---Yes.

And on the assumption that the interest paid to CFC was not taxable in the US - either because it either wasn't taxable or there was an amount brought to account as assessable and an amount deductable which set each other off - if you assume that, I'm suggesting to you that it was also in the interest - the commercial interests of the CVX group to have a higher level of debt rather than a lower level of debt?---If I take your assumption, yes.

Yes. And I'm suggesting to you that based on that assumption, it was also in the commercial interests of the CVX group to have a higher, rather than a lower, level of interest rate?---Yes.

...

[Mr Dalzell was shown an email dated 12 April 2002 from Mr David Krattebol to Ms Rhonda Zygocki.]

And it's an email that you were copied in on? David Krattebol was the CVX group treasurer, was he not?---He was.

And, ultimately, he was the person that called the shots on how the Australian business unit was to be financed?---Yes.

And I'm suggesting that the objective was, as he sets out in the second sentence of the email, to obtain the lowest cost of funding Mr Krattebol's objective?---Yes.

And that was - I'm suggesting that was achieved in this case by CFC raising commercial paper at 1.2 per cent?---Are you asking me?

Yes?---Yes. Yes.

Yes. And that obviously wouldn't be achieved - that is obtaining the lowest cost of funding for the group - would not be achieved by borrowing of Australian LIBOR plus 4.14 per cent, would it?---No.

... I suggested to you that the only recommendation that came to CAHPL or the Australian business unit from CVX treasury was the recommendation at page 921. Do you agree with that?---Yes.

And you will see the recommendation is contained in the second last paragraph?---Yes.

Continuing:

Consistent with our understanding of the capital structure of our peers and cash flow from Australian operations, for the most tax efficient corporate capital structure we recommend CAHPL incur $2.5 billion of total debt.

Do you see that?---Yes.

That's the only recommendation that CAHPL received from CVX treasury, is it not?---Yes.

Not phrased in the terms of sustainable debt at all?---No.

Now - and I suggest to you that it was the most tax efficient corporate capital structure because it gave rise to a large tax deduction in the hands of CAHPL, no tax was payable in the hands of CFC and the profit made by CFC was able to be returned to CAHPL by way of tax-free dividend. Do you agree with that?---Yes.

And I suggest to you that it was the most efficient - the most tax efficient corporate capital structure because, as I think you agreed with me before, a higher principal amount - rather than a lower principal amount - resulted in a higher commercial benefit for CAHPL?---Yes.

And a higher, rather than lower, interest rate resulted in a higher, rather than lower, commercial benefit for CAHPL?---Yes.

122. Mr Dalzell was taken to the Deutsche Bank November 2002 discussion materials. That document set out Deutsche Bank's understanding of ChevronTexaco Corporation's key objectives as follows:

  • - introduction of an appropriate/optimal capital structure
  • - maximise sustainable leverages within appropriate constraints with the objective to repatriate cash to the US
  • - no granting of security
  • - no guarantees from parent companies
  • - AUD denominated floating-rate obligation
  • - minimise covenants
  • - non-amortising

123. Mr Dalzell agreed that the effect of not granting security would have been to push up the interest rate rather than push it down; that providing no guarantee from the parent also had the effect of pushing the interest rate up rather than down, although he added that there would have been some sort of guarantee fee had that been a parent company guarantee but he did not believe that CFC paid CVX a guarantee fee. He agreed the effect of denominating the loan in Australian dollars as opposed to US dollars meant that in 2002 or 2003 the interest rate would be higher than if it had been denominated in US dollars; he agreed that minimising covenants on the loan would have the effect of pushing up the interest rate rather than pushing it down; and that one of the effects of having the loan as non-amortising would be to push up the interest rate rather than push it down. He also agreed that the advice from Deutsche Bank was to support an interest rate for transfer pricing purposes. Mr Dalzell also agreed that assuming a credit rating of BB would lead to a higher rate of interest than assuming a higher credit rating. Mr Dalzell also agreed that the loan agreement from CFC to CAHPL did not contain covenants that would be customary for similar publicly traded debt; that the loan agreement that was entered into was covenant-lite; and that having fewer covenants rather than more covenants would tend to push up the interest rate rather than push it down. Mr Dalzell also agreed that a non-investment grade rating gave a much higher interest rate than an investment-grade rating.

124. Mr Dalzell also accepted that Treasury recommended AUD LIBOR +4.14% and that is what he did. The following exchange occurred:

You don't produce an email or any document from treasury indicating that interest range?---No. It wasn't a range. Treasury recommended LIBOR - A dollar LIBOR plus 4.14.

Okay. So that's the recommendation you got from Treasury and that's what you did?---Yes.

You were told to do - you were told to borrow 2.5 billion and you were told to pay Australian LIBOR plus 4.14 per cent and that's what you did?---It was recommended that we do it, yes.

And that's what you did?---Yes.

125. Mr Dalzell accepted that the arbitrage was the cash benefit to CAHPL and the CVX group that arose from the payment of interest at AUD LIBOR +4.14% to CFC and on the assumption that CAHPL did not receive any dividends from CFC a borrowing of USD2.5 billion in AUD at an interest rate of 8.97% was not sustainable. He agreed that he had in fact assumed that the dividends would be available from CFC. Mr Dalzell also accepted that if CAHPL was borrowing from an independent party USD2.5 billion or the AUD equivalent at 8.97%, it would not be sustainable and that the borrowing at 8.97% was only sustainable because of the arbitrage received back.

126. Mr Dalzell agreed that at least as at November 2002 he would have been aware that


ATC 17822

one of the benefits of the loan being in Australian dollars was to create an interest rate margin and that the margin would not be subject to tax in the United States and that the dividends coming back would not be taxable in Australia.

127. I accept Mr Dalzell's evidence.


ATC 17823

Mr Callaghan

128. Mr Callaghan was an accountant, Manager, upstream accounting policy, employed by CAPL. From 2002 to 2008 he reported to Mr Dalzell. His evidence concerned how the functional currency of the applicant's accounts (Australian dollars) was determined; the structure of the ownership of the assets of the applicant's subsidiaries; and the nature of the revenue of the applicant's subsidiaries. He swore two affidavits in the proceedings.

129. In his first affidavit, sworn on 14 March 2014, Mr Callaghan said that for the year ended 31 December 2003, approximately 88% of the Chevron Australia Consolidated Group revenues from the sale of goods was denominated in USD. Those revenues were typically generated by around 150 to 200 USD sales of crude oil and LNG.


ATC 17824

The other 12% of the Chevron Australia Consolidated Group revenues from the sale of natural gas and LPG was denominated in AUD. In addition, approximately 97% of Chevron Australia Consolidated Group expenses - typically over 5000 expense transactions per year - were incurred in AUD, with another 1% of expenses being denominated in USD and the remaining 2% in other foreign currencies.

130. Mr Callaghan said that CFC issued commercial paper in the United States and on-lent the AUD equivalent of those funds to CAHPL and that CAHPL was liable to pay interest and to repay principle in AUD. He said that because CAHPL's functional currency was AUD, if CAHPL had borrowed in USD, any movement in the value of the USD as against the AUD would have had to be recorded in CAHPL's statement of financial performance. Because of the size of CAHPL's loan from CFC, if CAHPL's borrowing from CFC had been denominated in USD, CAHPL's statement of financial performance would have disclosed significant volatility.

131. Mr Callaghan also deposed that as the financial forecasting and reporting manager for Chevron Australia during 2002 to 2008, he was responsible for the coordination of annual reporting to CVX's head office on the "standardised measure of discounted future net cash flows related to proved oil and gas reserves" for Chevron ' s Australian interests. CVX filed Annual Reports with the SEC and in each Annual Report it disclosed the Standardised Measure for the Chevron group. He said that although the functional currency of the Chevron Australia Consolidated Group was AUD, in CVX's Annual Reports dollar values were expressed in USD. The discounted future net cash flows for the North West Shelf division in 2002 and 2003 were USD4,677,619 thousand and USD5,695,058 thousand and for the West Australian Petroleum Pty Ltd division in 2002 and 2003 were USD286,813 thousand and USD230, 676 thousand.

132. Mr Callaghan also set out the Standardised Measure for the Australian business unit, taking into account income tax payments. For the 2002 year the total was USD3,561,983 thousand and for the 2003 year was USD4,249,206 thousand.

133. In his second affidavit, sworn on 24 July 2014, Mr Callaghan deposed that virtually all of CAHPL's assets were held by CAPL and TAPL. He said that over 99.8% of the value of the property, plant and equipment indirectly acquired by CAHPL in the year ended 31 December 2002 was a result of its acquisition of the shares in CAPL and TAPL, and that CAPL and TAPL were acquired by CAHPL at the net fair value of their assets which was assessed by a PricewaterhouseCoopers reviewed Valuation of the Issued Share Capital of CAPL and TAPL as at 1 July 2002. CAPL held the Chevron group's one sixth non-operated interest in the North West Shelf Joint Venture and an interest in Scott Reef/Brecknock. Both CAPL and TAPL held interests in the Greater Gorgon fields, Barrow Island and Thevenard Island. Mr Callaghan also deposed that in the years ended 31 December 2003 to 31 December 2007, CAHPL derived most of its sales revenue from sales in South East Asia and Australia.

134. In cross-examination, Mr Callaghan accepted that, from a commercial point of view, had CAHPL borrowed from a third party in Australian dollars, the CVX accounts, the parent accounts, would have been exposed to a foreign exchange gain or loss either through the profit and loss statement or through equity.

135. I accept Mr Callaghan's evidence.

136. Mr Callaghan was taken to a number of annual reports, in particular to the CAHPL consolidated accounts for the years ended 31 December 2003, 2004, 2005 and


ATC 17825

2006. These questions appear to have been directed at the point that was ultimately abandoned by the respondent, the respondent accepting, in light of the note in CAHPL's 2007 accounts, that CAHPL's statutory accounts for 2003, 2004, 2005 and 2006 contained an error which was corrected in 2007 and that figure should have been approximately $3.7 billion. Thus the respondent accepted that the loan balance remained outstanding during the period in dispute. In consequence, the respondent accepted that there was a drawdown of the US dollars.

Ms Taherian

137. Ms Taherian gave evidence in relation to the background and actual pricing of the loan. She was employed by CVX and its subsidiaries in the United States between 1996 and 2010 and was responsible for communicating with Goldman Sachs and Deutsche Bank. She was the director of international financing in the Treasury group from the second half of 2001 until 2005 but was a relatively junior employee of the US Treasury of Chevron . Ms Taherian gave evidence addressing the engagement of Goldman Sachs and Deutsche Bank to advise on the quantum and price of the loan. The work done by those banks was not relied on by the applicant otherwise than to prove how the loan had been priced.

138. In her affidavit, Ms Taherian said that in around April 2002 she was asked to help the Australasian Business Unit obtain advice from Goldman Sachs and Deutsche Bank about the level and pricing of debt to be provided to the Australasian Business Unit, meaning CAHPL, its subsidiary CAPL and other subsidiaries as well. She understood, from her experience, that to raise money in the commercial paper market, CFC would need a guarantee from CVX in order to achieve attractive pricing in that market. She also expected, from her experience in arranging intercompany loans in the Chevron group, that the intercompany loan from CFC to CAHPL would not be guaranteed. She could not recall any intercompany loan that had a parent company guarantee. Ms Taherian said that in her initial discussions with Goldman Sachs and Deutsche Bank, she asked them to consider the appropriate amount of debt the Australasian Business Unit could raise. She informed both banks that they should initially assume that the borrower would be rated BBB on the S&P credit rating scale. The BBB rating was the lowest investment grade rating and as such seemed to Ms Taherian to be an obvious starting point. The higher the level of debt, the more likely it was that the Australasian Business Unit would be assigned a lower credit rating.

139. She understood that providing the BBB rating as an assumption to the banks could have the effect of limiting the amount of debt the Australasian Business Unit could raise. However, she expected that Chevron would refine the assumptions the banks were required to make as the work progressed. She recalled that both Goldman Sachs and Deutsche Bank informed her that based on their analysis the maximum amount of debt that the Australasian Business Unit could borrow if assigned a BBB credit rating would be less than USD2 billion. She was later told by Mr Sheppard of Chevron that she should ask the banks to consider the maximum amount of debt that the Australasian Business Unit would be able to borrow if it had a lower credit rating. In late November 2002 a representative of Goldman Sachs provided Ms Taherian with the conclusions that bank had reached in relation to the amount that could be borrowed by the Australasian Business Unit if it were assigned a lower credit rating of BB, as well as the interest rate at which that amount could be borrowed. The amount was a maximum of USD2.5 billion and the interest rate was AUD LIBOR BBA +4.14%.

140. On about 16 January 2003 Goldman Sachs emailed to Ms Taherian a copy of their discussion materials and memorandum. The proposal was that CFC would borrow in US dollars in the commercial paper market and lend to the Australasian Business Unit in Australian dollars. It followed, she said, that CFC would bear foreign currency exchange risk. As she understood the proposal, CFC would not enter into hedging arrangements to deal with the foreign currency exchange risk. She understood that the spread between the interest rate at which CFC would borrow in the public debt markets and the rate at which it would loan the funds to the Australasian Business Unit could give rise to a profit in CFC. However, that profit would be affected by any movement in the exchange rate between the US dollar and the Australian dollar. Ms Taherian said she was not involved in making any decision about the amount that the Australasian Business Unit borrowed or the interest rate applicable to the borrowing. She understood that the pricing exercise undertaken by the banks took into account the fact that CFC bore the exchange risk on the proposed loan. Nor was she involved in implementing the loan made by CFC to the Australasian Business Unit.

141. In cross-examination, Ms Taherian agreed that in mid-2002 she was aware that the proposed loan by CAHPL from CFC would reap significant tax benefits for CAHPL. She also agreed that in mid-2002 she was aware that the proposed loan by CAHPL from CFC would reap significant tax benefits for the CVX ( Chevron and Chevron Treasury) group. She agreed that the review of the capital structure of the Australasian Business Unit involved leveraging the Australasian Business Unit to create new debt and there was to be some refinancing. The Australasian Business Unit had some debt before the project started and part of the project was to increase that debt level.

142. She agreed that in 2002 and 2003 it was Chevron policy that CVX in California, which may have included legal and other sections, ultimately decided all matters concerning internal restructures. This included the extent to which subsidiaries were financed by debt or equity. On any external financing, Treasury would have a very strong decision. The board of that particular subsidiary also had to approve and review it. The policies of Chevron were such that no external financing could take place unless Treasury or CVX and another department of CVX approved it. Any such borrowing had to be in accordance with Chevron ' s policies and procedures. This was to ensure that any external borrowing was struck at the lowest or most efficient rate possible, consistent with CVX's foreign exchange policies. It was obviously the objective of the Chevron group to obtain the lowest cost of funding in its external borrowing.

143. Ms Taherian was aware that in 2002 and 2003 at least some of Chevron 's external funding came through the commercial paper program that CVX Treasury managed. CVX Treasury decided when the group needed to raise funds from the market to meet the group's cash commitments and on a whole Chevron group basis Treasury had to make sure that when any member of the Chevron group had to pay money to a third party there was cash there to do so. The cash was managed on a group basis. She was aware in about April or May 2002 that there was a proposal to raise funds on the commercial paper market by the company that became


ATC 17826

known as CFC. She was aware that the timing of the issue of the commercial paper, or at least some of it, by CFC was determined by CVX's needs in the United States to pay dividends and to repay commercial paper.

144. Ms Taherian was aware from an email on which she was copied that the first tranche of $1.45 billion of CFC commercial paper was issued on 6 June 2003 and that the funds would be loaned into Australia to be used by Australia to pay down their existing intercompany loans and that the proceeds in turn would be used to pay the dividend the following week, plus pay down "the Corp CP outstanding amounts". She also understood that the so-called Australian commercial paper program was intended to raise $2.5 billion.

145. She agreed that when Chevron went to the market to borrow either directly or through a subsidiary, it always tried to borrow by using its AA credit rating.

146. She was aware in late 2002 that the plan was that the money would be raised at an interest rate that was close to USD LIBOR and then on-lent to the holding company at a higher interest rate.

147. Her general impression in late 2002 was that any profit that CFC made on the lending of the funds to CAHPL would not be taxable in the United States and she was told that by Mr Lewis or Mr Wilkinson or someone in tax. She did not recall discussing or hearing about whether dividends paid by CFC to CAHPL would be tax-free in Australia. She was aware in the second half of 2002 that the funds raised by CFC were not required by the Australasian Business Unit for that Unit's external debt.

148.


ATC 17827

Ms Taherian was aware in November and December 2002 that the commercial paper that was intended to be issued would be in US dollars and she was aware there was some debate about whether the proposed loan from CFC to CAHPL would be in US dollars or Australian dollars. She was also aware that there was a desire on the part of the tax people in Australia for the loan to be in Australian dollars and she was aware that that gave rise to some concerns within CVX as to how any foreign exchange gain or loss in CVX should be dealt with.

149. She was asked whether she understood that because the proposed transaction between CFC and CAHPL was an internal transaction within the CVX group, the currency of the loan would have made no difference, from a foreign exchange risk point of view, to the CVX group as a whole. Her response was that even if you have the loan within a company, if it went across two legal entities in potentially different countries, they could have different foreign exchange ramifications that did not offset each other so it may not cancel out at the group level. Ms Taherian agreed that there was a possibility that CFC could make a gain or a profit in respect of the borrowing it had made and the on-lending of that money to CAHPL, and that any surplus could be used to repay commercial paper consistently with the rules of the SEC.

150. She recalled the debate about whether CFC should internally hedge the transaction and the debate came about because of the desire of the tax people in Australia to have the loan in Australian dollars. Ms Taherian was taken to an email dated 3 December 2002 she sent to Mr Lewis, amongst others, the subject heading being "Australia Financing: Need for CTFC Swap or Forward Contracts", which she agreed accurately set out her views and concerns at the time. It read:

It doesn't seem the SEC will accept [CFC] making investments/loans to other CVX companies.

One other alternative comes to mind: When [CFC] can't pay dividends, could [CFC] use its surplus cash to pay off CP [commercial paper]? From Treasury perspective, we want to make sure the surplus cash is used to pay off third-party debt, regardless of whether it's corp CP or [CFC] CP. ... Regardless of the size of the Aus CP program, [CAHPL] would continue to get tax deductions in Australia on the intercompany loan so we'd still reap the benefits of this structure.

...

151. Ms Taherian was taken to an email that she had sent on 2 December 2002 to Ms Ruth Modisette, who was a US attorney in US law and not an Australian tax lawyer or a tax expert, saying, in part:

The main reason we are considering a forward contract to hedge f/x exposure of CTFC is because we want to make sure CTFC can disburse its surplus cash, even if (sic) has negative earnings.

152. Ms Taherian accepted, by reference to an email dated 19 November 2002 from Mr Lewis, on which she was copied, that she was told that the profit in CFC from the interest rate margin within CFC, being a reference to the interest expense and the interest derived, would not be subject to tax either in the United States or in Australia.

153. As to hedging, the possible need for a hedge only arose because the borrowing was in US dollars and the loan was proposed to be in Australian dollars and Australian dollar interest rates were higher than US dollar interest rates at the time. Ultimately the decision was made not to hedge.

154. Ms Taherian was taken to the key objectives of ChevronTexaco set out in the Deutsche Bank Discussion materials dated 20 November 2002, set out at [122] above. She agreed that one of ChevronTexaco's key objectives was to maximise sustainable leverage. She also agreed that an objective was to repatriate cash to the United States: a general goal, as corporate treasury, was to centralise cash holdings in the United States because it was more efficient.

155. As to the loan having no security, Ms Taherian thought that that was an assumption made by Deutsche Bank but she agreed that she would have told them if they were wrong. She also agreed that the effect of not granting security was to make the interest rate on a loan higher rather than lower. She agreed that an objective was that there be no guarantee from CVX. She said in general no Chevron intercompany loans had CVX guarantees and agreed there would not be any need to guarantee an inter-company loan. As to the loan being denominated in Australian dollars, Ms Taherian thought that Chevron had probably given Deutsche Bank an assumption that the loan would be so denominated. She agreed that the effect of an Australian


ATC 17828

dollar loan over a US dollar loan was that the interest rate would be higher than for a US dollar loan, although she said she did not think that was the reason for an Australian dollar loan. She thought the reason was more that CAHPL was an Australian company so otherwise the borrowing would not be in the currency of the company.

156. As to the objective of minimising covenants, Ms Taherian thought that the goal there was to simplify the loan. It was an intercompany loan and for an intercompany loan Chevron did not think it was necessary to get to the level of complexity of a thousand pages of legal documentation. She said that minimising covenants might make it more difficult to do the loan but she did not think it had the effect of increasing the interest rate.

157. As to the credit rating, Ms Taherian was asked whether Chevron had given instructions to Goldman Sachs that CAHPL's credit rating should be assumed to be at least BA2 and BB. Her response was:

We approached them and said we needed guidance on what would be, sort of, market terms and interest rate. And they said that for their analysis they needed an idea of what credit rating would we be willing to support. And we originally said - well, it was a little bit a random response, we said arbitrarily "Well, it should be maybe investment grade." And they did some analysis based on that. And we - that's the information to both Deutsche Bank and Goldman Sachs. And I can't recall which one of them asked about credit rating, but it was something that they both needed. But upon further reflection it was very clear to us that an investment grade credit rating really isn't meaningful for a wholly owned subsidiary. We were never going to actually get issued a credit rating or be required to produce that credit rating. In fact, most of Chevron ' s subsidiaries don't have a credit rating. So that investment grade credit rating was quite arbitrary. And so in that case we said "Well, try for something lower but reasonable or respectable" and that's where this BB or BA2 ratings came from.

So - - -?---And as we saw, there's a lot of companies that were comparable with those kind of credit ratings. So those seemed a reasonable benchmark to pick.

...

... So the position is that in order to be able to sustain - I'm suggesting to you, in order to be able to get an interest rate for a $2.5 billion loan, you asked them to assume a rating of BB - of at least BB; is that right?---Yes.

Now, because, of course, an investment grade rating, based upon Goldman Sachs' analysis, would not have supported a loan of USD2.5 billion, would it?---It did not, based on their analysis.

158. Ms Taherian was asked whether Goldman Sachs did the credit metrics based upon a 5.38% interest rate and she was asked whether it was her understanding that to the extent that this might have been a sustainable debt it was based upon a 5.38% interest rate. On the analysis at page 14 of the Goldman Sachs document, she agreed.

159. I accept Ms Taherian's evidence.

Mr Oen

160. The applicant relied on an affidavit of Mr Oen, an LNG consultant. On 16 February 1998 he became a director of TAPL. On 10 October 2001, CVX and Texaco Inc announced the completion of their merger and after the merger Mr Oen became the General Manager Gas for Chevron ' s Australasian gas assets, including fields in the North West Shelf, the Greater Gorgon area and also in the Philippines. On 19 October 2001 he became a director of CAPL. On 7 November 2001, CAHPL was incorporated and he became a director of that company on its incorporation.

161. He described the Greater Gorgon area gas fields, the ownership and operation of Gorgon, the development phases, Gorgon project development difficulties, environmental concerns, disagreement between joint venturers, investment economics, LNG sales, and the CVX appropriation endorsement process.

162. As to the ownership and operation of Gorgon, Mr Oen gave evidence that there were four joint ventures holding the key Gorgon titles. Each of the joint ventures consisted of one or more of CAPL, TAPL, and Australian subsidiaries of Mobil, Shell and BP.

163. Mr Oen described a number of phases in the development of an LNG project undertaken by Chevron . He described the


ATC 17829

purpose of Phase Two as being to generate and select broad generic alternatives for the possible development of the opportunity. The alternatives were assessed at a high level to select the alternative which best met the goals of the participants in the project. He described Phase Three as considering in more detail the preferred alternative for what then still remained only a possible future development. If Phase Three was successful, it usually, but not always, resulted in the commitment to the investment in the project. He said that throughout the 1990s the Gorgon project had failed to progress from Phase Two, although it came close to proceeding to Phase Three a number of times, only to suffer setbacks. Phase Three of the Gorgon project commenced in early 2005, at which time it was hoped that the Final Investment Decision would be made by mid-2006. The Final Investment Decision was made on 14 September 2009. Mr Oen said that despite the significant size of the resource as it was understood in 2002 and 2003, the Gorgon Project remained in Phase Two and he held serious concerns as to whether the Gorgon project would progress to the end of Phase Three. Up to June 2003, the project had not secured State government approval, had not obtained environmental approval, which had yet to be sought, had not secured firm customers for the future purchase of gas, had not commenced preparation of a Field Development Plan needed to secure a production licence and had not entered into any contracts for construction of the project. He also said that disagreements remained between the joint venturers on many details of the proposed development. In particular, Mr Oen said, there were four specific issues in around 2002 and 2003 that were of great concern: environmental concerns were especially significant as the Gorgon project's success depended on approval to use Barrow Island, a Class "A" nature reserve, as the location of an off-shore LNG plant; the Gorgon joint venture participants were not in agreement on certain key elements of the development; the investment economics of the project were marginal; and the Asian financial crisis led to a downturn in the LNG market, which threatened the viability of the project.

164. Mr Oen was not required for cross-examination. I accept his evidence.

Mr Krattebol

165. Mr Krattebol was not called by the applicant to give evidence. He was the person who made the decision as to the amount CAHPL would borrow and had overall responsibility for the project of leveraging CAHPL. The respondent submitted that what CVX did, and why, could be inferred from the balance of the evidence and those inferences could be drawn more readily in Mr Krattebol's absence:
Jones v Dunkel [1959] HCA 8; 101 CLR 298 at 332. I agree. I infer from the evidence that the objectives of the leveraging project were to obtain the lowest cost of funding and to achieve merger synergy objectives, including the tax benefits that would arise from the gearing of the balance sheet of the Australasian Business Unit. I also infer from the evidence that CAHPL's debt level of USD2.5 billion was chosen because it was the most tax efficient corporate capital structure and gave the best after-tax result for the Chevron group. I further infer that if CAHPL were borrowing from an independent party, borrowing USD2.5 billion or the AUD equivalent at 8.97% would not have been sustainable.

Mr Martin

166. Mr Martin was called by the applicant as an expert. He was the Chief Executive Officer of Anawan Cliffs Capital Management and Advisory LLC. He had over 20 years' experience in the global leveraged finance industry. Most recently, from April 2010 to January 2013, he served as the co-head of Global Leveraged and Acquisition Finance at Morgan Stanley & Co in New York.

167. He was instructed to prepare an expert opinion on whether the interest rate of 1 month AUD LIBOR BBA +4.14% per annum which CAHPL agreed to pay under the Credit Facility Agreement exceeded the consideration that might reasonably have been expected to have been given for the provision of that Credit Facility if CAHPL had entered into that transaction with an independent party with whom it had dealt with at arm's length. For his second report, Mr Martin was asked whether he agreed with the reasoning and conclusions recorded by Mr Hollas in his report dated 4 June 2014 and, if there were any aspects on which he disagreed, to set out his reasons. For his third report, Mr Martin was asked whether he agreed with the reasoning and conclusions recorded by Mr McCormick in his responses to questions 2 and 3 in his report dated 5 June 2014, and if there were any aspects on which he disagreed, to set out his reasons.

168. Mr Martin accepted that he was not a specialist lender to the oil and gas industry and still less to the E&P sector of the oil and gas industry.

169. In his first report, annexed to his affidavit affirmed on 10 March 2014, Mr Martin said that the foundation for any view on pricing was the determination of a borrowing company's risk rating. Lending institutions performed their own independent credit analyses when assessing/determining whether to lend to a prospective borrower. As part of their independent credit analysis, they typically assigned an internal risk rating to the borrower and the loan itself. A borrower may be rated by one or more of the rating agencies. However, lending institutions did not rely on those ratings to make lending decisions or to price loans.


ATC 17830

170. Mr Martin performed a formal credit analysis of CAHPL focusing primarily on the sustainability and predictability of CAHPL's cash flow, the company's ability to pay interest costs and fully amortise the loan over the tenor of the Credit Facility or refinance the loan at or prior to maturity of the Credit Facility. Embodied in this analysis was also an assessment of the company and its prospects, projected financial performance over the life of the Credit Facility and the value of the enterprise and any collateral underlying the loan. Mr Martin did not perform any sensitivity analysis but said that did not impact on any of his conclusions. Lenders analysed credits with a conservative eye towards what could threaten repayment of the loan, rather than analysing the "upside" like an equity investor would. Lenders were most concerned with what could go wrong rather than what might go right.

171. Mr Martin rated CAHPL as a weak BB borrower, that is, with a weak BB rating. He noted the lack of typical BB covenant package/protections (e.g. absence of financial covenants); the unsecured Credit Facility which would substantially shrink market demand for it and thus drive up interest rates lenders and loan investors would accept; and the absence of undrawn commitment fees. He also noted that at USD2.5 billion equivalent in AUD, the Credit Facility would have been among the largest credit facilities in all of 2003; the depth of the Australian and Asia-Pacific loan markets was likely unable to accommodate the Credit Facility; the size of the Credit Facility coupled with the unsecured nature of the facility would have required upwards of 75-100 lenders and institutional loan investors, subjecting CAHPL to interest rates well in excess of that charged by CFC; for broadly syndicated BB credit facilities it was very common for issuers to have many participants in its credit facility as lending institutions had borrower concentration limits; and the lack of basic BB covenant protections and the unsecured nature of the Credit Facility, hold limits for institutions would be much lower than normal, thus requiring a much larger bank/investor group to achieve the size of the Credit Facility. Mr Martin also noted there was no explicit guarantee agreement between CVX and the borrower or any other type of letter of support; while lenders would view CVX as a corporate parent/sponsor with strong financial wherewithal, the transaction lacked any explicit support for CAHPL in the event it defaulted on its loan obligations; lenders would evaluate creditworthiness of the borrower on a "stand-alone" basis and would assign a credit rating based on analysis of the borrower itself (without assuming any support from its ultimate corporate parent); and banks would certainly recognise CVX's financial strength, but their credit risk assessment would not be affected as CVX was not obligated to provide any support in the event of default by CAHPL under its loan agreement.

172. Mr Martin's summary of his conclusion was based on the weak BB rating and his view of the markets in 2003. It was his opinion that the interest rate on the Credit Facility did not exceed what would reasonably be expected to be obtained in an arm's length transaction at the time. In fact, he estimated the interest rate would have likely been much higher than the rate charged by CFC.

173. In his second report, annexed to his affidavit affirmed on 25 July 2014, Mr Martin said that nothing in Mr Hollas' report had caused him to change the conclusions he expressed in his first report, and gave detailed responses to Mr Hollas' report. In his third report, annexed to his affidavit affirmed on 1 August 2014, Mr Martin said that nothing in Mr McCormick's report had caused him to change the conclusions he expressed in his first report, and gave detailed reasons for that conclusion.

174. In cross-examination, Mr Martin agreed that it was not part of his task to express any opinion on what terms would have been negotiated had CAHPL been dealing independently at arm's length with a third party lender. He also confirmed that in preparing his report he made an assumption that there was not just one lender. None of the assumptions he was given required him to consider the Credit Facility in the context of a single lender. The loan would not have been able to be entered into with just one lender. One lender would not take that size exposure to a non-investment grade entity. Any such lending by an E&P specialty lender would have required security, operational and financial covenants and strict terms for borrowings under the Credit Facility. In order to raise the full amount of the Credit Facility, many lenders would have been necessary. The loan would have had to have been with 75 to 100 institutions lending, the vast majority of whom would have been global institutional investors rather than commercial banks. He also agreed it was not part of his task to compare CFC's position with the position of the institutional investors to whom he referred.

175. Mr Martin was not aware of a single Term Loan B without security in 2003.


ATC 17831

176. Mr Martin accepted that his analysis depended upon a BB rating, such that if the notional rating of CAHPL was not BB then he simply did not address the pricing of the loan. Mr Martin was asked whether if the terms of the Credit Facility Agreement were different to those that in fact existed, his opinion as to pricing would be inapplicable. His answer was that if the terms were different it would certainly fall into a different pricing category, but it would depend on what those terms exactly were. If the terms were worse, it might be different. If the terms were better, it might be better. He agreed that if the loan which should be priced was a loan which contained financial covenants, then that would reduce the risk and, therefore, the interest rate on the loan. He also agreed that it was his opinion that the lack of financial covenants could add several hundred basis points to the average yield, at least in a "really bad market".

177. Mr Martin agreed that of the credit metrics which Mr Hollas determined and described as E&P specific credit metrics the only one he, Mr Martin, looked at was debt to proved reserves. The E&P companies' credit metrics which Mr Hollas looked at were: under the heading "Size Characteristics", production, proved developed reserves, total proved reserves and present value of proved reserves; under the heading "Leverage And Cash Flow Coverage", debt/PD boe reserves, (Debt + Future Development Capex)/Total Proved Reserves, (Retained Cash Flow - Sustaining Capex)/Debt and Debt to Present Value of Proved Reserves; under the heading "Reserve Replacement Costs", 3-year all-sources F&D and 3-year drillbit F&D costs including revisions; and under the heading "Other Financial Ratios", total debt/(total debt + equity). The relevant part of the cross-examination was as follows:

And what [Mr Hollas] deals with first of all is size characteristics; do you see that?---Yes.

And the first matter he looks at is the size of production?--- Mmm.

And what he does is compares the size of production to other E&P companies which are listed to the right hand side; do you see that?---Yes.

And in the next column, he looks at proved developed reserves and then compares them to other E&P companies?--- Mmm.

Then total proved reserves and then present value of proved reserves; you see that?---Yes.

And then he looks at debt to proved developed reserves?--- Mmm.

What's that an indication of? ---That's an indication of the debt relative to the assets that are proven and developed of the company.

And that's a matter that you regard as relevant?---That is one data point but again, we're not secure.

But you said in your report that you had regard to that?---Yes, I did. Yes. It is yes, I had regard to it. Apologies.

You said that in the fourth dot point under paragraph 12?---Yes.

And then the next item is debt plus future developed CAPEX to total proved reserves?---Correct.

And what does that assess?---I'm sorry. Say that again.

What is that looking at? Why is that credit metric relevant?---Debt to - say that again - .....

Sorry. It says debt plus future development CAPEX?---Yes.

To total proved reserves?---That's the amount of money they need to spend to get - at least at that point in time to get the reserves out of the ground.

Well, it's actually the cost of bringing proved undeveloped reserves to being proved developed reserves, isn't it?--- .....

Sorry?---I said - again, in my mind, it's spending the money to get them out of the ground. We may have just said it differently.

Well, I don't think we did say it differently, with respect. I think what it really shows is what the cost is of bringing proved undeveloped reserves to being proved developed reserves. Is that your understanding or not?---That is a metric that is not one of the ones I looked at, to be perfectly honest.

Right. Well, none of them is except for debt to proved reserves, is it?---Correct.

And then the next one is retained cash flow less sustaining CAPEX to debt; do you see that?---Yes.

And what does that deal with?---That's not a metric I'm familiar with.

Would you accept that it looks at the amount of cash flow left over after the required spend on necessary CAPEX as a proportion to debt?---At that point in time, I would say at a point in time, yes.

But that would be relevant to looking at the financial risk of an E&P company?---It would be one data point, yes.

And then underneath that is debt to present value of proved reserves; do you see that?---Yes.

Then there's a heading which says reserved replacement costs and there is an item three year all sources, F&D; do you see that?---Yes.

What do you understand that to be?---That's not a ratio that I'm familiar with.

It assesses how much the cost is to replace reserves which have been depleted through production, averaged over three years. F&D stands for finding and developing?---Yes.

And all sources refers to drilling and acquiring. You didn't look at that credit metric?---No.

And that's because you don't profess to be a specific E&P lender; is that right?---No, not at all, no. I don't.

And then the next one is three year drill bit F&D costs including revisions. Do you have any understanding of what that metric is?---Again, not ratios that I am super familiar with.

So after you prepared your first report, you did have access to this document?---Yes.

And you didn't feel the need to analyse these E&P credit specific metrics?---You know, again, it was an unsecured loan. In my mind, the cash flows of the company, both historical and projected, were really what carried the debt.

178. Mr Martin agreed that in preparing his reports he did not look at what E&P companies in 2003 undertook as borrowings. He looked at the broad oil and gas industry data. In preparing his Table 2 he looked only at the chart of Oil and Gas Credit Statistics prepared by S&P Capital IQ forming part of Exhibit 13.

179. Mr Martin said that he based his pricing opinion on Tables 6 and 7 of his first report but he relied predominantly on Table 6. The difference between Table 6 and Table 7 was that Table 7 was an analysis of oil and gas new issue spreads while Table 6 was the whole market.

180. Mr Martin asked himself: "How would a third party investor look at this business and how would they go about making the loan?" and analysed the situation by asking: "Would investors invest in this loan at a price?"


ATC 17832

181. I accept Mr Martin's evidence that lending institutions performed their own independent credit analyses when deciding whether to lend to a prospective borrower and that lending institutions did not rely on those ratings to make lending decisions or to price loans. However, I find that Mr Martin's evidence was not persuasive and I do not accept, on the basis of his evidence, his opinion that the interest rate on the CFC loan to CAHPL, the Credit Facility, did not exceed what would reasonably be expected to be obtained in an arm's length transaction at the time, or his opinion that the interest rate would have likely been much higher than the rate charged by CFC.

182. One matter going to the overall cogency of Mr Martin's evidence was that he had no relevant experience in or substantial knowledge of the E&P industry. Secondly, and as a consequence, his opinions as to the borrower's risk disregarded E&P specific credit metrics. Thirdly, Mr Martin did not address what the consideration would have been if it had been negotiated at arm's length by entities independent of each other. This, in my opinion, is what the statutory task under the ITAA 1936 requires, as set out at [87] above. Fourthly, Mr Martin proceeded on the basis that the transaction could not have occurred with one lender and assumed that there would have been between 75 and 100 lenders. Again, in my opinion, this is not what the statutory task under the ITAA 1936 involves. Fifthly, Mr Martin's opinion was based on a Term Loan B form of transaction which would not have occurred. Although he stated that the transaction "could be priced" as a Term Loan B, he agreed that all such loans were secured and did not give an example of a Term Loan B transaction without covenants.

183.


ATC 17833

More particularly, I accept the respondent's submission that Mr Martin's opinions as to the margin over LIBOR were unreliable because they were not based on an understanding of what the E&P industry was; his opinions were based on material obtained from S&P which he did not clearly understand. To the extent his opinions were based on reliable information, that data was not shown to reflect margins which might be achieved for a transaction involving an E&P borrower; and his methodology for determining the margin, by adding basis point increments to an unreliable starting position of 300 basis points, as set out in Table 6 of his first report, was itself unreliable.

184. In short, I am not persuaded by Mr Martin's opinions in relation to the relevant industry and the interest rate for the Credit Facility. His opinions were insufficiently related to the market in which the Credit Facility was made. I make further findings in relation to Mr Martin's evidence at [511]-[517] below.

Mr Gross

185. The applicant called Mr Gross as an expert. He had been the head of Global Investment Banking at the Bank of America between 1998 and 2002, the Group Managing Director of Huron Consulting Group, Chicago between 2003 and 2006, and a partner at Sandhurst Capital Partners focusing on consulting and advising private equity and hedge funds between 2006 and the present. He swore four affidavits in the proceedings.

186. His first report, annexed to his affidavit sworn on 18 April 2013, was addressed to the question whether the interest rate of 1 month AUD LIBOR BBA +4.14% per annum which CAHPL agreed to pay under the Credit Facility Agreement exceeded the consideration that might reasonably have been expected to have been given for the provision of that Credit Facility if CAHPL had entered into that transaction with an independent party with whom it had dealt with at arm's length. Mr Gross' opinion was that the correct pricing of the Credit Facility was 500 basis points over 1 month AUD LIBOR BBA and the payment by CAHPL of an interest rate of 1 month AUD LIBOR BBA +4.14% per annum did not exceed the consideration that might reasonably have been expected to have been given by CAHPL to an independent third party for the provision of the Credit Facility.

187. Mr Gross said that banks did not rely on S&P ratings or Moody's because they did not consider them to be reliable.


ATC 17834

188. For his second report, annexed to his affidavit sworn on 1 August 2013, Mr Gross was asked for his opinion on whether the three credit facilities identified by Mr Hollas in his report as "comparable uncontrolled senior loan transactions" were transactions that, in Mr Gross' opinion as a banker, would enable him to determine the question answered by him in his first report, and whether he agreed with the reliance placed by Mr Hollas, Mr Emmer and/or Mr Taylor in their reports on the relationship between CAHPL and CVX for the purposes of determining what consideration CAHPL might reasonably have been expected to have given to an independent third party for the provision of the Credit Facility. Mr Gross' opinion was that the three credit facilities identified in the Hollas Report would not be helpful in determining the question posed to Mr Gross in his first report. The three companies had fundamentals and credit metrics that were dissimilar to CAHPL. The loan facilities were dissimilar in structure, term and conditions. Also, he did not agree with the reliance placed by all three reports on the relationship between CAHPL and CVX for the purposes of determining what consideration CAHPL might be reasonably expected to have given to an independent third party for the provision of the Credit Facility. He relied only on the fundamentals of the obligor in the absence of a formal parent company guarantee.

189. For his third report, annexed to his affidavit sworn on 5 March 2014, Mr Gross was asked a number of detailed questions about Mr Hollas' report as well as a more general question. The more general question was: assuming that a lender raised funds with the benefit of the guarantee from a third party, to what extent, if any, would the existence of that guarantee affect the price of a loan to an independent party with whom the lender is dealing at arm's length? Mr Gross' answer to the general question was that the manner in which a lender raised funds, or whether the funds were raised with the benefit of the guarantee from a third party, was irrelevant to the arm's length lending relationship that the lender had with the borrower. Banks raised funds from multiple sources. The source of funds did not affect the arm's length lending relationship between lender and borrower. If the lender raised funds with the benefit of the guarantee from a third party, this also would not affect the lending relationship. A bank would always price the loan at the market rate that was appropriate for that type of loan taking into account credit risk. Mr Gross restated the conclusion that he had reached in his second report. He also said the methodology utilised in both of the reports of Mr Hollas was outside the mainstream of conventional analysis used in the market. He did agree with the approach taken by Mr Hollas in his reports not to factor in any implied parental support in his analysis.

190. For his fourth report, annexed to his affidavit sworn on 25 July 2014, Mr Gross was asked whether he agreed with the comments made by Ms Azarchs at specified paragraphs of her report and whether he agreed with the reasoning and conclusions recorded by Mr Hollas in his report of 4 June 2014. In short, Mr Gross said that Ms Azarchs' report ignored reality in that major multinational companies did not always support their subsidiaries and this affected lending decisions. In any event, Mr Gross said, Ms Azarchs' factual analysis was incorrect. Mr Gross agreed with Mr Hollas that an independent lender would consider the risk profile of the borrower in pricing a loan rather than considering its cost of funds, but he disagreed with Mr Hollas' basic underlying principles which were based on the comparable uncontrolled agreements ( CUT s) approach which had no bearing on actual procedures used by banks and investment banks in the capital market.

191. In cross-examination, Mr Gross agreed that he did not profess to be a specialist oil and gas lender and still less would he profess to be a specialist lender to the E&P sector of the oil and gas industry. In answer to the question whether he professed to have particular expertise in the credit metrics of E&P companies, Mr Gross said that the people in the oil and gas industry underwrote the loans or made the loans and had the relationship with the company, but any loan that was, say, over $200m and had to be distributed or syndicated came to his group which re-underwrote the loan and had to give a letter of opinion that they could sell it down to the requisite hold level. In order to do that they had to know a lot about the loan and a lot about the market for the loan, although they were not the group that originated it.

192. Mr Gross agreed that he had set out his opinion as to the interest rate which CAHPL would have paid on the assumption that the Credit Facility Agreement was in exactly the terms that they were, and approached his opinion on pricing on the basis that the borrower had provided no financial covenants, that there was no guarantee given by CAHPL or by a related entity of CAHPL, and that CAHPL provided no security for the loan. He agreed he was asked no questions about what the parties would have agreed in their agreement had they been dealing with each other on a commercial arm's length basis and he did not address that in his report. He analysed how an independent arm's length lender, such as Bank of America, would have made the loan and that was the process of reasoning Mr Gross adopted in order to reach his pricing conclusion. He concluded that CAHPL had a notional non-investment grade rating equating to a S&P rating of approximately B. He assumed that the lender did not have to be one single lender and could have been as many lenders as needed. He said the loan would need to have been syndicated and it would have been impossible to either syndicate a loan or issue a high-grade or high yield bond without restrictive financial covenants. There were buyers in the market, such as hedge funds, who would buy a loan of this type without covenants so there would be a market but it would be extremely expensive.

193. Mr Gross said that his approach was to start with a spread of 300 basis points as a minimal hurdle rate that the bank would demand for a loan of this risk rating. It was necessary to look at where the market said the rate was: if the market were higher then obviously you would take a higher rate, and if the market were lower you may not make the loan at all.

194. Mr Gross agreed that a bank loan to an E&P company could be structured as a reserve-based loan, that is, a loan which would require security over the assets of the borrower. He agreed that a reserve-based loan would require financial covenants: ratios such as maximum debt to equity, maximum debt to EBITDA, EBITDA interest coverage, negative pledge or not pledged to anybody else, cash flow measures and restrictive covenants such as not selling the assets, not changing the business model, not selling the


ATC 17835

company and not transferring assets or incurring further debt or paying dividends or returning capital above certain limits. These would be standard terms in a reserve-based loan. A bank loan might also be comprised of a senior unsecured loan and a senior unsecured loan would not be generally available to a sub-investment grade borrower. A senior unsecured loan would typically have financial covenants and would also require restrictive covenants, and would often be made in the context of the oil and gas industry to a head entity of a group of companies where the subsidiaries ran the operating assets and the subsidiaries would provide a cross-guarantee.

195. As to institutional loans, such as a Term Loan B, Mr Gross agreed that an institutional loan was generally a senior loan often arranged by a party or a number of parties and on-sold and entered into as part of a suite of borrowings, normally together with revolving credit facilities or Term Loan As. Term Loan As were not institutional loans but banking transactions. Term Loan Bs usually had 10 years or at least five years but typically around seven or eight, and the lifeblood of institutional loans or Term Loan Bs was leveraged buyouts, secured in the broadly syndicated market and, in 2003, always containing financial covenants and restrictive covenants.

196. Mr Gross accepted that in his first report he did not identify E&P companies and compare CAHPL's credit metrics with credit metrics of those E&P companies.

197. Mr Gross accepted that if one commenced the analysis with an assumed interest rate then to some extent you would be predicting the result, being some of the coverage numbers but not the leveraged metrics.

198. Mr Gross accepted that he did not point to a transaction in which the Bank of America made a lending to a B rated company under 300 basis points or to a transaction involving the Bank of America in which it lent at 300 basis points or over to a B rated entity. He was taken to documents indicating that as at 25 October 2002 Bank of America was


ATC 17836

prepared to extend credit on a maximum of 3%; and to a June 2003 facility where the margins were 2.25 to 2.5%; to the example of Bank of America lending to Gray Television, a B rated company, under 300 basis points.

199. Mr Gross also agreed that he did not point to any Bank of America loan made in 2003 to a B rated company with a lending margin of 5%.

200. Mr Gross agreed that CAHPL could not have entered into the Credit Facility with an arm's length lender in the absence of restrictive or negative covenants as in the absence of such covenants the company could effectively dispose of all of its value immediately after taking a loan.

201. In re-examination, Mr Gross said the reason for his not pointing to a transaction in which the Bank of America made a lending to a B rated company under 300 basis points in the relevant period of time. His main reason was that a Bank of America lending document should not be taken out of context because the reader would not know what relationship the Bank of America or any of the other banks had with the particular borrower, for example, whether the bank earned M&A fees, or whether it earned underwriting fees, because these other fee-related transactions were taken into account in calculating the 300 basis points. It was not permissible to look only at a revolving credit agreement because there were other factors that the bank took into account in pricing. In the case of CAHPL, Mr Gross assumed no relationship so he did not take those types of fees and income into account.

202. The respondent made a number of criticisms of Mr Gross' evidence. An overarching submission was that Mr Gross' evidence did not address the correct question, that is, to determine the consideration which would have been paid under an agreement between parties which were independent of each other and dealing at arm's length. The specific criticisms included the following.

203. First, unlike the other experts who gave as their opinion a rating of CAHPL as BB, Mr Gross' opinion as to the margin was based on his view that the appropriate rating for CAHPL was equivalent to a B rating, as emphasised by the data he relied on in his first report and his repeated evidence to that effect. Unless CAHPL's rating or notional rating was B - which was not advanced even by CAHPL - Mr Gross' opinion with respect to margin was irrelevant as he did not address what margin would have been appropriate for a company with a BB or higher rating or notional rating. Further, his opinions were based on material obtained from S&P which was of limited relevance to the loan in question or to the industry in which CAHPL operated.

204. Second, Mr Gross did not address what the terms of an agreement would have been if it had been negotiated at arm's length by CAHPL and CFC, being entities which were independent of each other.

205. Third, Mr Gross' opinions were based on a Term Loan B transaction which did not, and could not have, occurred: a single lender would not have been prepared to lend the full amount; no lender would lend with a complete absence of covenants; in 2003, Term Loan Bs always contained financial covenants and restrictive covenants. He agreed that CAHPL simply could not have entered into the loan it did without restrictive covenants. Mr Gross' opinion was that the borrowing could not have occurred without an express guarantee or unless it contained covenants.

206. Fourth, Mr Gross took as his starting point 300 basis points, which he arrived at


ATC 17837

by analysing what the Bank of America would charge and concluding that the Bank of America would charge a minimum of 300 basis points for a B rated company. He added increments to his starting point for various factors which were unpersuasive being the 25 basis points added for a five year tenor and amounts to reflect fees such as LIBOR floors. He added further amounts to reflect Tables 4 and 5 which were based on unexamined transactions and entities.

207. Fifth, Mr Gross' opinion with respect to the borrower's risk was inadequate as it was reached by reference to larger companies which were known not to be a valid comparison and it was reached by reference to metrics based on the interest rate in fact charged (AUD LIBOR +4.14%) without analysing the metrics if the interest charged was less. Further, there was no comparison with an E&P company, let alone one of any comparable size, in reaching the notional rating conclusion. He did not look at the range of loans, the averages made to particular industries, to see whether there was a correlation between the size of the loan and the margin. However, he agreed that a bank would compare pricing with a special emphasis on companies in the same industry.

208. Sixth, Mr Gross' view was that the ratios of larger companies could not be compared directly with those of smaller companies, although he did do that in his first report in conducting his risk analysis. In Table 2, Mr Gross compared CAHPL to CVX, ExxonMobil, RD Shell and Santos. He did not compare CAHPL to any E&P company, and less still to one of comparable size. Further, he conducted the analysis in Table 2 by adopting metrics based on CAHPL having revenue of $915m, which was understated by approximately $400m, and interest at the actual rate of AUD LIBOR +4.14% being tested which necessarily negatively affected the resulting EBITDA based metrics.

209. Seventh, notwithstanding that Mr Gross considered the "more appropriate metrics" for E&P companies to be a function of reserves and production levels, he stated that EBITDA metrics "aren't always helpful". Mr Gross did compare (in Tables 3 and 3A) CAHPL's EBITDA metrics to S&P medians. He agreed that was the purpose of Tables 3 and 3A and that he was using the S&P medians as a guide. However, the S&P ratios were industrial ratios, dealing with all industrials except for financial entities and utilities. Mr Gross did not refer to the pages from the S&P publication he had used for those medians which expressly stated that they should not be used as a guide at all, because they may be unreliable with respect to particular industries.

210. There is force in these criticisms, although I note that the criticism set out in [204] above should be limited to the question of consideration, as set out in [87] above. I set out my conclusions at [506]-[510] below. However, I accept Mr Gross' evidence that banks did not rely on S&P ratings or Moody's because they did not consider them to be reliable. I do not accept the respondent's submission that Mr Gross was not objective and his opinions were affected by his desire to advance his client's case rather than to assist the Court.

Mr Long

211. The applicant called Mr Long who had over 30 years of regulatory experience in bank supervision and policy related matters with the OCC. The role of the OCC was to supervise nationally chartered banks and federal savings associations to ensure that they operated in a safe and sound manner and in compliance with laws requiring the fair treatment of customers and fair access to credit and financial products. In his first affidavit, affirmed on 4 March 2014, Mr Long was asked to address the following questions: (a) Based on your experience with the OCC, do you consider that any allowance for affiliation in risk rating a credit is prudent or accurate? (b) In your experience, how did banks in the United States in 2003 evaluate credit and the pricing of loans? Mr Long said he was answering the questions and viewing the transaction as if CAHPL had entered into a credit facility with an independent bank with whom it had an arm's length relationship. He said that based on his experience, it was not accurate nor prudent to rely on the benefit of affiliation (or so-called implicit support) when assigning a credit rating, and OCC examiners were instructed not to do it. He said that in the case of CAHPL, in the absence of legally binding guarantor support from the parent company, prudent credit analysis would not warrant the inclusion of the parent's financial condition, or any level of "implicit support" from the parent, in the assessment of the creditworthiness of CAHPL. Mr Long was also asked whether he agreed with the views expressed by Mr Gross in his affidavit of 18 April 2013 in relation to implicit support and whether the report of Mr Gross reflected the way in which banks in the United States operated in 2003. Mr Long said that his experience as an OCC regulator closely reflected the experience demonstrated by Mr Gross, particularly as it related to the lack of a legally enforceable guarantee document. In the absence of a legally binding document, the financial strength of the parent company should not be considered as having a positive impact to the risk rating of its subsidiary.

212. In his second affidavit, affirmed on 29 July 2014, Mr Long was asked whether he agreed with the reasoning and conclusions recorded by Mr Gaskell in section 9 of his report dated 3 June 2014. Mr Long said he did not agree. He said that Mr Gaskell's statement overall missed the main reason why OCC examiners, and banks in general, did not solely rely on public ratings when they risk rated credits, and why they certainly did not rely on a rating agency as a means to assign implicit support to a credit facility. One of the key reasons that agency ratings were not, and should not be, solely relied upon by banks when risk rating credits was precisely because they may improperly give allowance for implicit support, even in the absence of a clear legal obligation, economic incentive, and demonstrated willingness of the parent to support the affiliate's debt.

213. Mr Long was not required for cross-examination. I accept his evidence. I note, however, that Mr Long, in answering the questions and viewing the transaction as if CAHPL had entered into a credit facility with an independent bank with which it had an arm's length relationship, was not, in my opinion, asked the question posed by Div 13 of the ITAA 1936.

Mr Rowland

214. Mr Rowland was called by the applicant as an expert. He was an oil and gas executive and from 1992 to 2010 he had


ATC 17838

been an executive officer and Chief Financial Officer of Chesapeake Energy Corporation ( Chesapeake ). He was asked for his opinion on the following questions:
  • 1. Based on your experience, in 2003 was it possible for an oil and gas company, with a credit rating that was below investment grade, to carry on business?
  • 2. If it was so possible, in your experience, how did such companies typically raise funds?

215. In his first report, annexed to his affidavit sworn on 13 March 2014, Mr Rowland said that in his opinion it was not only possible for an oil and gas company with a credit rating that was below investment grade to carry on business in 2003, but that it was also very common. He said Chesapeake was an oil and gas company with a credit rating below investment grade in 2003. That company issued $950 million of new senior notes in four tranches in 2002 and 2003 with interest rates ranging from 6.875% to 9.0% per annum with six different left-lead bookrunners of recognised international financial significance while rated below investment grade. The company also completed numerous public market equity issuances in 2003 and early 2004 in addition to the issuance of non-investment rated debt and maintained flexibility to exchange debt at advantageous terms.

216. As to the second question he was asked, Mr Rowland said that his experience at Chesapeake from 1992 to 2010 included the initial public offering for the company, financial transaction oversight of secured revolving credit facilities of up to $5 billion, secondary common stock issuances, perpetual preferred stock offerings, secured hedging facilities in excess of $10 billion, numerous multi-billion joint-venture sales and projects totalling in excess of $20 billion, asset purchases and sales and unsecured senior note issuance. In total his financial experience over 18 years at Chesapeake was well over $100 billion in transactions. He said that all of those methods of financing were available to Chesapeake, despite the non-investment grade credit ratings of the company. He said that one of the principal reasons that the company had so much financial flexibility despite its non-investment-grade credit ratings was the financial structure of the senior unsecured notes it issued. Those features included the unsecured nature of the security; no "maintenance" covenants with regard to liquidity or equity minimum requirements; and long-term at maturity was typically 7 to 10 years from issuance, with no amortisation or sinking fund requirements. He said these attributes were not unusual and in fact were standard for many non-investment rated oil and gas issuers at the time.

217. In his second report, annexed to his affidavit sworn on 29 July 2014, Mr Rowland was asked to respond to the reports of Mr Gaskell and to the report of Mr Hollas. In response to Mr Gaskell's 9 July 2013 report, Mr Rowland said he did not agree with the main conclusion and did not find that "each element is on its own sufficient to make the Loan commercially imprudent had it been from an independent lender ..." Not only, said Mr Rowland, did he disagree with Mr Gaskell that each "element" of the transaction would individually render the transaction "commercially imprudent", but, in his opinion, all the noted elements combined did not render the transaction "commercially imprudent" either. Mr Rowland gave a second report in relation to the reasoning and conclusions in Mr Gaskell's report of 3 June 2014. In response to Mr Gaskell's 3 June 2014 report, Mr Rowland agreed with Mr Gaskell that Chesapeake was able to operate as a successful oil and gas business without an investment-grade rate, but disagreed that the sub-investment grade restricted Chesapeake's ability to borrow as much funding as it needed and with other aspects of Mr Gaskell's opinion as to Chesapeake's access to unsecured Senior Notes issuance, and why a company would maintain a secured bank facility, with rates of interest lower than unsecured debt, and not use it. Mr Rowland noted that the ratings of Chesapeake in 2003 for its secured bank facility were Ba2, BB+, and BBB=, which were not the lowest investment-grade from all rating agencies. In his opinion, CAHPL would have been similarly rated for this size of facility as well, not materially changing their cost of capital.

218. In relation to Mr Hollas' report, Mr Rowland said that in 2003 there were significantly more non-investment grade rated and non-rated oil and gas companies successfully carrying on business in the United States and around the world than the few investment grade rated oil and gas companies. In his opinion, CAHPL would have been able to execute all of their business plans as the non-rated company they were in 2003, and been able to do the same had they been rated non-investment grade. Mr Rowland disagreed with Mr


ATC 17839

Hollas' statement that CAHPL did not have the same concerns regarding financial flexibility as those of a non-investment grade rated oil and gas company.

219. In Mr Rowland's opinion, viewed as a stand-alone non-rated but probably non-investment grade rated if rated, CAHPL had nearly identical concerns to most non-investment grade oil and gas companies. He said that CAHPL was similarly situated to many non-investment grade oil and gas companies. From a balance sheet perspective, and pro forma for the combination of the Chevron and Texaco assets and issuance of USD2.5 billion in long-term debt, the book equity to debt was approximately 50/50. This was not unusual, in Mr Rowland's opinion, for a company with approximately the equivalent of USD5.5 billion in assets. However, there was a significant concentration of assets that would have caused management to require more "margin for error" or financial flexibility than what might have been readily apparent. In 2003 there was a significant concentration of assets and cash flow issues that would have limited CAHPL from using any traditional, less expensive capital from secured commercial bank facilities regardless of covenants. Additionally, the issuance of even a tiny amount of secured debt would not only significantly limit flexibility but would increase the cost of unsecured debt liquidity. Mr Rowland said the use of the debt structure constructed as senior unsecured debt was absolutely prudent given the need to maintain maximum financial flexibility and minimum covenants, in the absence of a strong parent financial guarantee. Senior bank loans, if available, would have been limited in size, included many covenants that would have limited flexibility, caused the unsecured debt to be more expensive, and caused higher probability of not achieving the business plans envisioned.

220. Mr Rowland also gave oral evidence of where he disagreed with the opinion of Mr Gaskell in a report filed on 8 October 2014 during the course of the trial. He disagreed with Mr Gaskell's opinion that no one would enter into such an agreement without knowing the sources of capital for the entire funding of the project. Mr Rowland said that it was very seldom that one knew the source of funding, the capital, that would be provided for the entire length of the project. He disagreed with what Mr Gaskell had said as to substantial Australian dollar exchange risk. As to marginal cost of capital, Mr Rowland said that his assumption was that as a stand-alone entity and a subsidiary, the cost of capital for the subsidiary would be higher than the parent but the cost of the debt capital would be substantially a lot less than the cost of equitable capital, which was true for the parent, and he believed would have been true for the subsidiary as well. He disagreed with Mr Gaskell's opinion as to the credit rating for a subsidiary. Mr Rowland said that even if the subsidiary might have obtained a rating uptick of at least one, that would still not have been investment-grade, even though it was purely conjecture as to whether that could have happened or not. Lastly, Mr Rowland said that a lender's objective was to make money and generally the lender would have to take into account market risk and rates more than their own cost of capital. The lending rate that would be established by a lender would be based on the creditworthiness of the borrower and on the market rates and not on the lender's cost of capital.

221. In cross-examination, Mr Rowland accepted that Chesapeake was a domestic producer of oil and gas within the United States so that, in 2003, its income and its operating expenses were both in US dollars, it paid tax in US dollars, in 2003 its borrowings were all in US dollars and Chesapeake regularly hedged its position on oil and gas prices by forward selling and some derivatives. The hedging activities in relation to the oil and gas products did not involve use of foreign currencies. Chesapeake did not seek to protect itself against price fluctuations by taking a position in Australian dollars or some other currency.

222. In 2003 Chesapeake considered obtaining for itself an issuer credit rating but found it would be unnecessary to obtain that rating as it would pay fees for it and the company was obtaining ratings on its debt such that it was unnecessary to receive or to apply for an issuer rating.

223. Mr Rowland agreed that Chesapeake's senior unsecured notes were unsecured and covenant-lite. Chesapeake had been through a difficult period in the late 1990s and was still proving itself to potential investors.


ATC 17840

224. At the end of 2003, the largest single debt issue that Chesapeake had in place at that time was $728 million 8.125% senior notes due 2011. The senior notes were supported by guarantees given by subsidiaries as was the standard practice in the industry. The parent was borrowing and so the lenders, the senior unsecured note investors, demanded guarantees from the subsidiaries holding assets so that if the parent defaulted on the obligation, the creditors could have recourse to the subsidiaries. The notes were unsecured in the sense that no specific physical assets had been provided as collateral.

225. Mr Rowland agreed that he was not asked at any point to assume that CAHPL was an independent entity and when he expressed opinions about the transaction entered into by CAHPL he did not do so on a hypothesis that CAHPL was a freestanding entity except in considering the cost of equity as calculated by PricewaterhouseCoopers referred to by Mr Gaskell.

226. Mr Rowland accepted that if CAHPL, the subsidiary, had borrowed with the benefit of the guarantee from CVX, its parent, CVX as a major oil and gas company would have been capable of preventing any financial default by its subsidiary on the loan. On the assumption that CAHPL's projects had some significance to the overall operations of the Chevron group and assuming that CAHPL borrowed USD2.5 billion with the benefit of the guarantee from its parent CVX, Mr Rowland accepted that CVX would have an incentive to inject funds into CAHPL to prevent it from defaulting on its loan. Mr Rowland also agreed that CAHPL as a subsidiary of CVX had as much financial flexibility as CVX opted to give it.


ATC 17841

227. The respondent submitted that Mr Rowland's evidence was not relevant because it did not demonstrate how "the funding arrangements of non-investment-grade oil and gas companies are arranged". The position of Chesapeake in 2003 as described in Mr Rowland's evidence was quite different from CAHPL at that time. No inference was available that the debt channels available to Chesapeake would have been available to CAHPL or to a hypothetical company in CAHPL's position in 2003. The structure and operations of Chesapeake were not comparable to CAHPL or the CVX group as a whole and Mr Rowland made no assertion that they were. Mr Rowland "wasn't at CAHPL and is not in a position to answer questions about CAHPL". In any case, Chesapeake provided no "typical" case study of a non-investment grade borrower. This was because, first, Chesapeake's funding arrangements were quite different to CAHPL's in 2003. Secondly, Chesapeake was close to being investment grade. Thirdly, Chesapeake obtained capital as it did in order to retain financial flexibility, while CAHPL, as a subsidiary of CVX, had as much flexibility as CVX wished to give it.

228. In my opinion, Mr Rowland's evidence is of very limited assistance. I accept the submission of the respondent Commissioner that Chesapeake's borrowing practices, as described by Mr Rowland, do not provide any useful marker as to how CAHPL, if attributed a non-investment grade rating, would have transacted with an arm's length lender. I further consider Mr Rowland's evidence at [519] below.

Mr Wasow

229. Mr Wasow was a company director and a former Chief Financial Officer and Executive Vice President at Santos Limited. Mr Wasow had 23 years' experience in various financial roles at BHP, including senior roles such as Treasury Funding Manager; Assistant Treasurer, Americas; Vice President Finance BHP Petroleum, Americas; Vice President and Group Controller BHP Petroleum; and Vice President Finance, BHP. He held a Bachelor of Commerce, majoring in accountancy and financial management. He held a Graduate Diploma in Management. He was a Fellow of CPA Australia.

230. For his first report, annexed to his affidavit sworn on 18 April 2013, Mr Wasow was called by the applicant to give evidence on the question: "To the extent that a decision is made to finance E&P activities by debt, what factors would an independent Australian company in the oil and gas industry have taken into account in determining the mix or type of debt in 2003?"

231. Mr Wasow's opinion was that the principles of financing were common to all oil and gas companies in Australia, but their application was always specific to the company in question and its circumstances. He said that each form of capital had its own risk and return profile. Debt providers assumed less risk than the providers of equity capital and as a consequence demanded a lower rate of return. In addition, interest payments on debt were tax-deductible, while dividend payments to owners were not. Mr Wasow said that in enumerating a financing strategy, a company must determine how best to structure its debt liabilities within its tolerance for financial risk. He dealt with this under the headings: establishing the maximum quantum of debt; establishing the preferred maturity profile; establishing the appropriate liquidity buffer; establishing the preferred currency of debt; establishing the preferred interest rate structure; establishing acceptable covenants and security arrangements; and establishing the desirability of structured financing. The evidence in his first report was general as at that point he had not been told anything about the particulars of the matter. He had not been given any financial data for the company he was later told was CAHPL.

232. For his second report, annexed to his affidavit sworn on 27 May 2014, Mr Wasow was asked to assume that the CAHPL consolidated accounting group: primarily had operating expenses denominated in Australian dollars; derived primarily revenues denominated in US dollars; and had previously borrowed in Australian dollars and proposed to refinance much of this debt in 2003. He was asked to give his opinion on whether, having regard to the Annual Report of CAHPL for the year ended 31 December 2002, if CAHPL proposed to refinance in US dollars, its exposure to foreign exchange gains and losses would have increased, decreased, or not changed at all. He was asked to set out his reasoning.

233. His report proceeded from the view that the accounting standards were the authoritative means by which commonly used terms like "profit" were defined. Reporting companies like CAHPL would have produced Australian dollar financial statements in accordance with the standards, and any foreign exchange gain or loss would be reported in accordance with the measurement methodology prescribed in the relevant standard. Accounting Standard AASB 1012 "Foreign Currency Translation" dated November 2000 was the relevant standard for the purpose of his report. Applying the accounting standards to the CAHPL facts, Mr Wasow separated items of revenue and expense from monetary items. In his opinion, any large-scale USD borrowing would have given rise to a potential source of significant foreign exchange gains or losses. A USD revenue stream could have offset a USD principal repayment and reduced volatility, but it would not have reduced foreign exchange gains or losses. Moreover, he said, the future stream of USD revenues and the principal repayment schedule would have been mismatched in time reducing the ability of one to offset the other and thereby reducing the opportunity to affect volatility. His conclusion was that CAHPL was an AUD reporting entity and as such the refinancing of a large-scale AUD loan with a USD loan would have increased the foreign exchange gains and losses reported by the company. He said that in general, Australian companies in the circumstances of CAHPL at the time would have been financed by AUD borrowings specifically to avoid foreign currency gains and losses.

234. Mr Wasow's third report, annexed to his affidavit sworn on 28 July 2014,


ATC 17842

responded to the report of Professor Boymal, the report of Mr Gaskell and the report of Mr McCormick.

235. In relation to his response to Professor Boymal, Mr Wasow was asked to assume that CFC was not related to CAHPL. Mr Wasow's opinion was that on the assumption that CFC was independent of the CAHPL group, CAHPL would have borrowed in Australian dollars.

236. In relation to his response to the report of Mr Gaskell, Mr Wasow was asked to assume that Mr Gaskell had accurately described the evidence of Mr Dalzell, Mr Callaghan and Mr Lewis, and that CFC was not related to CAHPL. Mr Wasow said that Mr Gaskell's analysis proceeded from the assumption that CFC was not independent of CAHPL, and its results would therefore have been consolidated into the group's result with the effect that CAHPL effectively did borrow in USD. He said that Mr Gaskell's analysis could not stand if one proceeded on the assumption that he, Mr Wasow, was asked to make, namely that CAHPL and CFC were independent of each other. Further, he disagreed with Mr Gaskell's analysis of the "natural hedge". He said that Mr Gaskell's arguments for borrowing in USD were not consistent with the actual borrowing practices of the companies with which he, Mr Wasow, had experience.

237. In cross-examination, Mr Wasow agreed that his first report was a report as to how a company, such as Santos or Woodside, or a company having similar characteristics to Santos and Woodside, corporate groups with a wide range of public shareholders, would behave and that typically such a company would seek to establish and maintain an investment grade rating. He said that Woodside and Santos were reasonably comparable to CAHPL but there were distinguishing characteristics of CAHPL. Its only operating asset was the non-operating interest in the North West Shelf. Also, CAHPL did not have any public shareholders. He agreed that typically such a company would want to maintain an investment grade rating and would not take steps intentionally to give itself a sub-investment grade rating.

238. Mr Wasow was also questioned about preferred currency. He said that the rate of interest charged would depend on the currency and companies would typically hold borrowings on the currency with the lowest interest rate, subject to the qualification where it was possible to hedge the debt principal and interest payments against assets and cash flow. He was taken to the 2003 and 2004 financial reports in relation to Woodside and similar reports in relation to Santos.

239. Mr Wasow said that in all of his reports he regarded CAHPL as a single economic entity which, for the purpose of the question in his second report meant that if it had refinanced its debt in US dollars, it would have increased its exposure to foreign exchange gains and losses. He said that in the reports he had given he had been asked to assume both that CAHPL was independent of CVX and that CFC was independent of CAHPL. He distinguished what CAHPL did as a part of a much larger group from what an independent oil and gas company, operating in Australia, would have done in similar circumstances.


ATC 17843

240. I do not accept the respondent's general submission that Mr Wasow's evidence lacked the quality and degree of impartiality appropriate in an expert. In particular, I do not accept that Mr Wasow's responses to questions eliciting answers that he (apparently) considered did not suit the applicant's case theory tended to be evasive, equivocal or antagonistic, or that he presented his evidence in a partisan manner.

241. However, I regard the questions on which Mr Wasow was asked to give an opinion, and thus his answers, as of very limited utility. In my opinion, Div 13 of the ITAA 1936 did not pose the question in the terms of the assumption Mr Wasow was instructed to make, as set out at [239] above. I accept the respondent's submission that Mr Wasow's analysis of the features of Santos Ltd and Woodside Petroleum Ltd (and, later BHP Co Ltd), which were quite different in structure to CAHPL, lent no insight as to the suitability of CAHPL'S debt arrangements for the purposes of an arm's length dealing.

Mr Thieroff

242. Mr Thieroff was called by the applicant. He was a ratings witness. As will be apparent from the evidence I have accepted as given by Mr Martin and Mr Gross, I do not regard evidence of the practices of rating agencies to be relevant. Nevertheless, I shall summarise Mr Thieroff's evidence.

243. Mr Thieroff had worked as a credit analyst for over 20 years and, more particularly, in the oil and gas sector since 1994. His most recent position was as Senior Vice President, Capital Markets - Ratings Advisory for GE Capital Markets Inc in New York. Before that, his experience included working as a senior analyst for S&P with a focus on the energy and retail sectors. He was called "in the alternative" because the applicant's primary contention was that bankers and other commercial lenders did not use credit rating agency ratings. He swore four affidavits in the proceedings.

244. For his first report, annexed to his affidavit affirmed on 10 April 2013, he was asked whether the interest rate of 1 month AUD LIBOR BBA +4.14% per annum which CAHPL agreed to pay under the Credit Facility Agreement exceeded the consideration that might reasonably have been expected to have been given for the provision of that Credit Facility if CAHPL had entered into that transaction with an independent party with whom it had dealt with at arm's length. He was asked to disregard any benefit by way of implicit credit support CAHPL had by virtue of its ownership by ChevronTexaco Corporation. To answer the question, Mr Thieroff performed a credit ratings analysis of CAHPL on a stand-alone basis. He endeavoured to perform an accurate credit rating for CAHPL, taking into account his instructions. Based on the information provided and that available publicly, Mr Thieroff believed that if CAHPL had entered into a credit facility with an independent party with whom it had dealt with at arm's length, the company would have garnered a stand-alone "BB" rating. He said that his assessment was driven by the burden high financial leverage placed on the company's operations and the high amount of future spending necessary to monetise CAHPL's assets. Given the lack of expected debt amortisation, CAHPL would likely have needed to grow its reserves and production significantly to achieve the deleveraging necessary to warrant a higher rating.

245. For his second report, annexed to his affidavit affirmed on 5 August 2013, Mr Thieroff was asked for his opinion on the credit rating that might reasonably be expected to have been given to the Credit Facility if CAHPL had entered into that facility with an independent party with whom it had dealt with at arm's length. He was asked, to the extent that CAHPL had the benefit of any implicit credit support by virtue of its ownership by ChevronTexaco Corporation, not to disregard that benefit. He was also asked for his opinion as to the reasoning and conclusions recorded by Mr Emmer in his 3 July 2013 report and the reasoning and conclusions recorded by Mr Taylor in his 8 July 2013 report. The fourth question he was asked was whether the three credit facilities identified by Mr Hollas in his report as "comparable uncontrolled senior loan transactions" were transactions that, in Mr Thieroff's opinion, would enable him to determine the credit rating that might reasonably be expected to have been given to the Credit Facility between CAHPL and CFC.

246. Mr Thieroff's summary answer to the first question was that the issue rating would have been equal to that of the issuer. He based this assessment on the facts that the issue he was asked to rate was the only significant debt of CAHPL and that there were no structural enhancements to the facility that would benefit lenders beyond the underlying credit quality of the borrower. As a result, he referred to the issuer rating throughout his report. He disagreed with Mr Emmer's conclusion that ratings on CAHPL and its AUD3.7 billion Credit Facility "would fall in the 'A' category" and his reasoning in reaching that conclusion. He disagreed with Mr Taylor's conclusion that the rating on CAHPL's AUD3.7 billion Credit Facility "would fall in the range of 'A' to 'AA'" and with his reasoning in reaching that conclusion. In answer to the fourth question, Mr Thieroff did not believe that by evaluating the three credit facilities identified by Mr Hollas, he would be able to reasonably determine a credit rating on the Credit Facility.

247. For his third report, annexed to his affidavit affirmed on 7 March 2014, Mr Thieroff was asked for his opinion on the following questions. First, in issuing a corporate credit rating, what, if any, was the


ATC 17844

significance of moving from a sub-investment grade rating to an investment grade rating. Secondly, he was asked to describe the conduct of the credit committee meetings at S&P in 2002/2003. In particular he was asked how differences in opinion between credit committee members were resolved. His summary answer to the first question was that the gap separating a "BB+" rating (the highest non-investment-grade rating at S&P, or "Ba1", at Moody's) and a "BBB-" rating (the lowest investment-grade rating at S&P, or "Baa3" at Moody's) was much wider than the difference between any other contiguous rating pair. He said reasons for this, among others, included market perceptions, investment limitations regarding non-investment grade securities for certain significant investor classes and heightened reputational risk for rating agencies should an issuer default from an investment grade rating. He said he observed during his tenure at S&P that a ratings upgrade into investment grade from a non-investment grade rating faced greater committee scrutiny and deliberation than other upgrades, because of the weight given to an investment grade rating. His summary answer in relation to question two was that differences in opinion among rating committee members tended to be very small, usually one notch, occasionally two. Differences greater than two notches were rare. Generally speaking, differences among committee members were resolved through respectful deliberation. The opinions of senior analysts and managers tended to sway the votes of junior analysts. During 2002/2003 S&P had experienced a wave of negative criticism due to its role in the Enron Corp and WorldCom bankruptcies. As a result, a greater emphasis on conservatism prevailed in rating committees during that period.

248. For his fourth report, annexed to his affidavit affirmed on 29 July 2014, Mr Thieroff was asked whether he agreed with the reasoning and conclusions recorded by Ms Azarchs in her report dated 5 June 2014. He did not. Mr Thieroff's criticisms included the following. First, he queried Ms Azarchs' qualifications to comment on issues specifically relating to oil and gas companies. Secondly, he did not agree that Exhibit 1 ("S&P Ratings of Parents and Subsidiaries") to Ms Azarchs' report supported her claim that data from S&P showed that ratings of subsidiaries were most often equalised with those of their parents. Mr Thieroff said that at least 120 of the 171 subsidiaries listed in the data relied on by Ms Azarchs to prepare her Exhibit 1 did not support the point she made in her section 6.2. After eliminating the 120 subsidiaries there listed, there were 51 subsidiaries remaining of which 12 did not have their ratings equalised with the parent company; another 14 were still sub-investment grade even after taking into account any upgrade; and of the remaining 25 subsidiaries, 16 were not in the energy industry. Thirdly, Mr Thieroff took issue with Ms Azarchs' suggestions that defaults by wholly-owned subsidiaries of healthy parent companies, absent sovereign government interventions such as currency controls or confiscation, had been very rare, and that corporations generally viewed their subsidiaries as highly strategically important and/or capable of generating significant reputational risks that could have lasting negative repercussions. Fourthly, Mr Thieroff disagreed with Ms Azarchs' comments about common sources of capital in relation to parental support. Fifthly, Mr Thieroff criticised Ms Azarchs for applying criteria promulgated in 2013. Sixthly, Mr Thieroff disagreed with Ms Azarchs' statement that historically S&P had maintained that the standards for the difference in credit profiles of companies rated one notch apart was the same for all one-notch increments on the rating scale, including the increment from BB+. He said he had found nothing published by S&P to support the statement that S&P had maintained such a practice. Seventhly, Mr Thieroff criticised Ms Azarchs' statement that S&P analysts were instructed that the difference between BB+ and a BBB- rating should be no different in terms of standard applied than the difference between any other two contiguous ratings. In his experience, no such instruction was given.

249. In cross-examination, Mr Thieroff was asked questions about what would have been involved in coming up with an issuer credit rating for CAHPL, that is, a rating allocated to the entity rather than a specific obligation. Mr Thieroff was taken to an August 2000 document issued by S&P which stated that typically, affiliation with stronger family members will enhance a weaker entity's rating by at least one notch. He agreed that the document was a correct statement of the policy that S&P applied from the year 2000 at least until 2003 but qualified this answer by saying that publications in 2002 would suggest that an answer of zero notches was possible. He agreed that what was a strategic holding in the S&P document covered a spectrum from the bottom end of core down to potentially stand-alone rating and that his conclusion that CAHPL was a strategic holding opened up the question where in that spectrum between its stand-alone rating and its parent's rating it was to be placed. Its stand-alone rating in his assessment was DD and the parent's rating was AA.

250. Mr Thieroff was taken to Ms Azarchs' analysis which began with 171 subsidiaries from which he excluded 120 leaving 51, of which 12 did not have the ratings equalised with the parent company and 39 did have the ratings equalised. Mr Thieroff agreed that if one wished to use the data in order to reach a conclusion about S&P's behaviour in all cases where a non-guaranteed and wholly-owned subsidiary was given a credit rating, there should be included in the analysis some proportion of the subsidiaries which were assessed as being core. He could not say how many of the 24 should be added back in but the sample size would increase from 51 although not to 75. All of the companies added back in as being core had their ratings equalised with their parents. He agreed that S&P, since at least 2000, had published ratings, methodologies and criteria expressed to apply to companies across all industries, that S&P had always attempted to maintain consistency across industries in its ratings; and the judgment that needed to be made in an individual case about the likelihood of parental support depended on the same issues regardless of industry. Those issues went to the parent's ability to provide support and the strength of the incentives for it to provide support. He agreed it was not necessary to break the analysis down to particular industries.

251. Mr Thieroff was taken to documents he had been involved in preparing where S&P had raised the corporate credit rating by more than three notches in relation to Pennzoil-Quaker State Company, Coral Energy and Unocal Corporation. Mr Thieroff was also taken to instances referred to by him of highly rated parent companies curtailing investments in subsidiaries. He was also taken to a statement by S&P that a possible bankruptcy finding by Enron and its affiliates would not likely cause changes to the ratings or outlooks on almost all rated US E&P and integrated oil and gas companies. That was his own view in November 2001. He said that S&P did not expect a wave of defaults from E&P companies as a result of Enron. Mr Thieroff was taken to part of an October 2002 document headed "Key Issues Affecting Credit Quality" and agreed there was no suggestion in that part of the document that the failure of Enron or other energy traders was a large issue affecting the credit quality of integrated oil companies and said he did not imply that there was a concern with E&P companies as a result of the Enron bankruptcy.

252. The respondent criticised the evidence of Mr Thieroff as follows. The designation of CAHPL as "strategic" was merely a starting point for consideration of where, in the range between BB and AA, CAHPL's rating would sit. There was a continuum from the bottom end of core down to immediately above a stand-alone company. Within that continuum, there was a necessity to evaluate, based on available data, the impact that parental support had in a particular case. This depended upon the same issues, regardless of industry. In his report, Mr Thieroff said he had never experienced a situation where a non-guaranteed, non-core


ATC 17845

subsidiary of a multinational company was notched upward three or more ratings above its standalone rating". However, he was named as credit analyst on a number of S&P publications which indicated the "notching up" of non-guaranteed subsidiaries by three notches or more to reflect parental support. He therefore had experience in observing the notching up of subsidiaries that he disavowed in his report. If his opinion that CAHPL would only receive one notch of uplift to reflect implicit support was based on his claim to have never seen an increase of three or more notches upon a stand-alone rating of a non-guaranteed, non-core subsidiary of a multinational, that claim was shown to be wrong. As a result, the judgments that he had made as regards CAHPL's notching up should be rejected.

253. In addition, the respondent submitted that Mr Thieroff did not consider what CAHPL's credit rating would be if it were to borrow $2.45 billion in USD rather than in AUD. There was a degree of iteration between determining interest rate and credit rating. Although Mr Thieroff notched CAHPL's rating up by one notch to just below investment grade, he did not recalculate the impact that would have on CAHPL's credit ratios. The respondent also submitted that the Enron collapse did not affect E&P ratings. The Enron collapse was unrelated to CVX's operations and involved a case of corporate fraud: it did not increase the conservatism of rating agencies towards E&P companies generally. The energy companies referred to in Mr Thieroff's report as having had their ratings downgraded at around the time of the Enron collapse did not include a single E&P company.

254. If, contrary to my opinion that the practice of rating agencies is not relevant, I find that Mr Thieroff's evidence, on his instructions as set out at [244] above, is not useful in answering the statutory questions.


ATC 17846

Ms Esposito

255. Ms Esposito was a ratings witness called by the applicant. Again, I do not regard this evidence as relevant, but I shall summarise it.

256. In her affidavit Ms Esposito, said that she had over 25 years' experience at S&P, including 10 years as a Managing Director. She had 30 years of experience in global capital markets. She had retired from her position as Managing Director and global capital markets executive and was an independent researcher. During her time at S&P she focused on different industries, including energy, mining, metals fabrication and construction materials. The first question on which she was asked for her opinion was: "In issuing a corporate credit rating, what, if any, is the significance of moving from a sub-investment grade rating to an investment grade rating? In particular, do you agree with the reasoning and conclusions recorded by Ms Azarchs in her report dated June 2014? If there are any aspects on which you disagree, please set out your reasons." Ms Esposito was also asked to describe the conduct of the credit committee meetings at S&P in 2002/2003 and, in particular, how differences in opinion between credit committee members were resolved.

257. In her report, Ms Esposito said that in her analytical experience and recall of ratings practices from approximately 2004, there was much greater significance to the move of an obligor rating from non-investment grade to investment grade. She said, there was a crevasse between "BB+" to "BBB-". This was especially so in regards to obligors in oil and gas (O&G) E&P which was capital intensive and highly cyclical with reliance on globally determined prices. Ms Esposito said nowhere in her review or application of the S&P Corporate Ratings Criteria 2003 (or in the April 2003 S&P Global Credit Portal RatingsDirect: Industry Report Card: Global Oil & Gas) did she find analytical directions given or the policy cited by Ms Azarchs that "standards for the difference in credit profiles of companies rated one notch apart are the same for all one-notch increments on the rating scale, including the increment from BB to BBB-". In her experience, Ms Esposito said, the process of determining whether a firm was credit risk rated BB+ or BBB- focused on the individual obligor's credit factors and the weighting of these factors to come up with the credit risk rating. These factors included but were not limited to consequential differences in economic, competitive, operational and financial circumstances. She also said that the oil and gas E&P sector fell into the category that had above average industry risk. E&P companies particularly had an inferior credit quality to integrated oil companies. In addition, many of the E&P industry's concerns were beyond the control of the obligor's management. What she had observed in her years of rating experience was the application of qualitative and subjective assessments contributing to a more conservative approach in moving from non-investment grade to investment grade. Ms Esposito did not agree with Ms Azarchs view that rating gaps were the same for all one-notch increments, even between non-investment grade and investment grade. Oil and gas E&P observations on the lower "B" and "BB" levels could generate counterintuitive results due to, amongst other things, the dearth of their observations in the full study referenced by Ms Azarchs, anomalies due to non-E&P causes, and specificity concerns.

258. In cross-examination, Ms Esposito agreed that she did not cite in her report any specific examples of a company being rated BB rather than BBB because of the special test being applied at that boundary. Ms Esposito accepted that the ratings definitions did not suggest that there was no level of risk at all associated with investment-grade obligations, but she said it was an extremely low risk. Ms Esposito agreed that the investment/non-investment grade distinction had a particular significance in the markets and a corporation which was rated below the investment grade threshold usually had significantly restricted access to funding. Any serious adverse effects of a downgrade from BBB to BB by S&P were a matter for the market's decision and not something which was a consideration in the rating room. Ms


ATC 17847

Esposito accepted that the goal of the rating was to be a forward-looking exercise which attempted to give an indication of the creditworthiness of a company or of a debt issue. Ms Esposito was taken to a table showing the default rates of companies that had been rated. She agreed that the curve that resulted looked roughly like a graph of an exponential function. She added that she did not believe that the rate of increase in the default rate was at the same rate.

259. It was put to her that if there were a special test or a special conservative bias applied at the BBB/BB boundary, companies that would normally merit BBB could be seen to be put into BB. Ms Esposito responded that if the individual case facts showed a BBB then it was a BBB. However, if you were looking forward to things happening, as an analyst you would be less inclined to give that credit, so it would be in the BB category awaiting the fruition of the positive events to happen. The analyst would be thinking very critically and not giving the benefit of the doubt to things that had not yet happened or looked like they were about to happen: the analyst would rather wait to see them happen which was why, she said, she felt there was a larger crevasse between the investment-grade and non-investment grade. By conservatism she agreed she meant striving to get things right. As an analyst at S&P in 2002 or 2003 she would be hyper-focusing on the things that could go wrong and she would not be giving as much credit to the things that could possibly go right because they had not yet occurred. She said that the hyper-focusing occurred at the BBB/BB boundary and the other boundary she would see it would be at the A+/AAA boundary. Ms Esposito was taken to the curve showing corporate default rates by rating in the report of Ms Azarchs and said that that curve, if it were limited to E&P would be likely to show a kink or show the majority of the whole curve shifted more because two thirds of E&Ps were non-investment grade. Ms Esposito agreed that she had not referred to any specific examples of the analyst needing to have a very strong belief before they would move the company into a BBB category and if they were waiting for certain factors to occur then they would wait. She agreed she had not herself done a study of defaults by E&P companies and said she did not have the data and it was not the question she was asked. It was put to her that she could only speculate about what a study directed at those companies would show and she answered that she could speculate but she felt it was a strong speculation on the 26 years that she had been doing this work.

260. The respondent submitted that the "crevasse" described by Ms Esposito was not based on any reliable foundation. The basis for that submission included the following. First, Ms Esposito did not refer in her report to any statement in any S&P publication which pointed to a special test being applied by analysts in respect of companies at the boundary between investment grade and non-investment grade ratings. Secondly, the heuristic statements provided by Ms Esposito provided no persuasive basis for her opinion: Ms Esposito suggested that the non-investment grade definitions "cause analysts greater pause" because investments in non-investment grade entities were "speculative". That observation, even if it were accepted, provided no explanation for Ms Esposito's opinion in relation to the approach taken by analysts to companies lying towards the investment grade and non-investment grade boundary. As Ms Esposito (eventually) agreed, the term "speculative", as used in ratings descriptions, implied an increasing level of risk as one moved between ratings, and not any binary position whereby risk or speculation was present for non-investment grade entities but wholly lacking for investment grade entities. Thirdly, the suggestion that the asserted investment/non-investment grade "crevasse" existed for E&P companies or all companies, despite not being evident in the data more generally, would be inconsistent with the logic behind credit ratings. Fourthly, Ms Esposito's evidence was not borne out by data from S&P backward-looking studies of corporate defaults, which she accepted provided a methodology for assessing the effectiveness of S&P ratings as a predictive tool.

261. On the assumption that this rating agency evidence is relevant, I am not persuaded by the metaphor of the "crevasse" which I regard as a distraction. The real question is whether there is a greater conservatism on the part of rating agencies at the BBB/BB boundary. This is put as educated speculation or informed speculation based on many years' experience and I accept that there was conservatism although I am not persuaded that there was any greater conservatism at the BBB/BB boundary than at any other. I find, however, that there was caution exercised in moving from non-investment grade to investment grade.

Dr Becker

262. Dr Becker, a transfer pricing economist, was called by the applicant as an expert. He was the President of Precision Economics LLC. He said that in over 20 years as a consulting economist, he had prepared over 400 transfer pricing reports for taxpayers, law firms, and tax authorities. He had also served as a transfer pricing economic expert witness in a number of Australian and international transfer pricing cases. He swore four affidavits.

263. He was not put forward as a primary or pricing witness but to demonstrate the deficiencies in the work undertaken by Dr Horst and Mr Hollas to the extent that that was necessary.


ATC 17848

264. For his first report, annexed to his affidavit affirmed on 2 August 2013, Dr Becker was asked whether or not the three comparable agreements identified by Mr Hollas represented true CUTs to the agreement between CFC and CAHPL. He was also asked whether he disagreed with the report of Mr Hollas dated 8 July 2013 and whether he disagreed with any aspects of the report of Dr Horst dated 8 July 2013.

265. As to Mr Hollas' report, Dr Becker said that the three foreign agreements for comparables each had significant differences from the CFC-CAHPL loan. Those factors caused the Hollas report's arm's length range to be unreliable and biased. Faced with the difficulties in acquiring comparable agreements evidenced by the Hollas report and the relative ease of assessing other market data for other pricing methods, this form of a CUT approach did not appear to represent a precise or reliable method for pricing in the present case.

266. As to Dr Horst's report, Dr Becker's conclusion was that the Horst report set the interest rate of CAHPL's AUD loan by translating its estimate for its credit rating to an interest rate premium over AUD LIBOR. In this process, Dr Becker said, the Horst report: (a) concluded a single point stand-alone rating when a range of interpretations would be more appropriate; (b) opined for the high-end of implicit support range


ATC 17849

with little substantiation; (c) failed to address the large interest-rate impact (moving from non-investment to investment grade) of his notch increase concluded for implicit support; and (d) chose a low interest rate outlier from the range of interest rates implied by his analysis. The Horst report's use of an ordered probit (ordered probability unit) approach to set CAHPL's stand-alone rating had some support in statistical literature, but the Horst report did not provide a full interpretation for its own results as its ordered probit results would be interpreted by many economists and statisticians as supporting a BB- stand-alone rating in addition to Dr Horst's interpretation of a BB rating. This range fell in the non-investment grade of the ratings grades. Whether implicit support was strong enough to bridge the gap from non-investment to investment grade (mathematically) drove whether the ultimate interest rate opinion differed significantly from the rate paid by CAHPL. However, unlike the detail in the ordered probit, the Horst report provided almost no quantification for its opinion that implicit support would lower the interest rate by 268 basis points. Rather, the Horst report's citations made it clear that companies like CAHPL might not receive a bump at all. Using the range of implicit support bumps cited by the Horst report documents and the range for a stand-alone rating derived from the Horst report's ordered probit, most of the possible combinations yielded interest rates of 340 or more basis points above AUD LIBOR. The Horst report, by contrast, had not allowed for a range but concluded for a single point that represented the high-end of both ranges: BB rating plus to two notches (144 basis points premium).

267. For his second report, annexed to his affidavit affirmed on 6 March 2014, Dr Becker was asked whether he agreed with the propositions that for transfer pricing purposes the loan between CFC and CAHPL should be priced solely by undertaking an analysis of the risks and functions of CFC and, because CFC was able to raise its funds with the benefit of the guarantee from CVX, a margin between USD LIBOR +0.73% (equalling 2.05% as at 2 June 2003) and the 1.2% payable on the commercial paper it issued, amply compensated an independent lender in the position of CFC. Dr Becker said that the contention regarding the lender was not accurate as evidence showed that interest rates on loans similar to that between CFC and CAHPL were based upon the likelihood of default by the borrower. Dr Becker also said that the lender's rating/riskiness did not influence interest rates it charged when lending, regardless of whether it was guaranteed on its own borrowings. As such, he would not value the interest rate at issue based upon the characteristic/rating of the lender - with or without an external guarantee.

268. For his third report, annexed to his affidavit affirmed on 12 March 2014, Dr Becker was instructed that the drawdowns under the Credit Facility took place by the transfer of the USD equivalent of AUD2,180,451,128 on 6 June 2003 and AUD1,526,717,557 on 26 August 2003 to a bank account of CAHPL held with Citibank in the United States: those transfers were effected by crediting these USD equivalent amounts to that bank account and thereafter CAHPL paid interest and principal under the Credit Facility by transferring the USD equivalent of its AUD liabilities to bank accounts held in the name of CFC with Citibank and subsequently JP Morgan in the United States. He was asked whether these instructions changed any of his conclusions and to provide reasons. He was instructed to assume that the operating expenditure of CAHPL and its consolidated accounting group were at all relevant times predominately incurred in Australian dollars and the revenues of CAHPL and its consolidated accounting group were at all relevant times predominately in US dollars. Dr Becker concluded, from a transfer pricing economist's perspective, that the borrowing activity required CAHPL to pay back a predetermined level of AUD funding. The credits and debits for interest rates also followed the AUD principal, not the USD principal. While Citibank and JP Morgan credited and debited USD accounts at then prevailing rates (those rates differing at the time of loan and payback) for each of the transfers in these transactions, the borrower had no ability to pay back the loan in USD. CAHPL received credit of the AUD equivalent of USD2.45 billion but seven years later its principal repayment/debt obligation was not USD2.45 billion, nor the then-prevailing AUD equivalent thereof. Rather, it needed to payback USD3.75 billion or 152.9% of principal in USD terms. The agreement between and actions of the parties corresponded to the AUD form, that is, a repayment of exactly 100%, not 152.9%, of principal. His opinion, that the loan in question was in AUD, had not changed.

269. For his fourth report, annexed to his affidavit affirmed on 28 July 2014, Dr Becker was instructed to prepare a report in reply to affidavits of Mr Hollas sworn 5 September 2013 and 4 June 2014, affidavits of Dr Horst sworn 5 September 2013 and 3 June 2014, and to prepare a report regarding borrowing rates of interest. Dr Becker dealt with these matters in separate reports.

270. As to his analysis of the affidavit of Mr Hollas sworn on 5 September 2013, Dr Becker grouped his disagreements under the headings: "Use of Bloomberg Compiled Interest Rates and Conversion to Australian Rates"; "Hollas CUT Report Comparables Differ from the CAHPL Loan"; "Detailed Description of Problems with the Hollas CUT Report 2"; "Small Loans and Large Senior Debt", "New Comparable in Hollas CUT Report 2", and "Conclusion".

271. Dr Becker said the same general disagreements/concepts applied to his disagreements with the Hollas revised search, being the affidavit of Mr Hollas sworn on 4 June 2014. He grouped his disagreements under the headings: "Hollas Revised Search Comparables Differ from the CAHPL Loans"; "Detailed Description of Problems with the Hollas Revised Search"; "Small Loans and Large Senior Debt"; and "Conclusion". In his conclusion, Dr Becker said that the Hollas revised search used a methodology that was not apt for pricing this loan. A more appropriate methodology was to first determine the rating of the borrower (and its loan) and then to use that rating to apply interest rate information from recognised sources such as Bloomberg, DealScan, et cetera. The Hollas revised search located benchmark agreements to compare to CAHPL's loan in question. While comparisons of agreements was only one type of potential pricing approach it was also important to understand the universe of all debt vehicle agreements available for consideration as well as other approaches in pricing interest rates. In this case, Dr Becker said, the Hollas revised search opined that American companies agreeing to borrow in USD for short-term working capital, with financial covenants, would have similar rates to an Australian company actually borrowing in AUD for all of its long-term operating needs, without financial covenants; and the Hollas revised search offered no opinion on CAHPL's ratings, but implied that it would borrow at investment grade rates. With the gulf between the lowest investment-grade and highest non-investment-grade notch representing 196 basis points, this one ratings notch alone translated to approximately AUD360 million of income. Dr Becker said that his report presented other agreements that would be closer matches to CAHPL than the four comparable agreements used in the Hollas revised search. That is, Dr Becker said, the other agreements he used, loans and notes, actually took place (money was borrowed) and were structured in a term (versus revolver) formula. While he did not present those debt vehicles affirmatively to price CAHPL's loan, those transactions more closely matched CAHPL than the Hollas revised search agreements.

272. As to the affidavit of Dr Horst sworn on 5 September 2013, Dr Becker found several disagreements with the "Horst Rate Report 2" and he grouped them by topic areas: "Cited Evidence Contradicts the Horst Rate Report 2's opinions"; "Conclusion to be Drawn from Ordered Probit Results"; "Implicit Support"; "Sensitivity of Results"; "Re-Characterising the Loan into Foreign Currency"; and "Critique of What Was Not Done".


ATC 17850

273. As to the affidavit of Dr Horst sworn on 3 June 2014, Dr Becker referred in his criticisms of the Horst report to incomplete and inconsistent information; misstatements by Dr Horst of his own (Dr Becker's) assignment and opinions; and the transaction being priced.

274. Dr Becker's final report provided his opinion on the question whether he agreed that if the borrowing rate of interest in the USD market was considerably lower than the borrowing rate of interest in the AUD market, this in itself would have made a USD borrowing more attractive to CAHPL. He described why his opinion was "no" with facts and economic theory which he set out showing why companies like CAHPL did not always (or even typically) borrow in the currency with the lowest rates. His disagreement with the premise centred on the range of other considerations that, in addition to current interest rate differentials, comprised the basis for firm borrowing decisions. He said the attractiveness of a lower current USD market interest rate may well be offset by one or more of the considerations which he then noted.

275. In cross-examination, Dr Becker agreed that he regularly provided evidence in transfer pricing matters about the methodology which


ATC 17851

an economist may take to a transfer pricing question. Although he did not do it in this particular case, he also regularly provided opinions as to the arm's length price of particular transactions. Dr Becker agreed he did not profess to be a banking expert or an expert lender. He thought of himself as an economist with knowledge of lots of different industries, rather than as an expert in a particular industry. He did not claim expertise as an active participant in the bank or capital markets as a lender to the E&P sector.

276. One of the areas of disagreement or debate between Dr Becker and Dr Horst was whether Dr Horst's Case 1, which Dr Horst preferred, or Case 9 was the more appropriate Case to choose. Dr Becker said that the context was that if an economist were to have gone through the process that Dr Horst did, which Dr Becker did not believe a typical transfer pricing economist would have done, and had those ordered probit results, then the appropriate Case to choose would be the Case 9 rather than any of the other 17 Cases. Dr Becker said that ordered probit was a known analytical tool but it was certainly not a very common one compared to others: it was something that was in literature and in books on statistics. Dr Becker said Dr Horst's Case 1 used something called the total notch differential as the method to indicate what he considered to be the best outcome, whereas Dr Becker preferred the log likelihood as best indicating the best outcome. He said there were a number of measures or approaches based on the log likelihood, including the log likelihood and information criterion approaches, which were the best way to compare models. His understanding of the expression "log likelihood for the model" was the log of a likelihood function of how likely the prediction was to be correct, broadly speaking.

277. Dr Becker said that he had never heard of the term total notch differential and he had never seen it applied before so he did not know if it was Dr Horst's new method or of it was a shorthand method, but he said it was not a method recognised by statisticians. He said he first looked through statistics books, those that he had found and those that Dr Horst had cited, as well as doing a Google search but did not find the term "total notch differential" anywhere. Dr Becker understood that Dr Horst, when he used the term "notch", was talking about a notch between one rating and another rating. He understood the context to refer to the differences between particular ratings. Dr Becker was then taken to a number of documents using notches as a measure of result in the specific context of credit ratings.

278. In relation to the reports prepared by Mr Hollas, Dr Becker agreed that he had located agreements on DealScan for 70 companies in the energy sector, and his search was not specifically for E&P companies but was across the energy sector. He rejected a number of the agreements and one of his reasons for rejecting agreements was that they were revolver agreements.

279. Dr Becker agreed that a company with a credit rating of CCC will borrow, typically, at a higher yield than a company with a credit rating of BB. Dr Becker had already said in his report that the rates amongst energy firms varied significantly but followed a pattern. That is, investment grade companies engaging in the loans borrowed at consistently and noticeably lower rates than non-investment grade companies. Dr Becker's Table 12 was illustrative of the range.

280. Dr Becker agreed, in relation to methodology, that as a transfer pricing economist one needed to take the precise terms of the particular transaction in order to determine an arm's length price. His understanding was that the form of the transaction in the present case was that CAHPL was granted the use of funds without providing financial covenants, security or a guarantee. He agreed that if he were engaged to price the transaction then generally his approach would be to price it assuming that it was provided with no financial covenants, security or a guarantee. He agreed that the necessary consequence of the way he would approach the task as a transfer pricing economist was that related parties to a loan agreement could choose the terms of


ATC 17852

their agreement such that they operated to increase the interest rate payable. What he would be looking at was whatever the arm's length price was. So, in general, if the parties were operating to increase a rate beyond an arm's length level then that would come up in his analysis. It was put to Dr Becker that a consequence of his methodology as a transfer pricing economist was that related parties could reach an agreement which increased the interest payable by agreeing not to provide security or not to provide financial covenants. He answered that parties could structure their agreements in different ways. If they wanted to add risk to the lender, they could do that; if they did not want to add risk to the lender, they could choose not to. But the way he had always done his work was whatever risk was assigned between the two related parties could be priced, and as long as the price was at arm's length that tended to be the way his opinion went. So if you were adding risk to the lender, by definition, the lender should be paid for that. He said that if he were doing an analysis he would take the terms of the transaction and, assuming they could be priced and there was no impediment to him pricing those transactions, he would take the risks, and the functions, and the comparables, and find an arm's length price.

281. The respondent made detailed and extensive criticisms of Dr Becker's evidence. He was criticised in relation to what he said were more comparable or better matching loans to the CAHPL loan. It was submitted that his rejection of revolver loans was based on the false assumption that they were short-term facilities. The method of compilation and the content of his Table 12 were heavily criticised by the respondent. It was submitted that Dr Becker could not have genuinely considered the loan agreements on which he relied to be better matches. He was also criticised for a lack of relevant expertise both as to the meaning of various credit ratings and the effect of not having a credit rating, and his method of converting a margin over US Prime to a margin over LIBOR. Dr Becker was criticised for his suggestion that there was a 196 basis point difference between investment grade and non-investment grade loan interest rates.

282. The respondent made submissions critical of Dr Becker's criticisms of Dr Horst's ordered probit. It was submitted that Dr Becker and Dr Horst were at odds as to whether or not to select Case 1 or Case 9 of Dr Horst's ordered probit analysis. Dr Becker considered that most economists would choose Case 9 because it provided a better log likelihood. However, the respondent submitted, the real question was which one produced the most reliable prediction. Dr Becker conceded that, comparing Case 1 with Case 9, Case 9 did not result in the highest number of correct predictions with the lowest degree of incorrect predictions. It was submitted that although Dr Becker was highly critical of Dr Horst's reference to "Total Notch Differential", it was clear that Dr Horst was simply explaining the method he chose to assess the performance of his model. Dr Becker, it was submitted, did not criticise the "Total Notch Differential" until his seventh report where he said the method did not exist in any statistical content. This was contrary to a number of publications. The result of Dr Horst's analysis showed that without consideration of implicit support, CAHPL would have a rating of BB+. Even if one accepted Mr Thieroff's view that implicit support accounted only for one notch, CAHPL would be considered investment grade.

283. I do not accept the respondent's submission that Dr Becker's general approach was that of an advocate for CAHPL, hired to find fault rather than assist the Court in determining an appropriate arm's length consideration. I find, however, that Dr Becker's opinion was insufficiently related to the statutory task to be of use. As I have noted above, Dr Becker was not put forward as a primary witness but to demonstrate the deficiencies in the work undertaken by Dr Horst and Mr Hollas. This may explain the theoretical nature of Dr Becker's reports. I state my conclusions in relation to Dr Becker's evidence in more detail at [520]-[524] below.

Professor Walker

284. Professor Walker, Emeritus Professor of Accounting at the University of Sydney, was called as an expert by the applicant. Professor Walker was asked whether he agreed with the reasoning and conclusions of Professor Boymal in his report of 5 June 2014 that, in his view, the optimal currency of the loan was US dollars.

285. In his report, Professor Walker first addressed the first question asked of Professor Boymal which was, from the point of view of the CAHPL consolidated financial statements, in what optimal currency would CAHPL have borrowed USD2.45 billion or its equivalent in another currency on the assumption that CFC was not a subsidiary of CAHPL, CAHPL was independent of CFC and dealing with CFC at arm's length, and on the assumption that 88% of the CAHPL group revenue was in USD. He stated his opinion that Professor Boymal's discussion of the "optimal currency" in which CAHPL might have borrowed to fund its Australian operations did not consider a range of commercial considerations that would be relevant to this decision. Professor Walker listed a wider range of commercial considerations at [12] of his report and stated that these other commercial considerations may have included (a) the scale of planned capital expenditure in Australia - which was to be predominantly in AUD; (b) forecast operating expenditure in Australia - which was to be predominantly in AUD; (c) expectations about future movements in the price of goods and services to be purchased in Australia to undertake planned capital expenditure; (d) expectations about future movements in prices of commodities to be sold; and (e) expectations about future movements in exchange rates between the AUD and USD, and the possibility of future gains or losses arising from those movements. He said that a decision about financing arrangements would necessarily be based on plans about future capital investments (that would be controlled by CAHPL), and expectations about a range of future events (including events that would be beyond CAHPL's control).

286. Professor Walker also addressed the second question asked of Professor Boymal which was: "Do you agree with the accounting and related justifications provided by Messrs Dalzell, Callaghan and Lewis for the choice of AUD rather than USD for the CFC to CAHPL loan, assuming that 88% of CAHPL group revenue was in USD, that the loan was denominated in AUD and that the loan was in substance an AUD loan?" Professor Walker said, amongst other things, that Professor Boymal's observation about the elimination of an intercompany loan was inconsistent with the assumptions that he was asked to adopt, namely that CFC was not a subsidiary of CAHPL and CAHPL was independent of CFC and dealing with CFC at arm's length.


ATC 17853

287. Professor Walker was also asked whether he agreed with the comments made by Mr Gaskell in relation to "Australian Generally Accepted Accounting Principles and Australian equivalents of International Financial Reporting Standards" in specified paragraphs of Mr Gaskell's report dated 3 June 2014. He said he disagreed with Mr Gaskell's statement that CAHPL's consolidated accounts would have the same volatility as if CAHPL had itself borrowed in USD. He disagreed with Mr Gaskell about the application of AASB 121. Professor Walker concluded that Mr Gaskell's critique of Mr Callaghan's observations regarding the impact of borrowing in USD on CAHPL's financial statements was incorrect. The introduction of AASB 121 would still have required unrealised gains or losses on foreign currency exchange rates to be reflected in CAHPL's financial statements, had CAHPL borrowed in USD.

288. In cross-examination, Professor Walker accepted that the fact that interest rates in US dollars were lower than Australian dollar rates would be one factor that would tend to support a US dollar borrowing rather than an Australian dollar borrowing. He also accepted the fact that 88% of a company's inflows or revenues were in US dollars would be a factor that would support incurring outflows in the same currency. He agreed that those two factors on their own would support a decision to borrow in US dollars rather than Australian dollars. As to the other commercial considerations referred to by Professor Walker at [12] of his report, he agreed that the reference to planned capital expenditure in Australia predominantly in Australian dollars was an assumption by him. As to the forecast operating expenditure in Australia which Professor Walker said was to be predominantly in Australian dollars, he appeared to accept that the statement of cash flow to which he referred contained inconclusive reference to GST and thus to whether the expenditure was Australian, and that the statement did not say anything about the scale of the expenditure and whether the actual expenditure was in Australian dollars or some other currency. As to whether future capital investments would be controlled by CAHPL, Professor Walker accepted that the only way one could work out whether any members of the group were planning to incur capital expenditure was to have a look at the consolidated accounts and he accepted that it was incorrect to say that the capital investments could be controlled by CAHPL as a separate entity.

289. The respondent criticised Professor Walker's evidence in a number of respects, as follows.

290. Professor Walker was not asked to provide his independent opinion on any fact in issue in the proceedings. Rather, like Dr Becker, Professor Walker was instructed to find fault with the conclusions expressed by two of the respondents' experts; in Professor


ATC 17854

Walker's case, these were Professor Boymal and Mr Gaskell. The respondent submitted that Professor Walker's report should be afforded little weight. Professor Walker was extremely critical of Professor Boymal and Mr Gaskell in his report, which reflected poorly on his independence. However, in cross-examination, Professor Walker expressly agreed with the validity of the latter experts' reasons for opining that the hypothetical CAHPL loan would have been denominated in USD. Their reasoning should be accepted and Professor Walker's critiques dismissed as peripheral to the question to be decided.

291. For the purpose of providing his report, Professor Walker was asked to assume and did assume, in addition to CAHPL's independence from CFC, that CAHPL was not a direct or indirect subsidiary of CVX. The respondent submitted that if the Court found that, in the relevant statutory counterfactual, CAHPL must be considered as a company that was a subsidiary of CVX or a US-based multinational energy company in the position of CVX, Professor Walker's opinion would be wholly irrelevant.

292. The respondent submitted that Professor Boymal's analysis of optimal currency and consolidated accounts was to be preferred. Professor Walker's criticisms of Professor Boymal's report were revealed in cross-examination either to have been unjustified, based on the difference in the assumptions that he and Professor Boymal were asked to adopt, and/or based on a matter of little consequence.

293. As to the loan currency, Professor Boymal expressed the opinion that the optimal loan currency was USD because: interest in the USD market was at a considerably lower rate than the borrowing rate in the AUD market; and since 88% of the CAHPL group revenue was in USD, then group borrowings in USD would provide a natural hedge against the volatility in group sales revenue arising from changes in the USD/AUD exchange rate. Professor Walker agreed that both of those considerations were factors that would support a borrowing in USD. In particular, Professor Walker agreed that there were commercial benefits obtained by a USD borrowing providing a natural hedge (albeit imperfect) against forex exposure.

294. However, Professor Walker was unwilling to take the logical and natural next step of agreeing with Professor Boymal's view that the optimal currency for borrowing would thus be USD. Professor Walker was unable convincingly to identify the asserted "other matters" that would or may tend against a USD borrowing by CAHPL. Professor Walker's comment that "a prudent financial management team would need to monitor developments" in foreign exchange rates said nothing about the optimal currency of the CAHPL loan once analysed by a team with appropriate expertise.


ATC 17855

295. Professor Walker commented on Professor Boymal's statements in the latter's report in relation to the accounting and related justifications provided by Mr Dalzell, Mr Callaghan and Mr Lewis for the (purported) choice of AUD rather than USD for the Credit Facility between CAHPL and CFC. Professor Walker agreed that his primary disagreement with Professor Boymal as regards Mr Callaghan's evidence came down to the difference in assumptions that they were asked to adopt. Professor Walker's suggestion that Professor Boymal ought to have considered Mr Callaghan's evidence on the basis that CAHPL and CFC were independent was irreconcilable with the fact that Mr Callaghan's evidence related to the actual transactions which took place in the world in which CAHPL and CFC were related.

296. The respondent submitted that even if one accepted Professor Walker's evidence at face value, no part of it supported a hypothetical CAHPL borrowing in AUD: despite having expertise to opine on this issue, he was not asked to do so.

297. Professor Walker accepted that matching the currency of inflows to the currency of outflows of a company reduced the risk of adverse foreign exchange movements from a commercial point of view, and that mismatching the currency of inflows and outflows exacerbated a company's foreign exchange exposure. He acknowledged that a borrowing in USD would have provided a partial hedge and minimised the foreign exchange exposure for the group. In this respect, Professor Walker disavowed the suggestion that an AUD outflow by CAHPL would operate to reduce foreign exchange exposure against a USD inflow: "no, I'm not suggesting that at all." His evidence thus directly contradicted that of Mr Wasow and Dr Webber in respect of the purported natural hedge arising from a correlation between oil prices and exchange rates.

298. Professor Walker acknowledged that the CAHPL economic entity in fact made a decision to borrow in USD and that when one looked at the consolidated accounts of CAHPL, they showed foreign-exchange fluctuations as a result of the decision of CFC to borrow in USD. Professor Walker agreed that whether CAHPL borrowed in AUD or USD from CFC, which itself borrowed in USD, made no difference to the foreign-exchange risk incurred by the group as a whole.

299. The respondent submitted that Professor Walker criticised Professor Boymal for failing to explain his view in his (Professor Boymal's) report as to why information from consolidated financial statements may be more meaningful to a reader than financial information presented for parent companies on a stand-alone basis. That criticism was unwarranted. Professor Walker accepted that the object of AASB1024, which required economic entities to prepare consolidated accounts, was to ensure the provision of relevant and reliable financial information about related entities as a single reporting entity to reflect their operation as a single economic unit. It was in that context that Professor Boymal provided his unobjectionable opinion as to the meaningfulness of consolidated accounts.

300. The respondent submitted that Mr Gaskell's opinion on functional currency was to be preferred. It was put that Professor Walker had no basis to disagree with Mr Gaskell's evidence as regards the accounting treatment of CAHPL. Further, Professor Walker merely assumed that the functional currency of CAHPL and its subsidiaries was AUD. He conceded that he had not actually checked what the functional currencies of CAHPL's subsidiaries were.

301. I do not accept the respondent's submission that Professor Walker presented as a quarrelsome and partisan witness or that he was unduly critical of the reports upon which he was asked to comment. I do accept, however, that Professor Walker's criticism of the "directive" nature of the question posed to Professor Boymal was pedantic as was his criticism of the language used by Mr Gaskell in [4.4] of Mr Gaskell's report.


ATC 17856

302. I find that they can be no doubt as to the currency of the Credit Facility, which was in Australian dollars. I consider the issue of the currency of an arm's length loan or a loan between independent enterprises at [583] below.

Dr Webber

303. The applicant called evidence from Dr Webber, the Managing Director of Webber Quantitative Consulting Pty Ltd, a consulting firm specialising in quantitative and statistical modelling. He was also an Associate Professor at the University of Sydney. He had over 20 years' experience in statistical and econometric modelling. He had been at Qantas Airways Ltd, first as General Manager Economics and then as Chief Economist.

304. Dr Webber was asked to prepare a report setting out his opinion on the following question: "Is there any relationship between the value of the AUD (as against the USD) and the price of crude oil/LNG in Asia?" He was instructed to concentrate on the period January 2000 to June 2007. He said there were two potential components to the connection between Asian crude oil and LNG prices and the AUD rate. He used the word "potential" because this was the theoretical contention. To determine whether there was an actual connection, this needed to be evidenced in the statistical analysis. The two potential components were, first, a direct connection between crude oil and LNG prices and the AUD rate, which stemmed from the observation that crude oil and LNG prices were highly correlated with Australia's key commodity exports in a strong global economic growth environment; and, second, an indirect connection based on the negative impact that higher crude oil and LNG prices had on the US trade weighted index, and the negative relationship between the US trade weighted index and the AUD rate. To estimate these two effects both linear and logarithmic multivariate regression models of the AUD rate and the US trade weighted index were used to construct partial elasticities that determined the direct and indirect connections.

305. On the basis of the use of those regression models and the partial elasticities, first, the linear model found that each 10% increase in the crude oil price resulted in an appreciation in the AUD rate of 2.2% on average but this could rise to as high as 3.7% in a high oil price environment; and, second, the logarithmic model found that each 10% increase in the crude oil price resulted in an appreciation in the AUD rate of 2.2% on average. These estimates were based on applying multivariate regression techniques using monthly data between January 2000 and June 2007. Using bivariate regression analysis applied to monthly data between January 2000 and June 2007, Dr Webber found that each 10% increase in the Asian crude oil price coincided with a 4.4% increase in the Asian LNG price. This would imply, he said, that each 10% increase in the Asian LNG price coincided with a 5% appreciation in the AUD rate.

306. In his oral evidence in chief, Dr Webber said he disagreed with the affidavit of Dr Horst in reply to his own affidavit.

307. He said the first technical issue was that Dr Horst did not conduct a particular test called a co-integration test, and it was necessary to perform this test in order to generate a reliable relationship between the AUD/USD exchange rate, and the price of oil and the price of gas. Without conducting this test you could not get a true and reliable indication of the extent of the relationship between these two variables. Without an analysis of that test, or without understanding the findings of that test, it was difficult to rely on the estimated relationships that were presented in sections 2C and 2D of Dr Horst's second statement.

308. The second technical issue related to the functional form. Dr Horst's statement relied exclusively on using the double log functional form and there was no analysis using the linear functional form. And on that basis, without being able to compare those estimates on the two different forms, it was difficult to rely on the estimates in that section as well. Dr Webber also disagreed with Dr Horst in relation to an historical relationship between the exchange rate variable and gas price and oil price. He said that when you were trying to assess the future relationship between an exchange rate variable and a gas price and an oil price, there were a number of different tools that you could use to make that assessment. One of them was to look at an historical relationship between those two variables, but if you exclusively relied on the historical relationship there was a risk that that historical relationship was not representative of what you saw in the future. So, for example, if you were estimating this relationship from the viewpoint of June 2003, Dr Horst in sections 2C and 2D used data up to June 2003 to estimate a relationship between these variables: the exchange rate, the oil price and the gas price. Dr Horst was saying that this estimated relationship was the one that you should use to make an assessment of the relationship between these variables past June 2003. But what often happened in an analysis of this type was that the historical relationship was not a very good representation of what happened in the future, and that was certainly the case in this case.

309. Dr Webber said it was much more difficult to predict a particular value of the exchange rate than it was to predict the relationship between the exchange rate and other variables. In the same way, it was easier to predict the relationship between things like inflation and interest rates, between the oil price and economic activity, and between other variables that collectively impacted on a corporation. He said there was unambiguously a very strong relationship between the AUD/USD and the oil price, and what Qantas established when he worked there was that each time the jet fuel price or the oil price went up, the AUD went up by a certain amount. That constituted a natural hedge because when the oil price went up or the jet fuel price went up, that added to the airline's costs. But when the AUD went up at the same time, that reduced the airline's fuel cost because the airline had to swap AUD for USD to buy jet fuel. And so the higher cost due to higher fuel prices was offset by a stronger AUD and by using that relationship, they could build that into the hedging decisions. And, in fact, what it meant was that instead of using expensive financial instruments that were purchased generally from banks, they used this free relationship in order to avoid some of those costs, saving the company millions of dollars.

310. In cross-examination, Dr Webber confirmed that the period he analysed was January 2000 to June 2007. He agreed that he expressed the relationship in terms of partial elasticities and he said that by setting up a regression


ATC 17857

relationship that said variable A was driven by variable B you were implying a causal relationship between those variables. Dr Webber said there was, first, potentially a direct causal link between the Australian dollar and sets of commodity prices. The second potential correlation he identified was one between oil and gas prices and the US trade weighted index and that index would have a flow on effect to the AUD exchange rate. Dr Webber agreed that the reference in his Table B5 to the log of JLC gas prices in the body of the Table should be a reference to the crude oil price, and in the title of Table B5 there should be a reference to gas prices. Dr Webber was asked whether calculating the regression for oil first and then the third step of the relationship between gas and oil made the process more complicated. He said there was an extra step involved but that did not mean it was less accurate. Although it appeared in the present case that the elasticity was .29 rather than .46, that, he said, still demonstrated a very strong relationship between the variables. Dr Webber agreed as a general proposition that the period that one selected for analysis could have a highly significant bearing on the result of the analysis. He agreed that confined to the period up to June 2003, there was a negative elasticity for oil but using his longer time period the elasticity was positive. It was put to him that the result did not support the existence of a natural hedge of the kind that he had found when he looked to the different timescale. His answer was that the result told you that between January 2000 and June 2003 there was a relationship, such that an increase in either the oil or gas price coincided with a weaker Australian dollar. He agreed that by choosing that period of time the finding said that a higher oil price coincided with a weaker Australian dollar, which was inconsistent with what he found over the longer period and was inconsistent with the existence of a natural hedge.

311. Dr Webber was taken to another of Dr Horst's columns in Table 1 and agreed that again there was a negative elasticity for the period which either he would not rely on or was a result inconsistent with the existence of a natural hedge of the kind that he, Dr Webber, identified. It should be noted that Dr Webber disagreed with the timeframe used by Dr Horst because, he said, the commodity price and exchange rate world changed as we knew it from around 2002/2003 onwards. Any analysis that incorporated data prior to that point would generate an inaccurate assessment of the extent of any natural hedge we were seeing now and seeing after the period which was around 2003. He broadly disagreed with the choice of a period where, he said, we were operating in a different world for commodity prices and the exchange rate. Any period in this analysis that was prior to around January 2002 or in June 2002/2003 was not giving a true reflection of the extent of the natural hedge beyond that point.

312. The respondent submitted that Dr Webber's evidence did not assist the applicant. Dr Webber purported to identify positive correlations between oil and gas prices and the AUD/USD exchange rate, but did so by analysing a period which ended four years after the loan date; thereby identifying correlations which could not have been known to arm's length parties negotiating a loan in June 2003. His oral evidence confirmed, the respondent submitted, that economic conditions had changed fundamentally during the period that he analysed. Further, the respondent submitted, Dr Webber demonstrated no expertise in relation to the Asian LNG market into which the CAHPL group sold the majority of its products. He had initially undertaken his analysis by reference to Malaysian Tapis crude oil data before being instructed that North West Shelf crude oil was a more appropriate oil price benchmark. His evidence about the hedging strategy of Qantas in relation to jet fuel was referable to a time period after the entry into of the Credit Facility Agreement and it had not been established that revenues of CAHPL from the sale of LNG and other hydrocarbons bore any meaningful relationship to the price of jet fuel.

313. The applicant relied on Dr Webber to argue that at arm's length CAHPL would have borrowed in AUD but Dr Webber's evidence could not assist since there was no coherent evidence that anybody in the oil and gas sector was basing decisions on the existence of such a natural hedge in 2002 or 2003; none of CAHPL's reasons for its preference for denominating its borrowing from CFC in AUD involved this purported natural hedge; and there was no evidence of any company in the E&P sector borrowing at arm's length in AUD in any substantial amount. The respondent submitted that Dr Webber provided no basis for concluding that an economist analysing the available data up to June 2003 would have seen the link which Dr Webber retrospectively identified. The respondent submitted that Dr Webber conceded that the data to June 2003, to the extent that it demonstrated anything, was contrary to the existence of the natural hedge he propounded in his report.

314. In my opinion, whatever the technical merits of Dr Webber's analyses, his evidence was too abstract to be persuasive. Related to that conclusion is, I find, that Dr Webber used hindsight for his theory of the natural hedge. Although his opinion was that the type of behaviour in commodity prices and exchange rates was evident well before June 2002 to June 2003, I find that those relationships were not contemporaneously evident to those in the oil and gas or E&P industries. For example, when Dr Webber was asked for his experience which enabled him to make the comments he did as at 2002 and 2003 he referred to his experience at Qantas between 2004 and 2011. For these reasons , I find Dr Webber's opinion not to be of assistance.

Ms Silberztein

315. Referable to the ITAA 1997 issues was Ms Silberztein's report annexed to her affidavit affirmed 13 August 2014. The respondent objected to the entirety of this report on the grounds of relevance and that it was not in reply.


ATC 17858

316. Ms Silberztein, a partner of Baker & McKenzie in Paris, chaired the firm's global transfer pricing group. She was the head of the OECD Transfer Pricing Unit and Secretariat to Working Party No. 6 on the Taxation of Multinational Enterprises from 2001 to 2011, when she became a partner at Baker & McKenzie. She was asked for her opinion on the role and history of the "exceptional circumstances" referred to in paragraph 1.37 of the 1995 OECD Guidelines, as supplemented in October 1999. She was also asked to address the question of how the second of the "exceptional circumstances" in that paragraph was intended to operate both generally and in relation to loans, and a number of more specific questions about the second "exceptional circumstance" in paragraph 1.37. I ruled that, as to the objection on the ground of relevance and s 135 of the Evidence Act 1995 (Cth), part A down to and including paragraph 16 should be admitted as setting out the history of paragraph 1.37 of the OECD Guidelines, which may be relevant to the task of the Court under s 815-20 of the ITAA 1997. The balance of the report, paragraphs 17 through to paragraph 34, I excluded under s 135. I ruled that I would read that balance as a submission. As will appear, I do not find it necessary to rely on the matters addressed in Ms Silberztein's report.

Professor Rosenbloom

317. The respondent relied on an affidavit by Professor Rosenbloom who was asked to address the question whether under the law of the United States in the period from 2003 to 2009 Art 9 of the United States convention imposed tax. His conclusion was that under the law of the United States in those years Art 9 did not impose tax. The applicant submitted that it was important to take into account the practice of each treaty partner in relation to the United States convention by reference to uniformity of practice. The applicant submitted that the fact that the United States regarded the United States convention as not capable of imposing tax was an important fact in concluding whether it could do so from an Australian point of view. I have addressed this issue at [62] above.

Mr Schreyer

318. The respondent also sought to read an affidavit of Mr Schreyer who had been, between November 1981 and September 1983, the lead negotiator on behalf of the United States on its tax treaty with Australia. The evidence sought to be adduced from Mr Schreyer concerned the negotiating of the tax treaty between the United States and Australia. I rejected that evidence because nothing in the


ATC 17859

affidavit, or in the annexures to it, cast any light on the construction of Art 9 of the United States convention. The probative value of the evidence was substantially outweighed by the danger that the evidence might cause or result in undue waste of time within s 135 of the Evidence Act. I reached that conclusion notwithstanding that the respondent did not require Mr Schreyer for cross-examination.

The respondent's witnesses

319. The respondent called the following seven witnesses.

Mr Gaskell

320. Mr Gaskell was employed in the Royal Dutch Shell Group from 1969 to 2003. He was the Deputy Group Treasurer from 1997 to 2000 and Shell Group Treasurer from 2000 to 2003. He was a chartered accountant and had a degree in economics. Since 2013 he had been non-executive Director and Chair of the Audit Committee of Basin Holdings LLC and since May 2012 Chairman of the Martin Currie Global Portfolio Trust PLC and had held other positions since leaving the Shell group. He had joined the board of an oil and gas exploration company, Nobel Oil, in July 2014.

321. For his first report, annexed to his affidavit sworn on 9 July 2013, he was asked to prepare a report addressing a number of questions, assuming that CAHPL was a wholly-owned subsidiary of CVX and part of the global CVX group; that CAHPL was subject to the CVX Corporate Policies including those for internal and external financing; and proposed to borrow on or about 6 June 2003 from a wholly independent lender, dealing wholly independently, for the purpose of discharging intercompany debt (or a portion of it) as a consequence of the global restructure, and without an express guarantee from CVX. He was asked the following questions:

Question 1

If CAHPL and CFC were independent of each other would a corporation in the position of CAHPL, acting reasonably and prudently based on the presentation made to the Board of CAHPL on 4 December 2002, have agreed to borrow the AUD equivalent of USD2.5bn for five years at an interest rate of the Australian bank short-term rate for 3 months plus 4.14% (or at 1 month AUD-LIBOR-BBA as defined in the Credit Facility plus 4.14% p.a.), and otherwise on the terms of the Credit Facility, dated 6 June 2003. If not what borrowing is likely to have been adopted and why? In answering these questions, please identify the considerations which would be taken into account in determining an appropriate financing structure including an appropriate level of debt for a corporation in the position of CAHPL, acting reasonably and prudently.

Question 2

In your experience what factors are relevant in determining the credit rating of CAHPL for the purpose of the Loan and how would such factors influence the credit rating of CAHPL?

Question 3

If a corporation in the position of CAHPL, were able to borrow the AUD equivalent of USD2.5bn at the rate of 1 month AUD-LIBOR-BBA plus 4.14% from an independent lender on the terms of the Credit Facility would CAHPL or CVX corporate treasury acting reasonably and prudently have recommended that such a borrowing be undertaken by CAHPL on or about 6 June 2003? If not, why not, and what recommendations would have been made?

Question 4

If a corporation in the position of CAHPL, acting reasonably and prudently had borrowed the AUD equivalent of USD2.5bn in mid-2003 from a wholly independent lender would a corporation in the position of CAHPL, acting reasonably and prudently and the independent lender have entered into a credit facility in substantially the same terms as the Credit Facility? If not, please identify what would have been different and why?

Question 5

Please answer questions 1 to 4 on the basis of the following additional assumption; that the independent lender has raised the funds raised by CFC under a CP [commercial paper] program identical to that under which CFC raised its funding including the CVX guarantee of the CP program.

Mr Gaskell was also asked to provide his comments on the expert witness reports submitted by Mr Wasow and Mr Gross.

322. Mr Gaskell's opinion in relation to question 1 was that the Credit Facility did not include all the elements of the financing structure within which the loan was embedded and which were presented to the CAHPL Board when it approved the loan at its meeting on 4 December 2002. Without this financing


ATC 17860

structure, some aspects of which would not be available if CAHPL and CFC were independent parties, it would not have been commercially prudent for CAHPL to have agreed to the loan from an independent lender. The only elements of the financing structure present in the Credit Facility were that the loan would be the equivalent in AUD of USD2.5 billion repayable in full after a period of five years but with provision for early repayment at CAHPL's option, and CAHPL did not guarantee CFC and did not provide to CFC any security over its other assets. Other elements of the financing structure which were not present in the loan had the result that the loan as documented was for an excessive amount relative to forecast cash flow; carried a substantial AUD exchange rate risk; provided for a possible immediate repayment should CVX cease to have 100% control; and did not contain an express agreement for CVX to fund CAHPL's capital and exploration expenditure if its own cash flow was insufficient during the period of the loan. Examples of such elements were that CVX was expected to fund all capital and exploration expenditure in excess of CAHPL's net cash generation, after interest and tax, during the period of the loan; that CFC, a subsidiary of CAHPL, provided the loan in AUD but funded it in the US using a USD commercial paper program guaranteed by CVX; and that CAHPL recognised that the loan amount was about 47% of total assets but did not undertake to maintain this ratio.

323. Each element was, in Mr Gaskell's opinion, on its own sufficient to make the loan commercially imprudent had it been from an independent lender and, in combination, the elements introduced a high risk of future default by CAHPL such that the Board of CAHPL could not have reasonably agreed to an AUD loan equivalent to USD2.5 billion from an independent lender on the terms of the Credit Facility. Moreover, borrowing an amount in AUD for the term of the Credit Facility such that the loan was sub-investment grade would have been imprudent as the additional interest cost on the total loan would have resulted in a marginal interest cost of the additional borrowing in excess of the investment grade ceiling which exceeded CVX's cost of equity.

324. In relation to question 2, Mr Gaskell said that if CAHPL and CFC were independent parties dealing independently with each other, CAHPL would have sought to minimise the cost of any loan. In his opinion, as a first step CAHPL as borrower would have discussed a possible credit rating with at least one of the rating agencies. An investment grade rating would substantially reduce the cost of the loan and increase the range of possible financing sources to include the public debt markets. Even if banks were chosen as the source of financing, their own credit assessment would have been influenced by the credit rating. The option for CAHPL of public market debt would provide the competitive benchmark against which the banks would have had to price if they wished to be included in such a large and high profile syndicated loan.

325. In relation to question 3, Mr Gaskell said he did not believe that CAHPL Treasury, advised by CVX Corporate Treasury, would have reasonably and prudently been able to recommend accepting a loan in June 2003 from an independent lender on the terms set out in the Credit Facility if this meant that the loan was classified as sub-investment grade. The core reason for this would be that a five-year non-investment grade loan would have been a seriously uneconomic source of funding for CAHPL and CVX as well as being an inefficient financing structure from the perspective of the CVX group. The treasury recommendation should have been to ensure that the loan had a financing structure as close as possible to that actually used by CAHPL/CFC for the loan, and of such an amount that it achieved an investment grade rating.

326.


ATC 17861

In relation to question 4, Mr Gaskell said that if CAHPL had borrowed the equivalent of USD2.5 billion in AUD from an independent lender for a term of five years with full repayment on maturity, he did not believe that the terms could have been those set out in the Credit Facility. Lenders would have required full and binding contractual documentation for the main elements of the financial structure which were assumed by the CAHPL Board in approving the loan. If the loan were to be investment grade rated then he considered the loan would have had to have been in USD rather than AUD because of both the additional currency risk for CAHPL and the additional interest cash cost of an AUD loan.

327. In relation to question 5, if the independent lender were able to raise its funds in the same way as CFC by using a CVX guaranteed commercial paper program, as part of its agreement to lend the commercial paper proceeds to CAHPL in the form of a five-year term AUD loan equivalent to USD2.5 billion, some of the answers would be substantially changed because the range of potential lenders would be much increased.

328. For his second report, annexed to his affidavit sworn on 3 June 2014, Mr Gaskell was asked to set out his comments in response to the March 2014 affidavits of Mr Callaghan, Mr Lewis, Mr Rowland, Mr Long and Mr Dalzell. Mr Gaskell said that there were six reasons for the choice of the AUD as the currency of the loan given in the evidence by the three former CAHPL executives, Mr Callaghan, Mr Lewis and Mr Dalzell. Those reasons were:

  • (i) increased financial performance volatility;
  • (ii) impaired dividend paying capability;
  • (iii) tax payable on realised gains without the currency gains being reported in consolidated CVX accounts;
  • (iv) possible changes to Australian law to tax unrealised currency gains;
  • (v) uncertainty in thin capitalisation tax calculation; and
  • (vi) absence of a USD asset to hedge the USD loan.

329. Mr Gaskell said under the CFC structure for the loan none of the reasons given were relevant to the choice of the currency for the loan because they did not apply to CAHPL and its consolidated accounting group ( CAHPL-CG ) results for 2003 and 2004 which were effectively based on CFC's USD borrowing and because mitigation for the contingent risk of a tax law change would be readily available. If CAHPL had borrowed from an independent lender only two of the above reasons for not choosing the USD would have been relevant in that they would have had a different and potentially worse effect than that which the loan from CFC produced:

  • (a) after 2005 CAHPL-CG's profit and loss would have been more volatile although its balance sheet equity value would have been unchanged;
  • (b) the absence of a substantial USD asset in CAHPL's balance sheet would have produced a possible currency gain or loss in CAHPL's profit and loss but one which would have been mitigated by the reduced interest cost over the period of the loan and, in the CAHPL-CG accounts, by the offsetting effects on the revenues of CAHPL's other subsidiaries.

330. In Mr Gaskell's opinion, therefore it would have been unreasonable of CAHPL when the AUD borrowing was approved at the Board meeting in December 2002 not to have considered the additional cost of an AUD loan compared with that of a USD loan if the borrowing were from an independent lender. Had they done so, in his opinion, the AUD loan advantages of a probably more stable profit and loss for CAHPL-CG after 2005 and the possible absence of currency losses on repayment of the loan would have been outweighed by the actual disadvantages of the substantial additional interest cost and the loss of a cash flow currency hedge.

331. In his second report Mr Gaskell also commented on the evidence of Mr Rowland. He


ATC 17862

said that the sub-investment grade restricted Chesapeake's ability to borrow as much funding as it needed for its planned business development. The amount of debt that Chesapeake raised had to be supplemented by the issue of substantial equity. In CAHPL's position this equity issuance was not securely available as an option and had Chesapeake also not been able to issue equity then it would not have been able to operate its business as it wished. Also, Mr Gaskell said, as far as he was aware there were no cases of unsecured sub-investment grade issuance of the scale of the loan. The sub-investment grade debt market would therefore not have allowed issuance of the scale of debt that CAHPL needed to raise without a more complex syndicated loan structure and this would almost certainly have also involved some form of covenant or security arrangement which would restrict Chesapeake's flexibility. In relation to the funding methods that Chesapeake had used, Mr Rowland did not mention syndicated bank loans, convertible debt or production loans, all of which Mr Gaskell would have expected to be potential sources of finance for a sub-investment grade borrower.

332. Mr Gaskell also swore a third affidavit on 8 October 2014. He was asked to respond to the criticisms made by Mr Rowland in Mr Rowland's report annexed to his affidavit of 29 July 2014, which responded to Mr Gaskell's report annexed to his affidavit sworn on 9 July 2013. He responded to Mr Rowland's report under the headings Unfunded Capital Expenditures, Substantial AUD Exchange Risk, and Marginal Cost of Capital. In summary he concluded that Mr Rowland's evidence inaccurately addressed the part of Mr Gaskell's first report with which he disagreed relating to the loan being for an excessive amount relative to forecast cash flow; as to the AUD exchange rate risk, that Mr Rowland's evidence did not support his contention that there was no AUD risk; that Mr Rowland did not apparently address the point as to a possible immediate repayment should CVX cease to have 100% control of CAHPL; and that Mr Rowland's point about the documentation of CVX's intention to fund CAHPL's capital and exploration expenditure was not accurately based on Mr Gaskell's first report. Mr Gaskell therefore remained of the opinion that for CAHPL to approve the loan from an independent lender on the terms of the Credit Facility alone would have been commercially imprudent for the reasons described in his first report. Mr Gaskell also made comments on Mr Rowland's evidence in response to question two as to ratings. Mr Gaskell stated that in his experience, in particular for subsidiaries of oil and gas businesses, ratings were not based solely on the financial metrics considered by the rating agencies. His view was that the combination of financial metrics which might in isolation support a "good" BB rating (one within the upper half of the BB range) and the strongly positive qualitative factors to which the rating agencies would have regard, would raise the agencies' rating to investment grade, that is, by "at least" one notch.

333. In cross-examination, Mr Gaskell agreed that he had never worked for Shell in Australia and between his retirement from Shell and July 2014 he was not a participant working in a company that was an oil and gas E&P company. Although he had worked for an oil and gas company which raised debt without an investment grade rating, it was in the business of services to the E&P business rather than in the business of E&P of oil and gas. He had never practised as an accountant in Australia. As Shell Group Treasurer between 2000 and 2003 he was responsible for raising and managing Shell's borrowings from third parties and in that connection he had meetings with credit rating agencies in connection with the Shell group rating and in relation to the major subsidiaries that had ratings. As Shell Group Treasurer he was responsible for determining the extent to which wholly-owned subsidiaries of Shell were to be funded with equity and for deciding the mix of equity and debt in wholly-owned subsidiaries.

334. Mr Gaskell said that Shell Australia had a share ownership in Woodside but did not have control of what Woodside did in relation to the North West Shelf. It had an indirect economic interest. North West Shelf sold oil and gas and received US dollars for that in addition to some oil. Mr Gaskell believed that Shell Australia received US dollars for the sale of gas produced at the North West Shelf and certainly did so if the gas was sold outside Australia. Mr Gaskell appeared to agree that Shell Australia was funded by Shell in 2002 and that part of its borrowings included an AUD borrowing. The financial statements and reports of Shell Australia for the financial year ended 31 December 2002 stated that all borrowings, with the exception of lease liabilities, were unsecured.

335. In relation to his first report, Mr Gaskell agreed that if he had not been given the assumption that CAHPL was a subsidiary of CVX and if he assumed that CAHPL was an independent oil and gas company, that is, that it did not have a controlling shareholder, it would probably include some different comments and conclusions. He agreed that the expression "reasonable and prudent" which he had used in his paragraph 3.3.1 was one he had come up with as he believed that reasonableness and prudence were appropriate aspects of any company board directors' behaviour. The expression occurred in question 1 and his recollection was that he received a draft of that question, by telephone, before he received his formal instructions. In answering question 1 he did not take into account the 2003 CAHPL accounts but he had read those accounts, and the 2004/2005 accounts, before drafting his answer to the question. He interpreted question 1 to mean CAHPL as a wholly-owned subsidiary of CVX, in accordance with the assumptions he was given.

336. Mr Gaskell accepted that it was perfectly reasonable and prudent for CAHPL to be geared up to 47% subject to its cash flows being able to support that and subject to there being sufficient clarity about the funding of those cash flows if they were not supported.


ATC 17863

337. Mr Gaskell agreed that in CAHPL's consolidated accounts it would have been obliged to have restated the value of its USD revenues into AUD. He said that every cash receipt, on the day that it was received, would be translated into AUD at the rate applicable at that day. That would be the normal standard. If CAHPL had borrowed in USD then every year the AUD value of its USD principal liability would need to be translated into AUD: every reporting period end, the liability would be converted at the rate applicable into AUD. But that would be as a representation of the expected future cash outflow from that loan expressed in AUD as a best estimate, being the last known exchange rate. It would not necessarily be what the actual cash outflow would be. Mr Gaskell accepted that if he were assuming that CFC was independent of CAHPL then if CAHPL had borrowed in AUD it would not have achieved an effective USD annual interest cost and USD currency exposure.

338. Mr Gaskell agreed that his knowledge of the Australian GAAP and the Australian IFRS had been acquired by reading them in the course of preparation of his report and he had no other knowledge or experience of the standards outside of so reading them.

339. Mr Gaskell agreed that in his experience, at a sub-investment grade, the size of the loan would mean that the most practical type of structure would be a bank syndicated loan so that if one were hypothesising CAHPL actually borrowing a sub-investment grade loan, in his experience, it would do that via a bank syndicated loan. Mr Gaskell said those were the sorts of loans that he arranged, or was responsible for managing and arranging, inside Shell in the same sort of period. For those scales of loans and that type of risk the bank syndicated loan was the most practical way of doing it.

340.


ATC 17864

Mr Gaskell was asked whether he was suggesting that CVX would consider that the realised gain derived from the internal loan would be a benefit to the CVX group that would outweigh any tax cost in fact incurred by reason of that gain. He said that was true for the CFC subsidiary situation and agreed that, in the real world, in isolation the realised gain was of no consequence to the CVX group but paying tax was a real cash outgoing for the CVX group.

341. Mr Gaskell was challenged in cross-examination as to whether he was in a position to judge what was commercially prudent "Given the circumstances that you and Shell found yourself in when you were group Treasurer ..." He was taken to a 24 August 2004 notice from the Financial Services Authority stating that Shell announced false or misleading proved reserves and reserves replacement ratios to the market throughout the period 1998 to 2003. Mr Gaskell said that Shell announced reserves and reserves replacements which turned out to be false but said they were not, when announced, known to be false. Mr Gaskell was also taken to a SEC order. Mr Gaskell was asked to assume that Shell announced false or misleading proved reserves and it was put to him that that was not commercially prudent behaviour. His answer was that if Shell had known that the proved reserves were false or misleading then that would not have been commercially prudent behaviour.

342. The applicant submitted in relation to Mr Gaskell that whether a borrower in the position of CAHPL acting reasonably and prudently would have borrowed on the terms of the Credit Facility was a test generated entirely by Mr Gaskell himself in proposed writings to the legal team for the Commissioner. No particular content was given to the invented standard - it was not anchored in any legislative material or fleshed out by reference to identified industry custom. It was a general rubric. The cross-examination revealed both its imprecision and the danger of its adoption without clearly defined criteria. Ultimately, Mr Gaskell's evidence about prudence and reasonableness was, at best, evidence of Mr Gaskell's own judgment of these matters, and not evidence of some objective standard of reasonableness or prudence. Even then, it was given in respect of a narrow set of assumptions; namely, that the "financing structure" did not exist, and only by looking at the December 2002 Board Paper.

343. The applicant submitted that Mr Gaskell repeatedly attempted to recast both the currency and the quantum of the loan. His evidence was directed to the Gaskell Standard and that question did not arise in this case, even under the OECD Guidelines. His "response" to the evidence given by the lay witnesses, Mr Dalzell, Mr Lewis and Mr Callaghan was given on the basis that it was relevant and permissible to review the choice made by CAHPL to borrow in AUD.

344. The applicant submitted that Mr Gaskell criticised Mr Wasow for failing to take into account "any implied equity or other financial support consequent on ownership by a large and highly rated parent". Parts of his first report also relied on what he called the "implied support of the CVX [ Chevron Corporation] ownership". Mr Gaskell conceded in cross-examination that he would need to make some (unstated) different comments and draw some (also unstated) different conclusions if CAHPL was not a subsidiary of CVX. Critically, he agreed that he had not addressed the reasonableness or prudence of the loan on the basis of the actual facts, which included the receipt of dividend income from CFC. He had only addressed this issue on the basis of what had not in fact taken place - namely the absence of what he called the "financing structure". In relation to his second report, Mr Gaskell was instructed to assume that CVX was the ultimate parent of CAHPL, and he assumed in his answer that CFC remained a subsidiary of CAHPL.

345. The applicant submitted Mr Gaskell was asked to assume that the independent lender had raised funds under a commercial paper program identical to that under which CFC raised its funds, including the CVX guarantee. He erroneously concluded that this assumption would change his analysis "substantially".

346. The applicant submitted that Mr Gaskell had no experience of how the funding arrangements of non-investment grade oil and gas companies were arranged, unlike Mr Rowland who was the former chief financial officer of Chesapeake, a non-investment grade rated oil and gas company in the United States. Mr Gaskell conceded that he had no experience working for such a company anywhere and could not comment on the accuracy of Mr Rowland's report in relation to Chesapeake other than by reviewing that company's published documents. Nor had Mr Gaskell worked in the oil and gas industry in Australia, unlike Mr Wasow, who was the former chief financial officer of Santos Limited.

347.


ATC 17865

The applicant submitted that any evidence Mr Gaskell had purported to give in relation to the accounting impact of a USD borrowing on Australian accounts of CAHPL should be accorded no weight. Mr Gaskell testified that the first time he looked at Australian accounting standards was in preparation of his reports in this matter.

348. In my opinion, Mr Gaskell's evidence was of very limited utility given the questions he was asked to address. The real issues in the present case are not concerned with whether the parties to the Credit Facility were dealing with each other at arm's length, a question to which Mr Gaskell's report might be relevant, but rather the different question of the arm's length consideration as between independent parties. In any event, I regarded as a distraction to consider a corporation in the position of CAHPL and whether such a corporation would be acting reasonably and prudently in entering into the Credit Facility Agreement. It is to be noted that the "acting reasonably and prudently" formulation is to be found expressly stated in question 1, question 3 and question 4 put to Mr Gaskell for his opinion.

Mr Hollas

349. Mr Hollas was called by the respondent as an expert. He was the managing director and co-founder of CUFTanalytics.com, Inc, a transfer price consulting firm with headquarters in Calgary, Canada. He had over 17 years of corporate banking and corporate finance experience which involved credit risk assessment and negotiating pricing for all types of financial transactions while working in senior positions with major Canadian financial institutions. He had not worked for a bank since approximately 1999. He swore five affidavits.

350. For his first report, annexed to his affidavit sworn on 5 July 2013, he was asked the following question:

If CAHPL and CFC were independent parties dealing with each other at arm's length, what rate of interest would CAHPL have paid CFC in respect of the 6 June 2003 intercompany loan transaction assuming:

  • a) that the loan was in AUD; and
  • b) that the loan was in USD?

351. Mr Hollas determined the rate of interest by applying the comparable uncontrolled price ( CUP ) method which, in his opinion, was the most appropriate method for answering the question. He said he compared the loan between CFC and CAHPL with sufficiently comparable borrowings of sufficiently comparable companies. He identified three comparable uncontrolled senior loan transactions that were executed by E&P companies with sufficiently comparable credit metrics to CAHPL's. The companies were Devon Energy Corporation, Noble Energy Inc and Ocean Energy Inc. The relevant comparable uncontrolled credit agreements were filed with the SEC during a period of approximately one year prior to the execution of the CAHPL/CFC loan. Based on the application of the CUP method using the three comparable uncontrolled transactions, Mr Hollas said the lending margin (the credit spread added to lender's cost of funds) that CFC would have charged CAHPL if those parties were independent and dealing at arm's length was in the range of 144 to 150 basis points. If CAHPL borrowed in the primary loan market, the cost of funds would have been the unrelated lender's cost of funds, that is, LIBOR. Therefore, in Mr Hollas' opinion, the interest rate that an unrelated borrower, which was comparable to CAHPL in terms of its credit metrics, would be charged by an unrelated lender for an AUD denominated loan that was sufficiently comparable to the characteristics of the CFC-CAHPL loan was AUD LIBOR plus 144 to 150 basis points. For a USD denominated loan that was sufficiently comparable to the characteristics of the CFC-CAHPL loan the interest rate would be USD LIBOR plus 144 to 150 basis points.

352. For his second report, annexed to his affidavit sworn on 5 September 2013, Mr Hollas was asked to prepare a report answering the question: are there are any aspects of the reports filed by Dr Becker and Mr Gross that he disagreed with and if so to (a) identify those aspects, and (b) provide reasons for his disagreement. Mr Hollas disagreed with the lending margin conclusions in the Gross and Becker reports. His disagreements centred on the approach to determine CAHPL's credit risk and credit risk comparability; the comparable data to be relied upon to price the CAHPL intercompany


ATC 17866

loan; and the most appropriate methodology or approach to be used to determine the lending margin for the CAHPL intercompany loan.

353. For his third report, annexed to his affidavit sworn on 5 June 2014, Mr Hollas prepared a report setting out his comments in response to the affidavit of Mr Martin sworn 10 March 2014. Mr Hollas disagreed with Mr Martin's conclusion that in 2002 CAHPL would have a borrower risk rating of a weak BB or high single B based on Mr Martin's credit risk analysis which relied on only three credit ratios and a cursory review of other factors. He disagreed with Mr Martin's final loan pricing conclusion which was premised on CAHPL being rated as a weak BB and Mr Martin's view of the markets in 2003 that the interest rate on the Credit Facility did not exceed what would be reasonably expected to be obtained in an arm's length transaction at that time. Mr Hollas disagreed with Mr Martin's conclusion to his analysis of the secondary source of repayment of the CAHPL loan from the underlying asset (or enterprise) value as being weak. Mr Hollas agreed, in general terms, with Mr Martin's comments on the structural considerations of the CAHPL Credit Facility. Specifically, he agreed that a bank credit facility to a BB rated (and in fact any non-investment grade) E&P borrower would have (some) financial covenants and would not be provided on an unsecured basis (or without a borrowing base limit). As to Mr Martin's comments on local market considerations, Mr Hollas agreed that the CAHPL loan would have been one of the larger credit facilities in 2002 and early 2003 in the local loan market. However, there were seven "jumbo deals" (ones over $1 billion) that Mr Hollas was aware of that were done in the Asia Pacific market in 2002-2003. He agreed with Mr Martin that a lender would take sponsorship factors (or the shareholder) into consideration in its credit analysis and that CVX provided strong sponsorship.

354. For his fourth report, annexed to his affidavit sworn on 12 June 2014, Mr Hollas was provided with additional material in respect of Mr Martin's affidavit sworn on 10 March 2014 and in respect of Mr Gross' affidavit sworn on 18 April 2013. His review of that material did not cause Mr Hollas to change any of the opinions he expressed in the reports to his third affidavit (sworn 5 June 2014).

355. Mr Hollas' fifth report, annexed to his affidavit sworn on 17 October 2014, made a number of corrections to Appendix 6d - "Revised Borrower Comparability Analysis" - to his report annexed to his third affidavit (sworn 5 June 2014). He said in his fifth affidavit that the corrections did not change the conclusions in any of his evidence in the proceedings.

356. In cross-examination, Mr Hollas accepted that, although when he was with the National Bank of Canada in Calgary from 1994 to 1996, his responsibility for credit assessment and loan approval decisions included oil and gas lending, that work never involved a loan of the size of the present loan and that the loans made to the energy sector during his time in Calgary in aggregate represented only a very much smaller amount than the present loan. He accepted that in 1994 the aggregate of loans made to the mines, quarries and energy sector by the National Bank of Canada was $215 million, representing 0.6% of the loans that were made in total by that bank. Mr Hollas also said that his only exposure directly to energy lending was with the National Bank of Canada.

357. For the purposes of his first report, Mr Hollas examined the actual loan and formed the view that it was in the AUD equivalent of a USD amount: effectively an AUD denominated loan.

358. For the purposes of his first report Mr Hollas said he looked at CFC and did a functional analysis of CFC as well as CAHPL. He said the characterisation of both the borrower and the lender was important to his analysis and he would have looked at the characterisation of CFC.


ATC 17867

359. The methodology used for his report was to hypothesise CFC to be an unrelated lender. His analysis of CAHPL's credit metrics was determined on the basis of quantitative factors and he did not rely upon qualitative factors, such as implicit parental support.

360. To determine CAHPL's credit metrics he looked at and used CAHPL's 2003 results and not its 2002 results even though the actual 2003 results would not have been available to an arm's length party entering into the loan in June 2003. He assumed that the parties would have had access to forecast figures in June 2003 which he said would be the practice in any lending situation, and he made a second assumption that the actual 2003 figures were a reasonable proxy.

361. Mr Hollas accepted that there was no evidence that showed what the reserves for CAHPL were as of June 2003.

362. In analysing the credit metrics of possible comparable companies, Mr Hollas used the 2002 data for those companies. Mr Hollas said that the reason for the different dates was that, apart from Chesapeake and CAHPL where the agreements were 2003, all of the other companies entered into their transactions in 2002 so he was looking for the credit metrics that the lender would have looked at in coming to a credit decision and pricing the loan transaction that occurred at that time. He was looking at the credit metrics on which a lender would make a credit decision.

363. Mr Hollas said that when he created the credit metrics for CAHPL he blacked out the actual interest liability that it paid and reinserted a new interest liability of AUD LIBOR and 150 basis points. The lending margin there referred to was very similar to the lending margin that he decided was the arm's length lending margin in his first report and by putting in the lending margin of 150 basis points he necessarily affected some of the credit metrics of CAHPL because the interest rate had an impact on some of the financial ratios. The credit metrics that would be affected by selecting the lending margin of 150 basis points were the interest coverage, operation total debt and free cash flow metrics. Mr Hollas selected 150 basis points as the midpoint between 125 and 175 basis points being the range within the year prior to the execution of the CAHPL loan to a borrowing E&P company of comparable size to CAHPL. Mr Hollas assumed the lending margin based on his experience and knowledge of senior loans, his experience tracking the lending margins for those types of loans over a long period and looking at how lending margins changed over a period of time. He accepted that during 2002 he was not a banker but, he said, he had transfer pricing experience of looking at loans during that period of time, as an economist or a consultant and not as a banker.

364. It was put to Mr Hollas that he chose 150 basis points because that was what he ended up with as his answer to the arm's length rate question and he agreed that it turned out that way but he also did a subsequent sensitivity analysis.

365. Mr Hollas agreed that he ultimately found three sufficiently comparable transactions in his first report and in his revised report he found one more. He got his arm's length lending margin from the lending margins found in those loan documents. He selected the four loans based upon his searching of loans on SEC databases, looking for loans for a particular period of time contained on the SEC database specifically for the E&P sector. He agreed that he would have no knowledge of any loans that were not filed with the SEC. He did look for Australian comparables and came up with four possibilities but the actual loan agreement in each case was not filed and therefore the data he needed was not available.

366.


ATC 17868

Mr Hollas accepted that if you had a USD loan you used USD LIBOR, and if you had an AUD loan you used AUD LIBOR.

367. All four of the loans were entered into between companies in the United States; all four of the loans were for different sums of money being $525 million, $200 million, $1 billion and $350 million; and each of the four loans was a revolver loan, meaning that the borrower had the ability to draw down and repay up to an agreed ceiling. In relation to the first three of the revolver loans, during the 2003 year they were wholly undrawn. In relation to the fourth loan, the Chesapeake loan, it was only drawn down to a small amount. Each of the four loans contained some financial covenants and they were financial covenants which did not appear in the CFC-CAHPL loan. The Chesapeake loan was a secured loan while the CFC-CAHPL loan was not a secured loan. The four loans identified by Mr Hollas were all for a different tenor, two were one-year revolver loans with an option to extend and two were four-year revolver loans. In each case, each of the companies that had entered into the revolver loans had other borrowings to fund their businesses: three of the four companies issued notes during that period. Each of the four did not in fact use the revolver facilities to fund their businesses. Mr Hollas accepted that the loan entered into between CFC and CAHPL was a five-year term loan that was almost fully deployed to refinance existing intercompany debt. He said he made adjustments to the pricing on that basis. He agreed that some of the four companies he had used as a comparison had less leverage. Mr Hollas agreed that the primary source of funding for the four companies was not their revolver facilities but the notes they had on issue and other bank facilities.

368. Mr Hollas agreed that the methodology he had used to price the loan, the actual CUP approach, was a transfer pricing analysis and was not in that format used by banks to price a loan and was not used by any of the rating agencies to determine the actual credit rating of a particular issuance.

369. Mr Hollas agreed that if he had been asked to price the loan in 2003 he would have had no difficulty in using Bloomberg data, because that was the only data that was available at that time.

370. Mr Hollas said that the CAHPL loan was definitely a senior term loan but said that probably more important to him in looking on the SEC database was the tenor, the loan commitment period, because whether or not a loan was called a term loan or a revolver the pricing decision was not on the name but on what it was and if it was a four year tenor committed loan period then it would be priced accordingly. He accepted that because he characterised the CAHPL loan as a senior term loan in his reports he did not search for notes, bonds or commercial paper.

371. Mr Hollas accepted that the CFC-CAHPL loan was covenant-lite - there were no covenants of any sort - whereas the senior term loans he identified were not covenant-lite. Mr Hollas said that if you were looking for transactions that had no covenants, you would not find any transactions at all: there were no notes or bonds that were issued without restrictive covenants.

372. Mr Hollas accepted that a senior loan had senior priority in terms of claim on assets and cash flow, and it was not subordinated to someone else. He also accepted that if the other North West Shelf joint venturers had a first charge over CAPL's interest in the North West Shelf to secure CAPL's obligations to the joint venture then there was a first charge over the entire interest in the North West Shelf, and it would follow that the CFC loan was also effectively subordinate to the preferred rights of the joint venturers. If that were so, Mr Hollas' characterisation of the loan as senior, he accepted, could be wrong because then the loan would be junior to the rights of other parties to the assets: it would be effectively a second lien loan.

373. It was put to Mr Hollas that bankers would not characterise the CFC-CAHPL loan as senior because it would be open to CAHPL at any time to issue legally senior debt which would subordinate the CAHPL loan. Mr Hollas said he was not sure how the loan would be categorised because it was just not an instrument that you would find. He said he was characterising the loan with what he considered to be a necessary term, which would be the restrictive covenants, because if there were no restrictive covenants there was nothing to price.


ATC 17869

374. Mr Hollas accepted that you would not use S&P's Credit Model to undertake an actual rating of an actual issuance.

375. Mr Hollas agreed that in his methodology the four transactions which were comparable to the CFC-CAHPL loan were transactions entered into between banks and those companies even though CFC was not a bank.

376. In relation to notching up the stand-alone credit rating of CAHPL by reason of the existence of implicit support from CAHPL's parent, Mr Hollas maintained the opinions expressed in an article he had written with Mr Hands in 2014 that lenders were not likely to be using the exact notching methods employed by the rating agencies and would rely upon their own credit risk framework analysis and evaluation so the notching was, at best, an approximation of the credit risk analysis undertaken by arm's length lenders and he and Mr Hands found no market evidence that a lender would provide a lower interest rate to a borrower for the notched rating using the notching methods and therefore there was no market evidence of any potential economic benefit for notching.

377. The applicant submitted that Mr Hollas was not a specialist lender to the E&P sector of the oil and gas industry. Mr Hollas conceded that his lending experience to the E&P sector of the oil and gas industry consisted of two years at the Calgary office of the National Bank of Canada and that lending to the energy sector formed a tiny part of that bank's business and that the loans he was then involved in were of a much smaller amount. Mr Hollas also conceded that his lending experience was largely based on making reserve based loans, a form of secured loan based on the expected value of future production. To obtain an unsecured loan of this size CAHPL would not have consulted a lender with the limited experience of Mr Hollas, who had not been in the banking industry for the last 15 years, as a production loan would not have been appropriate.

378. The applicant's submissions in relation to Mr Hollas' evidence were that he adopted an inappropriate pricing methodology. The primary method adopted by Mr Hollas did not assess CAHPL's credit risk profile but instead searched for "comparable uncontrolled agreements" or "CUTs". This was not a methodology adopted by banks or lenders to price a loan nor was it used by credit rating agencies. Mr Hollas exaggerated his own experience in this respect. Mr Hollas' methodology was not used by the economic advisers to the respondent Commissioner or by Ceteris or by Dr Horst. Dr Becker said that Mr Hollas' methodology was neither precise nor reliable. Mr Hollas rejected well recognised sources of financial data and ratings data from Bloomberg and S&P (being the same sources relied upon by the Commissioner's other transfer pricing expert, Dr Horst, and which were relied upon by lenders in practice) in lieu of his own personal search procedure. Mr Hollas conceded that if he had conducted this exercise in 2003, he would have used Bloomberg data. His search procedure involved a number of decisions by him to accept or reject borrowers based on his own selected criteria. Mr Hollas calculated CAHPL's credit metrics based on 2003 results notwithstanding that: that information would not have been available at the time the loan was taken out; that he was aware that there was a significant difference for CAHPL between 2002 and 2003 as a result of a 12.5% growth in reserves; and that he used 2002 data for his comparable companies. Mr Hollas had previously opined that transfer prices should not be set on an ex post facto basis.

379. His search was entirely dependent on the data available on the SEC website which he conceded was difficult to use. Mr Hollas did not assess the credit profile of CAHPL by undertaking a credit rating analysis such as that which would be undertaken by a credit rating analyst. Rather, he examined a select number of credit metrics which he considered to be appropriate and which, the applicant submitted, were not applicable to a loan that was not a production or reserve based loan. He then compared selected credit metrics of CAHPL to credit metrics of other companies and applied his own subjective criteria in determining which other companies were comparable. The errors in Mr Hollas' determination of his credit metrics affected his comparability analysis.

380. The applicant submitted that the results of Mr Hollas' CUT approach should not be accepted in circumstances where he did not take into account the deed of cross charge which had the effect of subordinating the CFC loan. Mr Hollas characterised the loan as a "senior" loan and not subordinated, but other North West Shelf joint venturers had a first charge over Chevron Australia's interest in the North West Shelf. The essential flaw in using a CUT approach for pricing a loan was that the applicable interest rate was not just the product of selected credit metrics and a simplified characterisation of the loan as, for example, "senior term loan". It depended on all relevant facts, including the borrower's specific relationship with the lender or lenders and the particular terms of the loan. The loan agreements identified by Mr Hollas were not comparable to the CAHPL loan facility for the reasons identified by Dr Becker, Mr Gross and Mr Martin. Those reasons included that the loan agreements were for revolver (not term) loans and for a different tenor; those loan agreements were facilities under which no (or only small) advances were in fact made and which formed part of a much broader borrowing program; they contained covenants which the CFC-CAHPL loan facility did not contain; and they involved US rather than Australian borrowers and, in any case, the borrowers involved had very different credit profiles to CAHPL. Mr Hollas also used an alternative method of attempting to analyse CAHPL's credit by application of an S&P credit model but accepted that he would not use that credit model to undertake an actual rating of an actual issuance. Other flaws in using a model to analyse CAHPL's credit were identified by Dr Becker, Mr Gross and Mr Martin.

381. In my opinion, the "mistakes" in Mr Hollas' report did not affect his conclusion in the comparability analysis. However, as he accepted, Mr Hollas' exercise was an artificial one in that it was not an exercise done by either lenders or rating agencies at the time. I also accept the applicant's submission as to Mr Hollas' lack of qualifications as a lender in the relevant market. I find his evidence to be of no utility in answering the statutory questions.

Dr Horst

382. The respondent called Dr Horst as an expert. Dr Horst was the founder and one of seven Managing Directors of Horst Frisch Incorporated, an economic consulting firm based in Washington DC specialising mainly in transfer pricing analyses. Major areas of his work included transfer pricing of US and foreign-based multinationals. Earlier he had been a Director, International Tax Analysis, Deloitte Haskins & Sells. He swore three


ATC 17870

affidavits, and associated reports, in the proceedings.

383. For his first affidavit, sworn on 8 July 2013, he was asked to prepare a report addressing the question: "What rate of interest CAHPL might be expected to have paid, had it agreed to borrow an amount of up to USD2.5 billion (or the AUD equivalent) for a term of five years from an independent enterprise with whom it had dealt at arm's length." He described the key points in his analysis as applying credit-rating formulas to the financial data for CAHPL in 2002 but assuming a loan of USD2.5 billion yielded an estimated stand-alone credit rating for CAHPL of BB+; his explanation of why, if CAHPL had borrowed from an independent enterprise, it would have denominated its long-term debt in USD rather than AUD resulting in a considerably lower interest rate and CAHPL's interest cost being reduced by a significant amount; and his conclusion that the fixed margin to be added to USD LIBOR to calculate a floating interest rate on CAHPL's intercompany loan should be based on financial market conditions in early June 2003 shortly before the intercompany loan agreement was finalised, rather than December 2002 when CAHPL's Board approved the intercompany loan of USD2.5 billion with a 4.14% spread over AUD LIBOR. His ultimate conclusion was that the fixed margin on CAHPL's intercompany loan should be 100 basis points. Based on the 1 month USD LIBOR of 1.32% per annum as of 2 June 2003, CAHPL's initial interest rate would be 2.32% per annum.

384. In his second affidavit, sworn on 5 September 2013, Dr Horst addressed the question whether there were any aspects of Dr Becker's report dated 1 August 2013 with which he disagreed. Section 2 of Dr Horst's second report considered Dr Becker's assessment of whether CAHPL would have denominated its loan in USD or AUD. According to Dr Horst, Dr Becker proceeded on a fundamental misunderstanding which was that the first report of Dr Horst concluded that the AUD form of the loan would be beneficial to CAHPL. However, Dr Horst said, his first report concluded that the AUD form of the loan would be detrimental to CAHPL. Section 3 of Dr Horst's second report responded to Dr Becker's claims regarding the best formula for predicting CAHPL's stand-alone credit rating. Dr Horst disagreed with Dr Becker's conclusion that Dr Horst's Case 9 provided a more accurate formula for predicting CAHPL's credit rating than Dr Horst's Case 1. Dr Horst said that if the interest rate on CAHPL's debt was based on debt denominated in USD not AUD, CAHPL's stand-alone credit rating would be BB+ under both Case 1 and Case 9. However, if the interest rate was based on debt denominated in AUD, not USD, CAHPL's stand-alone credit rating would be BB under Case 1 and BB- under Case 9. Section 4 of his report addressed Dr Becker's claim that Dr Horst erred by not taking into account various other ratios or amounts that he could have used to predict CAHPL's stand-alone credit rating.

385. In his third affidavit, sworn on 3 June 2014, Dr Horst annexed a report setting out comments he had in response to the third affidavit of Mr Dalzell and to the second and third affidavits of Dr Becker. Dr Horst reaffirmed the conclusion of his first report that had CAHPL borrowed from an independent third party it would have borrowed in USD because its consolidated revenues were predominantly in USD.

386. Dr Horst agreed with Dr Becker that the starting point in transfer pricing analysis was to identify the facts and characteristics of the related parties and the transaction between them without recharacterising either the parties or their transaction, other than by assuming they were independent and dealing with one another at arm's length. Dr Horst said that if the actual restrictions on CFC's activities were taken into account there was no single arm's length result. Rather there was a range of arm's length prices having CFC's actual borrowing cost at the low-end and CAHPL's cost of borrowing from third parties at the high end. Because CFC could lend to no borrower other than CAHPL and CAHPL could have secured a below-market interest rate from no lender other than CFC, the inherent bargaining power of CAHPL was counterbalanced by that of CFC. If an agreement could be


ATC 17871

reached, each party could reap a substantial gain that it could not otherwise obtain. Accordingly, Dr Horst thought it reasonable to conclude that the interest rate that would result had CAHPL and CFC been negotiating at arm's length was at or near the midpoint of the range delimited by the minimum and the maximum interest rates, respectively.

387. Dr Horst swore a fourth affidavit on 14 October 2014 which annexed a report in relation to estimates of the elasticity of the AUD exchange rate with respect to changes in North West Shelf crude oil prices and JLC gas prices.

388. Dr Horst identified a difference of understanding between Dr Becker and himself as to the arm's length standard generally and the OECD Guidelines in particular. He agreed that the determination of arm's length transfer prices usually reflected the structure adopted by the taxpayer but said that determining an arm's length price sometimes required consideration of the structure the parties would have adopted had they been independent of each other and dealing with each other at arm's length. He concluded that under the facts and circumstances of the present case denominating the loan in AUD rather than USD created unnecessary foreign exchange risks for both CFC (whose borrowing costs were denominated in USD, not AUD) and CAHPL (whose operating revenues were predominantly in USD not AUD). Both parties, he said, could have avoided or substantially reduced their respective foreign currency risks by denominating the Credit Facility in USD. Parties dealing at arm's length did not structure transactions in ways that created unnecessary risks for both parties. Denominating the CFC-CAHPL Credit Facility in AUD, rather than USD, was not a commercially rational choice under CAHPL's and CFC's actual facts and circumstances. Accordingly, he continued to conclude that if CAHPL and CFC had been dealing at arm's length but otherwise had the same facts and circumstances, CAHPL's Credit Facility with CFC would have been denominated in USD, not AUD.

389. In cross-examination, Dr Horst agreed that he did not have specialised knowledge or training in the oil and gas industry nor any specialised expertise in the world of banks. His experience was as an economist who had spent his professional life in the United States.

390. Dr Horst explained that Annexure 5 to his first report involved an area of simultaneous determination between interest rates and credit ratings. One could not get all the way to determining a credit rating just based on the analysis of the 26 companies referred to in the Goldman Sachs and Deutsche Bank reports. One also had to look at what would be the interest rate implications of those credit ratings and one had to go back and forth between interest rates and credit ratings. The view that the stand-alone credit rating was based only on the 26 companies was something of an oversimplification. Section 7 of his report discussed the simultaneous determination of a company's credit rating and its interest rate. Dr Horst agreed that the credit rating agencies such as S&P and Moody's did not employ his method for determining the actual ratings of particular issuances. He accepted that in the real world credit rating agencies not only look at quantitative matters but they also look at qualitative matters. Dr Horst said that the notch up was for all factors that were not reflected in the quantitative analysis but


ATC 17872

was essentially the difference between what the rating would be if you looked just at the numbers versus what happened when you took into account qualitative factors which were not reflected in the numbers.

391. Dr Horst accepted that he did not independently seek for his report any additional comparable or possible comparable companies beyond the 26 from the combination of the Deutsche Bank and Goldman Sachs reports. He accepted that there were other companies that might reasonably be added to his group of comparables. He accepted that if there were possible other companies that might have been added to his analysis, it was possible that some case other than Case 1 would have been the best. It might have not necessarily led to a different inference about what the estimated credit rating was but which case you would get as being the best could change if you added other companies. It could have led to a different result as to what should be the stand-alone credit rating of CAHPL and, Dr Horst said, he would not be surprised if it had gone up rather than down. He did not know how either Deutsche Bank or Goldman Sachs chose their comparables. He agreed that about half of the 26 companies were investment grade and half were non-investment-grade. The credit ratings ranged from a high of A- to a low of B+. He used the 26 companies to construct a mathematical formula, in a broad sense, that he could use to estimate or predict what CAHPL's credit rating would be. Dr Horst used his model first to predict or estimate what the credit rating would be of the 26 companies based on the credit metrics so he would have alongside both a predicted credit rating and an actual credit rating. Then he would later also use that same formula that was used to generate the estimates to generate an estimated credit rating for CAHPL. He also used the notch differentials to pick what he thought was the most accurate formula.

392. Dr Horst agreed that under the methodology he had used in his first report, identifying the credit rating of the borrower was fundamental to the determination of the arm's length rate of interest it should pay: one got to the interest rate through a credit rating.

393. As to the exchange rate, Dr Horst agreed that one of the factors he took into account in forming the view that the loan that was in AUD should have been in USD was the Deutsche Bank forecast. Dr Horst looked at the conditions that existed as of the time of the loan, including the Deutsche Bank forecast of appreciation of the AUD. He did not perform an analysis of the likely movements in the USD LIBOR rate as against the AUD LIBOR rate over the five years of the loan. Dr Horst said that if you were an independent entity in 2003 and you had to make a decision about whether there should be a USD or AUD loan, leaving taxes aside, there were three factors that it seemed you would take into account: one was what were the relative interest rates; two was what were the natural hedges that existed, so that if your revenue was coming in USD to have a lower risk outcome you would prefer a USD denominated loan and if your revenues were coming in AUD then you would prefer an AUD denominated loan, based on that one factor; and three what were your best forecasts of the future movements of the exchange rate.

394. Dr Horst gave his understanding of the factors that as an economist he would look at. What mattered to him was the underlying economic relationship between two cash flows: the USD revenues and the USD liabilities or outflows. The hedging came from the cash flow consequences: the incoming cash flow was in USD and the outgoing cash flow, which was the repayment of USD and of both principal and interest were both denominated in USD. He said that if in 2003 he had before him a financial forecast that said that the USD LIBOR would greatly increase over the five years of the loan and exceed the AUD LIBOR, he would go into an interest rate swap to adjust for that rather than change the currency in which the loan was made. Dr Horst agreed that he had not looked to the consequences for Australian accounting standards for foreign currency gains and losses. Also he did not look at any financial reports beyond the 2002/2003 period. He said his reason for determining that at arm's length CAHPL would have borrowed in USD was based on a cash flow analysis of the hedging, not on the accounting treatment of it. He did not look at whether, following the introduction of IFRS, Australian companies in their accounts could only use their USD liabilities as a hedge against their USD denominated assets.

395. Dr Horst was asked to assume that in early 2003 there was an observable correlation between the exchange rate between the USD and the AUD and the price of oil. He agreed that that would have been a factor that would have made borrowing in AUD slightly less unattractive commercially.

396. In relation to questions of Australian tax law, Dr Horst said that he assumed Mr Lewis had legitimate concerns but in his judgment the uncertainty about exactly how foreign exchange gains and losses were going to be taxed were likely to be swamped by the before tax effects to which he had referred: the interest rate differentials, the hedging relationships and the expectations about where currency rates


ATC 17873

were going. If it were an internal company loan, Dr Horst said he was sure it would be advantageous to borrow in AUD but that was not the question he was investigating. He was investigating the question of what they would have done if they had been borrowing externally, and that had a different answer.

397. As to CFC's actual facts and circumstances, Dr Horst agreed that his view was that because of CFC's obligation to lend to CAHPL and no one else, its bargaining power was limited and so the price that would have been struck by CFC for this loan would be somewhere between its cost of funds and the rate Dr Horst had determined in his first report. It was put to Dr Horst that the limitation was not that CFC must lend to CAHPL but that it must lend to its sole shareholder, its parent. Dr Horst said the hypothetical he was trying to price was if you had the same restriction, just one borrower, but the one borrower was not a related party. It was put to him that he felt comfortable in changing the currency of the loan but not comfortable in going on to restructure the transaction even though it contained an artificial limitation. Dr Horst responded that he did not have the knowledge and expertise to know what the transaction would look like after the restructuring but he felt he had enough knowledge of Australian tax law to venture an opinion about what the currency of the loan should be. He said he was aware that the requirement that CFC's only business was that of raising finance for its shareholder was a requirement to do with withholding tax.

398. The conclusion Dr Horst reached using his statistical method was that using a USD loan CAHPL should have a stand-alone rating of BB+ and a non-stand-alone rating notched up two notches from a BB+ to a BBB. Using an AUD loan, his statistical model came up with a stand-alone rating of BB and a non-stand-alone rating two notches up from BB, giving BBB-.

399. In relation to the 26 companies Dr Horst used as comparables, Dr Horst said that "all our transfer-pricing disputes centre on what is the appropriate group of comparables to use". In relation to the OECD Guidelines and the series of factors concerning comparability that must be considered in order to determine whether a company or transaction was comparable, Dr Horst said that the relevance of the different factors depended on the context in which you were doing an analysis so that a factor that might be relevant if you were pricing a good or a service might be different from the factors you would take into account if you were trying to estimate a credit rating. In the present case he did not think business strategy was something that would feed directly into the analysis of comparability of the 26 companies.

400. It was put to Dr Horst that his Case 1 was not a particularly good way of predicting a credit rating in that it only got it right 65% of the time. Dr Horst responded that it was miles ahead of the competition and if you looked at the S&P credit rating model or the Moody's model or another one of the articles cited by Dr Becker, his model was almost as good as it gets. But Dr Horst accepted that you could not issue $1 billion worth of commercial paper based on an estimated credit rating: you would


ATC 17874

have to go and get a credit rating from S&P and/or Moody's and the way that they would go about it would not be the kind of model that he had used. Dr Horst did not accept that his process was necessarily a less accurate way of estimating a credit rating. When he made the same assumptions as Mr Thieroff made, Dr Horst got to the same result.

401. The applicant submitted in relation to Dr Horst's evidence, that he, along with other experts called by the respondent attempted impermissibly to reformulate the terms of the loan. Dr Horst's initial observation was that if CAHPL had borrowed from an independent enterprise, economic considerations would have resulted in a USD denominated loan. On that basis, he applied a USD denominated floating interest rate, which "resulted in a considerably lower interest rate". Such recasting of the currency of the loan was impermissible, and in any event failed to take into account all relevant facts. The error infected the whole of Dr Horst's first report, and his ultimate conclusion. Dr Horst maintained this error in his second report, and in his third report, Dr Horst sought to support his earlier reports by an argument, advanced for the first time, that a re-characterisation was authorised by paragraph 1.37 of the OECD Guidelines because the terms of the Credit Facility "differ from those that would have been adopted by independent enterprises behaving in a commercially rational manner". Dr Horst also attempted to take into account implicit credit support.

402. The applicant also submitted that Dr Horst, along with two other of the experts called by the respondent, impermissibly attributed to the independent lender the risks, functions and assets of CFC, and moreover, erroneously sought to price the loan from the economic perspective of the lender. Dr Horst accepted that ordinarily the functions, assets and risks of the lender were irrelevant to pricing, but said in his third report only that in this case there were certain extraordinary circumstances pertaining to CFC that should be attributed to the independent lender because they were matters that affected pricing. Erroneous reasoning (for example, that the conditions he relied upon in pricing the loan based on CFC's actual circumstances were non-arm's length conditions) led to a conclusion that the minimum interest rate was not, as Dr Horst had previously concluded, USD LIBOR plus 1.00% per annum, but rather CFC's out of pocket costs on its own borrowing in the commercial paper market (which were less than USD LIBOR). Furthermore, he attributed to the independent lender the benefit of a free guarantee, itself a non-arm's length dealing which in fact provided it with the actual credit support that was both necessary and enabled it to borrow at very low rates in the US commercial paper market. Tellingly, for something so "extraordinary", this analysis was not mentioned at all by Dr Horst in his first two reports.

403. The applicant submitted that Dr Horst failed to undertake a direct credit risk analysis of CAHPL but rather sought to do so indirectly by a statistical analysis of the credit ratings given to other entities, being 26 companies identified by Goldman Sachs and Deutsche Bank for the purpose of their exercise and drawing inferences therefrom to predict a credit rating for CAHPL. On his own admission, Dr Horst was not qualified to determine the rating of CAHPL applying credit rating policies, having never worked at a rating agency. The evidence of Dr Horst was of little weight in circumstances where the Court had the benefit of evidence from people with direct experience in assessing credit ratings (in particular Mr Thieroff and Mr Emmer) and direct experience in assessing credit risk from a lender's perspective (Mr Gross, Mr Martin, Mr Long and to a limited extent, Mr Hollas). Dr Becker's evidence was that a transfer pricing economist would not prefer a statistical model based on a selected analysis of the credit metrics over other forms of direct analysis. Dr Horst in his cross examination agreed that if a company were seeking to raise debt it would go and get a credit rating from S&P and/or Moody's and that they would not go about it with his kind of statistical model. Indeed, Dr Horst's model itself was predicated on an assumption that CAHPL was rated within a particular range. As Dr Becker explained, it was preferable to examine the data supporting that assumption rather than constructing an ordered probit model based on the data of other companies.

404. The applicant submitted that Dr Horst's ordered probit regression analysis, on Dr Horst's own admission, did not reflect that actually used by credit rating agencies. Dr Horst's methodology was inappropriate for determining a rating of CAHPL. It left no room for the factoring in of qualitative judgements which may result in entities with similar credit metrics being accorded different ratings. Dr Horst's conclusion on an appropriate interest rate depended heavily on his assumptions about notching for parental support which, even if permitted by the legislation, was not supported by the evidence. The ordered probit approach was inferior to other approaches given the small clustered observations to which it was being applied. Dr Horst's selection of his preferred regression result was based on a "total notch differential" method of selecting the best regression result which was not one generally recognised by statisticians or mentioned in statistical literature. A formulaic approach was neither accurate nor reliable and was indeed dependent upon a view being taken of CAHPL's likely credit rating before the model was even developed. In estimating CAHPL's credit metrics, Dr Horst recharacterised the borrowing made by CAHPL from an AUD loan to a USD loan. Dr Horst's methodology was conducive to producing a range of results rather than a specific outcome.

405. The applicant submitted that the evidence of Dr Horst in relation to the "hedge" between USD revenues and a proposed USD borrowing was an observation about the economic relationship between cash flows, it was not evidence that an accounting hedge would be available to CAHPL to address foreign exchange gains and losses. While Dr Horst referred to the 2002 accounts of Woodside, he was unaware that the introduction of IFRS gave rise to a change in borrowing practice, and he had not considered the 2005 accounts.

406. The applicant submitted that Dr Horst dismissed Mr Lewis' concerns about the income tax consequences of a USD loan without even seeking to quantify the potential tax liability the subject of Mr Lewis' concerns and on the basis of an unarticulated assumption that the Australian tax treatment was broadly similar to that


ATC 17875

in the United States in the late 1970s and early 1980s.

407. I find that Dr Horst's reports were not useful in addressing the statutory issues. One of the elements of the question he was asked for his first report was what CAHPL might be expected to have paid rather than what is, in my view, the correct statutory question. I also accept the applicant's submission that it is not part of the statutory question to attribute to the independent lender the characteristics of CFC. This approach also informed Dr Horst's report annexed to his third affidavit. As to his determination of the actual ratings of particular issuances, Dr Horst accepted that credit rating agencies did not employ his method and it was not suggested that lenders employed that method. Dr Horst did not have any specialised expertise in the world of banks.

Mr Emmer

408. Mr Emmer was a financial services and ratings industry consultant, called by the respondent. He had been employed in the financial services industry since 1969 and was employed by S&P for 35 years. He held a number of senior positions during the course of his career with S&P, including Managing Director of International Ratings, Executive Managing Director of Financial Institution Ratings, Executive Managing Director of Global Corporate & Government Ratings, and Executive Managing Director of Equity and Research. He had also served on and chaired the S&P Corporate Ratings Criteria Committee. As I have said, I have found the rating agency evidence to be irrelevant. Nevertheless I shall set out Mr Emmer's evidence.

409. For his first report, annexed to his affidavit sworn on 3 July 2013, Mr Emmer was asked to address the following questions:


ATC 17876

  • 1. What is the credit rating that might reasonably be expected to have been given to CAHPL as at 6 June 2003, on the assumption that it was proposing to borrow USD 2.5 billion (or the Australian dollar equivalent) for a term of five years:
    • i. at 1 month AUD LIBOR-BBA +4.14%;
    • ii. at such rate of interest as might be expected to have been payable to an independent lender with whom it had dealt with at arm's length.
  • 2. What is the credit rating that might reasonably be expected to have been given to a proposed borrowing by CAHPL of USD 2.5 billion (or the Australian dollar equivalent) for a term of 5 years:
    • i. at 1 month AUD LIBOR-BBA +4.14%;
    • ii. at such rate of interest as might be expected to have been payable to an independent lender with whom it had dealt with at arm's length.

410. Mr Emmer assumed that funds for the proposed borrowing by CAHPL of USD2.5 billion on 6 June 2003 were to be provided by a third party lender. He said his analysis recognised that CAHPL was a wholly-owned subsidiary of CVX at the loan date. He said he had endeavoured to use information that would have been available prior to the loan date, and in certain cases he had used documents prepared after that date when it appeared that the documents illustrated facts that were known to the public or Chevron management prior to the loan date.

411. A summary of his opinion was:

  • 1. The credit rating that might reasonably be expected to have been given to CAHPL at the loan date would fall in the "A" category (i.e., "A+", "A" or "A-") or possibly higher.
  • 2. The credit rating that might reasonably be expected to have been given to the CAHPL credit facility at the loan date would fall in the "A" category (i.e., "A+", "A" or "A-") or possibly higher.

412. For his second report, annexed to his affidavit sworn on 4 June, 2014, Mr Emmer was asked to set out his comments in response to the affidavit of Mr Thieroff affirmed 7 March 2014 and to state whether he agreed or disagreed with the statements made in it. Mr Thieroff had responded to the questions:

  • 1. In issuing a corporate credit rating, what, if any, is the significance of moving from a sub-investment grade rating to an investment-grade rating.
  • 2. Please describe the conduct of the credit committee meetings at Standard & Poor's in 2002/2003. In particular, how differences in opinion between credit committee members were resolved.

413. In cross-examination, Mr Emmer agreed that in 2003 he did not have any personal involvement as an oil and gas analyst. Also, by 2003 Mr Emmer was not voting in rating committees.

414. Mr Emmer said that he did a stand-alone credit profile of CAHPL excluding the ownership of CVX and he assigned a rating to CAHPL including the ownership of CVX: those were the two steps. He said unproved reserves associated with Gorgon were a factor in his stand-alone rating of BB and it was an important factor in the notching up also.

415.


ATC 17877

As to some of the matters referred to in his report postdating the loan, Mr Emmer said that the announcement of a strategic initiative was at the end of rather a long process so that the company was probably thinking about it six or nine months before the announcement. As to a strategic initiative being dependent upon successful approvals from, for example, the Western Australian government in relation to Barrow Island, Mr Emmer said that if the company did not have approval they would not have gone ahead with the project. He had assumed that the project would go ahead at some point.

416. Mr Emmer agreed that even after taking into account the many references to the Gorgon field, he concluded that CAHPL was a strategically important subsidiary, rather than a core subsidiary.

417. Mr Emmer was taken to his recommendation in a case in Canada in which he had given evidence where he had equalised the rating of the Canadian subsidiary with that of its US parent. He agreed that a portion of it was based on implied parental support but his main reason was that the company was in a confidence sensitive business and had they let GE Capital default on its obligations it was very likely that it could have resulted in the markets losing confidence in GE Capital, which was a financial institution.

418. At the time he prepared his report, Mr Emmer compared the assets of CAHPL with those of the ultimate parent group and he compared the leverage of the parent and the subsidiary, CAHPL being more leveraged than the ultimate parent: the leverage of CVX at the time being 27% and that of CAHPL after the loan being 47%. It was put to him that if he was rating CAHPL based upon implicit parental support, S&P would have asked for either a guarantee or a keep-well agreement if they were going to equalise the rating. Mr Emmer said he did not think that was necessarily the case or possibly the case because at the end of the day CAHPL was going to look like whatever CVX wanted it to look like. As to the assets of CAHPL being approximately 2.3% of the assets of the group, again Mr Emmer said that although the discrepancy was substantial and there was neither a keep-well agreement nor a guarantee, there was management control and that meant that CAHPL was going to look like whatever suited the needs of the Chevron group.

419. Mr Emmer was asked questions about his evidence in the case in Canada and he said that he took a very different approach to CAHPL versus GE Capital Canada. He agreed that the Canadian court did not agree with his assessment that GE Capital Canada would be rated the same as GE Capital.

420. Mr Emmer said that S&P distinguished between proved reserves and unproved reserves and only gave credit to proved reserves, with unproved reserves being given qualitative credit.

421. Mr Emmer said that in his notching up he placed some reliance on the Gorgon unproved reserves but it certainly was not the major factor in reaching his conclusion. He said that the section headed "Management Control" and the section headed "Shared Name" were considerably more important than Gorgon which was of secondary importance. Mr Emmer said that ultimately CVX's most important asset was its reputation, not Gorgon. If it were to walk away from a wholly-owned subsidiary such as CAHPL as a result of economic factors, allowing CAHPL to default on its obligations, it would severely impact its business beyond any possible short-term financial benefits, for example CVX's ability to attract joint-venture partners in the future. He said that was the most important reason for his giving a ratings uplift to CAHPL. If CVX could not be deemed as a reliable partner to companies that it was going to participate with in joint ventures, people would be reluctant to enter into agreements with them as being a reliable partner.

422. Mr Emmer agreed that using spreads was a proper way to determine an interest rate and that Bloomberg was a reliable source for that purpose. He also agreed that lenders were not encouraged by S&P to rely upon S&P ratings as a substitute for their own analysis and lenders proposing to lend to an entity would perform their own analysis. In 2003 Mr Emmer certainly would encourage bankers to do their own analysis.

423. Mr Emmer was asked about his second report dealing with the topic of moving from a non-investment grade rating to an investment grade rating and his conclusion that S&P analysts were not biased to provide lower ratings. He said the ratings were a continuum and there was no special hurdle to overcome. Mr Emmer agreed that interest rates would be higher and there would be potentially a


ATC 17878

narrower group of investors where a rating moved from an investment grade to a speculative grade. He agreed that there were advantages for a company that moved or was moved in its rating from non-investment grade to investment grade: it resulted in a broader audience of potential investors. He said that analysts looked carefully at moving ratings either up or down from any category, irrespective of the market consequences. He said he saw no evidence of conservatism moving from a non-investment grade to an investment grade rating. Mr Emmer accepted that there was a sense of conservatism on the part of raters in the energy trading sector but said the companies mentioned by Mr Thieroff were not oil and gas exploration companies.

424. The applicant's submissions in relation to Mr Emmer's evidence were as follows.

425. Mr Emmer was an S&P analyst until 1978 but since then had been in management at S&P. By the late 1990s or 2000 he had ceased voting at rating committee meetings. He was not at the "coal face" during the relevant period and his opinion on the work of analysts and committee meetings should, therefore, be accorded less weight than that of Mr Thieroff and Ms Esposito. Mr Emmer had not applied the S&P criteria to an E&P company since the 1970s.

426. Mr Emmer placed heavy reliance on so-called implicit credit support and/or otherwise assumed that in the hypothetical, CVX was the parent of the hypothetical "independent" borrower. Mr Emmer's first report placed significant emphasis on the "linkages between Chevron and Chevron Australia", and ultimately he notched up his assessment of the stand-alone credit rating of CAHPL by four to eight notches to give CAHPL an investment grade rating in the "A" category. In his second report, Mr Emmer attempted to defend his "notching" analysis. In his reports, Mr Emmer increased the rating four to eight notches to arrive at not just an investment grade rating, but a high investment grade rating, in the "A" category. The principal departure between the Thieroff reports and the Emmer reports was the emphasis to be given to "relationship factors" or "affiliation" (so called "implicit support") both generally and on the facts of this case.

427. Mr Emmer devoted a very large part of his primary report to the importance of Gorgon to the Chevron group, and he relied on a mix of annual reports, earnings calls and other similar documents that Mr Thieroff described as serving a "marketing function". With respect to those kinds of documents, the applicant submitted, it was important as a credit analyst to learn to separate the wheat from the chaff. It was not the case that one could assume consistency between what a company said in those documents and what it said to rating agents. Furthermore, many of the documents Mr Emmer cited post-dated the loan itself, and, however backward looking, would not have been available for consideration by a rating agent at the relevant time. The critical fact was that when the loan was made, essential governmental approvals were yet to be obtained. The company had announced that it would not proceed with Gorgon if that approval was not forthcoming. All observations about Gorgon prior to that time must be read in that light. Mr Emmer read Mr Oen's affidavit concerning Gorgon but did not appear to have taken it into account. He simply preferred to assume that Gorgon would go ahead anyway.

428. Mr Emmer placed significant emphasis on the disproportionately large holding of possible and probable reserves by CAHPL in order to support his "linkage" analysis to notch up CAHPL. That approach was contrary to the direct instructions of S&P in its publication "Oil & Gas Exploration and Production Keys to Success", which stated that "S&P attributes little value to these assets beyond their liquidation value in distress scenarios". Although he cited these publications in support of his approach, Mr Emmer acknowledged that they did not in fact support the taking into account of the reserves and was unable to point to any other S&P publication that did. Without taking those reserves into account Mr Emmer would have given CAHPL a BB+ rating.

429. Mr Emmer identified a small number of instances where subsidiaries had received multiple notch uplifts. However, four of the six companies he listed benefited from various levels of government support. Governments had an ability to provide support and run businesses at a loss for political reasons: thus, these four companies did not support Mr Emmer's proposition that such notching would or could apply to CAHPL. The other two were joint venture companies, and S&P treated joint ventures differently because partners in a joint venture had the ability to enforce financial obligations on each other: that is, they could compel each other to provide support. Another example quoted, Pennzoil-Quaker State, appeared to have been the beneficiary of a representation of support made by the parent directly to S&P, and its parent demonstrated its support for Pennzoil-Quaker State by tendering for its outstanding debt. Without that fact, S&P likely would have taken a different course. Other examples raised in cross-examination were explained by reference to the presence of guarantees, and by the designation of the subsidiary as a core subsidiary.

430. Mr Emmer had no trouble in undertaking a stand-alone assessment nor in agreeing that CAHPL's stand-alone rating was "BB".

431. In my opinion, there is force in the criticism that Mr Emmer had not been directly


ATC 17879

involved in rating for some decades. On that basis, if it were otherwise relevant, I would give Mr Emmer's opinion less weight. I also find that the questions Mr Emmer was asked, and therefore his answers, proceeded by reference to CAHPL itself rather than to an independent entity. I do not find his evidence to be useful.

Ms Azarchs

432. Ms Azarchs was called by the respondent Commissioner and gave evidence as an independent consultant. She was an independent researcher on matters affecting financial institutions. For most of her career, for the 28 years ending in 2010, she was a financial institutions analyst at S&P, spanning both equity analysis and credit rating analysis of financial companies. Between 1989 and June 2010 she had been the managing Director of the Financial Institutions Rating Department at S&P. She was asked to address the following questions:

Question 1 : Could or would S&P rate subsidiaries at the same rating as a parent absent a guarantee?

Question 2: Is the difference between a rating at the high-end of the non-investment grade (BB+) and the low end of the investment grade (BBB-), which nominally represents a one notch differential, greater than other one-notch differentials?

433. As I have said, I do not regard the approach of rating agencies as relevant. Nevertheless I shall set out Ms Azarchs' evidence. Her summary answer to question 1 was that S&P, as well as other rating agencies could, and in a preponderance of instances did, rate subsidiaries at the level of the parent group rating even in the absence of a guarantee or other less legally binding expressions of support. She said that logic and experience dictated that parent company management saw great incentives in supporting, or forestalling the default of, their subsidiaries. The parent may find that it is less costly to provide financial support than to suffer the consequences of the bankruptcy of that subsidiary, both in terms of the loss of the value of that subsidiary and in terms of reputation risk.

434. Ms Azarchs' summary answer to question 2 was that S&P analysts were instructed that the difference between a BB+ and a BBB- rating should be no different in terms of the standards applied than the difference between any other two contiguous ratings. She said that market perceptions could be different, so that a rating changed from or into investment-grade or non-investment grade may result in a greater change in the availability of funding or in its pricing than would be common for other rating changes.

435. In her report, Ms Azarchs was considering the general creditworthiness of CAHPL, not the general creditworthiness of the facility, the CFC-CAHPL loan.

436. In cross-examination, Ms Azarchs said that when she did her search in


ATC 17880

2013 all she could pull up were the 2004 S&P published criteria for subsidiary ratings as anything before 2004 that was published in paper form was not available electronically discoverable to the outside. She added that she had since looked at the 2002 publication and it was verbatim the same as the 2004. She agreed that for one section of her report she relied on the 2013 criteria because they were more systematic but not materially different, in her view.

437. Ms Azarchs agreed that a rating was in the end an opinion and the rating experience was as much an art as it was a science. She also agreed with the statement that a credit rating was S&P's opinion of the general creditworthiness of an obligor or of the creditworthiness of an obligor with respect to a particular debt security or other financial obligation, based on relevant risk factors. Ms Azarchs also agreed with the following underlined statements as published by S&P in 2003 when the CFC-CAHPL loan was made:

A weak subsidiary owned by a strong parent will usually, although not always, enjoy a stronger rating than it would on a stand-alone basis. Assuming the parent has the ability to support the subsidiary during a period of financial stress, the spectrum of possibilities still ranges from ratings equalisation at one extreme to very little or no help from the parent's credit quality at the other. The greater the gap to be bridged, the more evidence of support is necessary.

(The parent's rating is, of course, assigned when the parent guarantees or assumes subsidiary debt. ...)

(Italics in original.)

Ms Azarchs understood that the gap that needed to be dealt with in the present case was from BB to AA.

438. Ms Azarchs was taken to a S&P publication of August 2000 entitled "Global Credit Portal RatingsDirect" which included the following statements under the heading "Issuer Credit Ratings" ( ICRs ):

Typically, affiliation with stronger family members will enhance a weaker entity's rating by at least one notch. Multiple notch enhancements that seek to elevate a significantly weaker entity to the level of the consolidated group reflect the determination that the business is a core activity and will be supported to the fullest extent possible by available consolidated resources. ICRs for subsidiaries viewed as strategic can range anywhere between stand-alone and consolidated ratings (unless insulated), depending on explicit and/or implicit system support.

These passages were Ms Azarchs' understanding of the position as it would have applied in 2003 when the CFC-CAHPL loan was made.

439. Ms Azarchs was also taken to the 2013 criteria which stated the S&P Ratings Services was updating its methodology for rating members of corporate groups to align it with the criteria for members of financial institutions and insurance groups, and therefore was adding a section titled "Methodology: Corporate Groups". Ms Azarchs said that the broad principle did not change as to whether S&P would equalise the ratings of parents and subsidiaries when they were considered to be highly strategic or core. On that basis I admit that evidence.

440. Ms Azarchs was cross-examined on her statement that CVX had effectively guaranteed the debt of CAHPL in that the commercial paper issued by CFC and downstreamed to CAHPL was guaranteed by CVX. CFC had no other source of revenue than debt repayments by CAHPL, so CVX was effectively saying that either CAHPL would pay or CVX would cover CFC's obligation. She said that if there was a guarantee there would be no question that she simply would have rated CAHPL's debt at the same level, but where there was no guarantee then she was looking for evidence that might, at the margin, be some evidence of parental attitude towards CAHPL.

441. Ms Azarchs was cross-examined on her statement that she agreed that the fact that there was no reason to believe that CVX would allow a subsidiary to default, in contradiction to a long track record of minimal subsidiary defaults, was relevant. She said the defaults she was referring to were parents allowing subsidiaries to default when the parent continued, as at 2013. She said she was referring to subsidiary defaults in isolation from their group and she did not rely on the 2013 document to make that statement. Ms Azarchs was taken to an S&P publication of 17 December 2002 and agreed there was no reason to doubt the statement that the number of investment grade companies defaulting during 2002 was unprecedented and it was the result of parent companies removing support from subsidiaries.

442. Ms Azarchs was taken to a Moody's December 2003 document and to a heading "Established Subsidiaries With Lower Risk Profiles", and agreed that S&P would also commence with the intrinsic stand-alone rating and, depending on the facts and circumstances involved, notch the rating higher if the rating committee determined that the parent had significant financial incentives to provide support to the subsidiary, and that any "ratings lift" based on the parent's willingness and ability to provide support could be limited to only one or two notches above the stand-alone rating. She said that perhaps S&P were more bullish. She did not agree that at S&P there was the sort of distinction between financials and corporates referred to in the Moody's documents to the effect that the norm for "ratings lift" may be only one or two notches for non-financial corporations but it could be considerably more for financial corporations because of the role played by regulation and reputation among financial firms.

443. Ms Azarchs was cross-examined about her statement that the costs of allowing a subsidiary to default could be incalculable because, in part, the reputational damage to a company of being seen to allow a subsidiary to default could have repercussions on its own ability to fund itself in the marketplace, or to make future acquisitions. She was taken to a S&P 2004 document "Financial Institutions Criteria" and agreed with the statement in that document that banking was more likely than any other industry to be directly or indirectly affected by any sovereign default or other such crisis and that this vulnerability was due, amongst other things, to their dependence on confidence, which could disappear in a crisis. She also agreed with the statement in that document that maintaining confidence in banks was a key policy area for all governments, as any loss of confidence could cause a short-term liquidity crisis owing to the very high leverage in the banking sector.

444. In relation to her statement that other rating agencies, not only S&P could, and in a preponderance of instances, did rate subsidiaries at the level of the parent group rating Ms Azarchs agreed she did not have the data to support her statement about the other rating agencies. She agreed that whether or not the subsidiary would be rated equal with the parent or whether or not it would be notched up depended upon what the analysts and the committee decided based upon that particular company's circumstances in relation to carrying out the tasks set out in the criteria. Each case depended on its own facts and depended upon quantitative as well as qualitative matters. Qualitative matters were subjective.

445. Ms Azarchs was taken to her statement that of the 164 subsidiaries rated by S&P that belonged to 93 parent companies in the energy and minerals sectors globally, 146 subsidiaries were rated the same as the parent while only 10 were rated one notch lower, four were two notches lower, and four were three or more notches lower. She agreed that the energy and mineral sector globally was wider than the oil and gas sector and much wider than the E&P sector, and also agreed that the exercise was carried out with respect to data available from 2013. She agreed she did not know the stand-alone rating for any of the unidentified subsidiaries or whether there was a guarantee in place in respect of any of them, or whether there was a letter of comfort or a keep-well agreement in relation to any of them. She agreed she did not know the basis, in relation to any of those companies, of the notching up to the parent, where that had occurred. The following exchange occurred:

If I accept Mr Thieroff's enumeration of which were rated on their own merits, stand-alone rating, or on the basis of a guarantee, and if I exclude all of the partially owned subsidiaries, joint ventures and other things that would probably disqualify those subsidiaries from being called core, then what we get is a list of 84 non-guaranteed, non-government supported, not rated on their own merits, you still get 84 subsidiaries that are essentially deemed to be core, because their ratings are equalised with those of the parent and you get only seven that have been differentiated from the ratings of their parent.

So 84 out of 164, that's a good enough hit rate for you?---84 equalised versus seven notched down still supports the statement that the preponderance of subsidiaries are rated at the same level as their parent, absent a parent guarantee.

446. Ms Azarchs agreed that she did not know how many of the 164 companies were government related or benefited from government support; or how many were owned by joint ventures; or how many of them were guaranteed.

447. Ms Azarchs was taken to a number of errors as between her Exhibit 1 ("S&P Ratings of Parents and Subsidiaries") and the S&P data which she used to prepare it but maintained that Exhibit 1 was supported by that S&P data.

448. Ms Azarchs said that the markets behaved very differently with regard to companies in the investment-grade versus the non-investment grade range and that it was an important difference to the markets and to the companies because the ability to raise money as an investment grade as opposed to a non-investment grade company was quite different and the rates at which money could be raised were quite different. But she did not agree with Mr Thieroff that there was a practice of exercising caution in going from non-investment grade to investment grade.


ATC 17882

449. The applicant's submissions in relation to Ms Azarchs' evidence were that she placed heavy reliance on the so-called implicit credit support and/or otherwise assumed that in the hypothetical, CVX was the parent of the hypothetical "independent" borrower. The applicant submitted that Ms Azarchs had never rated oil and gas companies. The principal (although by no means the only) departure between the Thieroff reports on the one hand and, amongst others, the Azarchs report on the other, was the emphasis to be given to "relationship factors" or "affiliation" (so called "implicit support") both generally and on the facts of this case.

450. The applicant submitted that Ms Azarchs, a financial institutions analyst who had never worked in corporate ratings nor, in particular, oil and gas companies, managed to agree with Mr Taylor and Mr Emmer's reasoning that CAHPL was a "highly strategic" or even "core" subsidiary of Chevron without even reading their reports. Her evidence, the applicant submitted, was clearly influenced by her background in the financial institutions area, which was a predominantly investment grade industry (as opposed to E&P) and had greater emphasis on reputation and market confidence, two factors that were highly relevant to "implicit support". For that reason her opinion should be accorded little weight. The applicant submitted Ms Azarchs made representations about "other rating agencies" but she did not consider the relevant Moody's publication, which provided that the "norm" for a ratings lift was one or two notches for non-financial corporations.

451. The applicant submitted that Ms Azarchs referred to and applied a 2013 methodology that was clearly different to the relevant methodology. Her proposition that S&P data showed that "ratings of subsidiaries are most often equalized with those of their parents" was not supported by her Exhibit 1, for the following reasons. First, Exhibit 1 included a number of companies that were guaranteed by their parents and thus would have obtained "equalization" on that basis. Ms Azarchs was asked to opine on whether S&P could or would rate subsidiaries the same as the parent, absent a guarantee, but she took no steps to exclude guaranteed companies. Worse than being an "imperfect attempt", her Exhibit 1, fundamentally, could not answer the question she was asked. Secondly, Exhibit 1 included a number of other categories of companies that may have been "equalized" with their parent due to factors other than implicit support. They included companies that enjoyed government ownership or support, joint ventures, and companies that had stand-alone ratings equal to their parent (and therefore could not be said to have benefited from implicit support). In that respect, Ms Azarchs was completely unaware of the basis on which the subsidiaries referred to in her Exhibit 1 had been "equalized" with the parent. Thirdly, there were many discrepancies between Exhibit 1 and the S&P data used by Ms Azarchs that Ms Azarchs was not able satisfactorily to explain. Some companies were excluded deliberately, although the basis for doing so, and the number of exclusions, was unclear. What was clear was that Ms Azarchs made errors in relation to the number of subsidiaries in the S&P data, she identified some as being rated the same as the parent when in fact it was higher, she identified some as being the same as the parent when in fact it was lower, and in some cases she identified the rating difference (the number of notches) incorrectly. In light of the above, the applicant submitted, Ms Azarchs' statement that her conclusion was supported should be rejected by the Court.

452. As to the difference between the two contiguous ratings, the applicant submitted that the graph prepared by Ms Azarchs did not assist the Court on this question, any more than her other evidence; the relatively smooth exponential line resulted from the fact that it covered defaults by companies from all industries, which obscured industry specific risk factors. No E&P specific graph was put forward by Ms Azarchs; due to the high risk nature of that industry it was possible that such a curve could be "kinked" or even "displaced".

453. In my opinion, Ms Azarchs' evidence, if relevant, was not persuasive. On the basis of her evidence I would not conclude that S&P would rate subsidiaries at the same rating as a parent, absent a guarantee. Neither am I persuaded that, so far as concerns rating agencies, there was not a practice of exercising caution in going from non-investment grade to investment grade. I find that there was such caution

Mr Taylor

454. Mr Taylor, a credit rating advisor since 1996, was called by the respondent. As I have said before, I do not regard evidence of the practices of the rating agencies to be relevant and Mr Taylor adopted the


ATC 17883

methodology that he said would have been applied by the rating agencies. Nevertheless, I shall summarise Mr Taylor's evidence.

455. He prepared his first report, annexed to his affidavit sworn on 8 July 2013, to address the following questions:

  • 1. Mr Thieroff has prepared his report on the basis of the following assumption:

    "For the purposes of preparing your opinion, to the extent that CAHPL does have the benefit of any implicit credit support by virtue of its ownership by Chevron Texaco Corporation, please disregard that benefit.

    Do you consider this to be an appropriate assumption to make in answering the question:

    "What is the credit rating that might reasonably be expected to have been given to the credit facility entered into between CAHPL and [CFC] on 6 June 2003, if CAHPL had entered into that facility with an independent party with whom it had dealt with at arm's length?

  • 2. If not, what is the appropriate methodology for determining the credit rating that might reasonably have been expected to have been given to the credit facility?
  • 3. If you make the assumption that Mr Thieroff was asked to make, do you agree with Mr Thieroff's conclusions as to the credit rating of the credit facility that might reasonably be expected to have been given?
  • 4. What credit rating might reasonably have been expected to have been given to the credit facility if the methodology you consider appropriate were used?

456. His answer to question 1 was "no". He said that a credit rating must take into account any factor which would influence the creditworthiness of CAHPL and the facility. Therefore, in estimating a rating for CAHPL it was not appropriate to disregard either the implicit credit support that appeared likely to be available to CAHPL by virtue of its ownership by CVX or the actual support assumed to be received from CVX, such as management exchange and other operational support, such as Treasury and technology support. In answer to question 2, he said the appropriate methodology was the methodology that would be applied by the rating agencies in assigning a rating to the facility had they been approached to do so on 6 June 2003. He said that in his opinion Mr Thieroff's report was an incomplete assessment of CAHPL's creditworthiness because it uncritically neglected parent/subsidiary considerations set out in their corporate credit rating methodologies which were required to be evaluated to properly determine overall creditworthiness. He also said that, in his opinion, Mr Thieroff's report was an evaluation of the risk profile of CAHPL's assets, operations and that implied by its financial leverage without regard to CAHPL's corporate context. The summary of Mr Taylor's answer to question 3 was that Mr Thieroff had ignored all forms of parental support, but Mr Taylor agreed that Mr Thieroff's estimate of BB (non-investment grade) was reasonable in determining the creditworthiness of the facility if all forms of parental support and CAHPL's corporate context were ignored. Mr Taylor noted that the rating agencies factor all forms of support either provided or likely to be provided by a parent to a subsidiary into their analysis. This support would include day-to-day financial support as well as technical, management and treasury support. The summary of Mr Taylor's answer to question four was that, in his view, the Credit Facility would be likely to have received a credit rating in the range of A to AA, arriving at that conclusion through what he described for the purposes of explanation as a three-step process: first, commencing with a "stand-alone" creditworthiness assessment for the Credit Facility as determined by Mr Thieroff; secondly, allowing for day-to-day financial support as well as technical, management and treasury support; and thirdly, determining the likely level of financial support from CVX provided in extraordinary circumstances in order to ensure the Credit Facility continued to be serviced. Mr Taylor later explained that steps one and two were in reality combined as there was no meaningful concept of a pure stand-alone rating and what the agencies did was to ask what were the key characteristics of CAHPL from a credit sense and what was believed to be the benefit of the day-to-day support that CAHPL received as a result of its ownership by CVX.

457. For his second report, annexed to his affidavit sworn on 6 June 2014, Mr Taylor was asked whether he had any comment in respect of the matters raised in Mr Thieroff's affidavit affirmed on 5 August 2013 and, if so, to provide such comment and the reasons for any opinion he expressed. Mr Taylor provided his comments under the headings: "Absence of peer comparison"; "Absence of a range of possible credit rating outcomes"; and "Differences in interpretation of the meaning or significance of certain S&P criteria".

458. In cross-examination, Mr Taylor agreed that his relevant experience was as a ratings advisor at different investment banks and he had no particular expertise in any one industry sector: rather his experience was in relation to a broad range of companies in the industry sectors. He had never worked at a rating agency. He agreed that in 2003 it was usual to seek ratings from more than one agency and the two most popular were Moody's and


ATC 17884

S&P.

459. Mr Taylor agreed that it would be important for the rating agency to be told whether the loan was secured or unsecured as it could have a bearing on the actual rating of the loan, and whether it was senior, and whether it would be subordinated or unsubordinated.

460. Mr Taylor was asked whether it was the Credit Facility that he was shadow rating as opposed to the company itself and his response was that he looked at both and the rating would be the same for CAHPL and the facility.

461. Mr Taylor was questioned as to credit rating agency usage in 2003 and it was put to him that the agencies had no trouble using the terminology "stand-alone credit rating" in 2003 and it was not until 2009 that any publication by S&P referred to a "stand-alone credit profile". Mr Taylor said the nomenclature was not what he would call significant for the purposes of his report. He said he was applying the stand-alone credit rating (including actual parental support) to notch up the credit rating and then he would add what the agencies would attribute for implied support after that and that got from BB to BBB-.

462. Mr Taylor agreed that he should have set out as an assumption an undertaking by CVX in considering a hypothetical loan to an external third party. He assumed that were CAHPL to have taken out a loan issued to an independent third party that the Board of CAHPL would expect some type of undertaking. He incorporated that assumption in the day-to-day support.


ATC 17885

463. On the question whether CAHPL represented an economically material investment for CVX and the "general rule of thumb" to which Mr Taylor referred that assets representing more than 5% of total group assets were considered to be material by the rating agencies, it was put to him, and he accepted, that he was comparing revalued assets of CAHPL to historic cost assets of CVX. He said that if you were to revalue all of CVX's assets on that particular day the denominator could well be higher and therefore CAHPL's percentage contribution could be lower.

464. Mr Taylor said that structural subordination would be relevant to determining what the stand-alone credit profile or stand-alone creditworthiness of CAHPL would be, but it would not necessarily have an impact on the overall rating that the rating agencies would assign because ultimately they were trying to determine the support, the implied support, and the propensity that CVX would step in and support the debt issued by CAHPL.

465. Mr Taylor was taken to two examples, to which he had referred as supporting his conclusion about implied parental support in relation to CAHPL, those examples being Coca-Cola Enterprises and Schlumberger. He said he only selected them as a result of the presence of high gearing and that was the similarity between CAHPL and those subsidiaries. He said they were provided to consider whether they could provide any insights into how a highly leveraged subsidiary could be treated because there was a very big gap between CAHPL's stand-alone credit profile and CVX's. Mr Taylor said he found the two examples in the Moody's publication but said they were also rated by S&P. He was asked whether Moody's said that, special circumstances apart, the norm in 2003 for a ratings lift was one to two notches. The relevant passage from the Moody's document, which was read to Mr Taylor, was as follows: "While the norm for 'ratings lift' may be only one or two notches for non-financial corporations..." Mr Taylor said that he applied the methodology of S&P and used it as a result of a response to Mr Thieroff's analysis.

466. The applicant's submissions in relation to Mr Taylor's evidence were that he placed heavy reliance on so-called implicit credit support and/or otherwise assumed that in the hypothetical CVX was the parent of the hypothetical "independent" borrower. The applicant submitted that Mr Taylor (who had never worked at a credit rating agency) based much of his conclusion in his first report that CAHPL would have an investment grade rating in the "A" category or higher on his determination that CAHPL was of "high strategic importance to Chevron " and that Chevron CVX's investment in CAHPL was "highly significant". He defended that view in his second report.

467. The applicant submitted Mr Taylor denied that some subsidiary upgrades were influenced by parental representations made to S&P analysts. This observation was not consistent with the actual rating experience of Mr Thieroff and was made by someone who had never been a ratings analyst. For that reason, his opinion should be accorded less weight. Unwittingly, Mr Taylor also referred to Pennzoil-Quaker State, which appeared to have benefitted from precisely that kind of representation. Mr Thieroff, who worked as a ratings analyst at S&P, explained how these representations (or lack thereof) were taken into account and the language that was used in rating reports to indicate that a representation had been made. Further, the applicant submitted, Mr Taylor: 1) double counted implicit support by applying a notching up methodology "stand alone credit profile" that was not referred to in any S&P document until 2010; 2) acknowledged that his characterisation of what was in the CAHPL Board papers as an undertaking was inaccurate, and an assumption to that effect was not set out in his report; 3) compared CAPHL's assets (recently revalued) with CVX's assets (at historical cost) to calculate a proportion of 9%, which he said was significant as it exceeded a 5% "rule of thumb". When the inappropriateness of the comparison was pointed out and the true figure of under 3% revealed, he sought to minimise that factor; 4) raised comparisons which came from a Moody's document of notching up for implicit support, namely companies (Schlumberger and Coca-Cola) that were "financially, strategically and operationally intertwined" with their parent. Mr Taylor conceded that they could only be compared to CAHPL to a "limited extent" - by reference to leverage; 5) even though he had regard to the Moody's document referred to above, he did not apply the Moody's methodology in that same document which provided that the norm for a ratings uplift was one or two notches. In circumstances where two ratings would be obtained at the time, a Moody's rating being one of them, this methodology was relevant.

468. The applicant submitted that the notching undertaken by Mr Taylor - pushing CAHPL over the line into investment grade - appeared to have been undertaken without regard to what Mr Thieroff and Ms Esposito regarded as a cautious barrier between non-investment-grade and investment grade ratings.

469. I found that Mr Taylor gave his evidence cogently, but it is significant for present purposes that he had not been a rating analyst. If I had found the practices of rating agencies to be relevant, I would have given Mr Taylor's evidence less weight accordingly.

Professor Boymal

470. Professor Boymal, accountant, was called by the respondent. Part of his career had been as an auditor but he had also spent many years as the technical standards partner of Ernst & Young, which he left in 2003.

471.


ATC 17886

He was asked to prepare a report containing his answers to the two following questions:

Question 1 : From the point of view of the CAHPL consolidated financial statements, in what optimal currency would CAHPL (the separate company) have borrowed USD2.45b or its equivalent in another currency on the assumption that CFC was not a subsidiary of CAHPL, CAHPL was independent of CFC and dealing with CFC at arm's length, and on the assumption that 88% of the CAHPL group revenue was in USD?

Question 2: Do you agree with the accounting and related justifications provided by Messrs Dalzell, Callaghan and Lewis for the choice of AUD rather than USD for the CFC to CAHPL loan, assuming that 88% of CAHPL group revenue was in USD, that the loan was denominated in AUD and that the loan was in substance an AUD loan?

472. Professor Boymal's opinion was that the optimal currency in which CAHPL would have borrowed the USD2.45 billion was USD. His reasons were, first, based on the information provided in his instructions, the borrowing rate of interest in the USD market was considerably lower than the borrowing rate of interest in the AUD market and this in itself would have made a USD borrowing more attractive to CAHPL. Secondly, since 88% of the CAHPL group revenue was in USD, then group borrowings in USD would provide a natural hedge against the volatility in group sales revenue arising from changes in the USD-AUD exchange rate. CAHPL faced volatility in group sales revenue as a consequence of its revenue being primarily in USD. For example if the AUD was to strengthen against the USD, the comparative decline in group revenue stated in AUD arising from translating the USD revenue into AUD at the date of each sales transaction would be offset by the comparative gain arising from translating the USD borrowing into AUD at the balance date exchange rate, thus reducing the volatility in the overall reported result in the AUD group financial statements.

473. Professor Boymal did not agree with the accounting and related justifications provided by Mr Dalzell, Mr Callaghan and Mr Lewis for the choice of AUD rather than USD for the CFC to CAHPL loan, assuming 88% of CAHPL group revenue was in USD.

474. In cross-examination, Professor Boymal accepted that he had never prepared financial statements of a large company and the last time he audited them would have been approximately 1995. He did not profess to have any particular expertise in relation to the oil and gas industry. He said he did have a knowledge of hedging generally but no particular knowledge about hedging as it pertained to the oil and gas industry. He had never been responsible for determining the hedging strategy of any company.

475. Professor Boymal accepted that the question of a low interest rate was a commercial matter about reducing the costs of the organisation but said it was also an accounting


ATC 17887

matter insofar as it produced an accounting outcome. It was not applying any particular accounting rules but it was applying accounting in a general sense.

476. As to the question of the USD providing a natural hedge against the fact that a considerable portion of the revenue was in USD, Professor Boymal agreed that when he was speaking about a natural hedge he was talking only about a hedge in a commercial sense and not in terms of a technical hedge accounting sense: he was not applying hedge accounting, being a specific method of recording when hedging takes place.

477. Professor Boymal also agreed that his opinion as to the natural hedge contributed materially to his conclusion about the selection of USD as the currency of the loan.

478. The applicant's submissions in relation to Professor Boymal's evidence were that he was a former accountant but purported to give "commercial" evidence, having never worked for an oil and gas company and thus he did not have sufficient expertise to assist the Court in determining how Australian oil and gas companies should borrow. Also, it was far too simplistic to assume, as Professor Boymal did, that borrowers always borrowed in the currency with the lowest rate. If that were so, then all borrowers would borrow in the same currency. The fact was, the applicant submitted, that the selection of currency was a complex matter and depended on a number of factors, including the denomination of revenues, the denomination of expenses, and the functional currency of the company. It required knowledge of the industry in which the borrower operated, particularly if one was to make an attempt to take advantage of a "natural hedge". The applicant submitted that Professor Boymal attempted impermissibly to reformulate the terms of the loan as his observations in relation to the "optimal currency" of the loan, and his review of the evidence of the lay witnesses, was directed solely to the proposition that arm's length parties would have chosen a different currency. The hedging evidence of Professor Boymal was said to go to a "commercial hedge" rather than hedge accounting. It could not, therefore, touch the issue raised about volatility in the accounts. Nor could it touch the tax issue. And the "commercial" evidence given by Professor Boymal should be given little weight, given that his expertise was in accounting.

479. I find that Professor Boymal, as an accountant and with his experience had very limited qualifications to give the evidence in chief which he gave on the question of the optimal currency of the loan. I give his evidence little weight.

Consideration

Division 13 of the ITAA 1936

480. I approach this question in light of the reasoning of the Full Court in SNF 193 FCR 149 at [128] as follows, recognising that in that case the Commissioner had made a determination under s 136AD(4) whereas in the present case he has not:

Of course, it is true that s 136AD operates to engage the statutory fiction that the consideration paid was "the arm's length consideration" and that the existence of more than one such value may give rise to practical problems of administration. But it is precisely to situations of that kind that the Commissioner's power conferred by s 136AD(4), to determine the arm's length consideration, is apposite: "where, for any reason (including an insufficiency of information available to the Commissioner), it is not possible or not practicable for the Commissioner to ascertain the arm's length consideration ... [it] shall be deemed to be such amount as the Commissioner determines". Where there is more than one arm's length price (as often there will be), the Commissioner may determine which he will apply. Correspondingly, in review proceedings the taxpayer will be entitled to succeed if it shows that the prices paid by it were arm's length prices or less than arm's length prices. If it pursues the latter course there is no need for it to establish a particular price as the arm's length price. It will be sufficient to show that it paid less than an arm's length price.

(Original emphasis.)

481. The applicant submitted that for s 136AD(3)(c) to be satisfied, the amount of the consideration which the taxpayer gave or agreed to give in respect of the acquisition must have exceeded "the arm's length consideration in respect of the acquisition". The applicant submitted, first, that the subject of the inquiry mandated by s 136AD(3)(c) was the "consideration in respect of the acquisition": those words confined the inquiry to the actual acquisition. They were only concerned with an examination of the consideration in respect of the acquisition of the particular property in fact acquired. Here, the acquisition was of a particular bundle of rights conferred by the Credit Facility Agreement. Secondly, the applicant submitted, the phrase "consideration that might reasonably have been expected to have been given ... [by] ... independent parties" was one that necessarily required the removal of all the connections between the actual parties in the hypothesis or comparison mandated by the provision. Unless all the connections were removed, parties would not be independent of each other. That included the common ownership by a single ultimate parent. Thirdly, the applicant submitted, given that the only question posed by s 136AD(3) was whether the consideration in fact given was an arm's length one, the section warranted no investigation or consideration of motive or purpose for the actual acquisition.

482. As to the first of these matters, the applicant submitted CAHPL acquired rights to borrow a sum expressed in AUD. The cost of acquiring and exercising this right was CAHPL's promise to repay principal and interest, denominated in AUD. It was not open to the Commissioner to challenge the commercial choice to prefer AUD: CAHPL borrowed from CFC in AUD to refinance debts that had been incurred in AUD. It was the arm's length consideration for the rights and benefits in fact conferred by the Credit Facility Agreement, which must be determined. As to the second of these matters, in addressing the hypothetical enquiry required by s 136AA and in determining the "arm's length consideration", the legislation ordinarily required one to exclude the particular attributes of the parties in question, and to focus instead upon the intrinsic value of the property or services in question. As a consequence, s 136AD should be construed as requiring the specific circumstances of the taxpayer to be disregarded. Because the hypothetical was concerned with truly "independent" parties, ss 136AA and 136AD(3) required the negation of any connection between the actual parties in formulating the attributes of the hypothetical independent parties dealing at arm's length with each other. In the present case, the matters to be excluded were: (a) the fact that CAHPL and the lender (CFC) were in a parent-subsidiary relationship (that is, the ownership of CFC by CAHPL); and (b) the fact that CAHPL and the lender were each ultimately wholly owned by CVX, that is, the source of their common ownership. It was these "connections" which resulted in CAHPL and CFC not being independent and in order to hypothesise how independent parties would have transacted, one needed to eliminate both.

483. The applicant submitted that where the pricing of the supply or acquisition of goods or services was necessarily affected by the attributes of the purchaser, the general rule against introducing such attributes to the Div 13 hypothetical inquiry needed to be qualified. The provision of financial accommodation was an example of such a transaction. From an economic perspective the characteristics of the borrowing entity were of utmost importance in determining an interest rate on a loan, and it may be accepted that from the perspective of the hypothetical lender the creditworthiness of the purchaser of the financial accommodation (together with other objective attributes, such as the size of the loan and its currency) should affect the price for such a supply.

484. CAHPL accepted that the pricing of the loan between CAHPL and CFC must take into


ATC 17888

account the assets, risks and functions of CAHPL which informed the likelihood of CAHPL repaying both principal and interest. Assumptions about the assets, risks and functions of the purchaser of financial accommodation were needed in order to price such a supply. In that respect, the Commissioner agreed with CAHPL.

485. However, the applicant submitted, the assets, risks and functions of CAHPL to be attributed to the hypothetical inquiry were those of CAHPL as a stand-alone entity. This was the point of disagreement with the Commissioner. Assets, functions and risks in fact attributable to it by reason of its ownership by the Chevron group must be excluded from the hypothetical because of the need for independence mandated by the words of the statute. Those attributes resulting from membership of the Chevron group constituted "connections" between CAHPL and CFC and therefore did not form part of the required hypothetical. That would require the exclusion of what rating agencies refer to as "implicit support", namely the assumed existence of a willingness (or the perception of the existence of a willingness) of a parent of the borrower to provide the borrower with credit support in the event of default on the obligation to repay the loan, in the absence of any legally enforceable obligation to do so. The concept of "implicit support" or credit benefit obtained by reason of the taxpayer's affiliation with CVX was the very product of the non-arm's length relationship. To take account of so called implicit support required a preservation of a key aspect of the non-arm's length relationship between CAHPL and CFC, which was contrary to, and incompatible with, the concept of "independent parties" prescribed by s 136AA(3).

486. Turning to the actual circumstances of the present case, the applicant submitted that the task was to remove the aspects of the actual transaction that arose from the non-independence and non-arm's length dealing between the parties and to price that transaction. That was not a reconstruction of the transaction. It was an acceptance of the transaction but a rejection of all non-arm's length relationships between the parties to it.

487. The applicant submitted that, in effect, the pricing evidence of the expert witnesses it called was unchallenged, that evidence being in the reports of Mr Martin and Mr Gross as to how lenders would price the facility that was actually entered into by CAHPL and in the first report of Mr Thieroff which applied the credit rating methodology used by S&P contemporaneously with the Credit Facility Agreement to CAHPL as an independent entity on a stand-alone basis.

488. The applicant submitted that there were six features of the Credit Facility Agreement that went to the essential nature of the "property acquired" and constituted the "rights, benefits" conferred upon CAHPL. Those features were: (a) the size or quantum of the loan, being a loan facility that was the AUD equivalent of USD2.5 billion; (b) the term was five years; (c) the facility did not contain any onerous financial covenants; (d) the loan was unsecured; (e) the loan was non-amortising; and (f) the currency of the loan was AUD. The applicant submitted that the closest analogue was an institutional loan, known as a "Term Loan B". This was the best and closest comparator: a Term Loan B, like the CAHPL loan, typically required little to no principal amortisation, had a 5-7 year tenor, and had a bullet maturity. There were differences between a typical Term Loan B and the CAHPL facility, namely that the CAHPL facility was unsecured and lacked the financial covenants one would expect to find in a typical Term Loan B, and it provided for the borrower to prepay the loan without penalty. However, none of these differences would cause institutional investors any difficulty in pricing the loan. Provided it was appropriately priced, there would have been sufficient global market capacity to place an institutional loan of this size (AUD equivalent of USD 2.5 billion) in 2003 to a non-investment grade rated borrower without a parental guarantee. It followed that the CAHPL facility could be priced like an institutional loan, with adjustments for the differences between the CAHPL facility and typical institutional loans to companies with similar credit profiles to CAHPL. Those adjustments reflected the additional amounts institutional investors would charge to compensate them for the increase in risk. After making such adjustments, Mr Martin concluded that the price paid by CAHPL did not exceed the arm's length price.

489. The applicant submitted that the relevant evaluation process involved the following principles. First, in assessing credit quality, sophisticated lending institutions (banks and institutional investors) did not rely on agency credit ratings. They performed their own credit analysis in a manner that differed markedly from the credit rating agency methodology. Secondly, sophisticated lenders did not rely on so called "implicit credit support" in pricing a loan. Mr Martin said that lenders evaluated the creditworthiness of the borrower on a stand-alone basis without assuming any support from a corporate parent, in the absence of an obligation to provide support. Mr Gross' view was that "parentage is not worth much": what lenders relied on were enforceable parental guarantees. That view was supported by Mr Long. From the perspective of a lender, significant upgrades in rating based on such "implicit credit support" were "ludicrous". Thirdly, the CAHPL facility could be priced by reference to the institutional loan market with, as necessary, adjustments for differences with the CAHPL facility. Those differences would not disqualify CAHPL from being able to access credit in the amount and on the terms that it did.

490. Mr Martin performed a credit analysis of CAHPL, concluding that it would be rated a "weak BB", which is non-investment grade. He observed that the institutional loan market would be the market in which debt of this kind would be raised at arm's length, principally due to the size of the facility, which would require a large number of lenders (75-100), the lack of security, the lack of covenants and the fact that the loan was a non-amortising bullet loan. While many banks would have difficulty with such a loan, the institutional market was more flexible, and institutional investors were prepared to bargain an increase in price for the absence of typical features like covenants. Hence, relevant differences between the CAHPL facility and typical institutional loans could be dealt with by way of a pricing adjustment. Mr Martin began his pricing exercise by considering the "borrowing spreads" charged by institutional lenders. He then moved to consider more specific borrowing spreads, namely those in relation to institutional loans to oil & gas companies. Mr Martin made an adjustment to that spread to reflect the underwriting fees that would likely be charged, and he made adjustments for the lack of financial covenants and the lack of security. Further, he made an adjustment to reflect a "size premium": given the very large size of the facility it would have to be priced at a level that would appeal to the widest possible audience. This resulted in a range between 6.06% and 7.73%. As a cross check Mr Martin then considered borrowing spreads on High Yield Bonds in order to see how institutional investors would have priced unsecured credit in 2003. He found an implied high yield spread range of 4.34%-6.74%, and made an adjustment of 2.5%-3.00% for size and the lack of "yield enhancers".

491. Mr Gross also commenced with a credit analysis of CAHPL and settled on a risk rating of 7 (B equivalent). In this case, at a stretch the most that the parent company, CVX, would be taken to affect the risk rating of the subsidiary borrower, CAHPL, was one notch on the risk rating scale, taking it to between a "6" and "7" (B+ equivalent). Like Mr Martin, Mr Gross began with spreads on institutional term loans, which for B+ borrowers was 4.00%. He made an adjustment of 0.25% to take into account upfront fees and a discount for prepayment. A further adjustment of 0.75% was made for the lack of covenants. Mr Gross also conducted a cross check by reference to spreads on High Yield Bonds.


ATC 17890

492. Mr Rowland was the CFO of Chesapeake for 18 years, from 1992 to 2010. Chesapeake was a non-investment grade rated energy company, and for several years was the largest producer of natural gas in the United States. It raised billions of dollars of debt despite its non-investment credit rating, including by way of senior unsecured notes totalling USD2.1 billion in 2003. Those notes provided Chesapeake with flexibility that it would not have enjoyed if it had used other available forms of cheaper debt with more onerous covenants. From this one could conclude that a sub-investment grade rating did not affect an energy company's ability to carry on business and to raise debt in a growth phase. The applicant submitted that Mr Rowland displayed a nuanced understanding of a complex and uncertain industry from the perspective of an independent company and expressed himself in a clear manner. The evidence in the proceedings had demonstrated that upstream E&P activities were uniquely characterised by: (a) a high level of risk; (b) a long period of time before a return on investment is received; (c) a lack of any definitively predictable relationship between the magnitude of expenditures and the value of any resulting reserves; (d) a high level of government regulation, with the continuing prospect of major intervention including with respect to taxation and environmental matters; (e) unique cost-sharing agreements between ostensible competitors.

493. The applicant submitted that Mr Rowland was uniquely positioned to opine on the strategies for navigating those treacherous waters, particularly without the support of the capital of a supermajor. The Commissioner's attempt to characterise Chesapeake as a non-investment grade company with investment grade strength, was misplaced: in fact Chesapeake did have an issuer credit rating at the relevant time, and it was B+ (non-investment grade). Mr Rowland was exposed to an environment (unlike Mr Gaskell) where he, and those responsible for other companies in Chesapeake's position, would seldom know the source of the funding that would be provided for the entire length of a project. In managing a strategy for Chesapeake, Mr Rowland dealt with the reality that whilst he always hoped that capital markets would be available to the company, there were numerous cases where they were not and Chesapeake had to curtail its expenditure. His stress on the importance of flexibility and the boon of a secured, mostly undrawn "rainy day" fund to maintain the company's funding and ratings was self-evident in that context. Mr Gaskell accepted that expensive debt that avoided the constraints of covenants would be preferred if it was in the aggregate interest of shareholders.

494. Mr Wasow explained that a financing strategy for an oil and gas business was designed to facilitate its business strategy. While many independent oil and gas companies sought to maintain an investment grade rating, borrowers with non-investment grade ratings were able to access the high yield market. Companies would typically not borrow in a foreign currency to lower interest cost beyond their ability to hedge the liability by offsetting it against "natural hedges" such as foreign currency assets. The actual practice of Australian oil and gas companies in 2003 was only to borrow in USD to the extent that the liability could be hedged against USD assets. In that respect, CAHPL did not have any significant USD assets. Borrowing in USD would thus have been contrary to accepted practice and would have increased CAHPL's exposure to foreign exchange gains and losses.

495. The respondent Commissioner submitted that the Credit Facility Agreement was not one which would have been entered into between independent parties dealing with each other at arm's length. This was not a case where the Court was dealing with a commodity with an established market or an item such as listed shares which might have a value independent of the character of the parties to the relevant agreement. The applicant's case, the respondent submitted, was effectively that CAHPL was a considerably uncreditworthy borrower, one in the "speculative" category, and that this would lead any lender to demand a very high rate of interest, if indeed any lender was prepared to lend. The respondent Commissioner submitted that the correct approach was that all the facts and circumstances of the parties (other than their non-arm's length relationship) be imported into the hypothetical parties, thus avoiding speculation. The respondent Commissioner submitted [closing submissions at 49] that it would defy not only the language, but also the object and purpose of the transfer pricing legislation to permit only the price under a non-arm's length agreement to be altered despite the agreement's other non-arm's length terms. A multinational enterprise could insert various unattractive and unrealistic terms in its related party agreements, terms which are commercially meaningless to it because of the pre-existing relationship between the parties, and which serve only to increase the price for tax purposes.

496. The respondent submitted that if a company in the position of CAHPL were borrowing from an


ATC 17891

independent party, CVX or a company in the position of CVX would have provided a guarantee of the borrowing.

497. The respondent submitted that on the assumption that the borrower was "stand-alone", it would have provided covenants and security, and/or it would have borrowed less and obtained the balance of the USD2.45 (AUD3.7) billion by way of equity, in order to reduce the interest rate payable from an unsustainable AUD LIBOR +4.14%. It would certainly not have borrowed in AUD when the Deutsche Bank report had stated that the maximum leverage based on an AUD interest rate (Deutsche Bank assumed 9%) was USD2.1 billion, and Goldman Sachs had calculated the maximum debt of USD2.5 billion based on a USD interest rate of 5.38%.

498. The respondent Commissioner submitted that the interest rate which would be payable by the borrower if it borrowed at arm's length was:

  • (a) On the basis that the borrower was a member of a group like the Chevron group:
    • i. USD LIBOR plus the margin for a AA rated borrower, which on 2 June 2003 was 0.09%;
    • ii. Alternatively, USD LIBOR plus the margin for a borrower rated in the A range (conservatively using the A- margin), which on 2 June 2003 was 0.73%;
    • iii. In the second alternative, AUD LIBOR plus the margin for an AA rated borrower.
  • (b) If, which the Commissioner disputed, the borrower was deemed to be a stand-alone company:
    • i. USD LIBOR plus the margin for a BBB rated borrower (on the basis that CAHPL would have borrowed USD1.7 billion in order to obtain an investment grade credit rating), which was 0.95%;
    • ii. Alternatively, USD LIBOR plus 1.75%, being the upper end of the range of interest margins represented by the comparable uncontrolled transactions identified by Mr Hollas;
    • iii. In the second alternative, USD LIBOR +4.01%, which on 15 January 2003 equalled 5.38% and was the basis on which the Goldman Sachs analysis, which the CAHPL Board relied upon, was undertaken.

499. In my opinion, the approach in the present circumstances should go past shorthand expressions such as "reconstruction of the transaction" to address an agreement between two parties independent of each other, neither party being an actual party to the actual loan. I would add that, although the construct is hypothetical, this does not mean that the exercise should depart from reality more than is necessary for the hypothesis. In my view, the exercise, although hypothetical, should remain close to undertaking the actual loan. Thus I would give little weight to factors which a lender and a


ATC 17892

borrower, at arm's length to each other, in the circumstances would not take into account. I do not accept the Commissioner's submission that ss 136AD and 136AA ask what form an agreement might have taken if it had been negotiated by entities dealing with each other at arm's length. In my opinion, this construction does not centre on the identification of the property, as I understood how it was put, but on the use of the indefinite article in the expression "if the property had been acquired under an agreement between independent parties ..." In my opinion, "an agreement" does not, in context, mean that the hypothetical agreement is to have the property acquired as its only matter coincident with the actual agreement.

500. Also, in relation to the s 136AD issue, I give no weight to the opinions of transfer pricing economists where those opinions appear not to be founded in the statutory language which the Court must apply.

501. It is easy to see, at one end, that the loan is one made at a certain time, in a certain amount, for a certain period, at a certain interest rate and with or without a certain security. At the other end, it is more difficult to apply the hypothesis required by s 136AD to a hypothetical borrower in place of CAHPL. Here, it seems to me, the statutory hypothesis must include what has been shown on the evidence to be relevant in the market in question. For example, if the evidence showed that a lender would take into account in pricing the loan that a borrower independent from the lender was in a particular industry and that the creditworthiness of a borrower in that industry was affected by particular matters going to its capacity to repay the loan and the likelihood that it would do so, then those factors would be relevant. It must therefore be a factor that the borrower was in the oil and gas E&P industry.

502. Another related question is whether the statutory hypothesis permits or requires to be taken into account that a borrower, the hypothetical borrower not being the taxpayer, has at the time of the loan certain financial resources which the lender would regard as relevant to the pricing of the loan. In principle, the answer must be "yes". In the present case, does the fact that the non-arm's length nature of the Credit Facility Agreement stems from the common ownership of the borrower and the lender by CVX point to a different answer? Of itself, in my view the answer is "no". But that does not have the consequence that the particular relationship between CVX and CAHPL is determinative in the context of the statutory hypothesis. That would be to import the actual entity, CAHPL, into the hypothesis contrary to the decision of the Full Court in
SNF 193 FCR 149.

503. Applying these general considerations to the facts of this case means the following. In my opinion, the correct perspective is that of a commercial lender. A commercial lender would not approach the question of the borrower's creditworthiness in the same way as would a credit rating agency.

504. CAHPL put its positive case primarily by reference to the expert reports of


ATC 17893

Mr Gross and Mr Martin.

505. As to Mr Gross, it will be recalled that the question he was asked was whether the interest rate in the Credit Facility Agreement exceeded the consideration that might reasonably have been expected to have been given by CAHPL to an independent third party for the provision of the Credit Facility. His conclusion was that the correct pricing of the Credit Facility provided by CFC to CAHPL was 500 basis points over 1 month AUD LIBOR BBA. Therefore the interest rate in the Credit Facility Agreement, in his opinion, did not exceed the consideration that might reasonably have been expected to have been given by CAHPL to an independent third party for the provision of the Credit Facility.

506. In my opinion, there are a number of difficulties with Mr Gross' reasoning and, therefore, his conclusion. I mention first that the appropriate question does not necessarily involve CAHPL giving consideration to an independent third party for the provision of the Credit Facility, but two parties independent of each other. It was also, in my opinion, incorrect to frame the question for Mr Gross' opinion as whether the interest rate specified in the Credit Facility Agreement exceeded the consideration that might reasonably have been expected to been given by CAHPL to an independent third party for the provision of the Credit Facility. It would have been preferable, with respect, to have asked a question less closely aligned with the ultimate issue, the question taking the form of identifying the property acquired and the non-consideration terms and asking what consideration might reasonably be expected to have been given under an agreement between independent parties. The question asked led to Mr Gross calculating CAHPL's metrics on the basis of the actual margin in the credit agreement. I accept, however, that the answers given by Mr Gross to the question he was asked may be relevant to deciding whether or not the amended assessments under the ITAA 1936 were excessive.

507. I take into account that Mr Gross' conclusion was based on his prior opinion that, although he did not use rating agency ratings, the risk rating he gave to CAHPL was approximately equal to a B rating by S&P. This led him to consider loan spreads of B+/B loans and new issuances in the bond market for B to B+. Mr Gross said he looked for any E&P loans issued between January 2003 and May 2003 which were institutional loans and rated B or B+ and could not identify more than one or two over the whole period. However, the case put by the applicant was that the relevant rating was BB or BB+. Mr Gross' did not give an opinion as to the appropriate margin if CAHPL were not rated B or approximately B.

508. Further, I would be reluctant to place weight on an analysis, such as that of Mr Gross, where he accepted that there could not have been a single lender under an agreement between independent parties and no lender would lend absent financial covenants. Mr Gross wrote:

In fact, such a lender would not make a five-year bullet loan without restrictive financial covenants. It would have been impossible to either syndicate a loan or use a high-grade or high yield bond without those covenants.


ATC 17894

He also said that there were no loans or bonds in the market with such a lack of financial covenants.

509. Mr Gross also agreed that an absence of restrictive or negative covenants meant that CAHPL, and I would infer an independent borrower, could not have entered into the Credit Loan Facility.

510. I also find unpersuasive the interest rate analysis made by Mr Gross which appeared to begin, under the heading "An Acceptable Rate of Return", with the proposition that the Bank of America's expectation was a minimum fee of 300 basis points to earn the required hurdle rate. He said the minimum fee of 300 basis points would have to be supplemented by the other factors such as term, market pricing and conditions of the loan. In light of the cross-examination of Mr Gross by reference to actual borrowings from the Bank of America, I do not accept the 300 basis points minimum fee. I am also unpersuaded by Mr Gross' evidence as to the balance of these matters. I accept the respondent's submissions in this respect. The add-on by reference to the five-year term of the loan, was evidence which was not supported otherwise. Under the heading "Market Pricing for Loans with the Same Rating and Term", Table 4 related to all industries, did not deal with the average spreads of loans to particular industries such as the E&P sector and did not indicate the low point or the high point of the yields. Tables 5 and 6 were directed to B+/B loans. Under the heading "Terms and Conditions of the Loan", Mr Gross dealt with the fact that the Credit Facility was unsecured and contained no restrictive covenants and concluded that the terms and conditions of the Credit Facility had the characteristics of a high yield bond and should therefore be priced like a bond. Yet Mr Gross accepted that Term Loan Bs to non-investment grade borrowers had to be secured and that they always contained financial covenants and restrictive covenants. I consider the Term Loan B issue further below.

511. As to Mr Martin, the question on which his opinion was sought was very similar. His opinion was based on the weak BB rating he gave to the borrower CAHPL and his view of the markets in 2003. In his opinion, the interest rate on the Credit Facility did not exceed what would reasonably be expected to be obtained in an arm's length transaction at the time and he estimated the interest rate would have likely been much higher than the rate charged by CFC.

512. Mr Martin rated CAHPL as a weak BB borrower. Having discounted the prospect that a bank would participate in such a loan, Mr Martin focused on a Term Loan B form of transaction.

513. I do not accept the applicant's submission that a Term Loan B was an appropriate comparator for the hypothetical where, most significantly, the borrower would be an E&P company with a BB risk rating equivalent; there were no financial covenants for the borrowing; there were no restrictive covenants for the borrowing; and the borrower could repay the loan at any time such that the lender would have the risks but not the benefit for potentially a short period. It follows, in my opinion, that Mr Martin's evidence that the Credit Facility could be priced as a Term Loan B lacks a realistic foundation.

514. In addition, I find that Mr Martin did not sufficiently take into account the credit metrics specific to an E&P company, those credit metrics being relevant to the assessment of risk. The only such metric he did take into account was debt to proved reserves. A substantial reason for this was that he did not profess to be a specific E&P lender. I have set out at [177] above the relevant parts of the cross-examination of Mr Martin.

515. Further, I find Mr Martin's opinion as to the interest rate chargeable on a Term Loan B rested predominantly on Table 6 in his first report which was an analysis of BB new issue spreads versus CAHPL borrowing spread (institutional basis) and concerned the whole market. I accept the respondent's submissions and find as follows. Table 6 had an unreliable starting point against the item "June 2003 Ave BB New Issue Spread" said to be sourced in S&P Capital IQ. The average in June 2003 was not in fact 300 basis points but 278.57 basis points. At one point Mr Martin said that the figure was for the quarter ended June 2003. Mr Martin later said that what he was doing in Table 6 was giving an average for the first half of 2003. He said the difference between 278.57 and 300 basis points was not material, a bare assertion which is difficult to accept. He also said there was trading data for June 2003 which would have indicated the average spread was around LIBOR plus 500 on a trading basis. No foundation for that statement was evident or was provided. More significantly, Mr Martin's Table 6 had a methodological flaw in that he did not know and could not say the extent to which the basis points he added on, in total 258 to 425, were already included in the beginning point of 300. For example, Mr Martin added a premium of between 50 to 100 basis points "required for size of credit facility" although the average already included all large loans which did not reflect a size premium, and he added between 33 and 75 basis points for underwriting fees spread over an assumed average life of three years despite the all-in spread being one which included in it underwriting fees spread over an assumed average life of three years. I find that Table 6 provides an unreliable basis on which to assess the arm's length interest rate for the Credit Facility.

516. Mr Martin relied to a lesser extent on Table 7 in his first report for his opinion as to the interest rate chargeable on a Term Loan B and which concerned the oil and gas market. I accept the respondent's submissions and find as follows. Table 7 was entitled "Analysis of Oil & Gas New Issue Spreads vs. CAHPL Borrowing Spread". It showed a credit spread range of 606 to 773 basis points. It recorded a starting figure of 348 basis points


ATC 17895

which was originally designated "June 2003 Ave. New issue Spread" with the comment "Observed BB Spreads in US Institutional Loan Market June 2003 - 30 transactions". These statements required correction in Mr Martin's oral evidence in chief so as to refer to the full year 2003 rather than June 2003 and to 33 transactions rather than to 30. A further important change had to be made because the documents on which Mr Martin relied did not relate to BB borrowers. The letters "BB" needed to be changed to read "oil and gas". One effect of this change is that Table 7 related to the entirety of the non-investment grade rather than to BB specifically. Table 7 did not therefore assist in answering the question of at what rate a BB or higher rated borrower would borrow. As to the number of observations, whether 33 or 30, Mr Martin accepted that he did not know how the document from which he had taken the observations was prepared. I find also that the figure of 348 basis points used by Mr Martin as the starting point was not persuasive as it was not made clear what the range was from which the figure of 348 was derived. Mr Martin accepted that one end of the range could have been as low as 1.5%. Mr Martin did not know whether the spreads from which he had derived Table 7 were all-in spreads or straight spreads. Mr Martin added between 33 and 75 basis points for underwriting fees although he accepted that the starting figure of 348 basis points already included a weighted average of all the loans of 2003. I find that Table 7 provides an unreliable basis on which to assess the arm's length interest rate for the Credit Facility.

517. Mr Martin also, in his Table 9, sought to find support for his opinion in giving an "Overview of US High Yield Market - June 2003" and showing effective yield and implied yield spread. However, the percentages were the yield over the US Treasury rate and thus provided no valid basis for comparison: as submitted by the respondent Commissioner, it did not appear to me that it was valid to compare directly a yield over US Treasuries for bonds with a margin over LIBOR for institutional loans.

518. The evidence of Mr Long did not go directly to the question of the arm's length consideration but to the question of whether notching up due to implied parental support was appropriate. I refer to my finding on this issue at [606] below.

519. The evidence of Mr Rowland, formerly of Chesapeake, did not establish that Chesapeake was in a similar position to CAHPL or to a hypothetical company in CAHPL's position in 2003. In any event, his evidence went to how Chesapeake made its funding arrangements. His evidence did not provide an alternative basis on which to reach an arm's length consideration.

520. I refer also to the evidence of Dr Becker, a transfer pricing economist. He was not asked to provide an opinion as to the arm's length price of the Credit Facility Agreement. His evidence was primarily in response to the evidence of Mr Hollas and of Dr Horst and was directed to loan agreements which Dr Becker said were closer matching to CAHPL than those of Mr Hollas and in that sense essentially negative. He did, however, seek to identify from DealScan agreements for 70 companies in the energy sector, not limited to an E&P specific search, five loans made to non-investment-grade companies which he said were better or closer matching as a point of comparison with the Credit Facility Agreement. Three of the five loans had premiums over LIBOR substantially higher than the Credit Facility Agreement and the other two, referable to Planes Resources Inc, had lower amounts. As will appear, I accept the respondent's submissions in this respect and I do not find Dr Becker's evidence to be useful in addressing the statutory questions.

521. The first loan related to Abraxas Petroleum Corp but this was a CC rated company, that is, one highly vulnerable to non-payment. Plainly this rating would affect the interest margin. Further, the loan agreement was in relation to a USD4.2 million loan, there was a maximum revolving amount of USD50 million and the borrower could not repay the term loan until the revolving loan was repaid. The loan did not have a LIBOR option but was a base rate (US Prime) loan with a base rate margin of 4.5%. In my opinion, Abraxas as a company was not comparable to CAHPL and the loan was not comparable to the Credit Facility Agreement. Further, I do not accept the validity of Dr Becker's method of calculating the spread at which he arrived: the US prime rate comprised not only the cost of funds, but other expenses of the bank and therefore would be higher than a LIBOR rate which only comprises the cost of funds to which the bank then adds a spread to cover operating costs, credit risk and profit.


ATC 17896

522. The second loan facility was in respect of KCS Energy Inc. This was a USD40 million term loan facility, there was a rating for subordinated debt of C, and the company's financial position provided no realistic point of comparison to CAHPL: it appeared that KCS Energy was financially distressed. In my opinion, KCS energy as a company was not comparable to CAHPL and the facility was not comparable to the Credit Facility Agreement. Again, I do not accept Dr Becker's method of calculating the spread at which he arrived.

523. The third loan facility was in respect of Mission Resources Corp. This was an USD80 million facility with a senior debt rating of CCC+. Mission Resources was in financial difficulties and was not comparable to CAHPL.

524. The fourth and fifth loan facilities were in respect of Plains Resources Inc. One had an agreement date of December 2002 and was for a loan amount of USD45 million and the second had an agreement date of June 2003 and was for a loan amount of USD60 million. Both these loans were substantially smaller than the CAHPL/CFC loan. Plains Resources had proved reserves and revenues and income very much smaller than CAHPL. The earlier loan agreement was a margin loan for a specific purpose. As to the later loan agreement, Plains Resources was not rated.

525. In my opinion, therefore, the applicant has not shown that the consideration in the Credit Facility Agreement was the arm's length consideration or less than the arm's length consideration nor proved that the amended assessments under Div 13 of the ITAA 1936 were excessive. Division 13 does not permit reasoning that reaches a non-arm's length interest rate on the basis that the actual interest rate is as high as it is because of the rating attributed to the borrower or borrowing, which rating relies on the absence of arm's length consideration given by the borrower. This reasoning is an addition to, and independent of, my conclusion at [87] above. I also reject the applicant's submission, set out at [29] above, that the determinations made under Div 13 of the ITAA 1936 ceased to be operative once the 2012 amended assessments were made under the ITAA 1997. There is nothing in the statutes or in the facts which suggests or supports that contention.

Division 815 of the ITAA 1997

526. In light of my conclusion as to the 2010 amended assessments under Div 13 of the ITAA 1936, the following consideration of Div 815 of the ITAA 1997 proceeds in the alternative.


ATC 17897

527. It will be recalled that the determinations made by the Commissioner under s 815-30 of the ITAA 1997 were in respect of the tax years 2006, 2007 and 2008.

Is Subdivision 815-A constitutionally invalid?

528. The applicant's submissions that Subdiv 815-A of the ITAA 1997 was constitutionally invalid were set out in Appendix A of its written submissions and were not developed orally. The respondent's submissions on this issue were set out in his written submissions and were also developed orally.

529. The applicant's submissions were that Subdiv 815-A was retroactive and, as applied by s 815-1 of the Income Tax (Transitional Provisions) Act 1997 to the income years commencing on or after 1 July 2004, ss 815-10 to 815-30 of the ITAA 1997 were invalid because they imposed an arbitrary exaction and therefore did not answer the description of a law with respect to taxation for the purposes of s 51(ii) of the Constitution. The applicant's primary submission was that the arbitrariness of the exaction imposed by the retroactive operation of Subdiv 815-A flowed from the absence of ascertainable criteria with sufficiently general application as to whether an entity had received a transfer pricing benefit. Alternatively, the applicant submitted, an unduly retrospective exaction could, in the circumstances of its imposition, be arbitrary in character and thus beyond legislative power. Informing its arbitrariness, was the inability for a taxpayer to comply with its criteria because they remained unknown during the course of ordinary commercial discourse: they could not be pointed to at the time when the events, which subsequently gave rise to purported liability, were entered into. Nor could they be identified when a tax return was prepared by a taxpayer. Subdivision 815-A was unduly retrospective and thus arbitrary.

530. The applicant submitted that the combination of the features of ss 815-10 to 815-30 together with their retroactive operation by virtue of s 815-1 of the Income Tax (Transitional Provisions) Act, bringing past transactions or conduct over a period of some eight years within the scope of Subdiv 815-A, resulted in invalidity. The retroactive operation of Subdiv 815-A was confined to a finite group of taxpayers. The legislature had confined the retroactive operation of the Subdivision to the class of taxpayers who obtained a transfer pricing benefit at a time when an international tax agreement containing an associated enterprises article or a business profits article applied to them. The arbitrariness of the exaction imposed by the retroactive operation of Subdiv 815-A flowed from the absence of ascertainable criteria with sufficiently general application as to whether an entity had received a transfer pricing benefit.

531. The following combination of features, the applicant submitted, indicated that, in their retroactive operation, ss 815-10 to 815-30 did not impose liability to tax by reference to ascertainable criteria of sufficiently general operation:

  • (a) the provisions applied to taxpayers over a period during which the criteria for liability were neither specified nor ascertainable, in view of both the terms of the provisions and the reasoning in decisions of the Federal Court to the effect that relevant double taxation treaties did not by themselves confer a power of taxation;
  • (b) the provisions contemplated the assessment of tax by reference to a transaction that was never entered into and did not adopt the same structure as the transaction that was in fact entered into; and
  • (c) the provisions facilitated the imposition of liability on members of a limited class of taxpayers, who did not incur transfer pricing liability under Div 13 of the ITAA 1936.

532. The applicant did not dispute the power of the Parliament to pass retrospective legislation in conformity with the plenary grant of the other powers conferred by s 51 of the Constitution but submitted that here the nature of the power conferred by s 51(ii) was limited and the nature of one of those limitations itself raised fetters on the ability to pass legislation which was unduly retroactive. The applicant did not independently invoke s 51(xxxi) of the Constitution, concerning the acquisition of property on just terms, but submitted that the presence of s 51(xxxi) may nonetheless be relevant to the determination of the scope of the taxing power. "Unduly retroactive" in that context was not solely a feature of time but encompassed any situation where the amendment went beyond what could fairly be seen as protective of the revenue to effectively cross the line into an unjust acquisition. Merely calling such legislation a "tax" was not enough.

533. The applicant submitted that a retrospective tax may be "arbitrary" by reason of the specific nature of its application. The submission was in part inspired by case law in the United States, including
Brushaber v Union Pacific Railroad Company 240 US 1 (1916) and
United States v Carlton 512 US 26 (1994).

534. The applicant also submitted that even if Subdiv 815-A was a valid law of the Commonwealth, any tax assessed in accordance with its provisions was not imposed during the years of income. This was because Parliament did not amend the Taxing Act to impose the tax so assessed with retrospective effect. Section 4 of the Taxing Act provided that "the Assessment Act is incorporated, and shall be read as one, with this Act". This method of incorporation did not refer to the Assessment Act as it existed from time to time. As such, and at common law, only the ITAA 1936 as it existed in 1986 was incorporated into the Taxing Act. Section 10 of the Acts Interpretation Act 1901 (Cth) amended the common law proposition so that the reference to the "other Act" was to that Act as originally enacted and as amended from time to time. If the ITAA 1997 was incorporated into the Taxing Act on an ambulatory basis, it was necessary to consider at what time Subdiv 815-A became incorporated by reference into the Taxing Act so that tax could be assessed in accordance with the terms of Subdiv 815-A and then imposed by the Taxing Act on a taxable income so assessed.

535. In amplification of these submissions, the applicant submitted the Tax Laws Amendment (Cross-Border Transfer Pricing) Act (No. 1) was assented to on 8


ATC 17898

September 2012 and accordingly it came into operation from that day. If Subdiv 815-A was not part of the ITAA 1997 prior to 8 September 2012, then, the applicant submitted, it could not be incorporated into the Taxing Act prior to that date unless the Taxing Act was itself specifically amended to impose the tax with retrospective application. Accordingly, no mechanism existed for imposing tax on a taxable income which included amounts assessed pursuant to the provisions of Subdiv 815-A prior to the year of income which included 8 September 2012 (the 2013 income year). Consistently with s 7(2) of the Acts Interpretation Act, the amendment of the ITAA 1997 with effect from 8 September 2012 did not affect the previous operation of the Taxing Act read with the ITAA 1997, or any right, privilege, obligation or liability acquired, accrued or incurred under it. The provision purporting to give Subdiv 815-A a retrospective application, s 815-1 of the Income Tax (Transitional Provisions) Act, did not go so far as to retrospectively incorporate Subdiv 815-A into the Taxing Act but said only that Subdiv 815-A of the ITAA 1997 applied to income years starting on or after 1 July 2004. It was submitted that a further provision was necessary to incorporate an Assessment Act containing Subdiv 815-A into the Taxing Act with retrospective effect. In the years in dispute, the Taxing Act imposed tax by reference to the ITAA 1936 and the ITAA 1997 as then enacted. No provision had since been enacted by Parliament to undo that fact. In the absence of such a provision, Subdiv 815-A was only incorporated by reference into the Taxing Act prospectively from the commencement of the Income Tax (Transitional Provisions) Act on 8 September 2012, and income tax could not be imposed on taxable income assessed in accordance with the terms of Subdiv 815-A prior to that time.

536. The respondent submitted that the applicant's point was more accurately described as a challenge to s 815-1 of the Income Tax (Transitional Provisions) Act, as inserted by the Tax Laws Amendment (Cross-Border Transfer Pricing) Act (No. 1) which made Subdiv 815-A applicable to income tax years starting after 1 July 2004.

537. The respondent did not submit that the enactment of Subdiv 815-A did not change the law, including the law applicable to the assessment of tax for past years. However, the respondent submitted, it was inaccurate to describe Subdiv 815-A, as given effect by s 815-1, as "retroactive". A law was not retrospective, the respondent submitted, merely because it attached new consequences to past events. Subdivision 815-A did not have the effect of deeming the law at some time in the past to be different from what it in fact was and it did not affect vested or accrued rights that had arisen from past events.

538. As to the applicant's primary constitutional argument as to "arbitrary" exaction, the respondent submitted that references in the case law should not be understood as inviting consideration of the wisdom or fairness of the criteria by which liability to tax was imposed. The requirement that a tax not be "arbitrary" was merely a requirement that it be imposed "by reference to ascertainable criteria with a sufficiently general application". The three features of Subdiv 815-A identified by the applicant did not, individually or cumulatively, lead to the conclusion that the tax was an "arbitrary" exaction in the relevant sense. The validity of the legislation did not depend on whether the Executive accurately, comprehensively and publicly predicted its content before it was enacted. Further, there was no general barrier to retrospective tax laws:
Mutual Pools & Staff Pty Ltd v Commonwealth [1994] HCA 9; 179 CLR 155 at 209 per McHugh J. The respondent submitted the real focus of the applicant's complaint appeared to be that the decision as to whether a transfer pricing benefit had been obtained pursuant to s 815-15 depended on "nebulous criteria" dependent on what was said to be "vague" OECD Guidelines. Although the enquiry may be complex and the correct answer debatable, that did not deprive the applicable criteria of objectivity or generality, at least so long as the Commissioner's assessment was amenable to review and appeal. The respondent submitted that the provision for recourse to the OECD Guidelines did not result in arbitrariness in any relevant sense because:

  • (a) the operative provisions were ss 815-10(1) and 815-30(1). Section 815-15, which was definitional in nature, identified circumstances in which "an entity gets a transfer pricing benefit".
  • (b) within that statutory regime, s 815-20(1) required reference to be made, for two purposes, to a set of documents which included the OECD Guidelines. The first requirement was that whether an entity gets a transfer pricing benefit was to be worked out consistently with those documents to the extent that they were relevant. The second was that a provision of an international tax agreement was to be interpreted consistently with those documents to the extent that they were relevant.
  • (c) the words "to the extent that they are relevant" served to confirm that the Guidelines were adopted only for the purpose of giving content to the broad language of the statutory definition in its application to particular cases where they could properly do so: where adherence to the OECD Guidelines would be inconsistent with applying s 815-15 according to its terms, the OECD Guidelines had to give way to the section. The Commissioner's application of the statutory definition was subject to appeal to the Court once it was reflected in an assessment.
  • (d) where an Australian statute was expressed to depend for its operation on the effect of an international convention, that convention would normally be interpreted in accordance with international law principles of treaty interpretation, but there was no impediment to the statute modifying that interpretation or specifying some other interpretive rule or principle for the purpose of identifying the effect of the relevant provisions as incorporated into domestic law.
  • (e) the drafting technique used in s 815-20 therefore raised no issue as to the validity of Subdiv 815-A. If it did, that provision would clearly be severable.

539. As to the point made by the applicant that Subdiv 815-A in its application to tax years prior to 2012 applied to members of a limited class, the respondent submitted that this was no more than an inevitable consequence of providing for tax to be assessed on the basis of events that had already occurred. The Subdivision was not arbitrary in the sense of imposing criteria that were insufficiently "objective" or "general".

540. Counsel for the respondent referred to


ATC 17899


MacCormick v Commissioner of Taxation (Cth) [1984] HCA 20; 158 CLR 622 at 640;
Deputy Commissioner of Taxation (Cth) v Truhold Benefit Pty Ltd [1985] HCA 36; 158 CLR 678 at 684,
Roy Morgan Research Pty Ltd v Commissioner of Taxation (Cth) [2011] HCA 35; 244 CLR 97, 111 [38]-[39] and, in oral submissions, to
WR Carpenter Holdings 237 CLR 198 at [9] as follows (omitting citations):

First, for an impost to satisfy the description of taxation in s 51(ii) of the Constitution it must be possible to distinguish it from an arbitrary exaction. Secondly, it must be possible to point to the criteria by which the parliament imposes liability to pay the tax; but this does not deny that the incidence of a tax may be made dependent upon the formation of an opinion by the Commissioner. Thirdly, the application of the criteria of liability must not involve the imposition of liability in an arbitrary or capricious manner; that is to say, the law must not purport to deny to the taxpayer "all right to resist an assessment by proving in the courts that the criteria of liability were not satisfied in his case".

541. As to the applicant's alternative argument based on United States decisions, the respondent submitted that those decisions applied the Fifth Amendment, which had no analogue in the Australian Constitution. No general guarantee of "due process" constrained Commonwealth legislative power; nor had Australian courts adopted "harshness" or "oppressiveness" as tests of validity.

542. In my opinion, the challenge to the constitutional validity of the provisions in Subdiv 815-A fails. As will appear, I accept the respondent's submissions in this respect.

543. In
Commonwealth of Australia v SCI Operations Pty Ltd [1998] HCA 20; 192 CLR 285, at [57]-[58] McHugh and Gummow JJ explained, with reference to what Jordan CJ had said in
Coleman v Shell Co of Australia (1943) 45 SR (NSW) 27 at 30, that "there has been some ambiguity in the use of the word 'retrospective' ". There was a distinction between a statute which provided that as at a past date the law shall be taken to have been that which it was not, and the creation by statute of further particular rights or liabilities with respect to past matters or transactions. Their Honours held that s 269N of the Customs Act 1901 (Cth), the provision there under consideration, did not render a concession order retrospective in the sense that it provided that, as at a past date, being 1 September 1987, and thereafter, the law was to be taken to have been that which it was not. Rather, the result was that, on 3 June 1994, there were brought into existence fresh rights or liabilities in respect of matters or transactions which had occurred on or after 1 September 1987.

544. Further, in my opinion, the form of the legislation and the reasoning in


ATC 17900

MacCormick and Truhold Benefit upholding the validity of the legislation there under consideration is inconsistent with the applicant's submissions.

545. It is to be recalled that in MacCormick, liability to pay recoupment tax arose only where company tax had not been paid, and the amount of the recoupment tax was quantified by reference to the overdue company tax: that liability was separate and distinct from the liability of the target company to pay the company tax: see the judgment of the plurality at 635. Further, at 638, the plurality said that the recoupment tax was undeniably a new pecuniary obligation imposed upon vendors and promoters notwithstanding that the obligation arose only upon the failure of a target company to pay company tax. At 639, the plurality held the exactions were not arbitrary as liability was imposed by reference to criteria which were sufficiently general in their application and which marked out the objects and subject-matter of the tax.

546. In Truhold Benefit, the plurality, with whom Brennan J agreed, at 684 explained MacCormick as follows (citation omitted):

In


ATC 17901

M v. Federal Commissioner of Taxation it was held that the recoupment tax, for which the Act provides, answers the usual description of a tax. Amongst the characteristics which were said by the majority to bring it within that description was the fact that the tax is not arbitrary. This was, as the relevant passage shows, a reference to the fact that liability can only be imposed by reference to ascertainable criteria with a sufficiently general application and that the tax cannot lawfully be imposed as a result of some administrative decision based upon individual preference unrelated to any test laid down by the legislation. To say that a tax may not be arbitrary in that sense does not, of course, preclude the pejorative description of a tax as arbitrary in the sense that the criteria which are laid down for its application give it a harsh or unreasonable incidence with regard to either its subject-matter or objects. To describe a tax as arbitrary in the latter sense is to do so in a manner which does not go to its validity.

547. It follows, in my view, that the United States cases on which it relied cannot assist the applicant.

548. A further reason why those cases cannot assist the applicant is, as submitted by the respondent, that the applicant seeks to invoke the due process provision of the Fifth Amendment but there is no equivalent to that provision in the Australian Constitution. There is thus no occasion to apply to enactments of the Commonwealth Parliament, the standard applied by the Supreme Court in Carlton that "a tax statute's retroactive application must be supported by a legitimate legislative purpose furthered by rational means". In Australia, the mere effect of a taxing statute to impose a tax by reference to past transactions does not bespeak invalidity.

549. As to the contention that the legislation was unconstitutional by virtue of uncertainty, I see no difference in substance between the impugned provisions and those the subject of s 15AB of the Acts Interpretation Act 1901 (Cth). Section 15AB provides that in the interpretation of a provision of an Act, if any material not forming part of the Act is capable of assisting in the ascertainment of the meaning of the provision, consideration may be given to that material to confirm that the meaning of the provision is the ordinary meaning conveyed by the text of the provision taking into account its context in the Act and the purpose or object underlying the Act; or to determine the meaning of the provision when: (i) the provision is ambiguous or obscure; or (ii) the ordinary meaning conveyed by the text of the provision taking into account its context in the Act and the purpose or object underlying the Act leads to a result that is manifestly absurd or is unreasonable.

550. I also note that in
Momcilovic v The Queen [2011] HCA 34; 245 CLR 1 it was held that the task imposed by s 32(1) of the Charter of Human Rights and Responsibilities Act 2006 (Vic), which provided: "So far as it is possible to do so consistently with their purpose, all statutory provisions must be interpreted in a way that is compatible with human rights", was held to be not outside the scope of ordinary principles of statutory interpretation and not to confer a legislative function upon courts.

551. Further, in my opinion, difficulties of construction are not to be regarded as synonymous with legal uncertainty which only arises after the process of statutory construction has concluded: see by analogy the cases, of which
King Gee Clothing Co Pty Ltd v Commonwealth [1945] HCA 23; 71 CLR 184 is an example, dealing with alleged uncertainty of subordinate price-fixing legislation.

552. As to the applicant's argument that the tax was not imposed by reference to Subdiv 815-A during the relevant tax years, the respondent submitted that the argument proceeded from a misapprehension of the relationship between taxing acts and assessment acts. The amount payable by force of s 4-10 of the ITAA 1997 for each tax year was assessed by the Commissioner, subject to appeal or review, under relevant provisions of the ITAA 1936 which defined "this Act" to include, amongst others, the ITAA 1997: see s 6(1) of the ITAA 1936. The insertion of a new taxing provision into the ITAA 1997, expressed to apply in respect of a particular tax year, changed the amount of income tax payable under s 4-10 for that year and income tax in that amount was imposed by the Income Tax Act 1986 (Cth). That Act operated in an ambulatory fashion but it did not impose tax for a particular year only during the course of that year and in accordance with the Assessment Acts as they stood during that year. There was no need, in the case of a new taxing provision applicable to past tax years, for an additional provision retrospectively incorporating the Assessment Act as amended into the Income Tax Act. Any additional liability created by the insertion of a new taxing provision became "due and payable" in accordance with the former s 204 of the ITAA 1936 and s 5-5(7) of the ITAA 1997.

553.


ATC 17902

In my opinion, the contentions on behalf of the applicant in this respect misconceived the nature of the amendments made by s 815-1 of the Income Tax (Transitional Provisions) Act. The provision had the effect, according to its terms, that Subdiv 815-A of the ITAA 1997 applied to income years starting on or after 1 July 2004. I accept the respondent's submission that the Income Tax Act did not impose tax only for the particular year in which it was enacted and did not impose tax limited to the form of the Assessment Act as it stood at the time the Income Tax Act was enacted. Section 7 of the Income Tax Act provided that the tax imposed by s 5(1) "is levied, and shall be paid, for the financial year commencing on 1 July 1986 and for all subsequent financial years until the Parliament otherwise provides".

Preconditions to the making of the Subdiv 815-A 2012 determinations

554. The applicant submitted that one of the statutory preconditions for the making of the Subdiv 815-A determinations which was not satisfied was the requirement in s 815-10(2) that an international tax agreement containing an associated enterprises article applies to the entity. The applicant submitted that the requirement was not satisfied because the Treaty in the present case did not contain an "associated enterprises" article.

555. Under s 815-15(5) an "associated enterprises article" is defined to mean either Art 9 of the United Kingdom convention or a corresponding provision of another international tax agreement.

556. The applicant also submitted that the Subdiv 815-A 2012 determinations were invalid as, in making them, no proper attempt was made to determine whether CAHPL had obtained a "transfer pricing benefit" or to calculate the amount of that benefit. The Court was entitled to, and should, infer that the Commissioner issued the Subdiv 815-A determinations by simply nominating the same amount that had been specified in the corresponding Div 13 determinations and, accordingly, no real attempt was made to ascertain the proper taxable income of CAHPL. The applicant referred to
Avon Downs Pty Ltd v Commissioner of Taxation (Cth) [1949] HCA 26; 78 CLR 353 at 360 and to
R v Commissioner of Taxation (WA); Ex parte Briggs (1986) 12 FCR 301 at 308. The applicant also submitted that this aspect of the making of the Subdiv 815-A determinations was not shielded by s 177(1) of the ITAA 1936 as part of the "due making" of the Subdiv 815-A 2012 amended assessments. This was because the discretionary power in s 815-10(1) was not enlivened, the applicant submitted, unless the Commissioner had first identified a transfer pricing benefit.

557. The respondent submitted that the "supporting document" did not demonstrate anything other than a genuine consideration of whether the applicant had obtained a "transfer pricing benefit". Given the close parallels between s 815-15(1)(c) and (d)(i) and the closing words of s 136AD(3), no breach of any requirement could be inferred from the fact that consideration of the same transactions under Subdiv 815-A led to an identical adjustment to that which had been made pursuant to Div13. Further, the respondent submitted that the power to make a determination under ss 815-10(1) and 815-30(1) was not conditioned on the Commissioner having any particular state of satisfaction, and the obtaining of such a benefit and its quantum were objective matters which may be contested in proceedings under Part IVC. Thus the taxpayer bore the onus of proving that in fact no such benefit was received or that the actual benefit was less than the figure upon which the relevant assessment proceeded. The respondent also submitted that the authorities established that the making of a determination such as that provided for in Subdiv 815-A was an aspect of the "due making" of the assessment and errors affecting the determination did not have any relevance in proceedings relating to the validity or correctness of the assessment.

558. In my opinion, the challenge to the validity of the Subdiv 815-A determinations on the ground that no real attempt was made to ascertain the proper taxable income of the applicant fails at the evidentiary level. I would not draw from the "supporting document" the inference for which the applicant contends.

559. The reference to Avon Downs does not, in my opinion, in the circumstances of this case, assist the applicant. I assume that the applicant's reference was to the dictum of Dixon J at 360 that the conclusion that the Commissioner had reached may, on a full consideration of the material that was before him, be found to be capable of explanation only on the ground of a [legal] misconception, so that it could be seen that in some way the Commissioner must have failed in the discharge of his exact function according to law. But, in my opinion, the factual basis for this inference has not been established in the present case. The mere fact that the numbers are the same as in the Div 13 assessments does not suffice.

560. Further, Briggs is to be distinguished. In Briggs, it is to be recalled, one of the agreed facts was that none of the Deputy Commissioner of Taxation and three other officers of the Australian Taxation Office had made any attempt to ascertain the prosecutor's taxable income nor intended to undertake any relevant process of calculation prior to the issue and service of the notices of amended assessment and assessment, but the Commissioner issued the notices for the purpose of forcing the prosecutor to consult with him or his officers. A further agreed fact was that the Commissioner decided to issue the notices of amended assessment and assessment knowing that they did not reflect any rational assessment of a liability of the prosecutor or with reckless indifference to whether they did or did not reflect any such assessment. It was these agreed facts which the Full Court was referring to at 308 in saying that the respondent had admitted that the documents issued by them were not, in truth, assessments of taxable income, and it was a case of the respondents asserting that they had abused their powers. Their Honours distinguished this situation from the case where there had been a genuine attempt to ascertain the taxable income of a taxpayer, even if carried out cursorily or imperfectly.

561. The applicant also submitted that other statutory preconditions for the making of the Subdiv 815-A determinations were not satisfied.

562. First, as I have noted at [554] above, the applicant submitted the "treaty requirement" in s 815-10(2) was not satisfied because the United States convention did not contain an "associated enterprises article". That term was defined in s 815-15(5) to mean Art 9 of the United Kingdom convention or a corresponding provision of another international tax agreement. The applicant submitted that the United Kingdom convention did not contain any provision that corresponded to Art 1 of the United States convention. The construction of Art 9(1) of the United States convention was affected by the presence of Art 1. Furthermore, Art 9 was agreed in 1982, prior to the publication of the relevant OECD Guidelines and prior to the entry into force of Art 9 of the United Kingdom convention. It followed, the applicant submitted, that Art 9 of the United States convention did not "correspond" to Art 9 of the United Kingdom convention. In order to correspond, the provisions must, at the very least, convey an analogous or similar meaning: they must be similar in character and function. Where two provisions contained similar text, but one was substantially modified by its context, that test could not be satisfied. Where there was an ambiguity, the domestic legislation would be construed consistently with the Treaty. To read Subdiv 815-A as not applying to the United States convention was to favour a construction of a Commonwealth statute which accorded with the obligations of Australia under the international Treaty.

563. The respondent submitted that Art 9 of each convention was headed "Associated Enterprises". The language and structure of the two articles was the same save for some minor differences and the fact that the order of paragraphs 2 and 3 of the United Kingdom convention was reversed in the United States convention. The respondent submitted that Art 9 of the United States convention was amply within the description of a "corresponding provision" to Art 9 of the United Kingdom convention. The respective articles were in sufficiently similar language.

564. The respondent also submitted that, first, Art 1 of the United States convention did not alter the meaning of Art 9. Secondly, even if the effect of Art 1(2)(a) was to read into Art 9(1) a prohibition against Australia using Art 9 to increase a taxpayer's assessment whereas Art 9 of the United Kingdom convention could have that effect, the respective Arts 9 would still correspond to each other. The differences between the United States and United Kingdom conventions that the applicant contended for could not be relevant in the statutory context. Thirdly, the language of s 815-15(5) and the extrinsic material, being the explanatory memorandum in relation to the Tax Laws Amendment (Cross-Border Transfer Pricing) Bill (No. 1) 2012, paragraph 1.68, demonstrated that to be corresponding the relevant aspect which the articles must share was the subject matter to which they applied: they were each to be the article of the respective convention that applied to and defined "associated enterprises". This was Art 9 of the United States convention and Art 9 of the United Kingdom convention.

565. I accept the respondent's submissions in this respect, although I do not find it necessary to rely on the extrinsic material. In my opinion, "corresponding provision" does not focus on the detail but refers to another provision the gist of which is the same: see
New South Wales v Corbett [2007] HCA 32; 230 CLR 606 at [7] per Gleeson CJ, acknowledging that the meaning of "corresponding provision" depends on the context, see also


ATC 17904


Sackville-West v Viscount Holmesdale (1870) LR 4 HL 543 at 576 per Lord Cairns, discussed in
Samarkos v Commissioner for Corporate Affairs [1988] NTSC 10; 52 NTR 1 at 10, per Asche CJ. See also
Seaton v Mosman Municipal Council [1998] NSWSC 75; 98 LGERA 81 at 98 per Mason P with whom Meagher and Sheller JJA agreed.

566. In
Greenock Harbour Trustees v Greenock Corporation (1905) 13 SLT 367, the question was whether the Police Acts in force in Greenock had "corresponding" provisions to those in the Burgh Police Act of 1892. The Lord Chancellor, with whom Lord Robertson agreed in separate reasons (Lord Ashbourne dissenting), said at 368:

The words "corresponding assessment" are not technical, and it must be admitted that such a mode of referring to other sections of other Acts of Parliament is calculated to confuse and embarrass. Still, the very looseness of it is what makes it, I think, applicable here. I think one would say, colloquially, if you were referring to the two statutes in question, which are the two corresponding sections? And I think the answer would be favourable to the appellants' contention. They do correspond in object and purpose, though to some extent they may differ in machinery. The force of the Lord Ordinary's reasoning, I think, is applied to the difference between the sections of the General Improvement Assessment of the Act of 1892. It is true the 42nd and the 43rd sections [of the Greenock Local Act of 1877] only give limited powers of improvement and not general improvement. I cannot think that that prevents their being corresponding sections. I think it would be a very intelligible thing to say the corresponding sections differ in such and such particulars, and indeed the use of the word suggests that the language or even the substance of the enactment is not identical in omnibus, otherwise the simplest form would be instead of "corresponding" to say "identical".

567. In
Winter v Ministry of Transport [1972] NZLR 539 Turner J, for the Court of Appeal, said at 541:

[Counsel] confined his argument exclusively to the submission that the sections were not "corresponding" sections. This submission was founded upon the proposition that the 1970 provisions, taken as a whole were different from those of 1968. But this must be so whenever a new statutory provision is substituted for an old one. We read "corresponding" ... as including a new section dealing with the same subject matter as the old one, in a manner or with a result not so far different from the old as to strain the accepted meaning of the word "corresponding" as given in the Shorter Oxford English Dictionary - "answering to in character and function; similar to". The new [section] answers to the old one ... in character and function; it is similar in purpose, prescribes the same thing to be done, and is designed to produce the same result. We hold it to be a "corresponding section".

This passage was expressly approved by the Privy Council in
Vela Fishing Ltd v Commissioner of Inland Revenue [2004] 1 NZLR 313, 324.

568. In my opinion, Art 9 is a provision of the United States convention corresponding to Art 9 of the United Kingdom convention. It is sufficient that Art 9 of the United Kingdom convention deals with "associated enterprises" as does Art 9 of the United States convention and that the gist of each Article is the same. Therefore, Art 9 of the United States convention answers the definition of an "associated enterprises article" in s 815-15(5)(b) of the ITAA 1997. It is a different question, which I consider at [580]-[585] below, whether "the requirements in the *associated enterprises article for the application of that article to the entity are met" for the purposes of s 815-15(1)(b) for the purpose of deciding whether the entity "gets a transfer pricing benefit".

569. The applicant also submitted that even if Art 9 of the United States convention could be said to "correspond" to Art 9 of the United Kingdom convention, it was not the case that Art 9 applied to CAHPL (s 815-10(2)) or contained "requirements" which applied to CAHPL (s 815-15(1)(b) and (c)). The reason for this was the presence of Art 1 and Art 9(3) in the United States convention. If Art 9 of the United States convention did not and could not operate to increase the tax burden of taxpayers resident in Australia, it must follow that in the context of the proposed adjustment to increase the Australian taxable income of an Australian resident, Art 9 did not "apply" to that Australian resident; nor contain


ATC 17905

"requirements" for its application. The applicant submitted that even without Art 1, it may be doubted whether Art 9 of the United States convention here had any "requirements" capable of "application" to CAHPL, in the sense required by s 815-15(1)(b). That was because Art 9 did not contain any rules of law which were capable of applying to delineate the liability of a particular taxpayer. Rather, Art 9 established a broad principle, capable of being used as a basis for drafting such rules for domestic application. But, by its terms, it was too vague and general to constitute in and of itself, rules of law which could be used to impose tax upon a taxpayer.

570. The applicant submitted there was a further reason why Art 9 did not "apply" to CAHPL in this case. Article 9 of the United States convention must also be read in the context of Art 11. Article 11 dealt with the taxation of interest arising in Australia or the United States. It was Art 11(8) which permitted amounts that are interest to be re-characterised for the purposes of the United States convention. The applicant submitted that in the case of a payment of interest between a resident of Australia and a resident of the United States, Art 11 (and in particular Art 11(8)) supplanted the operation, if any, which Art 9 might otherwise have. It was Art 11 which governed the taxing rights of the Contracting States in relation to amounts of interest - it was, in that respect, an exhaustive code. To read Art 9 as permitting an adjustment to interest was to ignore its context and to offend the principle that a general provision is to give way if it is applicable to the same subject matter as a specific provision. It was also contrary to the last sentence of Art 11(8), which was an allocation mechanism that operated to permit other Articles of the United States convention to operate only in relation to the amount not treated as interest for the purposes of the United States convention. What followed from this was that Art 9 could not be said to "apply" to an Australian resident where what was sought to be done was the exercise specifically provided for in Art 11(8) with respect to an amount of interest. If any article of an international tax agreement could be said to "apply" to CAHPL, that article was Art 11. Furthermore, if there were any "requirements" to be met, those requirements were to be found in Art 11, not Art 9. It followed that Subdiv 815-A had no field of application to CAHPL in the context of the interest here in issue.

571. The respondent submitted that Art 9(1) of the United States convention, the associated enterprises article, contained a participation requirement and a conditions requirement. The former requirement related to CAHPL participating directly or indirectly in the management, control or capital of CFC or, in the alternative, the same persons at CVX participating directly or indirectly in the management, control or capital of CAHPL, an enterprise of Australia, and CFC, an enterprise of the United States. The latter requirement was that conditions operated between CAHPL and CFC in their commercial or financial relations which differed from those which might be expected to operate between independent enterprises dealing wholly independently with one another.

572. As to the applicant's argument based on Art 1 and Art 9(3), the respondent submitted that Art 9(3) of the United States convention was relevantly the same as Art 9(2) of the United Kingdom convention and it could not have been intended to exclude Art 9 of the United Kingdom convention in light of the express inclusion of that Article by s 815-15(5). Thus Art 9(3) of the United States convention did not deprive Art 9 of any "requirements". Secondly, the respondent submitted, Art 1 and Art 9(3) of the United States convention did not alter the meaning of Art 9(1) of that convention. Thirdly, even if the effect of Art 1 and Art 9(3) was to read into Art 9(1) a prohibition against Australia using Art 9 to increase CAHPL's assessment, it did not follow that Art 9 ceased to have requirements that could apply to CAHPL. The respondent submitted that under s 815-15, meeting the requirements of the relevant associated enterprises article was a precondition to making a determination under s 815-30 but it was the application of the provisions of Subdiv 815-A and those provisions relating to assessment or amending an assessment that caused the increase or decrease in the taxpayer's liability. Thus Art 1 of the United States convention would not be engaged even on the applicant's reading of the Treaty.

573. As to the applicant's alternative argument that Art 9 of the United States convention had no requirements for the purposes of s 815-15(1)(b) because Art 9 lacked rules of law which were capable of applying to delineate the liability of any particular taxpayer, the respondent again submitted that it was not the associated enterprises article of the United States convention that delineated the taxpayer's liability which instead came about through the operation of all of the provisions of Subdiv 815-A and the provisions of the ITAA 1936 relating to assessment. The respondent in any event disagreed with the proposition that Art 9 was too vague and general.

574. As to the applicant's contention that Art 11 of the United States convention was an exhaustive code in relation to interest and that this caused Art 9 to be outside s 815-15(1)(b) in the present case, the respondent submitted that Art 11 was not an exhaustive code for all taxation relating in some way to interest but rather it set out certain restrictions


ATC 17906

on Contracting States imposing tax on interest income. For example, Art 11 had no application in the present proceedings which were not about Australia taxing the interest income earned by CFC. Secondly, even if Art 11 were an exhaustive code it would not affect s 815-15(1)(b). That provision was a precondition which applied to all potential associated enterprises transfer pricing adjustments, not just those relating to interest. It went to the relationship between the parties. Art 9(1) of the United States convention applied in terms in relation to CAHPL. The precondition in s 815-15(1)(b) was met.

575. In my opinion, the applicant seeks to give the expression "the requirements in the associated enterprises article for the application of that article to the entity" too large an operation. It is clear that Art 9 of the United States convention contains requirements, one being the participation requirement, which includes an alternative, and the other the ultimate satisfaction of which is better to be considered in the context of whether CAHPL gets a transfer pricing benefit. To that extent that requirement is not a threshold issue. I reject the applicant's submission that, at the threshold, Art 9 did not "apply" to CAHPL, nor contain "requirements" for its application. In my opinion, any separate submission that Art 9 did not apply to CAHPL with reference to s 815-10(2) fails to recognise that that provision is concerned with whether the entity gets the transfer pricing benefit at a time when an international tax agreement applies to the entity. It is that time which is the focus of the provision and there is then a further requirement that the international tax agreement applying at that time contains an associated enterprises article.

576. I also reject the submission based on Art 1 and Art 9(3) in the United States convention. I am not persuaded that Art 1 has any present application. Neither it nor Art 9(3) alters the meaning of Art 9(1) or has the consequence that Art 9 no longer contains requirements for the application of that article to the entity. In my view, the applicant's submissions do not sufficiently recognise that the requirements referred to in s 815-15(1)(b) are requirements having that character, that is, for the application of that article to the entity.

577. I also reject the applicant's submission that because Art 9 did not contain any rules of law which were capable of applying to delineate the liability of a particular taxpayer and that therefore, as I understood the argument, Art 9 did not apply or its requirements to the entity were not met. Again, in my view, this is to seek to give Art 9 too great a substantive operation. As the respondent submitted, the taxpayer's liability comes about through the operation of all of the provisions of Subdiv 815-A and the provisions of the ITAA 1936 relating to assessment. It follows that I reject the applicant's submission that there were no criteria for liability under Art 9.

578. As to the applicant's Art 11 submission, this travels outside the terms of s


ATC 17907

815-15(1)(b) which relates only to the requirements in the associated enterprises article for the application of that article to the entity being met. Even if Art 11 were an exhaustive code in relation to interest this would not cause Art 9 to be outside s 815-15(1)(b). I reject the submission. I am also not persuaded, in any event, that Art 11 is an exhaustive code for all taxation relating to interest.

579. There remains to consider, again in the alternative given my conclusions in relation to the 2010 amended assessments under Div 13 of the ITAA 1936, the question whether CAHPL "gets a transfer pricing benefit". Before turning to that issue, which centred on s 815-15(1)(c) of the ITAA 1997, I should record that the parties agreed that if that paragraph was satisfied, s 815-15(1)(d)(i) posed the relevant question. That is, had the amount of profits so accrued to the entity CAHPL, the amount of the taxable income of CAHPL for an income year would be greater than its actual amount. In that case the amount of the "transfer pricing benefit" is the difference between the amounts mentioned in s 815-15(1)(d)(i): see the closing words of s 815-15(1).

580. The first issue arises from the terms of s 815-15(1)(c) which states that an entity gets a transfer pricing benefit if an amount of profits which, but for the conditions mentioned in the article, might have been expected to accrue to the entity, has, by reason of those conditions, not so accrued. The conditions are those mentioned in Art 9 of the United States convention and this refers to where "conditions operate between [CAHPL and CFC] in their commercial or financial relations which differ from those which might be expected to operate between independent enterprises dealing wholly independently with one another ..."

581. The applicant submitted that it was not apparent from the "conditions" set out in paragraph 45(b) of the respondent's appeal statement how the identified conditions were said to differ from those which might be expected to have operated between independent enterprises. The "arm's length" conditions had not been identified by the respondent Commissioner. Nor was it apparent, the applicant submitted, from paragraph 45(b) how some or all of the identified conditions were said to have impacted on the pricing of the loan to CAHPL and to have resulted in the non-accrual of profits.

582. In relation to this issue, the respondent submitted there were some eleven conditions which differed from those which might be expected to operate between independent enterprises dealing wholly independently with one another. These included: that CAHPL owned CFC and they both had a common parent, CVX; CVX Treasury decided how much debt and at what interest rate CAHPL should borrow from CFC; there was no bargaining or negotiation between


ATC 17908

CAHPL and CFC; the terms and conditions of the Credit Facility Agreement, including the terms in respect of the interest rate charged, the duration and the currency of the loan and the absence of covenants; the sole reason for CFC's incorporation, and the purpose of its commercial paper program, was to raise funds solely to on-lend to its parent CAHPL; that the credit profiles of CFC and CAHPL could be controlled by decisions made by CVX; that CFC profited from lending to CAHPL at a high interest rate; and the higher the interest-bearing loan from CFC and the higher the interest rate, the more profit CAHPL stood to make.

583. As to the currency of a loan, although it is not necessary to my conclusion, I am not persuaded that the condition as to the AUD currency which was operational between CAHPL and CFC differed from the condition as to currency which might have been expected to operate between independent enterprises dealing wholly independently with one another. I accept the applicant's submission in this respect that borrowings in AUD would avoid or limit foreign currency gains and losses. I am not persuaded Professor Boymal's opinion to the contrary.

584. I do not accept the applicant's implicit submission that for each identified condition it must be explicitly stated how it would differ from the condition which might be expected to have operated between independent enterprises or that the "arm's length" conditions must be explicitly identified by the respondent Commissioner. It is enough, in my opinion, for the respondent Commissioner to identify, as he has, which conditions operate between the two enterprises which differ from those which might be expected to operate between independent enterprises dealing wholly independently with one another.

585. Neither do I accept the applicant's further implicit submission that the Commissioner must identify how each identified condition was said to have impacted on the pricing of the loan and to have resulted in the non-accrual of profits. In my opinion, once the differential conditions have been identified the question then arises, under s 815-15(1)(c), whether an amount of profits which but for the conditions might have been expected to accrue to CAHPL has, by reason of those conditions, not so accrued and whether, here under s 815-15(1)(d)(i), had the amount of profits so accrued to CAHPL the amount of CAHPL's taxable income for an income year would be greater than its actual amount.

586. The applicant also submitted that if the Div 13 determinations remained effective, then s 136AD(3) operated to deem the consideration for the loan to CAHPL to be the consideration given by the taxpayer "for all purposes of the application of this Act in relation to the taxpayer". It followed that there could be no "transfer pricing benefit" within s 815-15 of the ITAA 1997.

587. In relation to this issue, the respondent appeared to accept that there was not a contrary intention so that the definition of "this Act" in s 6(1) of the ITAA 1936 did not apply to the ITAA 1997 but did not contend that the two determinations, one under the ITAA 1936 and one under the ITAA 1997, for each income year operated cumulatively. The respondent submitted that if the Div 13 adjustments were wholly upheld, the Subdiv 815-A determinations did not need to be considered. On the other hand, if Div 13 did not apply then the Div 13 determinations could have no effect on the availability of s 815-15.

588. I accept the respondent's submission in this respect. Indeed, as I have indicated above, these reasons consider the Subdiv 815-A issues in the alternative because, if the applicant's Div 13 case had succeeded, only then would it be necessary to consider Subdiv 815-A.

589. In any event, the applicant submitted, there was no amount of profits which, but for the conditions, might have been expected to accrue to CAHPL and therefore s


ATC 17909

815-15(1)(c) was not satisfied. The applicant's primary submission in support of that proposition was that Art 9 and the test in s 815-15 did not permit: a re-characterisation of the entire transaction or a rewriting of the terms of the loan; and ratings, and in particular the credit rating agency concept of so-called "implicit credit support", to be taken into account in determining the arm's length interest rate for the loan. In the alternative, the applicant submitted, when viewed in its totality, it could not also be said that there were any "profits" which did not accrue to CAHPL in its commercial or financial relations with CFC and thus the requirements of Art 9 were not satisfied. CAHPL paid interest on the loan to CFC and received dividend income from CFC. The respondent had ignored entirely the dividend income CAHPL received in the years in dispute, totalling $1,110,559,595. The applicant submitted that the word "profits" in Art 9 did not refer to taxable income but profits in its more generic sense.

590. The respondent Commissioner submitted that but for the conditions he identified, CAHPL would have derived an additional amount of profits. Its interest expenses would have been less, with the difference representing an additional amount of profits. The respondent submitted the precise amount of profits which CAHPL would have derived depended on whether CAHPL was treated as a subsidiary of CVX, or a group having the same characteristics as the CVX group, or as a stand-alone company, whatever that meant.

591. In my opinion, the question posed under s 815-15(1)(c) has a focus different from that which arises under Div 13. What is here required is the assessment of an amount of profits which might have been expected to accrue but for the conditions identified by reference to Art 9 but which, by reason of those conditions, have not so accrued. It follows, in my view, that the question of re-characterisation of the transaction in question or rewriting the terms of the loan does not directly arise. If, as I understand it to be, the applicant's proposition is that the statutory regime does not permit a departure from the actual "commercial or financial relations" between the parties which in this case was the relationship of debtor and creditor in respect of an AUD credit facility in the sum of AUD3.707 billion, then I do not accept the width of that proposition.

592. Neither do I accept the related submissions by the applicant that where the thin capitalisation rules (Div 820) were engaged, it was those rules that expressed the legislature's determination of the appropriate amount of debt with which an entity can be financed and the transfer pricing rules should not be taken to override or interfere with that limit. Section 815-25 modifies the transfer pricing benefit an entity gets, or apart from that section would get, in an income year in the circumstances there set out. One of those circumstances is if Div 820 applies to the entity for the income year. This shows that Div 820 does not stand outside Subdiv 815-A. It follows, in my opinion, that the applicant's reliance on what was said at [1.76]-[1.77] in the explanatory memorandum to the New Business Tax System (Thin Capitalisation) Bill 2001 is misplaced for the reason that it cannot qualify or contradict the terms of the statute, s 815-25, which remains paramount, and because the explanatory memorandum was not directed to Subdiv 815-A, a later enactment. Further, a reading of [1.78] of the explanatory memorandum suggests that the thin capitalisation rules do not always prevail. That paragraph states as follows:

However, the thin capitalisation rules do not have the same scope as Division 13 and comparable provisions of DTAs - the latter apply to a wider range of transactions. Further, there may be instances where the purpose of the application of the arm's length principle under Division 13 and comparable provisions of DTAs to a particular case is not the same as for applying the arm's length test under the thin capitalisation rules. In these cases, the arm's length principle articulated in Division 13 and comparable provisions of DTAs should apply. For example, the application of the arm's length principle to determine whether a rate of interest is greater than an arm's length amount can only be done under Division 13 and comparable provisions of DTAs.

I therefore reject the applicant's submissions that the thin capitalisation rules are an exclusive code or that the thin capitalisation rules confer an entitlement outside the terms of Subdiv 815-A.

593. A related submission made by the applicant was that a negative inference should be drawn from the existence and terms of Subdiv 815-B. The applicant submitted that there was no provision in Subdiv 815-A that authorised the Commissioner to substitute for the actual terms of the loan other terms (including security or financial covenants or currency). Section 815-25 did not authorise such substitution and to the extent the explanatory memorandum suggested otherwise, it was wrong. An express power to re-characterise could easily have been included, and was in fact drafted as part of the same programme of legislative reform: s 815-130 of Subdiv 815-B of the ITAA 1997 expressly permitted a reformulation of the commercial and financial relations between the taxpayer and the other entity. The contrast between the language of Div 13 and Subdiv 815-A, on the one hand, and the language contained in new Subdiv 815-B - which was not in issue in these proceedings - on the other, was instructive. Section 815-115 expressly substituted the "arm's length conditions" for the "actual conditions" that operated between the parties. The starting point or "basic rule" for determining the arm's length conditions was that one adopted the actual conditions, but those could be changed or re-characterised if one of the three express exceptions applied: s 815-130(2)-(4).

594. The applicant expanded on its submission that Art 9 of the United States convention and the test in s 815-15 did not permit a rewriting of the terms of the loan as follows. The applicant submitted a contextual analysis of Art 9 revealed that it did not permit a rewriting of the terms of a loan but only an examination of the "conditions". Further, there was no provision in Subdiv 815-A that authorised the Commissioner to substitute for the actual terms of the loan other terms (including security or financial covenants or currency). Further, there was nothing in the OECD Guidelines which permitted the Commissioner to substitute the actual loan for some other loan in determining the dealing that was to be priced. The circumstances in which the OECD Guidelines did contemplate the possibility of re-characterisation (assuming domestic legislation permitting this) were very limited. The applicant submitted, in applying these principles, that although the Commissioner disavowed re-characterising, his case was that the prohibition applied only to the characteristics of the parties and not to the terms of the transaction itself.

595. The applicant referred to the explanatory memorandum to the Tax Laws Amendment (Cross-Border Transfer Pricing) Bill (No. 1) 2012 and, I understood, particularly paragraph 1.109 and example 1.6. The applicant submitted that at least the final paragraph of example 1.6 was wrong. It was


ATC 17910

that Bill, once enacted, which inserted Subdiv 815-A into the ITAA 1997. For completeness I reproduce that material, so far as relevant:

1.109 The following examples illustrate the interaction of Subdivision 815-A and Division 820. They are intended purely to illustrate the respective fields of operation of Subdivision 815-A and the thin capitalisation rules and are not intended to suggest that a particular method for pricing debt must be applied to the circumstances of a particular case. Nor are the examples intended to preclude the use of other methods that produce an arm's length outcome.

Example 1.4: Thin capitalisation adjustment and transfer pricing adjustment

Aus Co is an Australian resident subsidiary company of For Co, a resident of the UK. Aus Co is an 'inward investment vehicle (general)' for the purposes of Subdivision 820-C.

For an income year, Aus Co has:

  • • a 'safe harbour debt amount', determined in accordance with section 820-195 of $375 million;
  • • 'adjusted average debt' determined in accordance with subsection 820-185(3) of $400 million, of which $200 million is borrowed from For Co at an interest rate of 15 per cent, and $200 million from an independent lender at an interest rate of 10 per cent; and
  • • equity of $100 million.

Aus Co's only debt deductions are for the interest incurred at a rate of 15 per cent on its $200 million related party debt, and 10 per cent on its $200 million debt from the independent lender, meaning that it has $50 million of debt deductions for the income year.

The Commissioner considers whether Aus Co has received a transfer pricing benefit under section 815-15. In doing so, the Commissioner has regard to the arm's length rate in relation to the debt interest (that is, the arm's length interest rate), applied to the actual amount of the related party debt.

Assume that the loan from the independent lender is sufficiently similar to the loan from For Co and the circumstances in which each amount of debt funding was provided do not present material differences that would affect the rate applicable to the debt interest or Aus Co's ability to obtain $400 million in debt funding (that is, the independent loan is directly comparable to the related party loan). As a result, the Commissioner determines that using a comparable uncontrolled price is the most appropriate method for determining the arm's length rate. In these circumstances it is commercially realistic for the Commissioner to determine that the arm's length interest rate is 10 per cent. In this case, Aus Co gets a transfer pricing benefit of $10 million (being the difference between an arm's length rate of 10 per cent applied to the debt interest arising from the loan from For Co ($200 million) and the actual interest rate of 15 per cent on the debt interest).

Further, to the extent that Aus Co has 'excess debt', Division 820 will apply to deny a corresponding proportion of Aus Co's debt deductions remaining after the $10 million reduction under Subdivision 815-A.

Example 1.5: Transfer pricing adjustment and no thin capitalisation adjustment

Assume the facts and circumstances are the same as in Example 1.4, except that Aus Co has $300 million of debt ($150 million from For Co and $150 million from an independent lender) and $100 million of equity, producing a safe harbour debt amount for Division 820 purposes of $300 million. The interest rate on Aus Co's debt to For Co is 15 per cent, so that, before applying Subdivision 815-A and Division 820, Aus Co has total debt deductions of $37.5 million.

As was the case in Example 1.4, the Commissioner determines that an arm's length interest rate of 10 per cent is to be applied to the debt interest from For Co. As such, Aus Co gets a transfer pricing benefit of $7.5 million (being the difference between the arm's length rate of 10 per cent applied to the debt interest from For Co ($150 million) and the actual interest rate of 15 per cent on the debt interest).

Example 1.6: Transfer pricing adjustment and no thin capitalisation adjustment

Assume the facts and circumstances are the same as in Example 1.5, except that the entire $300 million of debt is borrowed from For Co at an interest rate of 15 per cent. Aus Co's debt deductions for the interest incurred on its $300 million debt total $45 million for the income year.

Unlike the previous examples, there is no internal comparable uncontrolled price that provides an arm's length rate. As such, the Commissioner determines the arm's length rate of interest for the loan having regard to available data of market reference rates and the credit standing that the capital markets would be likely to give Aus Co. The market data shows that Aus Co's credit standing would allow it to borrow $250 million from independent lenders. Having regard to the information available, the Commissioner determines that the closest commercially realistic arm's length scenario at which a loan might reasonably be expected to exist between independent parties dealing wholly independently with one another is a loan of $250 million at 10 per cent.

In this case, the Commissioner is able to determine the amount of the transfer pricing benefit by reference to an amount less than the actual amount of the debt interest (being an arm's length amount). (The fact that Aus Co's debt amount is less than its safe harbour debt amount for Division 820 purposes is not relevant to determining the amount of the transfer pricing benefit.)

The Commissioner determines that Aus Co's transfer pricing benefit is $15 million (as required under subsection 815-25(2)). This is worked out by applying the 10 per cent arm's length interest rate to Aus Co's actual debt amount ($300 million), and comparing this to Aus Co's actual debt deductions of $45 million.

596. The respondent submitted in this respect that Subdiv 815-B provided no comfort to the applicant, first, because this was an inappropriate approach to construction and, secondly, because on the very terms of Div 13, Subdiv 815-A and Art 9 there was no warrant for an approach which set in stone all aspects of the non-arm's length agreement save as to interest rate.

597. As to the first of these matters, the respondent submitted that Subdiv 815-B was enacted after each of Div 13 and Subdiv 815-A and was for a different purpose, which was to apply the arm's length principle to relevant dealings between both associated and non-associated entities, looking to supplies and acquisitions under international agreements. Further the machinery of Subdiv 815-B was quite different from the determination provisions which appeared in Subdiv 815-A and Div 13. Therefore no expressio unius implication arose. Subdivision 815-B did not apply to the income years in issue in the present proceedings. Subdivision 815-A continued to operate in respect of income tax years after 1 July 2004 and until the commencement of the Tax Laws Amendment (Countering Tax Avoidance and Multinational Profit Shifting) Act on 29 June 2013 which repealed Div 13 and added Subdiv 815-B at the end of Div 815. In those circumstances that Act shed no light on the earlier provisions and judicial authorities recognised the difficulties in attempting to construe words used in a statute by reference to later amendments, let alone later separate legislation: see
Ajinomoto Company Inc v NutraSweet Australia Pty Ltd [2008] FCAFC 34; 166 FCR 530 at [92]-[99] and the authorities there referred to.

598. I agree with this submission made by the respondent. It is necessary to start and finish with the words of Subdiv 815-A, both as to what they do provide and as to what they do not.

599. As to the broader point, the respondent submitted that the applicant would have the Court approach the arm's length pricing analysis by reference to the exact terms of the Credit Facility Agreement, which would not have been agreed between arm's length parties, as opposed to the terms that might be expected to have been agreed between CAHPL and CFC had they been independent of each other. The respondent submitted that the provisions in Subdiv 815-A (as in Art 9) required that there be an examination of the "conditions" operating between the two associated enterprises - not just the price - and an evaluation of profits absent the identified "conditions". That is, the respondent submitted, Subdiv 815-A did not envisage that one should take a related party loan transaction exactly as one found it and price the interest payable on the transaction (if that was even possible). Rather, by focusing on "conditions" as the matter that was to be hypothesised on the counterfactual analysis, Subdiv 815-A recognised that associated


ATC 17911

entities may have dealings which are non-arm's length for reasons apart from the price that is attributed to them. It followed, in the respondent's submission, that the applicant's contention that the Commissioner or Court may not depart from any of the actual terms of the Credit Facility Agreement, other than interest rate, found no basis in the express terms of the provisions of Subdiv 815-A.

600. In my opinion, the applicant's submissions suffer from the use of a shorthand as to not "re-characterising" or not "rewriting" as a substitute for the statutory language and thereafter to describe what the respondent Commissioner has done as impermissible because it involved "re-characterisation" or "rewriting". On the other hand, the respondent's submissions seek to align Div 13 and Subdiv 815-A without sufficient regard to the different language of the two sets of provisions: Div 13 focuses on "consideration" whereas Subdiv 815-A focuses on the broader term "conditions". Having said that, I accept the respondent's submission that by focusing on "conditions", Subdiv 815-A recognised that associated entities may have dealings which are non-arm's length for reasons apart from the price that is attributed to them.

601. Another significant difference between the parties in construing the provisions was that the applicant submitted, under the heading "Implicit Support" that the terms of Art 9 meant that one must consider the conditions that one might expect to see between a lender and a borrower who are


ATC 17912

independent, and are dealing wholly independently with one another. In the applicant's submission, the relationship between the lender and the borrower must therefore be eliminated in order to undertake this task, and in a situation where the entities in question were sister companies, so too must the relationship between each of them and their common parent. If that latter relationship were permitted to subsist, then it could not be said that the lender and borrower were independent or were dealing independently. Their hypothetical dealing would be infected by the characteristics of each party, the borrower in particular, that were referable to ownership by the common parent.

602. The applicant submitted that a textual analysis of Art 9 supported the contention that the concept of "independent enterprises" was used in contradistinction to, and as the converse of, "associated enterprises". The OECD Guidelines provided that two enterprises were "independent" if they were not "associated enterprises". It followed therefore, in the applicant's submission, that all and any attributes that give rise to entities being "associated" within the meaning of Art 9 must be disregarded in determining the attributes of the independent parties. Within Art 9 there were two conditions that could result in parties being regarded as associated. The first was participation by one entity in the management, control or capital of the other. Negating this attribute required one to ignore the parent-subsidiary relationship that in fact existed between CAHPL and CFC. The second condition of association was the same persons participating in the management, control or capital of the two enterprises. Negating this attribute required one to ignore the ownership by CVX of each of CAHPL and CFC. The terms of Art 9 thus required one to hypothesise a stand-alone borrower and a stand-alone lender. There was no room for implicit parental support which of necessity derived from the common owner, CVX. This was further confirmed in the OECD Guidelines, which said that Art 9 required one to treat members of a multi-national group as if they were operating as separate entities rather than part of a single "unified business", and thus "attention is focused on the nature of the dealings between those members".

603. The respondent submitted that while the transfer pricing rules required the


ATC 17913

affiliation between the parties to the transaction to be ignored, there was no warrant for ignoring the affiliation between a party to the transaction in question and other members of the group of companies of which it formed a part. To do so, the respondent submitted, would be contrary to the natural language of the relevant provisions, their judicial interpretation and the object and purpose of the transfer pricing rules. Each of the relevant provisions focused on the relationship between the parties to the relevant transaction.

604. While I accept the applicant's submission that one must consider the conditions that one might expect to see between a lender and a borrower who are independent, and are dealing wholly independently with one another, which is the language of Art 9, it by no means follows that where, as here, the entities in question are sister companies, also to be eliminated is the relationship between each of them and their common parent on the basis that, otherwise, it could not be said that the lender and borrower were independent or were dealing independently. In my opinion, independent enterprises dealing wholly independently with one another may still be subsidiaries and may still have subsidiaries even if the enterprises are independent of each other. I therefore accept the respondent's submission insofar as he contended that there was no legislative warrant for ignoring affiliation between a hypothesised party to a transaction and other members of that party's group of companies. At the factual level, at [606] below, I have accepted the applicant's submission as to implied parental support.

605. This conclusion means that the applicant's high-level contention about "implicit support' also fails. "Implicit support" may be generally relevant when assessing a borrower's credit rating. The high-level contention fails because it relies on the proposition that the relationship between CAHPL and CVX and the relationship between CFC and CVX, CVX being the common parent, must be eliminated from the analysis.

606. The applicant's submission that the existence and worth of "implicit support" is a matter of fact remains unaffected. I accept the applicant's submission, that in the absence of a legally binding parental guarantee, implicit credit support had very little, if any, impact on pricing by a lender in the real world. This was the evidence of Mr Martin and Mr Gross and the conclusion of an article published in 2014 of which Mr Hollas was a joint author: "Intercompany Financial Transactions: Factors to Consider in Analysing the Impact of Implicit Parental Support".

607. As to the applicant's reliance on a differentiation between the language of "association" and "independence", it seems to me that that distinction involves a non sequitur: to say that a party is independent of another party does not mean or require that either party is independent of all parties.

608. A further broad submission put by the applicant was that Art 9 mandated a determination of the "profits" which might have been expected "but for those conditions" and thus proceeded to consider a hypothetical situation in which commercial or financial relations take place, but absent the operation of the identified conditions. By its terms, the applicant submitted, Art 9 did not permit the addition of new "conditions" but was, instead, an "annihilation" provision.

609. In my opinion, the correct approach is to identify the conditions mentioned in Art 9 and then ask if there was an amount of profits which, but for those conditions, might have been expected to accrue to the entity but which has, by reason of those conditions, not so accrued: s 815-15(1)(c). As I have set out above at [27], Art 9 involves a comparison between, here, conditions which operate between CAHPL and CFC in their commercial or financial relations and whether those conditions differ from those conditions which might be expected to operate between independent enterprises dealing wholly independently with one another. It seems to me a distraction, in that context, to speak about "annihilation" provisions: conceptually the comparison between the actual conditions and the conditions which might be expected to operate between independent enterprises dealing wholly independently with one another is straightforward although its application may not be. In my opinion, nothing is "annihilated" but I accept that what must be compared are conditions which operate.

610. Once the approach I have outlined is borne in mind, in my view, the applicant's submission: "By its terms [Art 9] does not permit the addition of new 'conditions'" does not assist. It follows that I do not accept the applicant's submission that "Article 9 negates non-arm's length conditions but does not supply any condition which is absent and leaves the commercial or financial relations (as opposed to its terms) exactly as it finds it". It follows that I reject the applicant's submission that the "requirements" of Art 9 permit only an adjustment to the price of a transaction (in this case an adjustment to the rate of interest) for the purpose of determining the quantum of profits which might have been expected to accrue, and they might justify, but go no further than, the elimination of other terms or conditions upon which CFC lent to CAHPL. As I have said, in the present case Art 9 involves identifying conditions which operate between CAHPL and CFC in their commercial or financial relations and seeing where they differ from conditions which might be expected to operate between independent enterprises dealing wholly independently with one another.

611. It is necessary to return to the applicant's submission, referred to at [589] above, that viewed in its totality, it cannot also be said that there were any profits which did not accrue to CAHPL in its commercial or financial relations with CFC and thus the requirements of Art 9 were not satisfied. The applicant's submission was that CAHPL paid interest on the loan to CFC and received dividend income from CFC. The Commissioner ignored the dividend income CAHPL received in the years in dispute. It may be that the Commissioner overlooked the dividends received by CAHPL because he did not consider them to be "profits" of CAHPL for the purposes of Art 9. He may have read "profits" to mean "taxable income" because of s 3(2) of the International Tax Agreements Act, set out at [22] above.

612. The applicant submitted that the word "profits" where first appearing in Art 9 did not refer to taxable income, but profits in its more generic sense. The applicant submitted that this was supported by the history of Art 9. The reference to profits in this part of Art 9 was the same profit referred to in the old Art 5 of the 1933 Draft Convention for the Allocation of Business Income between States for the purposes of Taxation: it was a diverted profit


ATC 17915

which must be allocated to "one of the enterprises". It would make no sense, the applicant submitted, to read the word "profits" in that phrase as meaning taxable income, as the enterprise which may get allocated those profits may not be resident in Australia. The profits which might be expected to accrue as mentioned in Art 9 were therefore to be taken to refer to profits generally. Once this condition of Art 9 was satisfied, namely that there were profits which might be expected to have accrued to one of the enterprises, the mechanism whereby such profits may be domestically taxed (i.e. included in the taxable income computation) followed in Art 9 with the concluding language of that Article: "may be included in the profits of that enterprise and taxed accordingly". It followed that the profits of CAHPL included the dividends it had received from CFC for the purposes of the first condition in Art 9 relating to profits. In that respect, it could not be suggested that because of the conditions operating between CFC and CAHPL, "profits" did not accrue to CAHPL which should have. In other words, there had been no diversion of profits to CFC, and CAHPL's expense, precisely because they had returned to CAHPL. The respondent Commissioner submitted that s 3(2) of the International Tax Agreements Act applied to the profits referred to in Art 9(1) and deemed them to be taxable income derived by CAHPL. Neither party referred to any authority on the point.

613. In my opinion, s 3(2) of the International Tax Agreements Act has a limited purpose, as set out in the explanatory memorandum to the Income Tax (International Agreements) Bill 1953, circulated by the Treasurer, the Rt. Hon. Sir Arthur Fadden:

The proposed sub-section (2.) is, subject to minor drafting variations, the same as the corresponding provision enacted in 1947 as sub-section (2.) of section 160F of the Assessment Act. Its purpose is to permit references in agreements to profits to be construed, unless the context requires otherwise, as references to taxable income. The provision is required because the Australian law imposes tax upon taxable income and not upon profits as such.

I do not, therefore, regard s 3(2) as having a substantive or deeming operation. The applicant's argument based on the former Art 5 seems to me to be unnecessary to reach this conclusion which involves construing the different language of the present Art 9. It seems to me that the words "and taxed accordingly" are included in the text of Art 9(1) so as to make it clear what the relevant Contracting State may do. Section 815-15(1)(c) has effect accordingly. I am not, however, persuaded of the correctness of the applicant's consequential argument that "profits" means that in the present case there can be a net profit position arising from the particular arrangements between the parties. What is being dealt with by Art 9 is profits which, but for the difference between the actual conditions operating and the conditions which might be expected to operate, have not accrued to, here, CAHPL. I therefore do not accept the applicant's submission that there were no profits which accrued to CAHPL.

614. Having rejected the applicant's submissions, primarily submissions as to the proper construction of Art 9 and of Subdiv 815-A, and having considered at [505]-[524] above the evidence of the applicant's main witnesses, I find that the requirements in the *associated enterprises article for the application of that article to CAHPL are met. I also accept the respondent's submission identifying conditions, set out at [582] above. I find that but for the conditions operating between CAHPL and CFC which differ from those which might be expected to operate between independent parties dealing wholly independently with one another an amount of profits might be expected to have accrued but has not so accrued. It follows that the applicant has failed to show that the assessments under the ITAA 1997 were excessive. As I have said, my consideration of these matters is in the alternative to my conclusion as to the assessments made under Div 13 of the ITAA 1936.USD

Penalties

615. The applicant submitted that because CAHPL had obtained no "scheme benefit" it was not liable to an administrative penalty under either s 284-145(1) or s 284-145(2) of Sch 1 to the Taxation Administration Act. As set out at [15] above, the content of "scheme benefit" is given by s 284-150 in Sch 1 to the Taxation Administration Act which provided that an entity gets a scheme benefit from a scheme if a *tax-related liability of the entity for an accounting period is, or could reasonably be expected to be, less than it would be apart from the scheme or a part of the scheme.

616. The applicant submitted that if, contrary to its submissions, CAHPL did obtain a "scheme benefit", no penalty was here payable pursuant to s 284-145(1) in Sch 1 to the Taxation Administration Act because it was not reasonable to conclude that CAHPL entered into the facility for the sole or dominant purpose of obtaining a "scheme benefit".

617. The central provision of s 284-145 was as follows:


ATC 17916

  • (1) You are liable to an administrative penalty if:
    • (a) you would, apart from a provision of a *taxation law or action taken under such a provision (the adjustment provision ), get a *scheme benefit from a *scheme; and
    • (b) having regard to any relevant matters, it is reasonable to conclude that:
      • (i) an entity that (alone or with others) entered into or carried out the scheme, or part of it, did so with the sole or dominant purpose of that entity or another entity getting a scheme benefit from the scheme; ...

618. The applicant submitted that it could not be reasonably concluded that CAHPL entered into the Credit Facility Agreement for the dominant purpose of obtaining a scheme benefit - its dominant purpose was to refinance its existing Australian dollar denominated debt. True it was, it obtained a deduction for interest incurred in respect of that debt. But that was a consequence of choosing to use debt funding. CAHPL's choice to use such debt was lawful. It was permitted by Div 820, and was not impugned by the Commissioner.

619. The difference between the parties, the applicant submitted, was that the applicant relied upon the majority decision in
Commissioner of Taxation (Cth) v Star City Pty Ltd (No 2) [2009] FCAFC 122; 180 FCR 448 for the proposition that a purpose had to be subjective, not objective.

620. The applicant submitted that by reason of s 815-10 of the Income Tax (Transitional Provisions) Act 1997, the penalty provisions in Subdiv 284-C in Sch 1 to the Taxation Administration Act - which included ss 284-145 and 284-150 - had no application. The respondent Commissioner agreed. Thus the only penalty that could arise would be by reference to Div 13.

621.


ATC 17917

The respondent Commissioner submitted that CAHPL would have obtained a "scheme benefit" because apart from the scheme, it was reasonable to expect that CAHPL would not deduct the interest under the Credit Facility Agreement but instead would have borrowed at an arm's length interest rate and deducted that lower interest expense. Accordingly, its liability to income tax would be correspondingly greater.

622. The Commissioner further submitted that a reasonable person could conclude that each of CAHPL, CFC and CVX had the relevant dominant purpose. He submitted the factual context amply supported the drawing of such an inference. The matters pointed to by the Commissioner included:

  • (a) The objectives of the leveraging project as articulated by Mr Krattebol, global Treasurer of Chevron , were "to obtain the lowest cost of funding and achieve the Finance function's merger synergy objectives". Mr Dalzell acknowledged that the "merger synergy" objectives referred to by Mr Krattebol included the tax benefits that would arise from the gearing of the balance sheet of the Australasian Business Unit;
  • (b) Mr Lewis said in November 2002 that delays to the transaction meant "there is a real risk that we will not meet the Corporations merger synergy deadline" and that "we are chasing a merger synergy of around US$50MM per annum. Furthermore, we are leaving in excess of USD100,000 cash and earnings on the table each day that this transaction is delayed." He also noted that one of the "benefits" of the CAHPL loan being in AUD was that it would "create an interest rate margin" which "would not be subject to tax in either the US ... or Australia". He also estimated that the USD commercial paper interest rate would be around 2% whereas the AUD interest rate payable by CAHPL would be around 8.5%;
  • (c) Mr Dalzell accepted that the merger synergy could be calculated by multiplying the interest rate margin (or uplift) earned by CFC each year by the Australian corporate tax rate of 30%, and accepted that a USD50 million merger synergy could only be achieved with an AUD interest rate whereas a USD interest rate would result in a USD30 million smaller merger synergy;
  • (d) The tax benefits of interest deductions for the CAHPL group were referred to on a number of occasions by officers of CVX;
  • (e) There was no bargaining or negotiation between CAHPL and CFC in relation to the Credit Facility Agreement.

623. As to the applicant's submission that its dominant purpose was to refinance existing Australian dollar denominated debt, the Commissioner submitted that the purpose of refinancing would equally be achieved if CAHPL refinanced by borrowing at an arm's length interest rate. It would also be achieved more cost effectively. Accordingly, refinancing could not be a purpose of CAHPL in entering into the scheme as opposed to borrowing at an arm's length interest rate. It was not CAHPL's choice of "debt funding" that gave rise to the relevant dominant purpose inference, it was its choice (and that of CFC and CVX) of an interest rate of AUD LIBOR +4.14%.

624. The respondent Commissioner did not dispute that the applicant had a reasonably arguable position.

625. In my opinion, which penalty rule applies depends on which transfer pricing rule is engaged, as in either case the Commissioner conceded that the position adopted was reasonably arguable. The matter for judgment therefore is whether the penalty of 25% has been made out, that is, the Commissioner's position on purpose. If it has not, the penalty of 10% would apply, the applicant accepting that the conditions of s 284-145(2) in Sch 1 to the Taxation Administration Act are satisfied and the Commissioner agreeing that this subsection was satisfied if the purpose requirement of s 284-145(1)(b)(i) was not met.

626. I accept the respondent's submission as to "scheme benefit", that is, that apart from the scheme, it was reasonable to expect that CAHPL would not deduct the interest under the Credit Facility Agreement but instead would have borrowed at an arm's length interest rate and deducted that lower interest expense. The question under s 284-145(1) is therefore whether, having regard to any relevant matters, it is reasonable to conclude that CAHPL (alone or with others) entered into or carried out the scheme, or part of it, with the dominant purpose of CAHPL getting a scheme benefit from the scheme.

627. In approaching the question of purpose I apply
Commissioner of Taxation (Cth) v Ludekens [2013] FCAFC 100; 214 FCR 149 at [243]. There the Full Court said:

In assessing the purpose and evaluating its importance, and whether it is dominant, one must appreciate that it is the scheme in question to which the enquiry is directed, not a general state of affairs other than the scheme. Persons engaged in trade and commerce do so for personal gain. The purpose of all commercial arrangements is, in a broad sense, the making of profit: cf, by way of example,
Federal Commissioner of Taxation v Hart (2004) 217 CLR 216 at [52] and the authorities cited and
Federal Commissioner of Taxation v Consolidated Press Holdings Ltd (2001) 207 CLR 235 at [96]. Here the respondents undoubtedly wished to make profits from the purchase of woodlots and from running a foreign exchange business. They chose the Plan to effect that. Integral to the Plan was that the entities acquiring woodlots on 30 June 2007 ... would obtain scheme benefits from the GST refunds from the purchase of the woodlots and that the Secondary Investors would obtain scheme benefits from tax deductions and tax refunds from their participation. Those are not two purposes: they comprise one purpose ...

628. I do not accept the applicant's submission that the end of the inquiry is that CAHPL's dominant purpose was to refinance its existing Australian dollar denominated debt. I accept the Commissioner's submission that refinancing was not the dominant purpose of the scheme as refinancing could be achieved by borrowing at an arm's length interest rate which CAHPL did not. To limit the scope of matters to be taken into account merely to refinancing is artificially to exclude from consideration the circumstances of that refinancing.


ATC 17918

629. In my opinion, it is reasonable to conclude that CAHPL entered into the Credit Facility Agreement for the dominant purpose of obtaining a "scheme benefit". I refer to my conclusion in [628] above. Further, I rely on some of the factors pointed to by the Commissioner in this respect, those factors being: that the "merger synergy" objectives referred to by Mr Krattebol, the global treasurer of Chevron , included the tax benefits that would arise from the gearing of the balance sheet of the Australasian business unit; that delays to the transaction meant "there is a real risk that we will not meet the Corporations merger synergy deadline" and that:

we are chasing a merger synergy of around US$50MM million per annum. Furthermore, we are leaving in excess of USD100,000 of cash and earnings on the table each day that this transaction is delayed.

He also estimated that the USD commercial paper interest rate would be around 2% whereas the AUD interest rate payable by CAHPL would be around 8.5%. I also rely on the evidence of Mr Lewis set out at [104] above and the evidence of Mr Dalzell set out at [121] above.

630. I add that, contrary to the respondent's submission set out at [622] above, I do not regard Star City Pty Ltd (No 2) as standing for the proposition that the relevant approach is that a reasonable person could conclude that each of CAHPL, CFC and CVX had the relevant dominant purpose: compare Star City Pty Ltd (No 2) at [74] per Dowsett J. I prefer directly to apply the statutory language, which raises the issue whether it is reasonable to conclude that an entity that (alone or with others) entered into or carried out the scheme, or part of it, did so with the sole or dominant purpose of that entity or another entity getting a scheme benefit from the scheme. By reason of that statutory language, I do not accept that the question is solely subjective but I do accept that the purpose of the entity is its subjective purpose: see Star City Pty Ltd (No 2) at [31]-[32] per Goldberg and Jessup JJ and Ludekens at [243] which I have set out at [627] above.

631. For these reasons, I find the penalty is 25% of the scheme shortfall amount, pursuant to s 284-160(a)(ii).

Rulings on evidence

632. There were a number of deferred rulings on objections to evidence on the ground of relevance. It will be apparent from these reasons which affidavits and reports I have found not to be relevant. I do not see it as necessary formally to rule on the balance of these objections as they turn on issues of statutory construction.

Orders

633.


ATC 17919

I shall direct that, within 21 days, the parties bring in agreed short minutes to give effect to my conclusions. Those short minutes are also to deal with costs. Failing agreement, within a further 7 days the parties are to file the competing orders for which they contend and any written submissions in support, the submissions on each side to be limited to three pages.


This information is provided by CCH Australia Limited Link opens in new window. View the disclaimer and notice of copyright.