FEDERAL COURT OF AUSTRALIA

Australian Securities and Investments Commission v Westpac Banking Corporation (No 2) [2018] FCA 751

File number:

VID 282 of 2016

Judge:

BEACH J

Date of judgment:

24 May 2018

Catchwords:

CORPORATIONS Bank bill market trading in prime bank bills Bank Bill Swap Reference rate (BBSW) bank accepted bill futures interest rate swaps cross-currency swaps market manipulation artificial price in traded BBSW referenced products financial market false or misleading appearance in a market false or misleading appearance with respect to price for trading contraventions of ss 1041A and 1041B of Corporations Act 2001 (Cth)

CORPORATIONS non-disclosure of rate set trading practice misleading or deceptive conduct false or misleading representations contraventions of s 1041H of Corporations Act contraventions of ss 12DA and 12DF of Australian Securities and Investments Commission Act 2001 (Cth) unconscionable conduct financial instruments referencing BBSW non-disclosure of rate set trading practice to counterparties unconscionable conduct within the meaning of the unwritten law statutory unconscionability contraventions of ss 12CA, 12CB and 12CC of Australian Securities and Investments Commission Act

CORPORATIONS financial services licence licence obligations whether financial services provided efficiently, honestly and fairly conflict of interest training of representatives contraventions of s 912A(1) of Corporations Act

Legislation:

Australian Securities and Investments Commission Act 2001 (Cth) ss 12BA, 12BAB(7), 12CA, 12CB, 12CC, 12DA, 12DB, 12DF, 12GF

Corporations Act 2001 (Cth) ss 5(2), 9, 760A, 761A, 762C, 763A, 763B, 763C, 763D, 767A, 912A, 1041A, 1041B, 1041E, 1041H, 1041I

Cases cited:

Aberdeen Railway Co v Blaikie Brothers (1854) 1 Macq 461; [184360] All ER Rep 249

Australian Competition and Consumer Commission v C G Berbatis Holdings Pty Limited (2003) 214 CLR 51

Australian Competition and Consumer Commission v Chats House Investments Pty Ltd (1996) 71 FCR 250

Australian Competition and Consumer Commission v Get Qualified Australia Pty Ltd (in liquidation) (No 2) [2017] FCA 709

Australian Competition and Consumer Commission v Leahy Petroleum Pty Ltd (2004) 141 FCR 183

Australian Competition and Consumer Commission v Olex Australia Pty Ltd [2017] FCA 222

Australian Competition and Consumer Commission v TPG Internet Pty Ltd (2013) 250 CLR 640

Australian Competition and Consumer Commission v Turi Foods Pty Ltd (No 4) [2013] ATPR 42-448; [2013] FCA 665

Australian Competition and Consumer Commission v Unique International College [2017] FCA 727

Australian Securities and Investments Commission v Astra Resources plc (2015) 107 ACSR 232; [2015] FCA 759

Australian Securities and Investments Commission v Australian Lending Centre Pty Ltd (No 3) (2012) 213 FCR 380

Australian Securities and Investments Commission v Australia and New Zealand Banking Group Limited [2017] FCA 459

Australian Securities and Investments Commission v Avestra Asset Management Limited (in liq) (2017) 348 ALR 525; [2017] FCA 497

Australian Securities and Investments Commission v Axis International Management Pty Ltd (No 5) (2011) 81 ACSR 631; [2011] FCA 60

Australian Securities and Investments Commission v Camelot Derivatives Pty Ltd (in liq) (2012) 88 ACSR 206; [2012] FCA 414

Australian Securities and Investments Commission v Cassimatis (No 8) (2016) 336 ALR 209; [2016] FCA 1023

Australian Securities and Investments Commission v Fortescue Metals Group Ltd (No 5) (2009) 264 ALR 201; [2009] FCA 1586

Australian Securities and Investments Commission v Narain (2008) 169 FCR 211

Australian Securities and Investments Commission v National Exchange Pty Ltd (2005) 148 FCR 132

Australian Securities and Investments Commission v Sino Australia Oil and Gas Ltd (in liq) (2016) 115 ACSR 437; [2016] FCA 934

Australian Securities and Investments Commission v Soust (2010) 183 FCR 21

Australian Securities and Investments Commission v Whitebox Trading Pty Ltd (2017) 251 FCR 448

Australian Securities Commission v Nomura International PLC (1998) 89 FCR 301

Australian Softwood Forests Pty Ltd v Attorney-General (NSW) ex rel Corporate Affairs Commission (1981) 148 CLR 121

Avoca Consultants Pty Ltd v Millenium3 Financial Services Pty Ltd (2009) 179 FCR 46

Bathurst Regional Council v Local Government Financial Services Pty Ltd (No 5) [2012] FCA 1200

Bilta (UK) Ltd (in liquidation) v Nazir (No 2) [2015] 2 WLR 1168

Briginshaw v Briginshaw (1938) 60 CLR 336

Campomar Sociedad Limitada v Nike International Ltd (2000) 202 CLR 45

Cargill Inc v Hardin (1971) 452 F2d 1154

Carly v Farrelly [1975] 1 NZLR 356

Casaclang v WealthSure Pty Ltd (2015) 238 FCR 55

Commercial Bank of Australia Limited v Amadio (1983) 151 CLR 447

Commonwealth Bank of Australia v Kojic (2016) 249 FCR 421

Demagogue Pty Ltd v Ramensky (1992) 39 FCR 31

Director of Public Prosecutions (Cth) v JM (2012) 27 VR 1

Director of Public Prosecutions (Cth) v JM (2013) 250 CLR 135

Ducret v Chaudhary's Oriental Carpet Palace Pty Ltd (1987) 16 FCR 562

Evans v Braddock [2015] NSWSC 249

Gardam v George Wills & Co Ltd (1988) 82 ALR 415; [1988] FCA 289

Given v C V Holland (Holdings) Pty Ltd (1977) 15 ALR 439; [1977] FCA 33

Hawksford v Hawksford (2005) 191 FLR 173; [2005] NSWSC 463

Gore v Australian Securities and Investments Commission (2017) 249 FCR 167

Ingot Capital Investments Pty Ltd v Macquarie Equity Capital Markets Ltd [No 4] [2006] NSWSC 90

International Litigation Partners Pte Ltd v Chameleon Mining NL (receivers and managers appointed) (2012) 246 CLR 455

Kakavas v Crown Melbourne Limited (2013) 250 CLR 392

Kimberely NZI Finance Ltd v Toreo Pty Ltd [1989] ATPR 46-054; [1989] FCA 400

Kowalczuk v Accom Finance Pty Ltd (2008) 77 NSWLR 205

Lennard's Carrying Co Ltd v Asiatic Petroleum Co Ltd [1915] AC 705

Meridian Global Funds Management Asia Ltd v Securities Commission [1995] 2 AC 500

Miller & Associates Insurance Broking Pty Ltd v BMW Australia Finance Limited (2010) 241 CLR 357

Mills v Federal Commissioner of Taxation (2012) 250 CLR 171

Monroe Topple & Associates Pty Ltd v Institute of Chartered Accountants in Australia (2002) 122 FCR 110

Moulin Global Eyecare Trading Ltd (in liquidation) v Commissioner of Inland Revenue [2014] HKCFA 22

Muschinski v Dodds (1985) 160 CLR 583

North v Marra Developments Limited (1981) 148 CLR 42

North v Marra Developments Ltd [1979] 2 NSWLR 887

Paciocco v Australia and New Zealand Banking Group Ltd (2015) 236 FCR 199

Phipps v Boardman [1967] 2 AC 46

R v Manesseh (2002) 167 FLR 44; [2002] NSWCCA 27

R J Elrington Nominees Pty Ltd v Corporate Affairs Commission (SA) (1989) 1 ACSR 93; [1989] SASC 1941

Re Ku-ring-gai Co-Operative Building Society (No 12) Ltd (1978) 22 ALR 621; [1978] FCA 107

SAP Australia Pty Ltd v Sapient Australia Pty Ltd (1999) 169 ALR 1; [1999] FCA 1821

Tesco Supermarkets Ltd v Nattrass [1972] AC 153

The Bell Group Ltd (in liq) v Westpac Banking Corporation (No 9) (2008) 70 ACSR 1; [2008] WASC 239

Thorne v Kennedy (2017) 250 ALR 1; [2017] HCA 49

Trade Practices Commission v J & R Enterprises Pty Ltd (1991) 99 ALR 325; [1991] FCA 23

Trade Practices Commission v Mobil Oil Australia Ltd (1984) 3 FCR 168

US v Regan, 937 F 2d 823 (2d Cir, 1991)

Wright v Optus Administration & Anor (No 5) [2013] NSWSC 1717

Avgouleas E, The Mechanics and Regulation of Market Abuse: A Legal and Economic Analysis (Oxford University Press, 2005)

Date of hearing:

23, 25, 30, 31 October 2017, 1, 2, 8, 9, 10, 13, 14, 15, 16, 17, 20, 21, 22, 23, 24, 27, 28, 29, 30 November 2017, 1, 4, 5, 6, 11, 12, 13, 14 December 2017

Date of last submissions:

9 January 2018

Registry:

Victoria

Division:

General Division

National Practice Area:

Commercial and Corporations

Sub-area:

Economic Regulator, Competition and Access

Category:

Catchwords

Number of paragraphs:

2539

Counsel for the Plaintiff:

Mr PD Crutchfield QC, Mr PW Collinson QC, Mr M Borsky QC with Mr C Archibald, Ms C Van Proctor and Mr CJ Tran

Solicitor for the Plaintiff:

Johnson Winter & Slattery

Counsel for the Defendant:

Mr MJ Darke SC with Mr JRV Williams and Mr J Burnett

Solicitor for the Defendant:

Allens

ORDERS

VID 282 of 2016

BETWEEN:

AUSTRALIAN SECURITIES AND INVESTMENTS COMMISSION

Plaintiff

AND:

WESTPAC BANKING CORPORATION (ACN 007 457 141)

Defendant

JUDGE:

BEACH J

DATE OF ORDER:

24 MAY 2018

THE COURT ORDERS THAT:

    1. Within 14 days of the date hereof the plaintiff file and serve proposed minutes of orders and short submissions (limited to 3 pages) to give effect to these reasons and for the further conduct of the matter.

    2. Within 14 days of receipt of the plaintiff's proposed minutes and submissions, the defendant file and serve proposed minutes of orders and short submissions (limited to 3 pages) in response.

    3. The further hearing of this proceeding be adjourned to a date to be fixed.

    4. Costs reserved.

Note:    Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011.

REASONS FOR JUDGMENT

BEACH J:

1    The Australian Securities and Investments Commission (ASIC) has brought these pecuniary penalty proceedings against Westpac Banking Corporation (Westpac) concerning its trading over the period 6 April 2010 to 6 June 2012 (the relevant period) in Prime Bank Bills in the Bank Bill Market allegedly to influence the setting of the Bank Bill Swap Reference Rate (BBSW).

2    ASIC's claims are framed:

(a)    first, as contraventions of ss 1041A and 1041B of the Corporations Act 2001 (Cth) (the Corporations Act) involving market manipulation, market rigging and creating a false or misleading appearance with respect to the relevant market(s);

(b)    second, as contraventions of ss 12CA, 12CB and 12CC of the Australian Securities and Investments Commission Act 2001 (Cth) (the ASIC Act), involving unconscionable conduct;

(c)    third, as contraventions of s 1041H of the Corporations Act and ss 12DA, 12DB and 12DF of the ASIC Act, involving misleading or deceptive conduct and misrepresentation; and

(d)    fourth, as contraventions of s 912A of the Corporations Act, involving various breaches of Westpac's financial services licensee obligations.

3    The BBSW is a key benchmark interest rate in Australian financial markets. The purpose and function of the BBSW is to provide an independent and transparent reference rate for the pricing and revaluation of Australian dollar derivative instruments, securities and commercial loans.

4    Trading in Prime Bank Bills on the Bank Bill Market informed the setting of the BBSW, which setting during the relevant period was dependent upon views submitted to the Australian Financial Markets Association Ltd (AFMA) by AFMA-designated BBSW panellists by 10.05 am on each Sydney business day in respect of the yields of the best bids/offers for Prime Bank Bills in each tenor at or around 10.00 am on that day. Westpac was both a BBSW panellist and an AFMA-designated Prime Bank. Almost all trading in the Bank Bill Market took place in the period between approximately 9.55 am and 10.05 am on each Sydney business day, which I will designate as the BBSW Rate Set Window, albeit that there is a dispute between the parties as to whether trading between 10.01 am and 10.05 am informed the setting of BBSW which strictly ought to have been based on relevant trading as at 10.00 am.

5    Bank accepted bills of exchange (Bank Bills) are instruments by which banks may either borrow or lend funds for a short term. By selling a Bank Bill that a bank has accepted, a bank borrows funds. By buying a Bank Bill, a bank lends funds. Bank Bills entitle the holder to receive the face value of the Bank Bill on maturity. But they are traded at a discount to their face value, with the size of the discount representing the amount of interest (or yield) payable on the Bank Bill. The higher the yield, the lower the price of the Bank Bill and vice versa. Prime Bank Bills are Bank Bills where the acceptor is a Prime Bank. I will explain these and other terms later, although for ease of reference I have annexed a glossary to these reasons.

6    During the relevant period, Westpac was a party to a large number of derivative instruments, lending transactions and deposit products (BBSW Referenced Products) pursuant to which:

(a)    an obligation to pay an amount of money was quantified when the BBSW in the relevant tenor was set on a particular day;

(b)    the amount, and in the case of interest rate derivatives whether the amount was payable by Westpac to the counterparty or by the counterparty to Westpac, depended upon the rate at which the BBSW in the relevant tenor set on that day; or

(c)    more generally, Westpac's or the counterparty's rights or liabilities thereunder were influenced by or derived from BBSW.

7    On each Sydney business day in the relevant period, the books in Westpac's Group Treasury included holdings of BBSW Referenced Products in relation to which:

(a)    an obligation to pay an amount of money would be quantified when the relevant BBSW was set on that day; and

(b)    that amount, and, in the case of some of the BBSW Referenced Products, whether the amount was payable by Westpac to the counterparty or by the counterparty to Westpac, depended upon or was influenced by the rate at which the relevant BBSW was set on that day,

and therefore the profit and loss of those books was affected by movement in the BBSW on that day (BBSW Rate Set Exposure). For the moment it is sufficient to note that although Westpac had a separate division, Financial Markets, which had its own BBSW Rate Set Exposure, when I am referring to BBSW Rate Set Exposure generally speaking I am referring to the Group Treasury BBSW Rate Set Exposure. This is because Westpac's traders in Prime Bank Bills had reference only to the latter exposure. Equally relevantly, the Group Treasury BBSW Rate Set Exposure can be taken as a proxy for Westpac's BBSW Rate Set Exposure in terms of considering the motivational influences on Westpac's traders trading in Prime Bank Bills. In these reasons I have also refrained from using terms such as "net" or "gross" exposure. But it should be appreciated that the Group Treasury BBSW Rate Set Exposure on a particular day is the sum of individual books' exposures, some of which will have a short exposure and some of which will have a long exposure on a particular day, such that in aggregate there will either be a long exposure or a short exposure. Let me elaborate.

8    On each trading day in the Bank Bill Market and prior to the BBSW Rate Set Window, Group Treasury was able to and did ascertain its BBSW Rate Set Exposure, which was either:

(a)    a long exposure, meaning that the aggregate of the net earnings of certain books managed by the relevant desks insofar as being referable to the BBSW:

(i)    would be increased in the event that the BBSW was set by AFMA at a higher rate on that day; and

(ii)    correspondingly, would be decreased in the event that the BBSW was set by AFMA at a lower rate on that day; or

(b)    a short exposure, meaning that the aggregate of the net earnings of certain books managed by the relevant desks insofar as being referable to the BBSW:

(i)    would be increased in the event that the BBSW was set by AFMA at a lower rate on that day; and

(ii)    correspondingly, would be decreased in the event that the BBSW was set by AFMA at a higher rate on that day.

9    On days when Group Treasury including its relevant desks had in sum a long exposure, it was receiving BBSW and so benefited by the BBSW setting higher. Conversely, on days when there was in sum a short exposure, it was paying BBSW and so benefited by the BBSW setting lower.

10    ASIC contends that there existed throughout the relevant period, including on specific contravention dates, a material profit-making opportunity for Westpac to use its trading in Prime Bank Bills in the Bank Bill Market in the BBSW Rate Set Window to advantage its positions and exposures under the BBSW Referenced Products. It is said that Westpac developed and pursued a practice in furtherance of enhancing the earnings generated from its BBSW Rate Set Exposure to the disadvantage of counterparties. In particular, ASIC says that on specific days in the relevant period (the contravention dates) and also from time to time during the relevant period, it was the practice of Westpac to trade Prime Bank Bills including to trade newly issued Negotiable Certificates of Deposit (NCDs), which I will include in the expression "Prime Bank Bills", in the BBSW Rate Set Window:

(a)    with the sole or dominant purpose of influencing the level at which the BBSW was set in a way that was favourable to its BBSW Rate Set Exposure; and

(b)    which therefore resulted in yields which did not reflect the forces of genuine supply and demand.

11    I will refer to this practice as the Rate Set Trading Practice. It is said that the Rate Set Trading Practice was in existence and implemented generally throughout the relevant period, and specifically on purchase contravention dates and sale contravention dates. Let me elaborate further.

12    It is said that on each of the purchase contravention dates (6 April (2 different Prime Bank Bill tenors), 30 April, 20 May, 20 September (2 different Prime Bank Bill tenors), 22 September, 1 December and 6 December 2010, 1 March, 4 March, 1 June and 6 June 2011, and 6 June 2012), Westpac knew that a substantial short BBSW Rate Set Exposure existed, and purchased Prime Bank Bills in the Bank Bill Market during the BBSW Rate Set Window with the sole or dominant purpose of lowering or maintaining the rate at which the BBSW for the relevant tenor(s) was set by AFMA on that day and thereby sought to increase its earnings by minimising its BBSW referenced payment obligations.

13    It is said that the purchase contraventions:

(a)    were undertaken by Westpac for the sole or dominant purpose of lowering or maintaining:

(i)    the yield at which Prime Bank Bills were trading at approximately 10.00 am on the relevant day; and

(ii)    the level at which the BBSW was set by AFMA on the relevant day;

(b)    therefore resulted in yields which did not reflect the forces of genuine supply and demand in the Bank Bill Market on the relevant day; and

(c)    had, or were likely to have, the effect of creating an artificial price for trading in three classes of traded BBSW Referenced Products, being relevantly for present purposes limited to bank accepted bill futures contracts (BAB Futures), cross-currency swaps and interest rate swaps.

14    Further, it is said that on each of the sale contravention dates (10 June 2010 and 9 June 2011) Westpac knew that a substantial long BBSW Rate Set Exposure existed, and sold Prime Bank Bills in the Bank Bill Market during the BBSW Rate Set Window with the sole or dominant purpose of raising or maintaining the rate at which the BBSW for the relevant tenor(s) was set by AFMA on that day and thereby sought to increase its earnings accordingly.

15    It is said that the sale contraventions:

(a)    were undertaken by Westpac for the sole or dominant purpose of raising or maintaining:

(i)    the yield at which Prime Bank Bills were trading at approximately 10.00 am on the relevant day; and

(ii)    the level at which the BBSW was set by AFMA on the relevant day;

(b)    therefore resulted in yields which did not reflect the forces of genuine supply and demand in the Bank Bill Market on the relevant day; and

(c)    had, or were likely to have, the effect of creating an artificial price for trading in the three types of traded BBSW Referenced Products referred to above.

16    ASIC contends that such conduct including the implementation of the Rate Set Trading Practice contravened ss 1041A and 1041B of the Corporations Act. It is said that the conduct had, or was likely to have, the effect of creating an artificial price for trading in the said traded BBSW Referenced Products, being BAB Futures, interest rate swaps and cross-currency swaps. It is also contended that such conduct also had, or was likely to have, the effect of creating a false or misleading appearance with respect to the market for, or the price for trading in, those derivative instruments. Because of statutory restrictions, the case concerning ss 1041A and 1041B has been confined to the financial products of BAB Futures, interest rate swaps and cross-currency swaps; the detailed reasons as to why this is so will become apparent much later in my reasons.

17    Further, ASIC contends that the trading on the contravention dates and more broadly the Rate Set Trading Practice, coupled with the non-disclosure of the practice, constituted unconscionable conduct and misleading or deceptive conduct in that such conduct had the actual or likely effect of influencing the calculation of payments required under BBSW Referenced Products to the detriment of counterparties to those instruments or put them at a risk of which they were unaware. I would note that this part of the case involves a broader set of derivative and financial instruments than just BAB Futures, interest rate swaps and cross-currency swaps.

18    It is said that counterparties did not appreciate that Westpac and other participants in the Bank Bill Market could manipulate the BBSW and that they were vulnerable to loss. Further, it is said that the counterparties reposed trust in Westpac to deal fairly with them in respect of the BBSW Referenced Products entered into between them and Westpac. In evidence before me, many of the counterparties deposed to their understanding that the BBSW was a benchmark rate that was not affected by actions, including trading during the BBSW Rate Set Window, intended to cause the BBSW to set at a level other than that which should have obtained under the forces of genuine supply and demand. The counterparties have said that had the true position been disclosed to them, they would have regarded Westpac's conduct in trading directed at influencing the BBSW to its advantage as unfair and contrary to the trust that they had reposed in Westpac.

19    Further, ASIC says that Westpac's failure to disclose its Rate Set Trading Practice to counterparties also gave rise in all the circumstances to a representation to counterparties that BBSW Referenced Products referenced a benchmark rate that was genuine, independent and transparent. It is said that by reason of Westpac engaging in its Rate Set Trading Practice such a representation was false or misleading.

20    Further, ASIC contends that Westpac as the holder of an Australian financial services licence breached statutory obligations as to honesty, fairness, compliance with financial services laws, the taking of steps to manage conflicts of interest and to ensure compliance by its employees with such laws, and also failed to ensure the adequate training and competence of its employees.

21    Now as to these four sets of claims, the present trial has proceeded on the basis of liability only at this stage. The case has been presented by the legal representatives for both ASIC and Westpac with notable efficiency.

22    Now as I have said, ASIC alleges that during the relevant period, including on the contravention dates, Westpac traded with the sole or dominant purpose of influencing where the BBSW set. It is said that I should so find having regard to the contemporaneous communications which directly evidence the existence of that practice, the fact that Westpac had the opportunity to engage in such a practice, the fact that Westpac's traders had a broad discretion and there were no hard limits on Prime Bank Bills trading, the fact that Westpac and its traders had an incentive to engage in such a practice, and the fact that Westpac knew that its trading could affect the BBSW. It is also said that objective trading and exposure data is consistent with the existence of this practice, and that Westpac's alternative explanations for its trading are not credible. Now proof of purpose against a corporation usually resorts to inferential reasoning based upon circumstantial evidence and the exclusion of reasonably open exculpatory explanations. But in the present case ASIC says that there is a significant body of contemporaneous evidence comprised of extensive written communications, including by informal means such as instant chat messaging or email, coupled with voice recordings of an array of internal Westpac communications, from which it is said that I can readily infer that the impugned trading was motivated by the alleged dominant manipulative purpose.

23    Now ASIC bears the onus of proof on each of the elements of the causes of action to the requisite civil standard having regard to the gravity of the matters alleged: s 140(2) of the Evidence Act 1995 (Cth) (the Evidence Act) and Briginshaw v Briginshaw (1938) 60 CLR 336 at 361 and 362. But I would note that ASIC's purpose case does not principally follow the conventional path of demonstrating manipulative purpose from the absence of any rational explanation for the impugned trading. In most cases it has downplayed the objective circumstances of the trading and asked me to principally infer manipulative purpose from the contemporaneous communications. But many of these communications are open to competing interpretations, particularly when understood in context. Now ASIC has sought to construe such communications in ways different from the explanations given by Westpac's witnesses. But in order for ASIC to discharge its onus, I am required to be comfortably satisfied that its interpretation is what was meant and intended by the parties to the communications. Accordingly, where there has been uncertainty as to what was said or what was meant or conveyed after considering the context and all relevant circumstances, I have resolved that uncertainty in Westpac's favour.

24    In summary, I have rejected ASIC's case under ss 1041A and 1041B of the Corporations Act. Although I accept that Westpac through its traders did on four occasions during the relevant period (6 April 2010, 20 May 2010, 1 and 6 December 2010) trade in Prime Bank Bills in the Bank Bill Market with the dominant purpose of influencing the level at which BBSW was set in a way that was favourable to its BBSW Rate Set Exposure, I do not consider that contraventions of ss 1041A and 1041B have been made out. I say that for the following brief reasons. First, the statutory provisions are focused upon effect or likely effect rather than on dominant purpose, although purpose in some circumstances can be used to infer effect or likely effect. Second, to the extent that ASIC has relied upon Director of Public Prosecutions (Cth) v JM (2013) 250 CLR 135 to use dominant purpose in essence and in context as a sufficient condition to establish effect or likely effect, that case is relevantly distinguishable in a number of important respects that I will expand upon later. It dealt with the relatively more straight-forward case of share trading on the Australian Stock Exchange where the product traded, the purpose for trading and the effect of trading were all in the same dimension. But in my case I am dealing with a three-dimensional construct. In the present case, it is important not to elide distinctions that need to be made between three dimensions: (a) the first dimension is the trading of Prime Bank Bills in the Bank Bill Market; (b) the second dimension is the setting of BBSW, which is not of itself a price but rather the trimmed average mid-rate of the observed best bid/best offer for Prime Bank Bills for certain tenors calculated by AFMA; and (c) the third dimension is the pricing of BAB Futures, interest rate swaps and cross-currency swaps in separate financial markets. For the purposes of s 1041A, in the present case the relevant "artificial price" that is to be considered is the price(s) for BAB Futures, interest rate swaps and cross-currency swaps. In simplistic terms, ASIC's case proceeds on the foundation of taking a product in the first dimension and its trading (Prime Bank Bills in the Bank Bill Market) to achieve a purpose in the second dimension (influencing where BBSW set) to produce or likely bring about an artificial price in the third dimension (pricing under BAB Futures, interest rate swaps and cross-currency swaps). For the moment, it is sufficient to say that I am not satisfied that the holding of the relevant dominant purpose on the said four occasions, together with the other evidence, establishes the effect or likely effect of creating or maintaining an artificial price for or under such derivative instruments. And as for s 1041B, I likewise do not consider that establishing such a purpose for trading in Prime Bank Bills establishes a false or misleading appearance with respect to the market(s) in or price for trading of such derivative instruments.

25    Second, I would reject ASIC's more general allegation concerning the existence of the Rate Set Trading Practice during the relevant period. I am not prepared to infer from the isolated instances on the specific four occasions that I have identified or from the totality of the evidence that there was a pattern or system such as to give rise to such a practice. Further, to characterise such isolated examples as in and of themselves constituting such a practice over the relevant period would be to prefer form over substance and to allow the pleader's construct to inappropriately distort the analysis.

26    Third, in my view, on the four occasions that I have identified, Westpac engaged in unconscionable conduct under s 12CC of the ASIC Act (as in force prior to 1 January 2012), that is, under the statutory construct rather than under the unwritten law. But in making this finding I have avoided descending into the "formless void of individual moral opinion" (Paciocco v Australia and New Zealand Banking Group Ltd (2015) 236 FCR 199 at [306] per Allsop CJ citing Deane J in Muschinski v Dodds (1985) 160 CLR 583 at 616 who in turn cited Carly v Farrelly [1975] 1 NZLR 356 at 367 per Mahon J). It is sufficient for me to say that applying Allsop CJ's analysis in Paciocco at [259] to [306] and without dwelling in the paradigm of moral obloquy, Westpac's conduct was against commercial conscience as informed by the normative standards and their implicit values enshrined in the text, context and purpose of the ASIC Act specifically and the Corporations Act generally.

27    Fourth, I have also concluded that by reason of inadequate procedures and training, Westpac contravened its financial services licensee obligations under s 912A(1) of the Corporations Act.

28    For convenience, I have divided my analysis into the following sections:

(a)    Financial and derivative instruments [30] to [142];

(b)    The Bank Bill Market and BBSW [143] to [271];

(c)    Westpac structure and governance [272] to [351];

(d)    Funding and liquidity activities [352] to [400];

(e)    Management of non-interest rate risks [401] to [417];

(f)    Management of interest rate risks [418] to [544];

(g)    Rationales for trading Prime Bank Bills [545] to [616];

(h)    Dynamics of the Bank Bill Market and trading strategies [617] to [715];

(i)    Opportunity and incentives to manipulate [716] to [793];

(j)    Susceptibility of the Bank Bill Market and BBSW to manipulation [794] to [856];

(k)    Reliability of Westpac's witnesses [857] to [924];

(l)    Contravention dates and alleged practice [925] to [1677];

(m)    Trading and price impacts [1678] to [1885];

(n)    Market manipulation (ss 1041A and 1041B) [1886] to [2126];

(o)    Unconscionable conduct [2127] to [2253];

(p)    Misleading or deceptive conduct [2254] to [2334];

(q)    Financial services licensee obligations [2335] to [2533];

(r)    Conclusions [2534] to [2539].

29    It has been necessary to deal with these topics in this sequence and with this cumulative development in order to avoid an analysis based only upon superficial snapshots of the key communications and trading behaviour referable to the specific purchase and sale contravention dates. That said, the linear sequence of these reasons cannot fully capture in text the iterative and interactive nature of the analysis which has been necessary to ensure that findings on some aspects are informed by my analysis and findings on other aspects.

FINANCIAL AND DERIVATIVE INSTRUMENTS

30    Let me at this point make some introductory remarks concerning various financial instruments and the use of BBSW as a reference rate. Any capitalised expressions have corresponding definitions set out in this section or in the glossary to these reasons.

(a)    Bank Bills

31    A Bank Bill is a bill of exchange, as defined in s 8 of the Bills of Exchange Act 1909 (Cth), which has been accepted by a bank and bears the name of the accepting bank as acceptor, and which obliges the bank to pay the face value of the bill to the holder of the bill on the date that it matures.

32    Bank Bills trade at a price calculated by discounting the face value of the Bank Bills to reflect an interest rate or yield paid by the Bank Bills. The price (present value) of a Bank Bill can be expressed by the following formula:

Where:

PV =

Present value (price)

FV =

Face value

r =

Per annum yield to maturity

t =

Time to maturity in days

33    Where the yield used to price a Bank Bill is BBSW (in cases say where the Bank Bill is bought or sold outside of the Bank Bill Market and it is separately agreed), the formula for the price (present value) of the Bank Bill is:

34    Further, as I have said, there is a category of Bank Bills known as Prime Bank Bills where the acceptor is a Prime Bank; a Prime Bank is one designated as such by AFMA as I will explain later.

(b)    3 month Bank Accepted Bill Futures (BAB Futures)

35    Under a BAB Futures contract one party agrees to buy a 3 month Prime Bank Bill at an agreed yield (and thus price) on the expiry date of the contract and the other party agrees to sell a 3 month Prime Bank Bill at an agreed yield (and thus price) on the expiry date of the contract. For convenience, in these reasons I will refer to the different tenors of Prime Bank Bills in month(s) rather than days. Likewise, references to the applicable BBSW will be described in month(s). The glossary to my reasons sets out any necessary conversion.

36    BAB Futures contracts are standardised contracts with the following terms:

(a)    the contracts relate to the delivery of 3 month Prime Bank Bill(s) with a face value of $1,000,000;

(b)    the contracts expire on one of four dates per year being the Thursday before the second Friday in each of March, June, September and December; the expiry date is known as the close out date;

(c)    the contracts are "deliverable", meaning that if the BAB Futures contract is held to expiry then one party must actually deliver the 3 month Prime Bank Bill(s) to the other party, who must buy the Prime Bank Bill(s); and

(d)    the contracts expire at 12.00 pm noon on the expiry day and the delivery occurs the following day (so delivery occurs on the second Friday in each of March, June, September and December); the delivery date is known as the settlement date.

37    BAB Futures contracts are traded on an exchange, which was known as the Sydney Futures Exchange (SFE) during the relevant period until 31 July 2010 and then from 1 August 2010 known as ASX24. A party can enter into BAB Futures contracts with a term (time left to the expiry date) of up to five years. BAB Futures contracts are quoted in terms of yield per annum (in percentage terms) deducted from 100. So a BAB Futures contract quoted at "96" means that the parties to the BAB Futures contract agree to buy (or sell) a $1,000,000 3 month Prime Bank Bill on the expiry date at a price that represents a yield of 4% (that is, the "price" of the 3 month Prime Bank Bill will be discounted to reflect a yield of 4%).

38    The BAB Futures contract price reflects:

(a)    the market's view of what the price of 3 month Prime Bank Bills will be at the date of expiry of the BAB Futures contract; and

(b)    the current interest rate that applies to the period until the BAB Futures contract expiry date, which determines the cost of holding the BAB Futures contract.

39    The profit or loss made on a BAB Futures contract can be understood by considering a BAB Futures contract which is held to expiry.

40    On the expiry date the party who has agreed to sell 3 month Prime Bank Bills will need to hold or obtain 3 month Prime Bank Bills in order to deliver these to the other party. If:

(a)    the price of Prime Bank Bills falls (and thus the yield on Prime Bank Bills, and the BBSW, increases) to a level which is below the level implied by the BAB Futures contract price at the time it was entered into (meaning that the Prime Bank Bill yield is above the original BAB Futures contract implied yield), then the selling party will be paying less for the 3 month Prime Bank Bills than they will receive selling them (delivering them) pursuant to the BAB Futures contract (making a gain); and

(b)    the price of Prime Bank Bills increases (and thus the yield on Prime Bank Bills, and the BBSW, falls) to a level which is above the level implied by the BAB Futures contract price at the time it was entered into (meaning that the Prime Bank Bill yield is below the original BAB Futures contract implied yield), then the selling party will be paying more for the 3 month Prime Bank Bills than they will receive selling them (delivering them) pursuant to the BAB Futures contract (making a loss).

41    On the expiry date, the party who has agreed to buy 3 month Prime Bank Bills will need to take delivery of the 3 month Prime Bank Bills once delivered. If:

(a)    the price of Prime Bank Bills falls (and thus the yield on Prime Bank Bills, and the BBSW, increases) to a level which is below the level implied by the BAB Futures contract price at the time it was entered into (meaning that the Prime Bank Bill yield is above the original BAB Futures contract implied yield), then the purchasing party will be paying more for the 3 month Prime Bank Bills than the market price they could sell them at or otherwise buy them at (making a loss); and

(b)    the price of Prime Bank Bills increases (and thus the yield on Prime Bank Bills, and the BBSW, falls) to a level which is above the BAB Futures contract price at the time it was entered into (meaning that the Prime Bank Bill yield is below the original BAB Futures contract implied yield), then the purchasing party will be paying less for the 3 month Prime Bank Bills than the market price they could sell them at or otherwise buy them at (making a gain).

42    Only approved Prime Bank Bills were eligible for delivery. Approved Prime Bank Bills had to, of course, be accepted by a Prime Bank, have a face value of $1 million, mature 85 to 95 days from the settlement date and be classified as "early" month paper; I will explain this concept later.

43    In practice, a significant majority of BAB Futures contracts are not held to expiry, instead they are "closed out". A party can "close out" a BAB Futures contract on the ASX24 exchange by entering into an equivalent BAB Futures contract, but taking the opposite leg (e.g. agreeing to buy the 3 month Prime Bank Bills rather than sell, or sell rather than buy). So, for example a BAB Futures contract which is purchased three months out from expiry can be closed out two months from expiry by entering into an opposite position in another BAB Futures contract which is also two months out from expiry.

44    Further, trading in BAB Futures contracts involves maintaining a margin account. On each day that the BAB Futures contract is held, the margin account is adjusted to reflect the change in the BAB Futures contract price (based on where other BAB Futures contracts with the same expiry date are trading). The effect is that the gain or loss from the BAB Futures contract is not all realised on the expiry date, but is instead reflected in daily changes to the margin account over the term of the BAB Futures contract.

45    This daily adjustment is referred to as the Daily Settlement Price. The Daily Settlement Price was during the relevant period determined under the applicable exchange Operating Rules and Procedures as follows:

(a)    from 6 April 2010 until about 31 July 2010, pursuant to rule 1.9 "Daily Settlement Price" of the SFE Operating Rules; and

(b)    from 1 August 2010 until 6 June 2012, pursuant to item [2500] "Determination of Daily Settlement Price" in the ASX24 Operating Rules Procedures.

46    ASIC says that the Daily Settlement Price on a final trading day (prior to expiry) is referred to as the "Expiry Settlement Price", but no provision in the Operating Rules made that express. I will return to the question of the significance of the Daily Settlement Price later.

47    Now although trading in BAB Futures is conducted anonymously on-market via ASX24, which functions as a clearing house for BAB Futures, trading in the BAB Futures market requires a participant to maintain a margin account (as I have said) with the ASX. As trading in BAB Futures is a leveraged transaction, parties do not pay (or receive) the full value of the contract at the time the transaction is entered into. Instead, both buyers and sellers of BAB Futures pay an initial margin and are also liable for daily variation margin calls:

(a)    The initial margin (paid by both the buyer and the seller) covers the maximum probable one-day move in the rate of the futures contract, as assessed by ASX Clear (a wholly owned subsidiary of the ASX). ASX Clear sets the initial margin for BAB Futures contracts with reference to the volatility of the yields for BAB Futures. It does not do so by reference to volatility in BBSW. ASX Clear uses the "SPAN" methodology to calculate margin requirements, which references historical data in relation to movements in the yields of BAB Futures contracts within the period of approximately the preceding three months.

(b)    The variation margin is an amount that either is paid, or received, by a party to reflect a movement in the value of its BAB Futures position on a mark-to-market basis. On each day that the BAB Futures contract is held, the margin account is adjusted to reflect that change in the BAB Futures contract rate (based on where other BAB Futures contracts with the same expiry are trading at 4.30 pm on each trading day). In times of high volatility, ASX Clear may also call intra-day margins.

48    The margin posted by a party entering into a BAB Futures contract will be returned when the party "closes out" its position. Now as I have said, a party can "close out" a BAB Futures contract by entering into an BAB Futures contract with the same expiry date but taking the other leg, that is, agreeing to buy a 3 month Prime Bank Bill rather than sell, or sell rather than buy.

49    Further, a party can hold a BAB Futures contract to expiry. If a party is "short" BAB Futures at the close out date, that party is obliged to deliver 3 month Prime Bank Bills on the settlement date to the party who is "long" BAB Futures on expiry. The 3 month Prime Bank Bill must be sold and purchased at the rate agreed when the BAB Futures were initially entered into.

50    Let me say something at this point in terms of the phenomenon of convergence, although I will return to it much later in my reasons. This phenomenon is relevant to the question of whether there is a causal relationship between changes in Prime Bank Bill yields and changes in BAB Futures rates, and if so what. That in turn is relevant to the question of whether the price for trading of BAB Futures was affected by changes in BBSW arising from trading in the Bank Bill Market, which itself is one of the questions that I need to consider under ss 1041A and 1041B for one of the traded BBSW Referenced Products being BAB Futures. For the moment, let me just note the following concerning convergence.

51    Typically, the close out date is the only point in time at which the rate for Prime Bank Bills and the rate for BAB Futures may converge. At any other time, the rate for trading BAB Futures is reflective of the market's view of where 3 month yields or perhaps BBSW will be at the close out date. While each market participant's view of this will be affected by whatever indicia they consider to be important, in general terms the market's view of where interest rates will be at the expiry of BAB Futures is usually informed by participants' expectations of future movements to the cash rate, which, in turn, is affected by a number of other factors including the Reserve Bank of Australia's (RBA) monetary policy stance, global economic events and credit expectations of banks. If the majority of participants trading in the BAB Futures market expect interest rates to decrease, then in the ordinary course the implied yield at which BAB Futures are trading decreases and therefore the rate for BAB Futures increases. Conversely, if participants in the BAB Futures market expect interest rates to increase, then in the ordinary course the implied yield at which BAB Futures are trading increases and therefore the rate for BAB Futures decreases.

52    Now while 3 month yields or perhaps BBSW and the BAB Futures rate should in theory converge on the close out date, there is often a discrepancy between the two due to the market dynamics around the close out date. This discrepancy can be caused by participants in both the BAB Futures market and the Bank Bill Market having a similar view on future movements on interest rates, but with a different result.

53    So, if foreign investors in BAB Futures take the view that the RBA will reduce interest rates as a result of an economic downturn, this will probably cause BAB Futures to trade at a lower implied yield (and a higher rate). However, at the same time, the Prime Banks in the domestic Bank Bill Market may seek to raise funds as a result of economic downturn by issuing Prime Bank Bills, and that issuance may increase the yield at which Prime Bank Bills are trading.

54    In addition, whereas 3 month BBSW is published at 10.15 am, BAB Futures expire at midday. So, a new piece of economic data or financial information may become accessible to the market between the time at which BBSW sets and the time the BAB Futures market closes which affects the rate at which BAB Futures trade and causes the BAB Futures rates to move away from the three month BBSW published at 10.15 am.

55    Further as to the question of convergence, Mr Simon Masnick, an officer in Westpac's Financial Markets division, gave evidence to the following effect:

(a)    At any point in time prior to the close out date, the BAB Futures rate should represent the market's consensus view of the likely yield of a Prime Bank Bill at the close out date. Typically, the market has greater certainty as to the likely yield of a Prime Bank Bill as the close out date draws nearer (particularly after the RBA's cash rate announcement in the month of the close out date). However, the close out date is the only point in time at which the rate for 3 month Prime Bank Bills and the rate for BAB Futures are likely to converge, and even this is subject to matters which may cause a mismatch at expiry.

(b)    The BAB Futures rate and the rate of the underlying asset (that is, a 3 month Prime Bank Bill) usually start to converge in the days leading up to the close out date. However, this convergence does not mean that, on any particular day, the level at which 3 month BBSW sets at 10.00 am determines (or even influences) the BAB Futures rate for that day. Momentary or daily movements in the level at which 3 month BBSW sets on any given day have little to no impact on the BAB Futures rate on that same day (particularly if the close out date is not within a few days of trading in BAB Futures). Primarily, this is because the market's view of future movements in the cash rate is the factor which most directly informs both the BAB Futures rate and the level at which 3 month BBSW sets on any given day. For example, Mr Masnick regularly observed instances where important economic information such as Consumer Price Index (CPI) figures or employment data was released at around 11.30 am (that is, after 3 month BBSW set on that particular day). When this occurred, the BAB Futures rate would change prior to, and independently of, any change in the level of BBSW, which would not take place until the following day.

(c)    In addition, the BAB Futures rate will at any given point in time have additional information built into it which is not yet reflected in, and not relevant to, the rate of Prime Bank Bills (and so the level of BBSW). That is because BBSW is the market's assessment of the correct rate for Prime Bank Bills today, while the BAB Futures rate is the market's assessment of the rate of Prime Bank Bills at a future date (that is, the close out date).

(d)    The most significant factors that influence the rate at which BAB Futures trade are the following. First, the market's perception of what the level of BBSW will be at the date of the BAB Futures expiry. Second, the term of the BAB Futures contract. That is, whether a participant is buying the BAB Future which expires in less than three months' time, or the following BAB Future which expires in less than six months' time. Third, market volatility which might affect the credit spread between the cash rate and BBSW. Fourth, whether a party is a buyer or a seller.

(e)    Further, Mr Masnick observed that the volume of trading in the BAB Futures market is significantly greater than the amount of trading in the Bank Bill Market. Typically, parties trading in BAB Futures do not intend to take delivery of, or issue (in the case of Prime Banks), Prime Bank Bills into the futures close out. Instead, many parties trade BAB Futures so as to hedge interest rate swaps or speculate on likely cash rate and BBSW outcomes into the future. Accordingly, he observed that the trading behaviour of some market participants in the BAB Futures market often had little or no regard to the level at which BBSW set on a given day. For example, speculative trading in significant volumes commonly had a material effect upon the rate at which BAB Futures traded irrespective of any daily movements in the level of 3 month BBSW.

(f)    Mr Masnick had never referred to nor heard of the 3 month AUD Bank Bill 5.00 pm New York closing time yield (the Bloomberg Closing Yield), on which ASIC's expert witness before me, Dr Tiago Duarte-Silva, relied as a key integer for his measurement of the change in Prime Bank Bill yields on any day. Specifically, Dr Duarte-Silva posited that the change in the yield in Prime Bank Bills on each date could be calculated by reference to the difference between the level at which BBSW sets and the Prime Bank Bill yield at the 5.00 pm New York (which equates to 7.00 am to 9.00 am in Australian Eastern time depending on the date). But ASIC has now abandoned any reliance upon the evidence of Dr Duarte-Silva. No doubt this was due to problematic aspects of his statistical methodology which were exposed during one of the three concurrent evidence sessions that I held. But in any event in Mr Masnick's experience, the Bloomberg Closing Yield was not a source of information used by traders to trade in BAB Futures or Prime Bank Bills. I will return to Dr Duarte-Silva's evidence later, notwithstanding ASIC's abandonment of it. It is forensically relevant as far as I am concerned to show what could not be proved.

(c)    Interest rate swaps

56    An interest rate swap is an agreement between two parties to exchange streams of single currency cash-flow based on a notional amount for a set period. The most common type of interest rate swap is a "fixed for floating rate swap". Under a "fixed for floating rate swap" there are two legs. One leg of the swap has payment obligations calculated with reference to a floating interest rate (most commonly the BBSW). The swap party taking this leg of the swap has obligations to pay to the other swap party cash-flows equal to the agreed notional amount multiplied by the agreed floating interest rate. The other leg of the swap has payment obligations calculated with reference to an agreed fixed interest rate (known as the "swap rate"). The swap party taking this leg of the swap has obligations to pay to the other party cash-flows equal to the agreed notional amount multiplied by the swap rate.

57    Rather than both parties making payments, the parties exchange cash payments based on the net difference between the two cash-flows (referred to as the "Swap Cash Settlement Amount"). These net interest payments are made at regular intervals (the payment days are referred to as "reset dates"). The intervals (also known as "interest accrual periods") may be set monthly, quarterly, annually, or at another time period agreed between the parties. Typically, the interest accrual period is the same as the relevant tenor of the floating rate. So, for example, an interest rate swap which resets quarterly will use the 3 month BBSW as the floating leg reference rate. Payments only reflect net interest payments. No "principal" payments of the notional amount are made by the parties.

58    As I have said, the "reset date" is the date on which the floating rate is observed, and which typically occurs at the beginning of the interest accrual period. The reset date is usually different from the payment date, which is the date on which the net payment for a particular interest accrual period is made.

59    There is usually a difference between the date that the parties agree to enter into the interest rate swap (the "transaction date") and the date the interest rate swap first resets (the "commencement date"). Often, the commencement date will be the day after the transaction date. This is referred to by traders as being "T + 1". If, for example, the commencement date was three days after the transaction date, this would be referred to as "T + 3". Typically, subsequent reset dates are determined with reference to the commencement date. For instance, if the interest rate swap resets monthly, the second reset date will be the date corresponding to the commencement date in the following month.

60    The formula for calculating the cash-flow obligations of both swap counterparties to a fixed for floating rate swap on a reset date can be expressed as follows:

Payment obligation of the party paying the fixed interest rate leg of the swap

Where:

Afix =

Amount required to be paid by the party paying the fixed interest rate leg of the swap

Notional =

Notional amount

d =

Day count (between reset dates)

Rfix =

Fixed rate

Payment obligation of the party paying the floating interest rate leg of the swap

Where:

Afloat =

Amount required to be paid by the party paying the floating interest rate leg of the swap

Notional =

Notional amount

d =

Day count (between reset dates)

Rfloat =

Floating rate

The net cash-flow payment (Swap Cash Settlement Amount)

Which can also be written as follows by expanding the equation:

Where:

Anet =

Swap Cash Settlement Amount

61    The floating rate in an interest rate swap is most commonly the BBSW. In these swaps:

(a)    the "Rfloat" input in the equations above is the BBSW;

(b)    the payment obligation of the party paying the floating rate is directly referenced to the BBSW; and

(c)    the Swap Cash Settlement Amount (Anet) is directly referenced to the BBSW.

62    As to pricing, a matter about which I will say something more later, ASIC contends that the price for trading, within the meaning of s 1041A of the Corporations Act, in an interest rate swap:

(a)    consists of the obligations to exchange periodical cash payments undertaken by both parties to the swap at the time of acquiring an interest rate swap; and

(b)    is quantified in monetary terms on each reset date using the formula to calculate the Swap Cash Settlement Amount set out above.

63    ASIC contends that where the BBSW is used as the floating rate in an interest rate swap, the BBSW is an input into the formula for calculating the Swap Cash Settlement Amount and therefore the BBSW is an input into the price for trading of the interest rate swap. I would note at this point that Westpac's thesis as to the "price" for trading of an interest rate swap focuses more on the agreement or stipulation of the fixed rate referable to the fixed rate leg.

64    In relation to exposure, for interest rate swaps which used the BBSW as the relevant floating rate, an exposure to the BBSW rate would be realised at each reset date. As set out above, the formulae for calculating the obligations of the swap party paying the floating rate and the Swap Cash Settlement Amount which both parties were obligated to pay directly referenced the BBSW.

65    As for profit or loss, the profit or loss to the parties to an interest rate swap is realised on the reset dates. As set out above, the formula to calculate the Swap Cash Settlement Amount on a fixed for floating swap is:

Which can also be written:

66    Where the floating rate used is the BBSW, this equation becomes:

67    Accordingly, where the floating rate in a fixed for floating interest rate swap is the BBSW, the rate at which the BBSW sets on the reset dates will determine which swap party needs to make a payment and the value of the payment (being the Swap Cash Settlement Amount). If the BBSW increases, the swap party paying the floating leg of the swap will have to pay more (or will receive less) and the party paying the fixed leg of the swap will have to pay less (or will receive more). If the BBSW decreases, the swap party paying the floating leg of the swap will have to pay less (or will receive more) and the party paying the fixed leg of the swap will pay more (or will receive less).

68    There are typically four types of counterparties with which one would enter into an interest rate swap relevant to the present context:

(a)    Other banks and financial institutions, including foreign banks, that primarily use interest rate swaps to hedge their exposure to movements in interest rates as a result of their borrowing and lending activities.

(b)    Corporate customers, that commonly enter into interest rate swaps as a means of cash flow management in order to convert a floating interest rate exposure (for example, under a finance facility with a bank) to a fixed rate exposure or vice versa.

(c)    Institutional customers, such as superannuation funds or life insurance funds, that use interest rate swaps to manage the interest rate exposure of their portfolios of assets and liabilities.

(d)    Speculative investors, such as hedge funds, that might use interest rate swaps to take an outright position in relation to the direction of movements in interest rates in order to generate a profit if that position is realised.

69    Let me turn to some other matters concerning the details of an interest rate swap. One aspect that is of particular relevance is the determination of the amount of the fixed rate on the fixed rate leg of the swap, which is relevant to an issue that I will discuss much later concerning the "price" of an interest rate swap.

70    An interest rate swap is a zero net present value contract on the date on which the transaction is entered into, with the fixed rate leg set so that the payment obligations of the party paying the fixed rate (and receiving the floating rate) is equal to the net present value of the floating rate payment obligations of the party paying that rate. For example, assume that a party enters into a one year "fixed for floating" interest rate swap with a notional principal of $1 million where the reference rate for the floating leg of the swap is 3 month BBSW. In this example, the value of the fixed leg of the swap will equal the net present value of the four quarterly payments on the 3 month BBSW on the notional principal of $1 million. Accordingly, while the payment obligations of the party paying the fixed interest rate are known with certainty and the floating rate is to be determined on the four future rate reset dates, the parties expect them (notionally at least) to have the same total net present value for the term of the swap.

71    After the commencement date of the swap, however, the respective payment obligations of the parties to the swap will change as a result of movements in interest rates. While the fixed rate leg does not change over the term of the swap, the parties' respective payment obligations over the term of the swap will fluctuate depending on movements in the level at which the floating reference rate sets on the relevant rate reset dates. Ultimately, movements in the level of the floating reference rate will affect the sum of the parties' payment obligations over the term of the swap. Similarly, throughout the term of the swap, the mark-to-market value of the swap will change to the extent of subsequent movements of expectation of the level of the floating reference rate. The mark-to-market value of the swap is recalculated on a daily basis and a party may choose to crystallise their position (either at a gain or a loss) by selling an interest rate swap at the mark-to-market rate.

72    In practice, the amount of the fixed rate leg of the swap is determined by reference to two components:

(a)    the swap yield curve, where the term of the interest rate swap will determine the relevant point on the swap yield curve; and

(b)    a margin or "spread", where the margin which is applied essentially reflects the credit risk of a specific counterparty plus other transaction costs.

73    Mr Masnick for Westpac gave evidence to the following effect. Westpac regularly reviews the shape of its swap yield curve so as to ensure its traders have the best tools available to trade interest rate swaps effectively. Westpac's swap yield curve is built using a number of sources of information. During the relevant period, the swap yield curve was built using the following sources of information:

(a)    swaps with a term of up to two years are set by reference to the BAB Futures yield;

(b)    swaps with a term of longer than two years are set by reference to the Commonwealth government bond futures yield plus a margin (which is referred to as the Exchange For Physical (EFP) margin);

(c)    the Overnight Indexed Swap (OIS) rate, which was used to discount future cash-flows on the interest rate swap;

(d)    a credit valuation adjustment to account for the credit risk of counterparties; and

(e)    Westpac's profit margin (including the cost of capital and other related expenses).

74    In addition to the swap yield curve, Westpac traders during the relevant period would commonly have regard to the following information when trading in interest rate swaps:

(a)    their assessment of the likely future movements in interest rates;

(b)    the trading in other financial instruments which are set by reference to market views in relation to future movements in the cash rate, such as interest rate futures and BAB Futures;

(c)    the current level of trading, and recent market trades, in interest rate swaps;

(d)    the particular credit worthiness of the specific counterparty with which Westpac was proposing to enter into the swap, which is determined by Westpac's credit risk team;

(e)    whether an exchange margin (or other credit risk mitigant) can be agreed between the parties; and

(f)    other factors specific to a particular counterparty (for example, whether the counterparty is a strategic client of Westpac).

75    Because of the variety of information that was taken into account when trading interest rate swaps, during the relevant period there was not, at any point in or period of time, one single rate at which Westpac's traders would enter into an interest rate swap. Each transaction involved considering the full range of factors listed above, which meant that the rate at which interest rate swaps entered into by Westpac at or around the same time would differ as a result of the application of these factors to each transaction.

(d)    Cross-currency swaps

76    A cross-currency swap is a variation on an interest rate swap. It is an agreement between two parties to exchange:

(a)    a principal amount and associated interest payments denominated in one currency; for

(b)    a principal amount and associated interest payments denominated in another currency.

The parties pay their respective interest payments on dates referred to as "reset dates" and typically exchange the principal amount at the swap creation and maturity date (either in the two currencies or in one currency based on an agreed exchange rate). This is, of course, different to an interest rate swap, where no principal amounts are exchanged.

77    There are numerous types of cross-currency swaps. A simple example is a cross-currency swap where the parties agree to swap:

(a)    an Australian dollar ("AUD") principal plus interest payments determined by an AUD interest rate (such as BBSW); with

(b)    a principal of another currency (currency X) plus interest payments determined by a X currency interest rate (either fixed or floating as agreed between the parties).

78    A common type of cross-currency swap called a "floating for floating" basis swap involves exchanging a floating rate in one currency (say, BBSW) for a floating rate in another currency (say, the London Interbank Offered Rate (LIBOR)).

79    Cross-currency swaps involving AUD commonly use the BBSW as the AUD interest rate.

80    The formula for calculating the cash-flow obligations of both swap counterparties on a reset date (other than the creation or maturity date) can be expressed as follows:

Payment obligation of the party paying the AUD leg of the swap

Where:

AAUD =

Amount required to be paid by the party paying the AUD leg of the swap

AUD principal =

AUD principal amount

RAUD =

Australian interest rate

d =

Day count (between reset dates)

Payment obligation of the party paying the foreign currency leg of the swap

Where:

AX =

Amount required to be paid by the party paying the X currency leg of the swap

X principal =

X currency principal amount (foreign currency amount)

RX =

X currency interest rate

d =

Day count (between reset dates)

81    The rate in a cross-currency swap where one leg uses AUDs is commonly BBSW. In these cross-currency swaps:

(a)    the "RAUD" input in the equation above is BBSW;

(b)    the payment obligation of the party paying the AUD leg of the swap is directly referenced to BBSW; and

(c)    if the payments on the reset dates are netted out (using an exchange rate to convert the non-AUD payment to AUDs) then the net payment amount is directly referenced to BBSW.

82    As for pricing, on which I will say something more later, ASIC contends that the price for trading, within the meaning of s 1041A of the Corporations Act, in a cross-currency swap:

(a)    consists of the obligations to exchange periodical cash payments undertaken by both parties to the swap at the time of acquiring a cross-currency swap; and

(b)    is quantified in monetary terms on the creation of the cross-currency swap and on each reset date, using (on each reset date) the formula set out above.

83    ASIC contends that where the BBSW is used as the Australian interest rate in a cross-currency swap the BBSW is an input into the formula for calculating the payment amount paid by the party making AUD payments on each reset date and the maturity date and therefore the BBSW is an input into the price for trading of the cross-currency swap. Contrastingly, Westpac says that the price for trading is more aptly focused, in the present context, on the margin added to the AUD floating rate (say, BBSW) where under the cross-currency swap one has a floating interest rate on each side, say on the AUD side, BBSW, and on the USD side, LIBOR. But where there is a fixed rate on one side, then the fixed rate would determine the price for the trade.

84    In relation to exposure, for cross-currency swaps which used the BBSW as the relevant floating rate on the AUD leg, an exposure to the BBSW rate would be realised at each reset date. As set out above, the obligations of the swap party paying the AUD interest rate and the Swap Cash Settlement Amount which both parties are obligated to pay (if netted off) directly referenced the BBSW. The amount used by Westpac to calculate its BBSW Rate Set Exposure arising from a cross-currency swap was the size of the notional amount upon which the swap obligations were based.

85    As to profit or loss, the profit or loss to the parties to a cross-currency swap is realised on the reset dates.

86    As set out above, the formula for calculating the cash-flow obligations of both swap counterparties (to a swap involving the AUD) on a reset date can be expressed as follows:

Payment obligation of the party paying the AUD leg of the swap

Payment obligation of the party paying the foreign currency leg of the swap

87    Where the AUD rate used is the BBSW, the rate at which the BBSW sets on the reset dates will determine the payment obligation of the party paying the AUD leg of the swap. Where the two swap payments are netted out (by applying an agreed exchange rate), which swap party needs to make a payment and the value of the payment (being the Swap Cash Settlement Amount) will also be dependent on where the BBSW sets on the relevant reset date.

88    Similar to interest rate swaps, there is also usually a difference between the date that the parties agree to enter into the cross-currency swap (the "transaction date") and the date the cross-currency swap first resets (the "commencement date"). Often, the commencement date will be the day after the transaction date. This is referred to by traders as being "T + 1". If, for example, the commencement date was three days after the transaction date, this would be referred to as "T + 3". Typically, subsequent reset dates are determined by reference to the commencement date. For instance, if the cross-currency swap resets monthly, the second reset date will be the date corresponding to the commencement date in the following month.

89    In terms of matters relevant to the trading of a cross-currency swap and the margin added to the AUD floating interest rate, where one has a floating interest rate on each side, the following may be noted.

90    Theoretically, the net present value of a cross-currency swap is zero because, on the date on which the transaction is entered into, the present value of interest payment obligations on a principal amount determined by an Australian interest rate equals the present value of interest payment obligations on the same principal amount determined by a foreign interest rate. In practice, a margin is added to the Australian interest rate which is due to supply and demand factors. Accordingly, the Australian interest rate of a cross-currency swap is expressed as a floating rate plus a margin. The margin may, inter-alia, reflect the particular credit risk of the individual counterparty. This credit risk is greater for cross-currency swaps than for interest rate swaps because the former requires the exchange of principal amounts whereas the latter involves only the exchange of the difference in the rates on the notional amount.

91    Cross-currency swaps are generally used in the Australian market by large corporate customers to convert interest rate risk from foreign currency raised in offshore wholesale debt markets (typically in the United States) into an exposure to domestic interest rates. For example, assume a party (such as a large Australian company) issues a USD 1 million bond in the United States with a term of two years and receives an equivalent amount of USDs, the floating interest rate on which is referenced to 3 month LIBOR. A cross-currency swap allows that party to convert a floating exposure to 3 month LIBOR to an exposure to 3 month BBSW. As a cross-currency swap involves the exchange of a principal amount, the large Australian company also receives the AUD equivalent of USD 1 million (calculated by reference to an agreed spot foreign exchange rate) with which it can meet its liabilities payable in AUDs. At the maturity date the parties re-exchange the initial principal at the same agreed exchange rate, with which the party receiving the USD 1 million can pay its liabilities with respect to the bondholders.

92    In Australia, the most common type of cross-currency swap entered into is an AUD/USD basis swap. It is common for Australian corporations to issue foreign currency debt into offshore bond markets which can absorb larger debt issuances at competitive levels. As a result, there is generally greater demand for USD floating reference cross-currency swaps in the Australian market, which affects the prevailing rate for AUD/USD cross-currency swaps. In this context, given the imbalance between payment flows, the basis point margin added to the Australian reference benchmark rate incentivises parties to take the USD floating rate leg of the swap because the basis point margin for entering into the swap is paid by the party paying interest determined by reference to Australian interest rates only.

93    During the relevant period, Westpac traders would have regard to the following information to determine the appropriate rate(s) and any relevant margin at which to enter into a cross-currency swap:

(a)    the rate for cross-currency swaps observed on broker screens;

(b)    the particular flows within a particular trading book of Westpac; and

(c)    Westpac's profit margin including a credit spread to reflect the particular credit risk of the individual counterparty.

(e)    Electronic platforms

94    During the relevant period, interest rate swaps and cross-currency swaps were either:

(a)    privately negotiated between two counterparties; or

(b)    conducted indirectly through facilitation by a broker.

95    During the relevant period, companies offering voice brokerage services such as ICAP Brokers Pty Ltd (ICAP) and Tullett Prebon (Australia) Pty Ltd (Tullett) facilitated cross-currency swap and interest rate swap transactions. That is, they provided the ability for two counterparties to negotiate a swap transaction. They did not provide an active market in which "live" bids and offers were shown on the brokers' screens. Rather, the rates displayed on the screens were only indicative of where a counterparty might be willing to transact. It was only possible to get a confirmed rate at which a party was willing to transact by contacting the counterparty directly or through a broker.

96    There were also various electronic platforms.

97    The electronic platform called BETSY operated by Bloomberg Tradebook Australia Pty Ltd facilitated transactions in interest rate swaps. According to the evidence adduced before me:

(a)    BETSY was only available to large financial institutions and, as a result, facilitated interest rate swaps to a very narrow section of the broader market for interest rate swaps. The vast majority of interest rate swaps were negotiated bilaterally, either privately or through voice brokering.

(b)    Only 10 trades in interest rate swaps that had a payment referenced to BBSW were facilitated by BETSY during the relevant period (one of which was a "test" trade). By way of comparison, during the relevant period, the annual market turnover for interest rate swaps and cross-currency swaps was around $6 trillion. During the relevant period, Westpac entered into approximately 10,000 swap transactions each year.

(c)    BETSY did not allow participants to post "live" bids and offers for interest rate swaps to customers. Rather BETSY allowed participants to propose interest rate swaps with specific terms or make requests for quotes, after which the details of any trade could be negotiated bilaterally between the parties. In this way, BETSY provided a similar service to those companies providing voice brokerage services such as ICAP and Tullett.

(d)    BETSY did not compel participants to make bids and offers available to customers at any time. In addition, dealers were permitted to ignore orders from customers made via BETSY.

(e)    The execution, settlement and clearance of any transaction facilitated by BETSY was at the sole discretion of the parties to the transaction and was subject to the parties' independent confirmation and documentation processes.

(f)    Parties were not contractually obliged to inform Bloomberg of instances where a counterparty did not proceed to settlement despite accepting a transaction via BETSY.

98    Further, the BGC Trader platform (an electronic trading facility operated by BGC Partners (Australia) Pty Ltd and BGC Brokers LP named "BGC Trader") facilitated transactions in various financial products including interest rate swaps and cross-currency swaps. From the evidence adduced before me the following may be noted:

(a)    BGC Trader was only available to large financial institutions and, as a result, facilitated interest rate swaps to a very narrow section of the broader market for interest rate swaps and cross-currency swaps. By way of comparison, around $6 trillion was traded in the broader market for interest rate swaps and cross-currency swaps during the relevant period. The vast majority of interest rate swaps and cross-currency swaps were negotiated bilaterally, either privately or through voice brokering.

(b)    BGC Trader's "Volume Match auction" did not allow participants to post "live" bids and offers for interest rate swaps for cross-currency swaps and did not compel participants to show bids or offers. The Volume Match auction for interest rate swaps operated once daily for certain floating rate tenors and maturities, and the auction for cross-currency swap transactions occurred less frequently than that.

(c)    Given that the bids and offers observed on BGC Trader were anonymous, any trade confirmation provided by BGC Trader was always subject to credit risk checks which was conducted independently by the relevant parties on a bilateral basis (or via their broker).

(d)    The execution, settlement and clearance of any transaction facilitated by BGC Trader was at the sole discretion of the parties to the transaction and was subject to the parties' independent confirmation and documentation processes.

(e)    Parties were not contractually obliged to inform BGC Trader of instances where a counterparty did not proceed to settlement despite accepting a transaction via BGC Trader.

99    I will say something more about these electronic trading facilities later in my reasons.

(f)    Forward rate agreements

100    A forward rate agreement (FRA) is an agreement to lend or borrow money at a specified price (agreed rate) on a future date for a set period. Expressed another way, a FRA involves two parties agreeing to exchange interest payments based on two different agreed rates, at a future date (called the "settlement date"), based on a specified notional amount.

101    FRAs typically reference a set fixed rate, which is agreed between the parties and a floating rate, which is BBSW. Under these FRAs the parties agree to exchange the difference between cash-flows based on the agreed fixed rate and cash-flows based on the BBSW, for a specified period, based on a notional amount.

102    The payment of interest based on a specified notional amount at a particular interest rate and for a particular period can be implemented by:

(a)    the lender paying to the borrower the specified notional amount, discounted according to the formula:

Where:

DA =

The discounted amount

NA =

The specified notional amount

r =

Applicable interest rate (expressed as an annual rate)

d =

Number of days (between reset dates)

and

(b)    the borrower paying to the lender the full specified notional amount.

103    Where such payments of interest are exchanged between two parties based on the same specified notional amount, and in respect of the same period, but at different interest rates, the respective payments of the full specified notional amount (as referred to above) offset each other precisely and can therefore be ignored.

104    On this basis, the formula for calculating cash-flow obligations and the net payment on the settlement date can be expressed as follows:

Payment obligation of the party lending at the fixed interest rate (therefore borrowing at the floating rate)

Payment obligation of the party lending at the floating interest rate (therefore borrowing at the fixed rate)

The net cash-flow payment

Which can also be expressed as:

Where:

Notional =

Notional amount

d =

Day count (between reset dates)

Rfix =

Fixed rate

Rfloat =

Floating rate

Afix =

Amount required to be paid by the party lending at the fixed interest rate leg of the FRA

Afloat =

Amount required to be paid by the party lending at the floating interest rate leg of the FRA

Anet =

Cash payment required on payment date

105    The floating rate in a FRA is commonly BBSW. In these FRAs:

(a)    the "Rfloat" input in the equations above is BBSW;

(b)    the payment obligation of the party lending at the floating rate is directly referenced to BBSW; and

(c)    the net payment amount (Anet) due on the settlement date is directly referenced to BBSW.

106    In relation to exposure, for FRAs which used the BBSW as the relevant floating rate, an exposure to the BBSW would be realised on each payment date. As set out above the obligations of the FRA party paying the floating rate and the net payment amount which both parties were obligated to pay, directly referenced the BBSW.

107    In relation to profit or loss, the profit or loss to the parties to a FRA is realised on the settlement date. As set out above, the payment amount due on the settlement date, can be expressed as:

108    Where the BBSW is used as the floating rate in a FRA, the rate at which the BBSW sets on the settlement date will determine which party needs to make a payment and the value of the payment.

(g)    Asset swaps

109    An asset swap is a form of interest rate swap in which one of the cash-flow streams relates to an underlying asset. One cash-flow stream of the asset swap is based on a non-asset rate (a financial interest rate), commonly the BBSW. The other stream is calculated based on the yield generated by an underlying asset, for example, a bond held by a party to the swap. In a similar manner to an interest rate swap discussed above, the parties exchange periodic cash payments based on the net difference between the two cash-flows at regular intervals. Usually, only net interest payments are made, and no "principal" payments.

110    The formula for calculating the cash-flow obligations of both swap counterparties on a reset date can be expressed as follows:

Payment obligation of the party paying the asset referenced rate

Where:

Aasset =

Amount required to be paid by the party paying the asset referenced interest rate leg of the swap

Notional =

Notional amount

d =

Day count (between reset dates)

Rasset =

Interest rate based on the yield generated by an underlying asset

Payment obligation of the party paying the financial interest rate

Where:

AfinIR =

Amount required to be paid by the party paying the financial interest rate leg

Notional =

Notional amount

d =

Day count (between reset dates)

RfinIR =

Financial interest rate

The net cash-flow payment (Swap Cash Settlement Amount)

This can also be written as follows by expanding the equation:

Where:

Anet =

Swap Cash Settlement Amount

111    The non-asset rate in an asset swap is commonly BBSW. In these asset swaps:

(a)    the payment obligation of the party paying the financial rate is directly referenced to BBSW;

(b)    the "RfinIR" input in the formula above is BBSW; and

(c)    the Swap Cash Settlement Amount is directly referenced to BBSW.

112    In relation to exposure, for asset swaps which used the BBSW as the relevant floating rate, an exposure to the BBSW would be realised at each reset (or fixing) date. As set out above, the obligations of the swap party paying the floating rate and the Swap Cash Settlement Amount which both parties are obligated to pay, directly referenced the BBSW.

113    In relation to profit or loss, the profit or loss to the parties to an asset swap is realised on the reset dates. As set out above, calculation of the settlement amount where the financial interest rate is a floating rate can be expressed using the following formula:

114    Accordingly, where the non-asset interest rate used in an asset swap is the BBSW, the rate at which the BBSW sets on the reset dates will determine which swap party needs to make a payment and the value of the payment being the Swap Cash Settlement Amount.

(h)    Interest rate options (caps, floors and collars)

Interest rate cap

115    An interest rate cap is an agreement between two parties in which the buyer purchases from the seller one call option, or a series of call options, on future borrowing rates (each individual call option is called a "caplet"). In exchange for payment of a premium, the buyer acquires the right (but not the obligation) to require the seller to compensate it on prescribed reference dates if the agreed floating interest rate is greater than an agreed strike rate (known as the "cap rate"). The floating interest rate used in interest rate caps is typically BBSW. If so:

(a)    the parties agree that when the BBSW exceeds the cap rate, the seller will pay to the buyer the difference between interest on the notional amount calculated using the BBSW and interest on the notional amount calculated using the cap rate;

(b)    a payment is only required if the BBSW exceeds the cap rate on the prescribed reference date(s); and

(c)    if the BBSW exceeds the cap rate, the payment the seller must make to the buyer is based on the difference between the BBSW and the cap rate, the term of the period ending on the relevant reference date, and the contract's notional amount. Payments can be made "discounted in advance" or "non-discounted in arrears".

116    During the relevant period, the AFMA Interest Rate Options Conventions provided that "[w]ith caps and floors, settlement payments by the seller to the buyer will be on a non-discounted in arrears basis or on a discounted in advance basis payable on the reference date following the day on which the reference rate is set, or in the case of the final reference date, on the maturity date." (AFMA, "Interest Rate Option Convention", September 2009, [5.2.3]; AFMA, "Interest Rate Option Convention", March 2012, [5.2.3]). The AFMA Interest Rate Options Conventions set out the formulae for calculation of the settlement amount for caps on a discounted in advance, and a non-discounted in arrears, basis. The formula for calculation of the settlement amount for caps using the BBSW on a non-discounted in arrears basis does not need to be set out.

Interest rate floor

117    An interest rate floor is an agreement between two parties in which the buyer purchases from the seller one put option, or a series of put options, on future interest rates (each individual put option is called a "floorlet"). In exchange for payment of a premium, the buyer acquires the right (but not the obligation) to require the seller to compensate it on prescribed reference dates if the agreed floating interest rate is less than an agreed strike rate (known as the "floor rate"). The floating interest rate used in interest rate floors is commonly the BBSW. If so:

(a)    the parties agree that when the BBSW falls below the floor rate, the seller will pay to the buyer the difference between interest on the notional amount calculated using the BBSW and interest on the notional amount calculated using the floor rate;

(b)    a payment is only required if the BBSW falls below the floor rate on prescribed reference date(s); and

(c)    if the BBSW falls below the floor rate, the payment the seller must make to the buyer is based on the difference between the BBSW and the floor rate, the term of the period ending on the relevant reference date, and the contract's notional amount. Payment can actually be made "discounted in advance" or "non-discounted in arrears".

118    During the relevant period, the AFMA Interest Rate Options Conventions set out the basis upon which floor settlement payments will be calculated. The formula for calculation of the settlement amount for floors on a non-discounted in arrears basis does not need to be set out (see: AFMA, "Interest Rate Option Convention", September 2009, [5.2.4]; AFMA, "Interest Rate Option Convention", March 2012, [5.2.4]).

Interest rate collar

119    An interest rate collar is an interest rate cap and an interest rate floor combined. The effect of an interest rate collar which uses the BBSW as the agreed floating rate is an agreement that:

(a)    when the BBSW exceeds the cap rate on the reference dates, party one will pay to party two the net difference between interest on the notional amount calculated using the BBSW and interest on the notional amount using the cap rate; and

(b)    when the BBSW is lower than the floor rate on the reference dates, party two will pay to party one the net difference between interest on the notional amount calculated using the BBSW and interest on the notional amount using the floor rate.

General

120    With respect to interest rate caps, floors or collars which reference the BBSW as the floating rate, whether a payment obligation arises depends on the BBSW. And if a payment obligation arises, its value is calculated with reference to the BBSW (BBSW being an input in the formula).

121    In relation to exposure, for interest rate options which referenced the BBSW, an exposure to the BBSW might be realised at each reference date. The enlivening of the payment obligation (and thus the exposure) and the value of the payment required, were both dependent on the BBSW.

122    Let me say something about profit or loss.

123    The profit or loss to the parties to an interest rate cap is realised on the reference dates. The formula for a payment under an interest rate cap which uses BBSW as the reference rate can be expressed as follows:

Caps (if BBSW > strike):

124    Accordingly, where the reference rate used in an interest rate cap is the BBSW, the rate at which the BBSW sets on the reference dates will determine whether the interest rate cap is exercised and if so the value of the payment.

125    The profit or loss to the parties to an interest rate floor is realised on the reference dates. As set out above, the formula for a payment under an interest rate cap which uses BBSW as the reference rate can be expressed as follows:

Floors (if BBSW < Rstrike):

126    Accordingly, where the reference rate used in an interest rate cap is the BBSW, the rate at which the BBSW sets on the reference dates will determine whether the interest rate floor is exercised and if so the value of the payment.

127    The profit or loss to the parties to an interest rate collar is realised on the reference dates. The calculation of settlement payments for interest rate collars is the same as the formulae used for calculation of settlement payments for interest rate caps and interest rate floors, detailed above. However, for an interest rate collar there are two strike rates; one for the cap and one for the floor.

(i)    Swaptions

128    A swaption is an option contract in which, in exchange for a premium, the seller grants the buyer the right, but not the obligation, to enter into an underlying interest rate swap. The terms of the separate interest rate swap are agreed at the time of entering into the swaption. The underlying interest rate swap operates as set out above. Once a swaption is exercised, the BBSW is relevant to the underlying interest rate swap as set out above. Whether a swaption is exercised will also depend on the BBSW.

129    In relation to exposure, if a swaption was exercised, the parties would enter into an interest rate swap in consequence and would have exposure on each reset date as described earlier.

130    In relation to profit or loss, once a swaption is exercised and the swap entered into, the profit or loss to the parties to the underlying interest rate swap is realised on the reset dates and is as set out above.

(j)    Inflation swaps

131    An inflation swap is a form of interest rate swap in which two parties agree to exchange streams of cash-flow for a set period, with one party's payment obligations based on a specified notional amount and a financial interest rate (commonly the BBSW) and the other party's payment obligations based on another notional amount and an inflation related index (such as the CPI). The parties agree to exchange periodic "net" payments, being the difference between the two cash-flows on specified reset dates.

132    The separate payment obligations of each party to an inflation swap can be expressed as follows:

Payment obligation of the party paying the inflation related amounts:

Payment obligation of the party paying the financial interest rate

And the net payment formula (Swap Cash Settlement Amount) can be expressed*:

Where:

Ainflation =

Payment required from party paying the inflation referenced leg of the inflation swap

AfinIR =

Payment required from party paying the financial rate leg of the inflation swap (the financial interest rate)

Anet=    

Swap cash settlement amount

Notionalinflation =

Notional amount for the inflation related payments

NotionalfinIR =    

Notional amount for the interest rate payments

d =    

Day count (between reset dates)

RfinIR =

Financial interest rate

Rinflation =

An amount determined by reference to an inflation related index such as CPI (with the precise relationship depending on the particular product in question)

*As these Swap Cash Settlement Amount equations have the financial interest rate leg subtracted from the inflation leg, a positive result means the party paying the inflation leg makes the payment and a negative result means that the party paying the financial interest rate leg makes the payment.

133    The financial rate in an inflation swap is commonly the BBSW. If so:

(a)    the "RfinIR" input in the equations above is the BBSW;

(b)    the payment obligation of the party paying the financial rate is directly referenced to the BBSW; and

(c)    the Swap Cash Settlement Amount ("Anet") is directly referenced to the BBSW.

134    In relation to exposure, for inflation swaps which used the BBSW as the relevant financial (non-inflation) interest rate, an exposure to the BBSW would be realised at each reset date. As set out above, the obligations of the swap party paying the floating rate and the Swap Cash Settlement Amount which both parties were obligated to pay, directly referenced the BBSW.

135    In relation to profit or loss, a profit or loss to the parties to an inflation swap is realised on the reset dates. As set out above, the formula to calculate the Swap Cash Settlement Amount on an inflation swap is:

136    Accordingly, where the financial rate used in an inflation swap is the BBSW, the rate at which the BBSW sets on the reset dates will determine which swap party needs to make a payment and the value of the payment being the Swap Cash Settlement Amount.

(k)    Floating rate notes

137    A floating rate note (FRN) is a debt instrument which pays a floating interest rate (often referred to as the "coupon") on specified dates over the term of the debt (e.g. quarterly or semi-annually), as well as repaying the principal on the maturity date. The floating rate used is often the BBSW plus a margin (for example, "BBSW plus 50 basis points"). Where the floating rate used in a FRN is the BBSW plus a margin, the payment obligation of the issuer of the FRN on each payment date is directly referenced to the BBSW.

138    In relation to exposure, for FRNs which used the BBSW (or BBSW plus a margin) as the applicable interest rate, an exposure to the BBSW would be realised on each interest rate reset date. As set out above, the value of the obligations of the issuer of the FRN (borrower) to pay interest for the period would directly reference the BBSW.

139    In relation to profit or loss, the value of payments made under a FRN (and thus profit or loss to a bank once costs are accounted for) are realised on the reset dates. Where the FRN is set with the BBSW as the floating interest rate, the rate at which the BBSW sets on the coupon reset dates will determine how much interest the holder of the FRN will be paid by the issuer for that particular coupon period.

(l)    Commercial loans

140    Many commercial loans have a floating rate of interest which resets on specified reset dates. Commonly, the floating rate is the BBSW plus a margin (for example "BBSW plus 250 basis points"). Where the floating rate is the BBSW plus a margin, the interest payment obligation of the borrower is directly referenced to the BBSW.

141    In relation to exposure, for commercial loans which used the BBSW plus a margin as the applicable interest rate, an exposure to the BBSW would be realised on each interest rate reset date. As set out above, the value of the obligations of the borrower to pay interest for the period ahead, and the income stream the lender would receive from the loan, directly reference the BBSW.

142    In relation to profit or loss, where the rate applied to a commercial loan is set using the BBSW, the amount of interest paid for a particular period depends on the level at which the BBSW sets on the relevant reset date.

THE BANK BILL MARKET AND BBSW

(a)    Nature of Prime Bank Bills

143    During the relevant period, Australia's big four banks, National Australia Bank Limited (NAB), Commonwealth Bank of Australia (CBA), Australia and New Zealand Banking Group Limited (ANZ) and Westpac, the Australian branch of JP Morgan Chase Bank NA (to 30 November 2011) and the Australian branch of BNP Paribas (to 24 February 2012) were designated by AFMA as "Prime Banks".

144    Banks, including Prime Banks, could borrow funds for a short term of usually no longer than 12 months by issuing NCDs or by accepting Bank Bills. I have already given a brief description of Bank Bills. Let me say something further about NCDs.

145    An NCD is a certificate issued by a bank evidencing an interest bearing deposit with the issuing bank for a fixed term, which entitles the holder of the certificate to payment of a specified sum (face value) by that bank on the date that the certificate matures.

146    The descriptions that I have given for NCDs and Bank Bills are broadly consistent with the definitions in clauses 2.1 to 2.3 of AFMA's Negotiable / Transferable Instruments (NTI) Conventions.

147    In the relevant period, Prime Bank Bills (or Prime Bank paper) were NCDs issued and Bank Bills accepted by a Prime Bank for tenors of one, two, three, four, five and six months. Prime Bank Bills were recognised by participants in the short dated securities market as trading as a homogenous asset class and having the highest quality with regard to liquidity, credit and consistency of yield.

148    Prime Bank Bills generally traded at a discount to their face value, because the face value would not be received by a holder until the maturity date. The discount was calculated on the basis of simple interest according to the formula that I have previously set out, but for convenience I will set out again with the components described:

149    As is apparent, the price of a Prime Bank Bill is inversely proportional to its yield. In other words, if two Prime Bank Bills had the same face value and maturity date, the one with the higher yield would trade at a lower price.

(b)    AFMA

150    AFMA is a financial markets industry association. Since June 2006, it has operated as a public company limited by guarantee. During the relevant period, it was responsible, amongst other things, for maintaining the conventions for trading in the Bank Bill Market and for the overall management of the BBSW, including the procedure for setting and publishing the BBSW. Evidence about the structure and operation of AFMA was given in this proceeding by Mr David Love, Director of Policy and International Affairs at AFMA during the relevant period.

151    Some of AFMA's objectives as set out in clause 3 of its constitution included:

(a)    to represent the common interests of members in dealing with issues relevant to the good reputation of the wholesale banking and financial markets in Australia and the regulation of them by Government and market authorities;

(b)    to promote and facilitate the development and maintenance in Australia of efficient and competitive markets for financial instruments and other related transactions; and

(c)    to encourage responsible self-regulation by establishing efficient and ethical market practices and conventions and preparing and maintaining standard documentation.

152    AFMA had a secretariat with a permanent staff of about 28 people and over 110 members, including Australian and international banks, brokers, securities companies, state government treasury corporations, fund managers, energy traders and industry service providers.

153    There were two categories of membership of AFMA:

(a)    "Financial Market Members": organisations that participated directly in Australia's financial markets, including financial institutions.

(b)    "Associate Members": organisations generally comprising of advisory and legal firms and government agencies.

154    Throughout the relevant period, Westpac was a Financial Market Member of AFMA. As a member of AFMA and pursuant to clause 6.5 of its constitution, Westpac agreed to abide by:

(a)    AFMA's constitution;

(b)    AFMA's Code of Ethics and Code of Conduct;

(c)    the "Standards and Guidelines for Australian OTC Financial Markets"; and

(d)    any other rules and conventions adopted by AFMA at general meetings from time to time.

155    Under clause 11.4 of the AFMA constitution, Financial Market Members were entitled to nominate a director to AFMA's board. Westpac's representative on AFMA's board was Mr Robert Whitfield, Head of Westpac Institutional Bank.

156    During the relevant period, AFMA's board was advised by a number of committees, including the Market Governance Committee (MGC). The MGC developed and maintained market protocols designed to facilitate and promote the efficient and orderly running of over the counter (OTC) markets in Australia. Ms Joanne Dawson, Managing Director of Group Treasury, was Westpac's representative on the MGC during the relevant period.

157    The MGC was established by clause 15 of the AFMA constitution and met on a quarterly basis. Its aims and objectives included:

(a)    the development and maintenance of market protocols and operational standards that facilitate and promote the efficient and smooth running of the financial markets in Australia;

(b)    advising the board on issues that affect the reputation of financial markets and investor confidence; and

(c)    the promotion of high ethical standards in the financial markets within Australia.

158    The MGC played an important role in approving market conventions, standards and guidelines.

159    The MGC in turn oversaw a number of market committees that met regularly with the objective of developing consensus in the market on technical matters such as transaction documentation, standards, trading conventions and market data. These market committees included the NTI Committee and the BBSW Committee.

160    The NTI Committee was responsible for maintaining the conventions for trading in the Bank Bill Market and the implementation of the rules governing the list of Prime Banks. The table below sets out the dates on which the NTI Committee met during the relevant period, the name of the Westpac representative in attendance and the capacity in which they attended.

Date of Meeting

Westpac Representative

Capacity

26 May 2010

Ms Sophie Johnston

Deputy Chairperson

26 August 2010

Ms Sophie Johnston

Chairperson

25 November 2010

Ms Sophie Johnston

Chairperson

8 February 2011

Ms Sophie Johnston

Chairperson

10 May 2011

Mr Colin Roden

Alternative

19 August 2011

Ms Sophie Johnston

Chairperson

8 November 2011

Ms Joanne Dawson

Alternative

2 February 2012

Mr Colin Roden

Alternative

10 May 2012

Ms Sophie Johnston

Chairperson

161    The BBSW Committee was responsible for the overall management of the BBSW, rates directly related to the BBSW, the procedure for setting and publishing the BBSW, and the resolution of any disputes among AFMA members involving the BBSW and the rate set mechanism.

162    Although the BBSW Committee was a technical committee under the MGC, its role took on greater prominence over time, especially from May 2010 after AFMA received a letter from the RBA noting concerns expressed about BBSW and suggesting that the behaviour of BBSW warranted further investigation.

163    The table below sets out the dates on which the BBSW Committee met during the relevant period, the name of the Westpac representative in attendance and the capacity in which they attended:

Date of Meeting

Westpac Representative

Capacity

26 May 2010

Ms Sophie Johnston

Member

30 August 2010

N/A

30 November 2010

Ms Sophie Johnston

Member

10 February 2011

Ms Sophie Johnston

Member

12 May 2011

N/A

2 September 2011

Ms Sophie Johnston

Member

15 November 2011

Ms Sophie Johnston

Member

20 March 2012

Ms Sophie Johnston

Member

30 May 2012

Ms Sophie Johnston

Member

NTI Conventions

164    From time to time during the relevant period, AFMA encouraged self-regulation by preparing and publishing market conventions with a view to providing guidance to AFMA members. Market conventions were usually concerned with operational and technical matters and provided agreed approaches to doing business. These practices were agreed by AFMA members for the purpose of helping the markets work efficiently.

165    The AFMA NTI Conventions:

(a)    described products traded in the Bank Bill Market, including Bank Bills and NCDs;

(b)    outlined the criteria by which Prime Banks were designated; and

(c)    outlined the conventions relating to trading, confirmation and settlement processes in the Bank Bill Market.

166    The July 2009 version of the NTI Conventions applied at the start of the relevant period and the NTI Conventions were updated during the relevant period in November 2010, November 2011 and February 2012.

AFMA Code of Ethics and Code of Conduct

167    The AFMA Code of Ethics and Code of Conduct (AFMA Code of Ethics) set out the minimum standards, principles and rules of behaviour applicable to AFMA members in the conduct of their business. The AFMA Code of Ethics required AFMA members to:

(a)    conduct their trading activities in a fair and orderly manner;

(b)    maintain the integrity of financial markets; and

(c)    not carry out trading that:

(i)    would interfere with the normal supply and demand factors in the market for a financial product;

(ii)    had the potential to create artificial markets or prices; or

(iii)    was not based on a genuine trading or commercial intention.

168    The AFMA Code of Ethics also required that members establish policies and procedures to avoid and manage actual and potential conflicts of interest, with respect to both business and personal conflicts. As an AFMA member, Westpac agreed to use this Code, which in terms encouraged members to use the document as the basis for developing and implementing more detailed internal procedures to effect the objectives of the Code.

Eligibility criteria for Prime Banks

169    In the relevant period, AFMA specified and administered the process for the election and recognition of banks designated by AFMA as Prime Banks, which process was set out in the NTI Conventions. The list of Prime Banks was determined through confidential AFMA surveys of institutions deemed by the NTI Committee to be significant Prime Bank Bill investors. AFMA invited all banks meeting the eligibility criteria to nominate for consideration in the annual Prime Bank surveys. In order to be eligible, a bank had to:

(a)    be an APRA (Australian Prudential Regulation Authority) authorised deposit taking institution and classified by APRA as an Australian owned bank, a foreign subsidiary bank or a branch of a foreign bank that was authorised to carry on banking business pursuant to the Banking Act 1959 (Cth) (as amended) or comparable legislation in its country of origin;

(b)    satisfy a credit rating benchmark, the specifics of which changed during the relevant period; and

(c)    have securities eligible for use in the RBA's open market operations and standing liquidity facilities.

170    Clause 3.7 of the NTI Conventions specified the eligibility criteria for Prime Banks from time to time, as well as the protocols governing the selection of Prime Banks, the ongoing requirements for maintaining Prime Bank status, Prime Bank reporting requirements and the contingency plan should a major bank or banks lose Prime Bank status.

171    The annual survey process for appointing Prime Banks involved the following:

(a)    On or about the following anniversary of the prior election, AFMA would invite all banks (including existing Prime Banks) satisfying the above eligibility criteria to nominate for inclusion in the Prime Bank survey.

(b)    Following the receipt of nominations, AFMA would survey institutions deemed by the NTI Committee to be significant Prime Bank Bill traders and investors. The survey asked respondents to indicate which nominated bank or banks they believed should be designated as Prime Banks. A weighted voting system, agreed amongst members of the NTI Committee and which sought to ensure that the most active market participants had a proportionally higher involvement in the process than the least active member, applied to individual survey respondents.

(c)    Each bank nominating for Prime Bank status was provided with the list of survey respondents in advance of the survey.

(d)    An 80% response rate of eligible survey respondents had to be obtained in order for a survey to be valid.

172    Ordinarily, at least 70% of the weighted survey vote was required to remain a Prime Bank and 80% of the weighted survey vote was required to be appointed as a new Prime Bank, although alternative criteria would apply if the above process resulted in the selection of less than six Prime Banks.

173    Prime Banks were expected to display an ongoing commitment to the short-term money market, including endeavouring to maintain a minimum issuance level of $5 billion and providing consistent price support for all Prime Bank paper in the domestic short-term securities market.

BBSW panellists

174    Throughout the relevant period, the BBSW panel was comprised of up to 14 organisations that would input rates for the calculation of the BBSW rate set (BBSW panellists) who were listed in Appendix 1 to the Bank Bill Swap (BBSW) Reference Rate Procedures (BBSW Procedures) from time to time. Prime Banks were automatically appointed to the BBSW panel and other organisations were elected to the panel under the ballot procedure described in clause 1.1 of the BBSW Procedures.

175    The BBSW panellists throughout the relevant period were:

(a)    ANZ;

(b)    Bank of Scotland plc, Australia Branch (trading under the name Lloyds TSB Bank plc from 15 November 2010);

(c)    BNP Paribas, Australia Branch;

(d)    Citibank NA, Australia Branch;

(e)    CBA;

(f)    Deutsche Bank AG, Australia Branch;

(g)    HSBC Bank Australia Limited (HSBC);

(h)    JP Morgan Chase Bank NA, Australia Branch;

(i)    Macquarie Bank Limited;

(j)    NAB;

(k)    UBS AG, Australia Branch; and

(l)    Westpac.

176    The following entities were BBSW panellists during part of the relevant period:

(a)    Société Générale SA, Australia Branch (to 28 May 2010);

(b)    RBS Group (Australia) Pty Limited (from 11 March 2010 to 1 May 2012);

(c)    Royal Bank of Canada (a reserve BBSW panellist to May 2010 and a BBSW panellist from 28 May 2010); and

(d)    Suncorp-Metway Limited (a reserve BBSW panellist until April 2012 and a BBSW panellist from 1 May 2012).

I should also note that ABN AMBRO Bank NV, Australia Branch, was a BBSW panellist up to 10 March 2010, shortly before the commencement of the relevant period.

(c)    Operation of Bank Bill Market

177    In the relevant period, Prime Bank Bills were traded either in direct trades between counterparties or through interdealer brokers ICAP and Tullett. For the purposes of these reasons, I will use the term "Bank Bill Market" to refer to the market for trading in Prime Bank Bills operated by and through these interdealer brokers principally during the BBSW Rate Set Window; it will be appreciated that the traded product, Prime Bank Bills, was essentially homogenous, putting to one side collateral questions such as deliverability under BAB Futures. ICAP and Tullett acted as intermediaries by quoting prices at which their respective clients would offer to sell and bid to buy Prime Bank Bills to and from other participants in the Bank Bill Market, including Prime Banks, non-Prime Banks and non-banks. It was typical for each participant in the Bank Bill Market to have a broker at each of ICAP and Tullett.

178    Trading in the Bank Bill Market was primarily conducted by reference to the yield and tenor of Prime Bank Bills. The substantial majority of all trading in Prime Bank Bills, including by or on behalf of Westpac, took place between approximately 9.55 am and 10.05 am on each Sydney business day during the BBSW Rate Set Window.

179    In the relevant period, participants trading Prime Bank Bills in the Bank Bill Market did not use electronic platforms to execute trades, but rather executed trades by "voice broking", that is by participants giving verbal instructions to brokers and the brokers verbally executing transactions between themselves (broker to broker). Participants usually communicated and gave instructions to brokers via telephone lines which connected each participant to his or her individual broker(s). The telephone lines through which participants communicated with brokers were often connected to speaker boxes, known as "squawk boxes", at both the participant's and the broker's end. As I have said, typically a participant trading Prime Bank Bills in the Bank Bill Market would have a broker at each of Tullett and ICAP.

180    At or shortly after the time of participants making bids and offers for Prime Bank Bills through ICAP and Tullett, ICAP and Tullett posted the bids and offers for Prime Bank Bills of all tenors on separate electronic systems (that is, one system showing bids and offers through ICAP and one system showing bids and offers through Tullett). This information, once posted, became accessible by, and visible to, participants in the Bank Bill Market with access to the said electronic systems.

181    Let me elaborate further at this point in terms of the positions and perspectives of ICAP and Tullett. I will discuss later the operation of the Bank Bill Market from the perspective of Westpac and its traders; see the section of my reasons "Characteristics and Dynamics of Trading on Bank Bill Market".

182    In relation to ICAP and its operations relevant to the Bank Bill Market, the following evidence was given.

183    The dealing room, where the various "Desks" were located, was one large room. The dealing room contained the following ICAP Desks: the Short End Desk, the Long End Desk, the New Zealand Dollar Swap Desk, the Fixed Income Desk, the Forward Foreign Exchange Desk, the Energy Desk and the Futures Desk (the Futures Desk was reintroduced at ICAP in December 2011). The size of each Desk varied, depending on the number of brokers who worked on that Desk. The Short End Desk was located at a physical desk of approximately three metres by two metres. The brokers who worked on the desk sat beside each other and across from each other at each long side of the desk. Computer monitors sat in the middle of the desk between the brokers who were opposite each other.

184    There was a concentrated period of trading in Prime Bank Bills on the Short End Desk between about 9.55 am and 10.05 am (the concentrated trading period), when the vast majority of daily trading in Prime Bank Bills in which ICAP was involved occurred. Trading of Prime Bank Bills was especially concentrated approximately 2 minutes either side of 10.00 am. During the relevant period, ICAP's brokers arranged trading for, inter-alia, Westpac.

185    In relation to the Short End Desk, bids, offers and acceptances were generally communicated verbally between the brokers on that Desk, ICAP of course being an interdealer broker.

186    In relation to a broker communicating with a trader at a particular bank, each broker had a communication device known as a dealer board, which had a telephone keypad and, generally, two handsets, with a series of telephone lines. Certain lines were set up as lines that required pre-dialling whereas other lines were lines that were dedicated as direct connection with particular clients, being generally traders at banks. The dealer board had numerous buttons which could be configured to correspond with one or more of these dedicated lines. Most dealer boards also had speakers and a microphone, which stood up from the dealer board on a microphone stand that the brokers could use instead of the handsets.

187    A broker could communicate verbally with a trader in one of two ways. First, they could use one of the telephone handsets to activate the dedicated line to that trader. Using this method, the broker would hear the trader through the earpiece of the handset, and the trader would hear only what was picked up by the mouthpiece of the broker's handset. Second, the broker could use the speakers and microphone (also known as the "squawk box"). In order to do so, the broker would hold down the button corresponding with the dedicated line to that trader (referred to as "clicking in" to the line). The broker could also use this method to communicate with multiple traders at once by holding down a button that had been configured to correspond to multiple dedicated lines. Using this method, the trader(s) would hear everything picked up by the dealer board's microphone (including the noise generated by other brokers in the dealing room), and the sound from the trader(s) would be emitted from the dealer board's speakers. If a broker did not want the client to hear his communication with the brokers who acted for other market participants, he could release the button on the squawk box (referred to as "clicking out").

188    Brokers on the Short End Desk regularly clicked in and out of lines. For periods of intense trading such as during the concentrated trading period, a broker might have a line constantly open to a client so that the broker could hear instructions as soon as they were provided.

189    In addition to the dealing board, each broker on the Short End Desk had a computer with several monitors at their desk. A broker could use their computer to communicate with client traders through instant chat messaging applications and other electronic forms of communication. However these methods of communication were primarily used during less intense periods of trading.

190    Each Short End Desk broker also had access to a screen displaying ICAP's real time internal data on Prime Bank Bill bids and offers from the ISIS computer system, ICAP's in house pricing, data distribution and order management system, as well as screens from information vendors Reuters and Bloomberg displaying information on bids and offers of Prime Bank Bills through ICAP. Each of these screens could be displayed on the broker's computer monitors. It was a matter of preference for each individual broker what they chose to monitor on the screens to which they had access, including whether the Reuters screen or the Bloomberg screen was generally showing on one of the broker's personal monitors.

191    The Reuters and Bloomberg screens were "private" screens in that only those subscribers authorised by ICAP could access them. In his time at ICAP Mr Glenn McMurdy (who was chief operating officer from 2005 to 2007 and from 2013 to date) did not recall ever seeing an ICAP broker with a Reuters or Bloomberg screen of Tullett data displayed.

192    During the concentrated trading period, a number of brokers from desks other than the Short End Desk would provide additional coverage on the Short End Desk to ensure all client voice broking lines were covered. The categories of ICAP personnel involved in the trading and recording of trades in Prime Bank Bills were:

(a)    Brokers responsible for arranging trades in Prime Bank Bills;

(b)    Keyboard operators, whose responsibilities included entering into the ISIS system, in real time, the tenor, price and volume of the offers and acceptances of trades in Prime Bank Bills; and

(c)    Trade support assistants, whose responsibilities included acting as the keyboard operators during the concentrated trading period as well as recording the trades in Prime Bank Bills executed by the brokers into the Graphical User Interface Back Office System or GUIBOS system, which is ICAP's back office and trade confirmation system.

193    During the relevant period the general practice of the Short End Desk during the concentrated trading period was as follows:

(a)    Brokers and keyboard operators assembled around the Short End Desk prior to 10.00 am each business day, usually around 9.50 am;

(b)    Brokers took verbal orders (over the squawk box or over the telephone) or written orders (via other electronic means such as messaging apps or emails) from traders at their respective client banks to buy, sell or accept the offer of Prime Bank Bills of a particular tenor at a particular volume and yield;

(c)    Brokers quoted offers to buy (bids) or sell (offers) Prime Bank Bills of a particular tenor at a particular volume and yield from their respective banks to one another verbally, by calling them out;

(d)    Keyboard operators entered the tenor, price and volume of bids or offers into ISIS as they were made and called out by brokers;

(e)    ISIS automatically displayed the details of the bids or offers as entered by the keyboard operators so that those bids and offers were visible to all brokers, provided the brokers had opened and were viewing the ISIS screen on their personal computers;

(f)    ISIS automatically fed the price and volume data of bids and offers to Reuters or Bloomberg where it was displayed on a specific Reuters or Bloomberg page, visible to people that subscribed to the page when they were accessing the page;

(g)    Brokers accepted bids or offers, which had been called into the room verbally or which the brokers had seen displayed on the ISIS screen, on behalf of their clients after a trader had communicated the acceptance to their broker either down the squawk box or over the handset or other electronic means;

(h)    If a bid or offer was accepted in its full volume, the relevant bid or offer would be manually removed by the keyboard operator from ISIS with the next available price being displayed. Similarly, if a bid or offer was accepted but only in lesser volume, the bid or offer would be manually adjusted in ISIS by the keyboard operator to show the remaining volume.

194    The details of the bids and offers as displayed within ISIS to the brokers on the Short End Desk and as displayed on the Reuters and Bloomberg screens listed each tenor horizontally by month. Running beside each tenor were two columns, one headed "Early Bills", which reflected bids and offers made for Prime Bank Bills expiring in the first half of the month (up to and including the 15th of the month) and the other headed "Late Bills", which reflected bids and offers made for Prime Bank Bills expiring in the second half of the month (from the 16th of the month). I will elaborate on this "early" and "late" question later.

195    During the concentrated trading period, each Short End Desk broker kept a running personal tally of bids, offers and acceptances which they had arranged. This personal tally was generally recorded in writing on paper.

196    During the relevant period, the general practice at ICAP after the daily concentrated trading period closed was as follows:

(a)    Brokers checked and reconciled the trades recorded on their personal tally with each other for accuracy.

(b)    If there was confusion about transactions or disagreements as to the trading that had taken place, the brokers would seek to resolve any discrepancies, if necessary after speaking with their respective trader/clients. This could extend where necessary to reviewing tape recordings of the communications made through the brokers' dealer boards.

(c)    Brokers received details of the Prime Bank Bills they had traded from each selling bank's trader, which details included the identity of the bank that was the issuer or acceptor of the Prime Bank Bills the subject of the trade and the exact maturity date of the bills which had been traded.

(d)    Brokers wrote down the trade details (aggregated in some instances as described in subparagraph (e) following) on a document known as a "scratch ticket" recording details of the parties to the trade, the tenor and dollar value of the Prime Bank Bills traded.

(e)    When filling in scratch tickets, brokers could aggregate the trades by consolidating parcels of trades made between the same parties in the same tenor at the same price. They could also aggregate trades between the same parties in the same tenor but at different prices, by averaging the prices of the individual trades for the consolidated trade. This aggregation was done at the discretion of the individual broker, and with the consent of the relevant clients.

(f)    Brokers passed the scratch tickets to the trade support assistants once they had received details of the Prime Bank Bill issuer/acceptor from the selling bank via a broker.

(g)    Trade support assistants entered the details of the trades into GUIBOS on the basis of the details set out in the scratch tickets. The "Deal Time" recorded in GUIBOS was the time the trade was entered by the trade support assistant.

(h)    When entering data into GUIBOS, trade support assistants could aggregate the trades by consolidating parcels which traded at the same rate. So, for example, two separate trades of $20 million each at a yield of 4.95% may have been recorded as a single trade of $40 million at a yield of 4.95%. This aggregation was done independently of any aggregation conducted by the brokers, and was also done at the discretion of the individual trade support assistant.

(i)    After receiving the details of the Prime Bank Bill issuer/acceptor from the selling bank, the brokers would confirm the details to their buying clients (usually over the phone) and would then usually also provide email confirmation, depending on the preference of the particular client.

197    The trade support assistant would typically enter the data into GUIBOS on the same day if the trade occurred prior to midday, and the following day if the trade occurred after midday.

198    In relation to Tullett and its operations relevant to the Bank Bill Market, the following evidence was given.

199    Tullett is also a voice interdealer broker which carries on the business of facilitating transactions between financial institutions. During the relevant period, Tullett's business was divided into desks, which dealt in different securities. One of these desks was the short term interest rate desk (Short End Desk) which traded in short dated securities. The other desks comprising Tullett's business were the Long End Desk and the FX Desk.

200    During the relevant period, Tullett employed brokers on the Short End Desk who were responsible for arranging and facilitating trades in Prime Bank Bills, as well as other short dated securities and short dated derivatives. During the relevant period, Tullett's brokers arranged trading for, inter-alia, Westpac.

201    During the relevant period, Tullett also employed, as part of the Short End Desk, persons (inputters) who were responsible for recording in real time, in Tullett's order management system, the tenor, rate, counterparty name and volume of the offers and bids of trades in Prime Bank Bills arranged by the brokers. The inputters were also responsible for recording and updating the details of trades after the period of active trading in Prime Bank Bills. Depending on how busy they were, brokers on occasion entered trade data into Tullett's system themselves without the involvement of an inputter.

202    During the relevant period, the practice of the Short End Desk was that:

(a)    each broker worked mainly with a trader from one particular major bank when brokering the trading of the Prime Bank Bills; and

(b)    a minimum of 3 inputters were assigned to record bids, offers and trades of Prime Bank Bills which had 1, 3 and 6 month maturities.

203    The Short End Desk endeavoured to record the trading in Prime Bank Bills in real time. The Short End Desk was located in an area of approximately 30 square metres within a larger open plan broking room that also housed Tullett's other Desks. That area contained an array of adjoining desks, with the brokers assigned to particular desks sitting beside one another. Inputters were positioned at the desks interspersed between the brokers.

204    During the relevant period, each broker on the Short End Desk had, on their desk, a communication device referred to as a "dealing board". This device provided brokers with access to direct telephone lines to individual traders representing different banks, as well as a series of dedicated direct open lines to particular clients with speaker functionality. The dealing board consisted of two adjacent black boxes. The first component had numerous buttons as well as a keypad similar to that found on an ordinary telephone. The second component, which had a speaker, a microphone and a panel of switches activating dedicated direct open lines, was also referred to as a squawk box.

205    There were different ways in which a broker could use the dealing board to communicate with traders representing banks.

206    First, a broker could communicate via a bilateral phone line with a particular trader. This could be done in one of two ways. The first method of bilateral communication via phone line from the dealing board was by using a pre-programmed direct phone line on the dealing board to communicate with an individual trader. The audio input and output for this communication would be exclusively via an audio handset connected to the dealing board. The left component of the dealing board did not have an in-built microphone or speaker. In order to open a direct phone line to an individual trader by this method, the broker would push the appropriate button. The second method of bilateral communication via phone line from the dealing board was to dial a particular phone number.

207    Second, a broker could use the squawk box component on the right side of the dealing board to communicate with one or more individual traders using an in-built microphone and side by side speakers via dedicated lines. These lines could be opened by depressing and holding a switch or by switching up which would cause the line to stay open until the switch was manually moved back to the neutral position. The line controlled by each of these switches corresponded to a different individual trader at an individual bank or client. While a button was either switched up or held down, the sound picked up by the in-built microphone at the broker's end would be transmitted to the individual trader corresponding to that button. It was not necessary for the trader to accept or answer in order to activate the transmission from the broker's end.

208    Each of the various individual traders at banks could similarly control if and when they transmitted sound from their end back to the broker. When they did so, it would emerge from one of the in-built speakers in the dealing board, and would be audible not only to the broker but also to anyone else within earshot. When on the broker's dealing board there were lines to multiple traders open at the same time and one trader spoke, that trader's voice would be transmitted from one of the speakers in the dealing board, be picked up by the microphone, and be heard by all other brokers on the desk.

209    The close proximity of the brokers enabled them to hear not only the voice that emerged from the speakers of their own dealing board but also the voice that emerged from their neighbours' dealing boards. The position of the inputters enabled them to hear what the brokers were calling out, and also what was coming out of the brokers' dealing boards.

210    In addition to a dealing board, each broker had several computer monitors on their desk which were used by brokers and inputters to access the electronic systems "TP Contributions" and "Ausmarker". The monitors were also used to access, depending on the individual's preference, data screens available via subscription from Reuters, Bloomberg, or both.

211    Trading in Prime Bank Bills through Tullett occurred on each Sydney business day between 8.30 am and 4.30 pm. Trades executed before midday on a particular day were settled that day, while trades after midday were settled on the following business day.

212    There was a peak of activity in the trading of Prime Bank Bills during the relevant period at around 10.00 am on each business day. On average, approximately 90% of trading in Prime Bank Bills on a given day occurred during this period of high activity. The extent and precise time and duration of this peak of activity varied from day to day. During a period of high activity, there might be trades of several billion dollars of Prime Bank Bills. The period of trading would always be at around 10.00 am, but could be on either side of it.

213    During the relevant period, trading during the period of high activity around 10.00 am predominantly involved trading in early bills during the first half of the month and trading in late bills during the second half of the month; I will elaborate on the "early" and "late" question in a moment.

214    The brokers who traded in Prime Bank Bills on any given trading day assembled around the Short End Desk at about 9.45 am each business day together with the inputters.

215    An overview of the practice of the Short End Desk in arranging and facilitating a trade of Prime Bank Bills in the relevant period during the period of high activity in Prime Bank Bill trading around 10.00 am is as follows:

(a)    A broker would take instructions from a trader representing a particular bank trading Prime Bank Bills to bid or offer Prime Bank Bills of a particular tenor at a particular volume and yield. The trader would usually give such instructions through one of the two methods of communication made available by the broker's dealing board, that is, verbally over the squawk box or via the telephone.

(b)    The broker would then quote the bid or offer of Prime Bank Bills of the specified tenor, volume and yield to the other brokers on the desk verbally, by calling it out.

(c)    The other brokers would communicate the bid or offer to their trader(s) verbally through their dealing boards using one of the two methods described above. At the same time, the bid or offer would be inputted into an internal electronic system maintained by Tullett called "TP Contributions". Brokers would often also take handwritten notes of bids and offers from their clients, although there was no uniform system and each broker had their own practice in this regard.

(d)    Another broker would receive new bids or new offers, counter bids or counter offers, or directions to give or take existing bids and offers from other traders. Again, these instructions were received via one of the methods of communication made available by their dealing board. This broker would process the communicated information by matching any bids and offers, inputting counterbids or offers or adjusting existing orders. When transacting, commonly the broker would say words to the effect of "taken" or "mine" when accepting an offer, and words to the effect of "give" or "yours" when accepting a bid.

(e)    Acceptance of the bid or offer would then be inputted into the order management system TP Contributions. This would cause the bid or offer to become briefly highlighted in yellow on the TP Contributions screen before disappearing from this screen. Executed trades then appeared on the deal management system called "Ausmarker". The brokers would also inform their respective client traders of the executed trade, again via their dealing boards.

(f)    During this period of concentrated trading there was a person on the desk who monitored the public screens operated by Reuters and Bloomberg to make sure that the data from TP Contributions was being fed through consistently.

(g)    Following the concentrated period of Prime Bank Bill trading on a given day, brokers on the Short End Desk would typically:

(i)    confer amongst themselves to ensure that their respective trades aligned or "squared up" with each other;

(ii)    obtain information from traders on the selling side of each trade in relation to the Prime Bank Bills that were to be delivered in settlement of the trade (such as the identity of the issuer/acceptor and the precise maturity date), and input this information into the Ausmarker system (or have inputters do so);

(iii)    aggregate (or have inputters aggregate) the individual trades in the Ausmarker system entered into between the same clients in the same tenor into a consolidated trade for convenience;

(iv)    report back to traders (usually via their dealing boards) as to the trades arranged on their behalf that day, which allowed traders to conduct a process of reconciliation and to confirm that they accepted the details of the trades as accurate.

216    Tullett was not involved in the actual settlement of trades between clients.

217    In the event of a dispute between clients as to the terms of trades arranged between them through brokers, Tullett would attempt to negotiate an agreed trade between the respective clients.

218    Further, it was possible where a broker was taking instructions from more than one client, which was common, for the same broker to receive and process instructions from the trader-clients on both sides of the same trade. When this occurred, the trade was still inputted into Tullett's systems and handled in the same manner as described above.

219    Tullett invoiced clients in respect of its brokering services on a commission basis calculated by reference to the trades recorded in its systems. So unless a completed trade was executed into TP Contributions (and hence Ausmarker), Tullett would not generate an invoice. This invoicing was done by Tullett's Accounts team at the end of each month, on the basis of all trades arranged on behalf of a given client over that month. There was no systematic practice of providing a written summary of trades to clients on a more frequent basis.

220    The accuracy of the data contained in TP Contributions (and ultimately fed into Ausmarker and the public screens of Reuters and Bloomberg) depended on the accuracy of the inputting performed by brokers and inputters.

221    In addition to the communications referred to above in relation to specific bids, offers or trades, brokers would occasionally provide general information to traders via their dealing boards about the state of the market for Prime Bank Bills at a particular time. This information included, for example, the yield at which each tenor of Prime Bank Bills was trading at that time.

(d)    BBSW and the rate setting mechanism

222    During the relevant period, the BBSW was the trimmed, average mid-rate of the observed best bid/best offer for Prime Bank Bills for certain tenors on each Sydney business day published by AFMA. The BBSW was calculated and published by AFMA for the 1 month (30 days), 2 month (60 days), 3 month (90 days), 4 month (120 days), 5 month (150 days) and 6 month (180 days) tenors. It was set on the basis of views submitted by the BBSW panellists as to the yield at which Prime Bank Bills in each tenor were trading at 10.00 am or approximately 10.00 am. The precise time is a matter of dispute, which I will return to later. Westpac was a BBSW panellist throughout the relevant period.

223    Conceptually, the BBSW represented the rate of interest at which banks borrowed funds from one another for a short period of usually no longer than 12 months. It therefore represented a near risk-free interest rate, and naturally provided a benchmark, or otherwise supplied an important variable, for a wide range of transactions within Australia, if not more broadly.

224    In the relevant period, the BBSW was a key reference and benchmark rate in the financial markets in Australia, the independence and transparency of which was an important factor in ensuring the efficiency and integrity of those markets. It was the Australian equivalent of LIBOR and SIBOR (Singapore Interbank Offered Rate) in terms of usage as a benchmark or reference interest rate. The BBSW was, however, defined and calculated in a way that differed from LIBOR and SIBOR.

225    AFMA published the BBSW Procedures, which outlined the responsibilities and membership of the BBSW Committee, the BBSW calculation mechanism and the contingency rate setting procedure. The BBSW Procedures also included procedures for corrections to published and displayed BBSW rates. The BBSW Procedures applied throughout the relevant period and were updated regularly throughout the relevant period, including in April 2010, January and December 2011 and February, April and June 2012. I would note for completeness that Mr Love, Director of Policy and International Affairs at AFMA during the relevant period, relied upon a version of the BBSW Procedures dated April 2010 when describing the submission process.

226    I would also note that ASIC tendered as part of Mr Love's evidence the following additional documents:

(a)    AFMA document titled "Rate Contribution Best Practice Principles", dated March 2013;

(b)    an email from Mr Damian Jeffree of AFMA to BBSW panellists (including Ms Dawson of Westpac) requesting copies of their internal guidelines for staff making submissions to the BBSW rate set dated 17 October 2012;

(c)    documents relating to AFMA's "MGC BBSW Working Group" dated September and December 2012; and

(d)    a version of the BBSW Procedures dated February 2013.

But all of these documents fell outside the relevant period. For the moment, nothing more need be said about them.

227    In the relevant period, AFMA calculated and published the BBSW on each Sydney business day for each of the 1, 2, 3, 4, 5 and 6 month tenors. As I have said, the BBSW was the trimmed, average mid-rate of the observed best bid/best offer for Prime Bank Bills for each tenor. It was set on the basis of views submitted by BBSW panellists as to the yield at which Prime Bank Bills in each tenor were trading.

228    The BBSW Procedures determined the maturity date of a particular tenor on a "modified" following" basis. This meant that if the straight date for the tenor fell on a non-Sydney business day (e.g. a weekend or a public holiday), the maturity date would be rolled over to the next business day. The exception to this convention was when the rollover would push the maturity date over the mid-point of the month, because of the earlies/lates dynamic that was present in the market, on which I will say something more later. In such cases, the Sydney business day preceding the straight date would become the maturity date.

229    BBSW panellists made their submissions through an IT system that was used by AFMA to collect, calculate and disseminate market data, known as the AFMAdata system. Evidence as to the operation of that system was provided by Mr Jason Brown, Senior Information Systems Manager at AFMA during the relevant period. BBSW panellists inputted BBSW mid rates for each tenor quoted to two decimal places into the AFMAdata system. Not every BBSW panel member submitted their view of the mid-rate for each tenor on every Sydney business day. Rates ineligible for inclusion in the BBSW calculation process included those that failed to update by 10.05 am and those that failed to update for each tenor.

230    Mr Brown gave evidence that minimal change took place in the AFMAdata system and procedures relating to the BBSW submission process during the relevant period. Each BBSW panellist nominated one or more staff members as submitters and provided their details to AFMA. Submitters were provided with a unique username and password for accessing the submissions screen on the AFMAdata website. The mid-rates for each tenor could be entered into the submissions screen only between 10.00 am and 10.05 am on each business day (Submissions Period).

231    For the avoidance of doubt, it is appropriate to make a distinction between the Submissions Period and the BBSW Rate Set Window. The former lasted from 10.00 am to 10.05 am and related to the physical input of BBSW submissions. The latter lasted from 9.55 am to 10.05 am and is the shorthand description for the period during which the majority of Prime Bank Bill trading occurred.

232    Let me elaborate on the procedures that were prescribed for the calculation of the BBSW on each Sydney business day during the relevant period:

(a)    As I have said, AFMA received submissions from the BBSW panellists, who inputted their submissions into the AFMAdata system. The AFMAdata website included a submissions screen, accessible with submitters' login details, into which the mid-rates for each tenor could be entered during the Submissions Period, but not at other times. Each submitter was able to make multiple entries into the submissions screen during the Submissions Period. When a submitter made multiple submissions during the Submissions Period, AFMA used only the last submission as part of the BBSW calculation process. In order for a submission to be made, it was necessary for the submitter to enter mid-rates for all tenors; otherwise, the submission was not accepted by the AFMAdata system. When a submitter made multiple submissions during a single Submissions Period, it was necessary on each occasion for the submitter to enter a mid-rate for all tenors; otherwise the submission was not accepted.

(b)    AFMA required each BBSW panellist to make a submission to AFMA by 10.05 am Sydney time on each Sydney business day identifying that BBSW panellist's view of the mid-rate of the yield for Prime Bank Bills at 10.00 am, expressed to two decimal places, on that Sydney business day for each tenor. BBSW panellist's submissions on any particular day would not be considered by AFMA unless the BBSW panellist provided a submission for each tenor by 10.05 am Sydney time.

(c)    Following the receipt of submissions from the BBSW panellists, at 10.06 am on each business day the AFMAdata system automatically calculated the BBSW for each tenor, based on the submissions received during the Submissions Period. The highest and lowest mid-rates submitted for each tenor were eliminated, starting with the highest, until a maximum of eight submissions remained.

(d)    The BBSW for that day was calculated or "set" by AFMA by eliminating the highest and lowest submissions for each tenor and ascertaining, to four decimal places, the average of the remaining mid-rates for each tenor.

(e)    The calculated BBSW for each tenor, and the submissions made by each BBSW panellist, were automatically published by the AFMAdata system via XML feed to assigned information vendors including Thomson Reuters and Bloomberg. This occurred shortly after the assigned calculation time. The relevant information was published through the distinct output pages identified with the following codes:

(i)    "BBSW" (an output page for calculated BBSW mid-rates);

(ii)    "BBSWAV" (an output page for calculated BBSW mid-rates, BBSW average bid and BBSW average offer);

(iii)    "BBSWHIST" (an output page for calculated BBSW average mid-rates for the previous 10 business days); and

(iv)    "BBSWCONT" (an output page for the submissions made by BBSW panellists).

(f)    The calculated BBSW mid-rates were uploaded onto AFMA's public website (www.afma.com.au) at approximately 10.00 am on the following business day.

233    Under cross-examination, Mr Brown conceded that the above BBSW calculation methodology involved two elimination processes, described in subparagraphs (c) and (d) in the preceding paragraph, such that when eight or more BBSW panellists made valid submissions, the AFMAdata system calculated BBSW as an average of six mid-rates. This concession was made after it was shown, by reference to the submissions of BBSW panellists on a particular day for a particular tenor, that the elimination of submissions to six rather than eight was the only arithmetic possibility given where BBSW set that day in that tenor.

234    For completeness, this is consistent with reading sections 2.5 and 2.6 of the BBSW Procedures to be describing two separate elimination processes: first, the elimination of the highest and lowest rates in each individual tenor until a maximum of eight mid-rates remain to be displayed; and second, the elimination of the highest and lowest of those mid-rates before calculating BBSW.

235    Further, the BBSW Procedures also explicitly stated that:

(a)    if five, six, seven or eight contributions were displayed for a tenor, only three, four, five or six respectively would be used to calculate BBSW for that tenor; and

(b)    if there were fewer than five eligible input rates for a particular tenor, no contributions would be displayed and no calculation for BBSW would be performed for that tenor.

236    Now although the BBSW procedures required BBSW panellists to make submissions identifying the mid-rate of the yield for Prime Bank Bills at 10.00 am, submissions could be made or updated until 10.05 am. Westpac submitted that submissions were made based on trading at 10.00 am. Contrastingly, ASIC submitted that BBSW panellists in fact submitted their view of the mid-rate at approximately 10.00 am on each trading day. As will be shown below, the evidence of BBSW panellists revealed a variety of approaches to determining and inputting their submissions. Some of the BBSW panellists took into account trading behaviour that they observed after 10.00 am.

237    Let me deal with some other matters before discussing the specific evidence given by panellists.

238    Mr Brown gave evidence that the submissions made using the AFMAdata system were monitored by Mr Brown or another AFMA officer as they were made, for the purpose of checking whether there appeared to be a mistake in the data input due to a "fat finger" entry, meaning an obvious keyboard input error. If such an error was detected, it was Mr Brown's practice to ring the submitter and ask them whether they were comfortable with the data input. Mr Brown did not recall doing this very often during the relevant period.

239    A submitter was also able to call or email AFMA if they could not access the AFMAdata system, for example due to connectivity issues. As the administrator, Mr Brown could enter data on behalf of the BBSW panellists into the AFMAdata System. Mr Brown recalled that it was very rare for him to submit rates on behalf of BBSW panellists during the relevant period.

240    Further, in the relevant period, the AFMAdata system was programmed to automatically publish the BBSW from approximately 10.10 am on each Sydney business day via XML feed to assigned information vendors. The rates were made available to subscribers to the BBSW service by the information vendors, which in the relevant period included Thomson Reuters and Bloomberg. As a part of this publication process, all BBSW panellists were able to view, at about the same time as the publication of the BBSW, the last submissions of all BBSW panellists made on each business day. If a BBSW panellist had not submitted rates, a blank row would appear next to that panellist's name.

241    Further, on each business day, at approximately 10.00 am, the AFMA website was manually updated by Mr Brown or one of the other IT staff at AFMA to record the previous day's BBSW output. This information would then be available to the general public.

242    Further, AFMA also published the average bid and offer values for each tenor referred to as BBSY, but calculated using a consistent spread of five basis points either side of the BBSW.

(e)    Evidence from BBSW panellists

243    ASIC filed evidence setting out the practices of several BBSW panellists during the relevant period. Their evidence demonstrated a variety of submission practices during the relevant period on issues including:

(a)    whether submissions considered trading that occurred after 10.00 am;

(b)    whether submissions were updated and re-submitted during the BBSW Rate Set Window;

(c)    how individual submitters accounted for and weighed against one another the volumes and yields of trading; and

(d)    how individual submitters determined their submissions for illiquid tenors.

244    The parties are in dispute as to whether I should find that post-10.00 am trading had an effect on BBSW submissions, a consideration which is relevant inter-alia to the question of the likely effect on the BBSW of Westpac's trading on the purchase and sale contravention dates, which will become clear much later in these reasons. ASIC submits that such trading was taken into account, relying upon the evidence of BBSW panellists outlined below as well as comments by Westpac's witnesses. Contrastingly, Westpac contends that I should accept that panellists did what AFMA required them to do, i.e. relied on 10.00 am trading only. I will address this issue again later, although I am inclined to say at this point that ASIC's position is more consistent with the evidence at least in relation to some of the panellists' practices. But I would also note that given the two elimination processes that I have described earlier, the non-conforming submissions may in essence have had little significance on where BBSW set.

HSBC

245    Mr Graeme Bell, an Associate Director in Balance Sheet Management at HSBC throughout the relevant period gave evidence that he was the main person at HSBC making submissions to AFMA during the relevant period. Mr Bell used the HSBC general user ID and password to log in to the AFMA webpage. His general practice was to base submissions on the middle rate of the bids and offers displayed on each of the ICAP and Tullett broker screens at 10.00 am. The broker screens showed the last bid and offer in each of the tenors. Sometimes the two brokers displayed different rates for some tenors and at other times they displayed the same rates. Mr Bell said that he watched the broker screens for the whole Submissions Period and, if there were changes in the rates on the brokers' screens, he would re-submit the mid-rate between the bids and offers on the brokers' screens into the AFMA webpage. It was not unusual for multiple submissions to be made by Mr Bell during a Submissions Period.

246    If there was no trading in a particular tenor and no major changes in economic events overnight, Mr Bell would base his submission on the previous day's rate set and would make an arithmetical adjustment to make the submission for that tenor.

247    Mr Jonathan Hensler was an Associate and later an Associate Director in Balance Sheet Management at HSBC. He was another person who made submissions to AFMA on behalf of HSBC during the relevant period. Mr Hensler estimated that he made the submissions on approximately 30% of the days during the relevant period and Mr Bell made the submissions on approximately 70% of those days. Mr Hensler would log into the AFMA webpage prior to 10.00 am and would watch the ICAP and Tullett broker screens for the whole of the Submissions Period. The information Mr Hensler had regarding the trading of Bank Bills on any day was via the last bid and offer made as displayed on the brokers' screens. Mr Hensler did not know whether the displayed bids and offers actually traded. If there was a difference between either the bids, or the offers, between the two screens, Mr Hensler would try to calculate the mid-rate between them for submission to AFMA based on where he thought the Bank Bills in each tenor were trading at the time. If there were no bids or offers for less liquid tenors (2, 4 and 5 month tenors) on the broker screens, Mr Hensler would make an arithmetical adjustment to calculate the rate to input for those tenors. During the relevant period, Mr Hensler made multiple submissions into the AFMA webpage almost every day that he made the submissions on behalf of HSBC. His first priority at 10.00 am was to make sure that the rates for each tenor were submitted and it was then his practice to re-submit rates during the next five minutes, to 10.05 am, based on any subsequent trading he observed in the Submissions Period on the brokers' screens.

Société Générale

248    Société Générale was a BBSW panellist from the start of the relevant period until 28 May 2010, a period of around one month. Mr Daniel Cortis and Mr Jeffrey Weiss were both employed in Treasury at Société Générale in Australia until approximately September 2010, when it ceased its Australian-based Treasury operations.

249    Mr Cortis and Mr Weiss have not been employed in the banking industry since that time and their evidence is based on their recollections of making submissions to AFMA on behalf of Société Générale in the first half of 2010. Their recollections were not strong.

250    Mr Cortis recalled that between about March and September 2010, his role at Société Générale included submitting Bank Bill rates to AFMA by logging into an AFMA website. If he was away from the office or otherwise occupied, one of his colleagues on the money market desk, usually Mr Weiss, made submissions to AFMA instead. Broker screens were the primary source of information on which Mr Cortis relied in making submissions to AFMA. Mr Cortis recalled looking at the Tullett broker screen and at one or two other broker screens, although he could not recall the names of any other brokers whose screens he watched. Mr Cortis also listened to an open loud speaker phone that was installed on the desk, which conveyed the voices of brokers who were making bids to purchase and offers to sell Prime Bank Bills. This line was open all day. Mr Cortis calculated the mid-point of the bid and offer for each of the tenors and entered this into the AFMA webpage at or around 10.00 am. He recalled that he then continued to observe further trading activity in Prime Bank Bills until approximately 10.10 am. If the mid-point changed for any one or more of the tenors during this time, Mr Cortis would update the data he initially entered into his submission or would make another submission to AFMA. If a flurry of Prime Bank Bill trading during the Submissions Period pushed the rate for a particular tenor above or below the rate Mr Cortis had submitted, he would update and re-submit the midpoint rate for that tenor. Mr Cortis tried to ensure that the mid-points he submitted to AFMA took into account the trading activity he observed and heard occurring in the Bank Bill Market up until the end of the Submissions Period. If trading was quiet, his practice was to submit the rates close to, or at, 10.00 am. If it was busy, he would submit the rates later in the Submissions Period. Where there was no trading of Prime Bank Bills in a particular tenor on a given business day, Mr Cortis determined the rate to submit to AFMA by using a yield curve. The yield curve extrapolated the price of other tenors by reference to trading data for the 1 month, 2 month and 3 month tenors, and data from the futures market.

251    Mr Weiss recalled that in the period between March and September 2010 he had a computer screen at his desk that displayed Prime Bank Bill trading information provided by ICAP, Tullett and possibly another broker, BGC Partners Australia Pty Ltd. These were live trading screens which recorded the latest trading information supplied by those brokers, being the best bids to purchase and the best offers to sell Prime Bank Bills in each tenor. There was also a live loud speaker phone in the office space in which the Treasury team worked. During the morning of each business day when Mr Weiss was present in the office, he heard, through the loud speaker phone, brokers calling out bids and offers to buy and sell Prime Bank Bills at various yields. He observed that, at any given point in time, the bids and offers announced by various voices through the loud speaker phone for any given tenor were always, or almost always, the same as the bids and offers that appeared on the broker screens for that given tenor. To the best of Mr Weiss' recollection, submissions had to be made to AFMA between 10.00 am and approximately 10.08 am (in fact, it was 10.05 am) through an AFMA website, for which he was given a user name and password. Mr Weiss said there was no formal arrangement as to which Treasury employee would submit information to AFMA on behalf of Société Générale on any given day. Mr Weiss estimated that he was the submitter on less than half of the business days between March and September 2010 and that Mr Cortis was the main submitter during that period. In making submissions to AFMA on behalf of Société Générale, Mr Weiss would log into the AFMA website at approximately 10.00 am, observe the bids and offers on the broker screens, listen to the bids and offers on the loud speaker phone, calculate the mid-point of the bids and offers for each tenor, type those mid-rates into the appropriate typing space on the AFMA website and submit the rates to AFMA. Mr Weiss estimated that this process for all tenors of Prime Bank Bills would take him approximately 10 to 30 seconds in total. Mr Weiss recalled continually updating his mid-rate submissions during the remainder of the Submissions Period by making a fresh submission each time that he observed or heard a change in the mid-rate for any given tenor of Prime Bank Bills. He would do this right up until the end of the Submissions Period, when it was no longer possible to make further submissions to AFMA.

Suncorp-Metway

252    Suncorp-Metway (Suncorp) was a reserve BBSW panellist from around 2006 and a BBSW panellist from 1 May 2012 to the end of the relevant period, a period of around one month. As a reserve BBSW panellist, Suncorp was on a list of banks that could be asked to fill in for any bank that withdrew as a BBSW panellist, but Suncorp did not make BBSW submissions prior to 1 May 2012. Mr Rory O'Connor and Mr Darryn Long gave evidence of the AFMA submissions process on behalf of Suncorp.

253    Mr Long had been the Team Leader of Suncorp's Interest Rate Trading Desk since July 2010. He began working at Suncorp in 2004 and had traded Prime Bank Bills since around July 2006. From around 1 May 2012 to the end of the relevant period, the Interest Rate Trading Desk was responsible for making BBSW submissions to AFMA on behalf of Suncorp. Submissions were generally made by Mr O'Connor under the supervision of Mr Long. From time to time, Mr Long made the submissions to AFMA himself, generally when Mr O'Connor was away or unavailable. Mr Long gave evidence that trading in Prime Bank Bills generally began in earnest at around 10.00 am and lasted until 10.05 am, when BBSW submissions were due to be made to AFMA. Mr Long understood that submissions made to AFMA were meant to reflect the rates at which Prime Bank Bills were trading during the Submissions Period. Mr Long's general practice in making submissions was to watch the ICAP and Tullett screens and listen to the squawk box during the Submissions Period. Based on the bids and offers that he saw and heard, Mr Long made a judgment of where the bulk of trading occurred in each tenor during the Submissions Period, to the nearest basis point. Mr Long's practice was to consider the bulk of the trading in each tenor during the Submissions Period as a whole and to base his BBSW submissions on the rate at which Prime Bank Bills traded most frequently during this period. For example, if the rate moved quickly and steeply after 10.00 am, Mr Long's submission would reflect the higher rate traded after 10.00 am. As he was not able to discern the volume of Prime Bank Bills traded, generally his practice was to make a BBSW submission at the level at which the trading had stayed for the longest time during the Submissions Period.

254    Mr O'Connor has been employed at Suncorp since 2006 and worked there as an Interest Rate Trader between 2011 and 2014, reporting to Mr Long. From around 1 May 2012 to the end of the relevant period, Mr O'Connor was the primary Prime Bank Bill trader at Suncorp and the primary submitter to AFMA on behalf of Suncorp. Mr O'Connor gave evidence that he understood BBSW submissions were required by AFMA to reflect each panellist's view of the mid-rate of the yield at which Prime Bank Bills in each tenor were trading each business day between 10.00 am and around 10.05 am. To the best of his recollection, submissions were due to be made to AFMA by 10.05 am or 10.06 am. Mr O'Connor's general practice in making submissions was the same as that of Mr Long, as described above, except that Mr O'Connor made a judgment of where the bulk of trading occurred in each tenor at about 10.00 am, rather than between 10.00 am and 10.05 am as described by Mr Long. Mr O'Connor considered that the busiest time of the day for trading Prime Bank Bills was from around 10.00 am to a few minutes after. On occasions when the rate for various tenors did not move significantly during this time, it was Mr O'Connor's general practice to base his BBSW submissions on an observable traded rate close to either side of 10.00 am. On occasions when there was a significant movement in the market, it was Mr O'Connor's practice to submit the rate which he considered best reflected the trading as a whole, being the rate at which most of the trading occurred. Mr O'Connor did not quantify a "significant movement", but rather said that it was a movement of a size sufficient to change his view of the rate that accurately reflected the market from around 10.00 am to a few minutes after.

255    Where no trading occurred in a particular tenor, Mr Long and Mr O'Connor gave evidence that they would either base their submission on:

(a)    the OIS for the corresponding tenor plus a margin; or

(b)    an interpolation of the rate between two other observable tenors (for example, the rate for the 2 month tenor could be determined based on the rates for the 1 and 3 month tenors).

Royal Bank of Canada

256    The Royal Bank of Canada (RBC) was a reserve BBSW panellist until May 2010 and a BBSW panellist from 28 May 2010. Mr Darren Ryper worked at RBC from May 2000 to November 2014. During the relevant period, he was the head of Treasury Management Services (TMS), a team responsible for the funding and liquidity of the Treasury branch.

257    TMS was responsible for making submissions to AFMA on behalf of RBC. The primary person responsible for this within TMS was Ms Thu Tran, a Money Market Dealer. Mr Ryper himself would only make BBSW submissions to AFMA if Ms Tran was sick or on leave. Mr Ryper estimated that he made BBSW submissions to AFMA on approximately 25 days each year.

258    Mr Ryper did not give Ms Tran instructions on how to make BBSW submissions because he observed that she was familiar with AFMA's guidelines on this subject. Mr Ryper would generally discuss BBSW submissions with Ms Tran only if she had any uncertainty about where to submit the rate in a particular tenor. Mr Ryper was able to observe how Ms Tran was making BBSW submissions to AFMA as they sat close by. Based on these observations and their discussions, Mr Ryper believed that Ms Tran's practice in making BBSW submissions was the same as his.

259    Mr Ryper gave evidence that he would monitor Prime Bank Bill trading from about 9.55 am to 10.05 am and calculate the mid-rate of the bid and offer for each tenor to submit to AFMA based on the tradeable prices he observed during the Submissions Period. Mr Ryper recalled that most of the trading in Prime Bank Bills took place during a five minute period, but could not remember whether this was from 9.58 am to 10.03 am, or from 10.00 am to 10.05 am, or some other five minute window between 9.55 am and 10.05 am. In order to make BBSW submissions, Mr Ryper would log into the AFMA website before 10.00 am using RBC's username and password. He would then look at the ICAP and Tullett broker screens and listen to live trading on two separate voice boxes. Sometimes he would call a broker to clarify whether a bid or offer had traded. At the start of the Submissions Period, Mr Ryper would key in the rates for each tenor based on the most recent tradeable prices available, but would not usually submit those rates straight away. Rather, he would continue to monitor the trading during the Submissions Period on the broker screens and the voice boxes and update his submission to reflect any changes in the rates at which Prime Bank Bills were trading. It was not common for Mr Ryper to make multiple submissions to AFMA because he would usually only submit his rates towards the end of a Submissions Period, that is, close to 10.05 am. He would only change a submission if he had pressed submit and then saw a tenor physically trade for a rate different to what he had submitted, or if he saw a higher volume trade occur. After the end of the Submissions Period, Mr Ryper would log off from the AFMA website and visit the Thomson Reuters BBSW page at approximately 10.15 am to view the BBSW rates as set by AFMA for that day. Mr Ryper's main concern when making a submission to AFMA on behalf of RBC was to ensure that the submission reflected the mid-point during the Submissions Period, based on the most recently traded prices for Prime Bank Bills. In deciding on the mid-point, Mr Ryper would place more weight on the prices of trades that were for higher volumes. Mr Ryper recalled that it was rare for there to be no bids or offers on the broker screens for a particular tenor, but when this happened he would look at trading in other derivatives products and could also use a computer program to do an adjustment from the BBSW for the day before.

General

260    In my view, it would seem that some submitters had regard not only to the bids and offers at 10.00 am, but also to the bids and offers up to the end of the Submissions Period, and would either hold off on submissions or update their submissions accordingly. Messrs Bell, Cortis, Hensler and Weiss updated their submissions up to the end of the Submissions Period, such that their submissions reflected the rates that they observed towards the end of the Submissions Period. Mr Ryper also based his submissions on the most recent bids and offers in the Submissions Period, although he also placed more weight on trades that were for higher volumes. Mr O'Connor based his submissions on the rate at which most of the trading occurred during the Submissions Period. Similarly, Mr Long based his submissions on the rates at which the trading occurred for the longest time during the Submissions Period.

261    It would seem that participants in the Bank Bill Market being other panel banks held the view that submissions were based on where the market was trading at around 10.00 am and up to a few minutes later.

262    Further, the evidence of Mr Colin Roden, an employee of Westpac and whose role I will expand upon in a moment, was that submissions by some submitters might take into account trading after 10.00 am, including trading up to 10.05 am. And Mr Masnick also said that trading in Prime Bank Bills at or shortly after 10.00 am might affect the rate at which BBSW sets.

263    I will return to this question later.

(f)    Other matters

264    Before proceeding further, I should say something by way of an introduction to two topics that have relevance to my later discussion.

Accounting for "early" and "late" paper differences in liquidity and yield

265    To promote liquidity in the Prime Bank Bills Market, maturities were concentrated and traded in half month periods, generally referred to as the "early" and "late" pools. Prime Bank Bills that matured on a business day between the 1st to the 15th day of a month were designated early, while instruments which matured on a business day between the 16th to the last day of a month were designated late.

266    Intra-month liquidity was viewed as deepest within the pool that represented the current maturity period, which was also the pool against which BBSW was set. For example, if on 3 December 2010 an investor wanted to buy 3 month Prime Bank Bills, it would normally have been bidding for early March 2011 Prime Bank Bills, i.e. it would have acquired Prime Bank Bills whose maturity could have been on any date between 1 March 2011 and 15 March 2011. The investor could buy late paper as well if it desired to do so, but market liquidity would be concentrated in the early paper and prices shown in platforms such as broker screens would typically be for early paper. However, once trading moved into the new period the stock purchased in the prior period generally became less liquid.

267    Market participants who held Prime Bank Bills for trading or for risk management purposes may have wanted to sell Prime Bank Bills when they were about to become less liquid. For instance, if for trading purposes a bank held Prime Bank Bills that were early Prime Bank Bills, it may have wanted to sell them on or before the 15th of the month. In other words, the bank may have wanted to reduce holdings of Prime Bank Bills over from the early pool when the trading started in the late pool, as those bills would become less liquid. Liquid instruments were preferred in situations where the management of risk had to be dynamic, as it may have required rapid changes in positions.

268    The desire to hold more-liquid assets was a factor that may have informed trading decisions for Prime Bank Bills. For example, if traders felt that it was important to have some amount of more liquid assets in the balance sheet because of concerns over market developments, they may have been more willing to sell existing early positions when transitioning into a late period.

Engaging in arbitrage between Prime Bank Bills trading and BAB Futures trading

269    During the relevant period, there was a very active futures market on Prime Bank Bills, with BAB Futures contract trading. As I have already said, the BAB Futures contract was physically settled, meaning that the seller of the contract, if not closing out its positions, had to deliver approved Prime Bank Bills at the expiration of the contract. Approved Prime Bank Bills had to be accepted by a Prime Bank, have a face value of $1 million, mature 85 to 95 days from the settlement date, and be classified as "early" month paper.

270    Prime Banks such as Westpac tended to have sizable inventories of qualifying Prime Bank Bills or could directly issue Prime Bank Bills to be delivered at the expiration of the BAB Futures contract, and were therefore in a better position to take advantage of any arbitrage opportunities that might have arisen between the different markets.

271    Arbitrage opportunities resulted from the fact that if either the BAB Futures contracts or the Prime Bank Bills were mispriced, it was possible to create a portfolio using the other instrument that would have the same payoff to take advantage of such mispricing. For example, one could borrow to purchase Prime Bank Bills that had a maturity that made them deliverable on the BAB Futures contract and at the same time sell a BAB Futures contract that promised to deliver the Prime Bank Bills at maturity of the contract. If the value of the Prime Bank Bills to be delivered taking into account the cost of borrowing was less than the BAB Futures price, the BAB Futures price was too high (yield too low) relative to the price of Prime Bank Bills and it was preferable to buy the Prime Bank Bills and sell the BAB Futures contract. Conversely, if the price of the BAB Futures was considered too low (its yield too high), a market participant could sell Prime Bank Bills (or issue them if a Prime Bank), invest the proceeds in a risk-free instrument, buy the BAB Futures, and then use those proceeds at the expiration of the BAB Futures contract to receive an exactly offsetting Prime Bank Bill. This strategy had minimal risk. If the futures price traded at a level that was considered too high close to expiration, i.e. at a low implied yield, Westpac could either use its existing inventory or issue Prime Bank Bills and deliver them into the BAB Futures contract, taking advantage of its perceived low yield.

WESTPAC STRUCTURE AND GOVERNANCE

(a)    The Board

272    Westpac is, and during the relevant period was, the holder of an Australian financial services licence and an authority to carry on banking business under s 9 of the Banking Act 1959 (Cth).

273    During the relevant period, Westpac's Board of Directors had ten members, of which only one, the CEO, was not considered independent. The Westpac Board formally delegated part of its functions to its Board committees and to its Executive management. The Board retained oversight of these functions through its committee reporting structures, approved policies and Executive reporting structures. Board committees reported directly to the Board.

274    One of the Board committees in the relevant period was the Board Risk Management Committee (BRMC). Its purpose was to oversee Westpac's risk profile, set risk appetite, approve frameworks, policies and processes for managing risk, and accept risks that were beyond the discretion provided by the Board to the Executive. The types of risk considered by the BRMC included, but were not limited to, capital, credit, liquidity, market, operational and reputation risk. It was also a key reporting line between the Board and the Group Treasury division of Westpac.

(b)    The Executive

275    The Westpac Executive was made up of individual division heads and officers with responsibility across the entire bank. In particular, two members of the Executive during the relevant period were important reporting lines for relevant divisions of the bank: Mr Phillip Coffey (Chief Financial Officer of Westpac) and Mr Robert Whitfield (Group Executive of the Westpac Institutional Bank).

276    In the relevant period, the Executive operated a number of Executive committees, including the following:

(a)    The Asset and Liability Committee (ALCO), which oversaw Westpac's funding and liquidity risk profile and liquidity risk management framework (including key policies) and identified emerging funding and liquidity risks and appropriate actions to address those risks. In particular, ALCO was a key reporting line between the Executive and the Group Treasury division.

(b)    The Group Market Risk Committee (MARCO), which oversaw Westpac's market risk profile and risk management framework (including key policies) and identified emerging market risks and appropriate actions to address those risks. MARCO approved separate "value at risk" sub-limits for the trading activities of Financial Markets and Group Treasury divisions.

(c)    The Group Credit Risk Committee (CREDCO), which oversaw Westpac's credit risk management profile and risk framework (including key policies) and identified emerging credit risks and appropriate actions to address those risks.

277    Figure 1 outlines the formal delegations of authority by the Westpac Board and the operation of committees by the Executive as described above.

Figure 1 Westpac corporate governance at Board and Executive levels

(c)    Relevant divisions

278    When describing Westpac's corporate structure during the relevant period, it is necessary to distinguish between the customer-facing business divisions that operated under multiple brands, and internal corporate divisions. The governance relationships between the two types of divisions seem to have overlapped. For example, policies generated by Westpac Institutional Bank, referred to as WIB, a customer-facing business division to which the Financial Markets division belonged, were applied to Group Treasury, the internal and otherwise separate division that traded Prime Bank Bills.

279    Westpac's operations during the relevant period comprised the following customer-facing business divisions, which operated under multiple brands:

(a)    The retail, business banking and wealth operations in Australia, including the businesses of Westpac Retail and Business Banking, St George Bank and BT Financial Group;

(b)    WIB, which delivered a broad range of financial services to commercial, corporate, institutional and government customers. As noted above, Mr Whitfield was its Group Executive; and

(c)    Westpac New Zealand, which was responsible for sales and services of banking, wealth, and insurance products for consumer and small to medium business customers in New Zealand.

280    In some respects St George Bank was separate from the rest of Westpac. On 1 December 2008, Westpac acquired St George Bank. But from then, Group Treasury's rate exposure formally included St George treasury's rate set exposure, although data about that exposure was not automatically and immediately migrated to Westpac's computer systems. From about August 2010, the St George rate set data (in the "Kondor" system) was migrated across to Westpac's trading data system called "Murex", so that any rate set exposure arising out of transactions entered into by the St George treasury function was automatically captured in the Murex system. Prior to that time Mr Patrick Stokes manually incorporated St George data into Westpac's database.

281    Group Treasury was an internal division of Westpac. It was generally responsible for the aggregate management of Westpac's balance sheet incorporating funding, liquidity and capital management. Because Westpac makes more loans (assets) than it has deposits from customers (liabilities) to fund those loans, Group Treasury's roles were to:

(a)    raise additional funds necessary to meet the gap between deposits and loans;

(b)    manage the financial risk associated with Westpac's banking operations including:

(i)    exposure to movements in relevant interest rates and their effect on the cost of funding and the net value of assets and liabilities;

(ii)    exposure to movements in the market in the short, medium or long term, which might affect operations or the costs of funding;

(iii)    exposure to currency movements;

(iv)    concentrations of risk in relation to significant counterparties;

(v)    the risk that sources of funding might become constrained or expensive (on either a short or longer term basis);

(vi)    the positioning of Westpac's operations so that it was an attractive investment opportunity for the key institutions that provide its funding; in part, that involved demonstrating that Westpac was strong and able to withstand potential adverse market developments; and

(c)    conduct such operations as efficiently as possible with a view to assisting Westpac to make a return on its shareholder capital.

Importantly in the present context, in carrying out these roles Group Treasury traded financial products and entered into transactions involving financial products, including BBSW Referenced Products and BAB Futures.

282    Financial Markets was another internal division within WIB and as such reported to Mr Whitfield. It contained sub-divisions including Debt Capital Markets and Global Capital Markets. It had books which from time to time dealt in certain BBSW Referenced Products and BAB Futures. Further, it included a foreign exchange desk which from time to time dealt in FRAs, interest rate swaps, BAB Futures and cross-currency swaps. Financial Markets did not buy or sell Prime Bank Bills in the Bank Bill Market, but it did have exposure to BBSW rate set risk, which it managed by trading in derivatives including BAB Futures and FRAs.

Relationship between Group Treasury and Financial Markets

283    Since at least around 2001, Westpac's balance sheet and associated risk (i.e. the risk in relation to the assets and liabilities held as part of its banking operations and managed by Group Treasury) have been separate from the trading books that capture the transactions entered into on behalf of customers and the risk associated with those positions. The customer positions were managed by Financial Markets, and the associated customer risk and information concerning those matters was kept separate from Group Treasury.

284    There was an information barrier between Financial Markets and Group Treasury, but the effectiveness of this barrier was a topic of dispute between the parties. At this point, I will make the following uncontroversial points.

285    The information barrier was implemented so that Group Treasury could independently manage Westpac's affairs. As such, Group Treasury was not to interact with Westpac's customers and traded only on the Bank's behalf. The information barrier was not defined in any written Westpac policy, but relevant Group Treasury staff were informed about it when they commenced with the division.

286    The information barrier was implemented structurally through reporting lines and management and operationally through the separation of relevant systems. Financial Markets employees could not access trading data of Group Treasury and vice versa. The information barrier policy did not prohibit contact between employees in those two divisions but it did prohibit the sharing of trading information which could give rise to any conflict of interest between Westpac's position and the position of Financial Markets' customers. Now the barrier did not stop discussion regarding general themes in the marketplace, but it did preclude sharing of specific information regarding positions, client information etc.

287    Around the same time as Westpac implemented the structural separation between Group Treasury and Financial Markets, Group Treasury took complete control of Westpac's Prime Bank Bill trading. Since that time, Financial Markets has not been able to buy or sell Prime Bank Bills, although it has from time to time requested that Group Treasury conduct such transactions on behalf of its customers.

288    The movement of Prime Bank Bill trading to Group Treasury was prompted by a range of factors, but primarily by the fact that Group Treasury was a much more significant user of Prime Bank Bills for both funding and liquidity management purposes. Consequently, it was logical for Group Treasury to have a direct interface with the Bank Bill Market. It also made better sense so as to avoid intra-bank pricing issues between the two divisions.

289    Now the information barrier prohibited some exchange of information between the divisions, but not every exchange. ASIC says that contemporaneous documents show that there was discussion between Group Treasury and Financial Markets about trading in the Bank Bill Market, where the BBSW would set or had set, and their respective positions in the market, and that such discussion often involved Mr Adam Parker, a swaps trader in Financial Markets. I will return to this question later, although I would note at this point that Mr Parker was not called as a witness.

290    In summary, Figure 2 shows how these divisions operated within Westpac's corporate structure, including relevant reporting lines and the operation of the information barrier.

Figure 2 Westpac corporate governance at division level

(d)    Group Treasury

291    As I have indicated, the core objectives of Group Treasury in the relevant period were to:

(a)    ensure that Westpac had sufficient funding to meet its obligations and business needs;

(b)    ensure that Westpac had sufficient holdings of liquid assets that could be used to raise funding quickly in the event that longer term funding options became constrained;

(c)    manage Westpac's exposure to liquidity risk, interest rate risk and other risk associated with its funding; and

(d)    conduct such activities in the most efficient and where possible, profitable manner.

292    Figure 3 shows the structure of the Group Treasury division, including its subdivisions and the relevant books. Figure 3 also identifies relevant individuals within Group Treasury.

Figure 3 Structure of Group Treasury

293    Group Treasury contained a number of sub-divisions including Portfolio Risk Management (PRM, also referred to as TRL), Asset Liability Management (ALM), Capital Management, Structured Finance, and Global Funding. In the present case, ASIC's allegations particularly concern the activities of the PRM sub-division.

294    Group Treasury had a number of reporting lines within Westpac. First, Group Treasury submitted monthly reports to ALCO and quarterly reports to the BRMC. The monthly reports to ALCO were in turn sent to APRA. Second, Group Treasury also undertook an annual funding review that outlined the funding strategy for the coming year (trends in global markets, peer analysis, wholesale funding capacity, expected funding requirements and a funding risk analysis). The annual funding strategy was reviewed and supported by ALCO prior to its approval by the BRMC. Third, Group Treasury reported in the relevant period to both Westpac CFO Mr Coffey and to the head of WIB, Mr Whitfield. Neither Mr Whitfield nor Mr Coffey were subject to the information barrier as neither of them engaged in trading on behalf of Group Treasury or on behalf of customers in Financial Markets. Further, whilst Group Treasury was an independent unit of Westpac, it was subject to the risk management and compliance framework, policies, and procedures that applied to WIB.

295    In the relevant period, the head of Group Treasury was Mr Curtis Zuber, Group Treasurer of Westpac. Group Treasury's trading activity represented dealings that included the management of interest rate, foreign exchange and credit spread risk associated with wholesale funding, liquid asset portfolios and foreign exchange repatriation.

296    Group Treasury was comprised of a "Trading Book" and a "Banking Book". The Banking Book is an accounting and regulatory concept that refers to assets and liabilities on Westpac's balance sheet that are expected to be accounted for on an accrual accounting basis. In the relevant period, the main sub-division (or portfolio) that comprised the Banking Book was ALM. ALM was the aggregate of Westpac's transfer priced assets, liabilities and associated hedges, with a duration of typically between one month and five years, primarily variable rate mortgages and customer deposits. The aggregated risk from ALM's assets was transferred to STIRR, which I will explain in a moment, and managed in the Trading Book.

297    The Trading Book is an accounting and regulatory term that refers to transfer pricing of Westpac's assets, liabilities and associated hedges and direct external products (for example, NCDs and US commercial paper) that are marked to market daily for accounting purposes. The "Treasury Eagle" document, which I will also elaborate on later, was prepared each day for this purpose and to assist Westpac to closely monitor and manage its exposure to interest rates.

298    In the relevant period, PRM was responsible for managing Group Treasury's Trading Book risk on a day-to-day basis and comprised books including the following:

(a)    Short Term Interest Rate Risk (STIRR), which book executed Westpac's short-term wholesale funding in the domestic AUD market, managed credit decisions regarding Westpac's holding of short-dated liquid assets (including Prime Bank Bills), centralised short-term interest rate mismatch and managed associated interest rate risk. STIRR consolidated this short-term interest rate risk by issuing and trading in Prime Bank Bills including, of course, NCDs directly in the short-term funding market, taking on ALM's short-term interest rate risk and trading certain products to manage risk. Further, offshore fundraising that was converted into AUD by the International Money Market (IMM) book was transferred to STIRR.

(b)    Long Term Interest Rate Risk (LTIRR), which book managed Westpac's long-term capital needs and associated interest rate risk (that is, risk of two years or longer). It was responsible for managing both the BBSW Rate Set Exposure and the outright interest rate risk associated with those holdings. The majority of its positions were long term interest rate risk positions transferred from ALM. The remaining positions arose from issuing long-dated AUD fixed-rate bonds. LTIRR traded derivatives, and if it wanted to manage its BBSW Rate Set Exposure it could enter into a FRA. But it did not have other ways to manage its risk.

(c)    IMM, which book raised short-term foreign currency funding offshore (predominantly USD) and managed the associated interest rate risk, and managed the underlying Bills/LIBOR book through trading in cross-currency swaps associated with long-term funding programs. Short-term foreign currency funding raised would be converted into AUD and loaned through to STIRR. The role of IMM within PRM included: raising short-term foreign currency (predominantly USD) funding offshore and managing the associated interest rate risk; where necessary, converting USD (or other foreign currencies) into AUD, including by entering into FX swaps; managing the underlying Bills/LIBOR book through trading in cross-currency swaps associated with long-term funding programs; managing Treasury's international branch's risk and liquidity; and managing Westpac's collateral position. It included managing BBSW Rate Set Exposure and trading in derivatives to reduce or offset that risk. IMM did not trade in Prime Bank Bills.

(d)    Foreign Exchange (FX), which book managed the conversion of foreign currency funding raised offshore into AUD. Risk on foreign currency was then consolidated within STIRR, having been converted into AUD.

(e)    Liquid Asset Management (Liquids), which book was primarily responsible for managing Westpac's credit decisions with respect to its holding of long dated liquid assets. These assets included long term semi-government bonds, long term government bonds and residential mortgage backed securities.

The STIRR Desk

299    As should be apparent from what I have said, management of short term interest rate risk within the STIRR book was an important part of PRM's activities, and several of its books conducted trading that fed into STIRR for the management of interest rate risk. Employees and officers of STIRR were authorised to deal, and dealt, in financial products, including FRAs, interest rate swaps, BAB Futures, basis swaps, and Prime Bank Bills including NCDs. Mr Roden, whose role I will expand upon in a moment, stated that that he "regularly traded, or supervised other STIRR employees trading, in BAB Futures and interest rate swaps".

300    Westpac's short term interest rate risk was centralised in STIRR. The most significant risk was the mismatch between Westpac's variable rate housing loan portfolio (in relation to which Westpac received interest) and Westpac's wholesale funding (in relation to which Westpac paid interest) and the resulting "basis risk" between its variable rate housing rate and the underlying wholesale funding rate.

301    Short term interest rate risk is the risk that interest rates for one day up to 12 months move in a way that adversely affects Westpac's balance sheet. The STIRR Desk primarily focused on managing that kind of interest rate risk within Westpac's risk appetite, while trying if possible to profit from the risk position at the same time. In addition, the STIRR Desk was tasked with Westpac's short term debt issuance of up to 12 months and its investment in short-dated liquid assets.

302    Those activities were to be conducted as part of managing Westpac's short term interest rate risk and in accordance with Group Treasury's funding and liquidity objectives. In order to carry out the functions of the STIRR Desk, traders on that desk were the only individuals that traded Prime Bank Bills on behalf of Westpac. Its traders could issue or sell Prime Bank Bills in order to raise short term funding.

303    The STIRR Desk assisted in building Westpac's holdings of short dated liquid assets by purchasing and holding Prime Bank Bills (amongst other assets). Prime Bank Bills were short dated liquid assets because:

(a)    they could be sold in exchange for cash with same day settlement;

(b)    they were, generally, readily tradeable on the Bank Bill Market, particularly in the one, three or six month tenors and maturing within the prevailing maturity period;

(c)    they were readily tradeable in bilateral deals with counterparties outside of the Bank Bill Market;

(d)    regardless of their tenor, Prime Bank Bills could be "repurchased" or "repo'd" with the RBA for immediate cash; and

(e)    alternatively, their short tenor meant they could be held to maturity at which point Westpac received cash.

304    In the relevant period, STIRR's funding objectives were a sub-set of Group Treasury's objectives to fund Westpac. Ms Sophie Johnston, an employee of Westpac and whose role I will expand upon in a moment, said that she was aware of Westpac's weekly, monthly and longer term funding objectives from the following sources:

(a)    the weekly Group Treasury funding meetings, at which members of Group Treasury would discuss the overall funding outlook for Westpac, along with updates on the balance sheet forecast and discussions about proposed issuances, performance in various funding markets and liquidity generally;

(b)    the Weekly Maturity Forecast report; and

(c)    informal meetings with the Group Treasurer and more senior Group Treasury staff.

The Cash Desk

305    The Cash Desk managed Westpac's intra-day liquidity and daily cash position. It monitored Westpac's daily cash flows, ensured there was sufficient cash to meet outgoing payments throughout the day and was responsible for ensuring that Westpac's AUD cash balance in its exchange settlement account (ESA) with the RBA each day was not short, but also that the position was not overly long. The Cash Desk would try to minimise the surplus cash in the ESA as those funds received interest at a rate 25 basis points under the RBA target cash rate. In practice, Westpac's ESA at the end of each day was generally positive.

306    It was difficult to predict Westpac's intra-day cash flows due to unexpected or unknown transactions. Westpac's officers on the Cash Desk would monitor the cash position and keep track of forecast known movements and other flows as advised by Group Treasury or the wider Bank as information became available. In particular, the Cash Desk would seek to identify the known cash position at the commencement of each day and then during the course of the day, estimate the cash position by updating the original cash position with new cash flows that it became aware of. At the end of the day, the Cash Desk would be able to identify Westpac's exact cash position, which would differ (sometimes significantly) from the running estimate during the day. The next morning, the Cash Desk would reconcile the difference between the estimated and actual cash position from the prior day. At the beginning of each day, the estimated cash position would have moved as a result of movements in global markets overnight, Westpac's activities in overseas jurisdictions, customer transactions and asset maturities. Westpac's estimated cash position was separate to any amount in Westpac's ESA.

307    The Cash Desk was responsible for funding shortages in the daily cash position as they arose, without holding excess amounts of cash. The Cash Desk would generally do this by borrowing cash in the overnight market from other banks at the cash rate, or through overnight FX forwards that relevantly converted the currency of Westpac's cash.

308    The key functions of the Cash Desk included managing Westpac's intra-day liquidity by ensuring cash was ready at hand to be deployed by Westpac, and ensuring the ESA was funded each day. To carry out these functions in the relevant period, the Cash Desk would:

(a)    enter into intra-day repurchase agreements ("repos") with the RBA as part of the RBA's "standing facilities";

(b)    enter into overnight AUD/USD FX forwards to exchange USD cash for AUD cash;

(c)    enter into bilateral cash loans with other banks;

(d)    enter into overnight repos with other market participants (although the market did not trade in this way for all of the relevant period); and

(e)    frequently discuss the Bank's cash position with STIRR and other portfolios of Group Treasury in order to identify the best source of short term funding or the best way to invest cash (as required each day). For example, if the Cash Desk was low on cash and funds needed to be raised:

(i)    the trader in STIRR could issue or sell Prime Bank Bills to raise funds;

(ii)    the trader in STIRR could enter into a repo agreement with the RBA; or

(iii)    the trader on the IMM desk could convert USD funds raised offshore into AUD through short to long term FX forwards or cross-currency swaps in order to add AUD cash to the position.

309    While the Cash Desk worked closely with other portfolios in Group Treasury on short end funding and liquidity management, it was ultimately the Cash Desk's responsibility to ensure that Westpac had adequate funding each day to meet its obligations. The traders in the STIRR portfolio would rely on the Cash Desk to manage intra-day cash flows, and to square the cash position at the end of each day having regard to any trading conducted by STIRR that impacted that position.

(e)    Personnel within Portfolio Risk Management

310    I have already mentioned Mr Zuber, the Westpac Group Treasurer. Mr Zuber did not trade Bank Bills in the Bank Bills Market. There are other individuals within Group Treasury who need to be identified in order to properly understand the evidence adduced by ASIC. I will focus on a few major personalities, noting that several officers within Group Treasury represented Westpac on AFMA committees before, during and after the relevant period.

Mr Colin Roden

311    Mr Roden held the position of Executive Director of PRM from at least the start of the relevant period until 31 December 2010. He was then Managing Director of PRM until the end of the relevant period. Mr Roden had oversight of the PRM books and day-to-day responsibility for managing the short term interest rate risk that sat within STIRR. During the relevant period, Mr Roden reported generally to Ms Dawson, Managing Director of PRM. While Ms Dawson was on maternity leave for some of the relevant period, he reported directly to Mr Zuber.

312    In addition to his responsibilities within PRM, Mr Roden represented Group Treasury on a number of internal Westpac committees, including the ALCO. Further, he has represented Westpac on a number of AFMA committees. Mr Roden attended meetings of AFMA's Negotiable/Transferable Instruments Committee (NTI Committee) as a member (10 March 2009, 3 June 2009 and 1 December 2009), and as an alternative (10 May 2011 and 2 February 2012). He attended a meeting of AFMA's BBSW Committee on 13 March 2009 as a member. He attended a meeting of AFMA's MGC on 8 June 2012.

313    During the relevant period, Mr Zuber had a large number of conversations on a regular basis with the STIRR team and, in particular, with Mr Roden. Mr Zuber frequently expressed his directive to Mr Roden to increase Westpac's holding of liquid assets, including short-dated liquid assets. However he gave Mr Roden the authority to decide the timing of when those assets should be bought, the nature of those assets and the strategy for holding them. In essence, Mr Roden's task was to manage the acquisition and management of Westpac's Prime Bank Bill portfolio as efficiently as possible, recognising that the holding of liquid assets invariably came at a cost. Mr Roden had discretion as to what days he sold or issued Prime Bank Bills or purchased Prime Bank Bills

Ms Sophie Johnston

314    Ms Johnston held the position of Associate in PRM from at least the start of the relevant period until 31 December 2010 and was then Senior Associate in PRM until the end of the relevant period. Ms Johnston assisted with Westpac's management of STIRR's book within PRM.

315    Ms Johnston worked on the Cash Desk from around October 2008 to July 2009. From about July 2009, while she was still on the Cash Desk, she began to get more closely involved in the funding activities conducted by STIRR. By August 2009 when she formally moved to the STIRR Desk, she had been transitioning into STIRR by assisting on each desk. She was assigned to the STIRR Desk from around August 2009 to around March 2011. When she joined the desk, she reported to Mr Roden. During that time, she did not trade Prime Bank Bills on all days and was away for periods of time. For example, she was in London in around March and April 2010 for about six weeks to provide cover for another Group Treasury officer, Mr Satruhan (Zac) Sharma.

316    Initially, Ms Johnston's role on the STIRR Desk involved:

(a)    assisting in the funding requirements conducted by the STIRR Desk communicated from Financial Markets;

(b)    assisting in the trading conducted by the STIRR Desk in Prime Bank Bills, and in particular, trading in 1 month Prime Bank Bills;

(c)    assisting in the funding conducted by the STIRR Desk in the RBA's daily open market operations;

(d)    participating in discussions concerning the management of the Bank's short term liquid asset portfolio and short term funding through the Bank Bill Market;

(e)    monitoring counterparty credit limits and inventory management;

(f)    tasks relating to Westpac's position as an AFMA panel bank, such as providing Westpac's BBSW submissions to AFMA; and

(g)    undertaking the administration associated with the settlement of Prime Bank Bill transactions, including the logistics involved in booking and settling the trades that occurred in the BBSW Rate Set Window.

317    By 2010, her responsibilities expanded to include, from time to time, trading in the other tenors of Prime Bank Bills and making decisions about short end funding with other members of Group Treasury in Mr Roden's absence. She also assisted Mr Roden with hedging and managing STIRR's interest rate risk.

318    She stopped working on the STIRR Desk in around March 2011 and moved to the Liquids desk. She subsequently returned to working on the STIRR desk in around September 2012.

319    Ms Johnston represented Westpac on AFMA's NTI Committee as a member (25 February 2010), as deputy chairperson (26 May 2010), and as chairperson (26 August 2010, 25 November 2010, 8 February 2011, 19 August 2011 and 10 May 2012). She also attended meetings of AFMA's BBSW Committee as a representative of the NTI Committee (26 May 2010) and as a member (30 November 2010, 10 February 2011, 2 September 2011, 15 November 2011, 20 March 2012 and 30 May 2012).

Other personnel

320    Other relevant personnel within PRM included:

(a)    Ms Joanne Dawson Managing Director (with the exception of two periods when she was on leave in 2010 from 24 May 2010 to 13 August 2010 and in 2012 from 27 February 2012 to 18 May 2012). Ms Dawson attended a meeting of AFMA's NTI Committee on 8 November 2011. She was also a member of AFMA's MGC and attended its meetings on 11 December 2009, 12 March 2010, 15 December 2010, 4 March 2011, 9 September 2011, 9 December 2011 and 24 February 2012. She did not give evidence for Westpac.

(b)    Mr Daniel Park Director from at least the start of the relevant period until 31 December 2010, then Executive Director until the end of the relevant period. Mr Park was involved from time to time in managing the IMM book. In addition, Mr Park attended a meeting of AFMA's MGC on 31 August 2012.

(c)    Mr Zac Sharma Associate Director. Mr Sharma was involved from time to time in managing PRM's Foreign Exchange book. He is presently Executive Director, Group Treasury.

(d)    Mr Patrick Stokes Associate Director. Mr Stokes was involved from time to time in managing Group Treasury's ALM book. He traded Prime Bank Bills on the STIRR Desk during the period October 2002 to mid-2005. He is presently Director, Group Treasury.

(e)    Mr Bryan Duignan Senior Associate from the start of the relevant period until 3 May 2012, then Associate Director until the end of the relevant period. Mr Duignan assisted from time to time with the management of the FX and IMM books. He is presently Executive Director, Markets, Risk and Liquidity, Group Treasury.

(f)    Mr William Hosie Associate from around October 2010 until the end of the relevant period. He is presently Director, Markets, Risk and Liquidity, Group Treasury.

(g)    Mr Richard Conway an interest swaps trader within Financial Markets from around January 2006 to April 2010. On 19 April 2010 he moved to Group Treasury and worked on the LTIRR desk. He is presently Director, Markets, Risk and Liquidity, Group Treasury.

321    A number of employees and officers in STIRR, including Ms Johnston and Mr Hosie reported from time to time to Mr Roden and other employees and officers in Group Treasury during the relevant period. A number of employees and officers in PRM more generally, including Mr Sharma, Mr Stokes and Mr Duignan, also reported to Mr Roden from time to time during the relevant period.

Performance of divisions and remuneration of personnel

322    Group Treasury had an earnings target and there was a general expectation that Group Treasury would seek to make a profit. PRM was the only division in Group Treasury that had a profit/loss statement in the relevant period, and both PRM and ALM had profit targets.

323    Further, during the relevant period, Group Treasury officers had performance frameworks including financial objectives and non-financial objectives. Remuneration for Group Treasury officers included fixed pay and variable components. Mr Zuber had primary responsibility for allocating performance-based payments within Group Treasury. He considered an individual's contribution to the overall effective operation of the Group Treasury function and its contribution to Westpac's broader financial success.

324    With respect to Mr Roden in particular, Group Treasury's financial performance over the preceding 12 months was a factor taken into account in relation to the assessment of performance and the level of any variable incentive remuneration. The bonus system allowed Westpac employees to receive substantial performance payments.

325    The operation of remuneration policies is relevant because ASIC has submitted that financial incentives payable to bank officers, particularly Mr Roden, were a relevant influencing factor on their trading behaviour. I would agree with that submission in the generality with which it has been put, but at the end of the day I do not consider that it takes the matter far. No doubt he had an incentive to trade profitably, but that is not saying much. I will return to this question later.

Staff training

326    Group Treasury employees undertook training, including training offered by AFMA in the form of the Diploma of Financial Services. In addition, STIRR employees learned about trading in Prime Bank Bills on the job from more experienced employees, including Mr Roden.

(f)    Conduct policies

327    Appropriately for an institution of its size, Westpac established and maintained a number of policies and procedures at different levels within the business during the relevant period. These included:

(a)    Westpac's Code of Conduct, which applied to all Westpac Group employees and contractors and was supported by the Board;

(b)    Westpac's Risk Governance Framework, which applied equally to Westpac and its subsidiaries;

(c)    the Dealing Room Policies for Trade Risk: Market Risk Management;

(d)    the Dealing Room Policies for Risk in Financial Markets and Treasury;

(e)    the Management Frameworks for Market Risk;

(f)    the Management Frameworks for Liquidity Risk;

(g)    the Trading Book Policy Statement;

(h)    the Interest Rate Risk in the Banking Book Summary of Key Methodologies;

(i)    the WIB Chinese Walls and Physical Access Controls; and

(j)    the Conflicts of Interest Policy.

328    As part of Westpac's Code of Conduct (versions dated 10 December 2009 and 10 August 2011), Westpac required all staff to:

(a)    abide by and comply with Westpac's conduct policies;

(b)    act with honesty and integrity;

(c)    respect and comply with all relevant laws and regulations;

(d)    notify superiors or compliance representatives as soon as possible of any breach of the law or any Westpac policy by any colleagues;

(e)    respect confidentiality and not misuse information;

(f)    only use Westpac property, including information technology resources, for proper purposes;

(g)    take responsibility and be accountable for the decisions made in performing duties;

(h)    maintain and improve the skills, knowledge and competencies required for the position;

(i)    consider the broader impact of decisions on customers and the community; and

(j)    act fairly and reasonably towards customers and potential customers in a consistent and ethical manner.

329    Professor René M Stulz, an expert called by Westpac, described in some detail the structure of Westpac's governance framework, which evidence was not challenged by ASIC. Professor Stulz is the Everett D. Reese Chair in Money and Banking at The Ohio State University, where he is also Director of the Dice Center for Research in Financial Economics. He completed a PhD in economic at the Massachusetts Institute of Technology in 1980, and has published widely in finance and economic journals. He has taught risk management for a number of years to MBA students and to executives, consulted and advised regulatory bodies on issues of risk management and is responsible for a worldwide certification examination for risk managers. He has written a textbook on the topic of risk management and is a trustee of the Global Association of Risk Professionals. I found his evidence on the topic of risk management and regulation useful and reliable. What follows is a description of the framework and policies, based upon Professor Stulz's evidence and the documents in evidence. The effectiveness and operation of particular aspects of the framework or of the framework in general is something I will address later in these reasons.

General risk management "three lines of defence"

330    During most of the relevant period, Westpac had a "Risk Governance Framework" in place that established "three lines of defence"; the relevant Risk Governance Framework documents applying were dated 30 April 2010, 28 April 2011 and 5 March 2012. Professor Stulz stated that such an approach was common in large financial institutions. It was introduced at Westpac early in 2010.

331    The first line of defence consisted of having each business unit identify, assess, and manage the risks it was exposed to in light of its pre-existing risk limits and policies. Each division was responsible for ensuring it had sufficient resources to effectively monitor its risk and ensure it complied with firm-wide and business-specific risk limits and policies.

332    The second line of defence was to have established risk management policies and frameworks for monitoring risk at the business level, operating in an independent fashion. In turn, it contained three layers:

(a)    The first layer contained executive risk committees comprised of both risk and business representatives, whose ultimate goal was to optimise the risk-reward relation for each relevant risk factor. They provided advice on the development of divisional risk appetite statements, risk management frameworks, limits, and policies, and monitored the levels of risk of the businesses to guarantee that they were aligned with existing risk appetite limits and policies. At Westpac these committees included the Operational Risk & Compliance Committee (OPCO), CREDCO, MARCO and ALCO. As already described, ALCO was responsible for overseeing Westpac's funding and liquidity risk.

(b)    The second layer consisted of a "Group Risk" function, independent of the business side of the bank and reporting directly to Westpac's Chief Risk Officer (CRO). The CRO reported to the CEO and had direct access to the chairs of the Board and of the BRMC. At Westpac, key responsibilities of Group Risk were centred around the development and maintenance of (1) group-wide risk management frameworks, policies, models, and procedures, and (2) group-wide risk estimates and risk capital models. Group Risk was also responsible for monitoring the quality of the risk management information provided to senior executives and Board members.

(c)    The third layer consisted of the specific risk units of each business division, which were independent of the business and reported directly to the CRO. They were responsible for developing the relevant risk management policies, procedures, monitoring and reporting systems, and controls at the business division level, in line with Westpac's general Risk Appetite Statement (RAS), divisional RASs and the wider risk management framework.

333    The third line of defence contained internal and external auditors and advisors, who provided an independent assessment of Westpac's risk management framework, policies, procedures, and controls.

Operation of the "risk appetite statement" in Group Treasury

334    As noted above, a number of the compliance policies formulated by WIB were applied to the Group Treasury division, despite Group Treasury being an independent division of the Bank.

335    As part of the second line of defence described above, each divisional risk area was responsible for developing specific divisional RASs, policies, controls, procedures and monitoring and reporting capabilities. These were to align to the general RAS issued by the Board and the specific risk management frameworks approved by the BRMC.

336    Over the relevant period, WIB was governed by a RAS that was updated yearly and that also applied to Group Treasury. The design of WIB's RAS incorporated WIB-specific factors. For example, it aimed to account for WIB needing to be an attractive option for Westpac to invest capital and funding. WIB's risk appetite was proportionally higher than Westpac's overall appetite, reflecting the inherently higher volatility of its institutional banking business compared to retail banking.

337    WIB's RASs dated August 2010, August 2011 and April 2012 record in relation to compliance and regulatory risk that Westpac:

(a)    operated within a culture of compliance based on the proactive management of identified and potential risks;

(b)    aimed to meet both the letter and spirit of legal regulatory measures;

(c)    had zero tolerance for intentional regulatory breaches;

(d)    would notify the regulator where a material regulatory breach was identified; and

(e)    was required to have in place compliance plans, setting out obligations and controls, for each business area, which were to be regularly monitored.

338    WIB's RAS discussed the key strategic risks facing the institution and the division across different areas: regulatory, business, country, and people risk. It then identified the most important dimensions of risk and the limits associated with each of them. Beyond that, the RAS laid out specific risk appetite parameters across different dimensions: rate of return, core and non-core activities, volatility of results, credit and concentration risk, equity risk, market risk, liquidity risk, operational risk, regulatory risk, reputational risk, sustainability, and capabilities.

339    Certain limits discussed within the RAS are relevant to the present case. The Operational Risk and Regulatory and Compliance Risk sections established zero tolerance for regulatory breaches and compliance plans that were not approved or monitored. WIB's RAS also included liquidity metrics, and the requirement that WIB monitor its liquidity gap position under short-term stress scenarios. The latter required analysis of maturity mismatches between WIB's loans and deposits as well as a detailed understanding of all potential contingent liquidity requirements. WIB's RAS also served as a cornerstone to the "Market Risk Framework" and "Liquidity Risk Framework" documents, which set out Westpac's approach to managing these risks in the context of its overall risk appetite and strategy.

340    During the trial, I was taken to the August 2010 and August 2011 WIB RASs and would make the following additional observations. In relation to the August 2010 WIB RAS:

(a)    It contained a dimension relating to liquidity risk and established liquidity metrics relating to target deposits to loan ratio, total committed and uncommitted lines, and liquidity gap positions in short-term stress scenarios.

(b)    It contained a further dimension relating to operational risk, which set out that WIB had "zero tolerance" for breaches of Westpac's Code of Conduct.

341    The August 2011 WIB RAS went into further depth in relation to operational risk, with subsections addressing, inter-alia:

(a)    operational risk capital;

(b)    operational risk related loss incidents and non-lending losses;

(c)    operational risk profiles and risk maps;

(d)    internal and external fraud;

(e)    audit and credit risk review issues; and

(f)    risk relating to the development and use of modelling.

Operation of the "compliance plan" within Group Treasury

342    Westpac's Risk Governance Framework required all staff in the business unit "originating" a risk to:

(a)    retain primary accountability and responsibility for the risks accepted and their on-going management;

(b)    establish effective risk controls appropriate to their business that met risk management framework requirements for the relevant risk-stream and also satisfied relevant regulatory and legislative obligations; and

(c)    identify existing or emerging gaps in, or concerns with, risk management policies or controls.

343    As part of the risk management requirements at the divisional level, Westpac established the WIB Compliance Committee, with jurisdiction over all WIB operations and also Group Treasury. It was comprised primarily of officers from WIB's Executive and its senior management.

344    In line with Westpac's requirement that each business division maintain a compliance plan describing existing obligations and controls, WIB had its own "WIB Compliance Manual", and sub-divisions within WIB had their own compliance plans. The WIB Compliance Committee reviewed and endorsed these compliance plans, as well as monitoring new business approvals, dealing with breaches of compliance policies, reporting to OPCO, adequately resourcing compliance, overseeing training programs, and reviewing compliance effectiveness.

345    The WIB Compliance Manual from 2009 established non-negotiable policies on legal and regulatory compliance. Although WIB Compliance (a "section" of relevant individuals) was the owner of this manual, it worked with WIB's business units to ensure that these policies functioned across the various environments and units. Further, the Manual established clear requirements within WIB to determine the facts of any breach, report it to the WIB Compliance Committee, and advise on any appropriate action.

346    In relation to the individual compliance plans of sub-divisions within WIB, business unit supervisors were selected to take responsibility for compliance in relation to specific products and to meet reporting obligations, and their performance in connection with the compliance plan was regularly assessed. Accordingly, Group Treasury had its own compliance plan, and, within Group Treasury, the various sub-divisions had a compliance plan(s). According to the "ALM and PRM Compliance Plan", specific employees were responsible for developing, maintaining, and implementing their designated sections of the plan, especially the detailed processes to ensure conformity with compliance procedures. This plan was divided into 20 subcategories of functions, each with a distinct set of corresponding obligations and responsible employees. For each of the respective functions, the ALM and PRM Compliance Plan outlined relevant goals, risks associated with these goals, procedures by which to meet these goals, and frequency of monitoring depending on associated risk.

347    The ALM and PRM Compliance Plan also described the compliance duties associated with each of these functions, as well as the procedures and monitoring tasks that were required to meet those obligations. The obligations were ranked according to risk, which was determined based on, among other things, "the seriousness and frequency of any reported breaches, the results of compliance monitoring and internal audit reviews, changes in compliance obligations, and any changes in business." High risks were assessed quarterly; medium risks were assessed bi-annually; and low risks were assessed annually. Westpac required the reporting of potential or actual compliance breaches immediately to the WIB Compliance Committee.

348    Market misconduct was one subcategory of the ALM and PRM Compliance Plan. The plan classified risks as high, medium, or low, requiring self-assessment quarterly, bi-annually, or annually, respectively. Incidents involving market misconduct were high risks and required quarterly reporting to the Compliance Officer. Monitoring for market misconduct involved multiple parties and steps. WIB Compliance checked personal dealing records and conflicts of interest databases to determine non-compliance and to check whether there were any staff trading securities in their personal accounts that PRM also traded. Moreover, there were processes in place to alert for delays in the processing of orders or suspicious incidents that might suggest market misconduct and to ensure that such concerns were reported to WIB Compliance.

349    Now I was taken to a version of the ALM and PRM Compliance Plan dated 10 May 2010 and would note the following matters:

(a)    The plan contained categories of compliance procedures relating to "Conflicts of interest / Insider trading" as well as "Market misconduct". Both of these categories were classed as "High" risk.

(b)    The plan also contained categories relating to "Reporting breaches" and "AFMA Code of Ethics/Code of Conduct". The latter referred to "[r]ecording of phone calls and instant messaging" as one of its procedures.

350    Additionally, many employees in ALM and PRM were required to participate in a continuing education program of compliance courses for a set amount of hours. Apparently Westpac also provided some training around insider trading and market manipulation.

351    Westpac's "Dealing Room Policies for Trade Risk: Market Risk Management" provided further guidance on policy and practice for staff in Financial Markets and Group Treasury. The policy applied broadly to all areas of Westpac involved in the management of traded and non-traded risk. This included relevantly PRM and ALM. All staff bound by this policy were required to communicate to either the Head of Market Risk or the Senior Compliance Manager Financial Markets, any communication, conversation or market rumour (whether official or unofficial) with an external party where concern was expressed about Westpac's dealing room activities. It stated:

Any communication, conversation, market rumour, etc official or unofficial with an external party where concern is expressed about Westpac's dealing behaviour, transactional volumes, exposures, pricing, creditworthiness, etc must be communicated to either the Head of Market Risk or the Senior Compliance Manager, Financial Markets (FM) who will then, after informing the Managing Director, Debt Markets / Managing Director FX, Commodities, Carbon & Energy / Managing Director Equities / Group Treasurer (as appropriate) and Stakeholder Communications, independently investigate and report the matter to the Trading Risk Committee (TRC) or the Banking Book Risk Committee (BBRC) (as appropriate) and communicate any concern(s) / issue(s) to the Chief Risk Officer.

FUNDING AND LIQUIDITY ACTIVITIES

352    I have set out the structure of Westpac and particularly Group Treasury, the relevant personnel, and the reporting processes that were in place. That static picture is useful so far as it goes, but it is necessary to also understand the dynamics of Westpac's funding and liquidity activities and how Group Treasury operated on a day-to-day basis during the relevant period.

353    The key sources of funding for Westpac were (and remain) deposits from customers, and the issuance of debt across the spectrum from short term (overnight to twelve months) to long term (up to 30 years), both in domestic and offshore markets.

354    For every dollar lent by Westpac, approximately 70 cents was funded from deposits and approximately 30 cents had to be raised from wholesale funding markets. As a result, Westpac was heavily reliant on wholesale funding raised from both domestic markets and offshore. The funding team in Group Treasury funded the gap between loans and customer deposits. However, Westpac's funding profile had to be consistent with its funding risk appetite.

355    During the relevant period, short term domestic funding was executed directly by the STIRR Desk, primarily in the Bank Bill Market. For funding longer than twelve months, the funding team (that is the Global Funding and Liquidity Desk) handled those processes. Short term offshore funding was handled by the IMM Desk.

(a)    External pressures on funding and liquidity (2007 to 2009)

356    Mr Zuber gave evidence regarding the following matters that influenced his strategic decisions regarding Westpac's funding. From around 2007, he was particularly focused on financial pressures affecting world markets and, in particular, funding issues. He was eager where possible to build Westpac's holding of liquid or liquefiable assets due to external factors affecting global funding markets including the collapses of Bear Stearns in March 2007 and Lehman Brothers in September 2008, the onset of the Global Financial Crisis (GFC) and the extreme volatility in financial markets globally.

357    Additionally, on 1 December 2008 Westpac acquired St George Bank, which was also heavily reliant on raising wholesale funding in both domestic and offshore markets. After Westpac acquired St George, it went from being the fourth largest bank in Australia to the second largest.

358    In response to the contraction in global funding markets and concerns in relation to their stability, a number of countries across Europe and the US announced government guarantees to support debt issuance by their banks. That step caused a structural imbalance between offshore banks with the benefit of a government guarantee and those that did not have such a guarantee. This could conceivably have reduced the attractiveness of Australian banks from an investment perspective. In response, the Australian Government announced a guarantee for Australian deposit-taking institutions in around October 2008, which remained in place until around March 2010 (the Government Guarantee).

359    In 2009, the persisting effects of the GFC on financial markets were compounded by the European sovereign debt crisis. There was widely reported market concern that the fiscal positions and debt levels of a number of European countries were unsustainable and by May 2010, Greece required Eurozone assistance.

(b)    Lessons from global financial instability and their effect on Group Treasury activities

360    In 2009, partly in response to the difficulties it along with most other financial institutions had been experiencing in obtaining appropriate funding as a result of the issues outlined above, Westpac remained heavily focused on strengthening its liquidity position and building a robust balance sheet that would assist in carrying it through any future funding crisis or difficulty. To that end, it was a core focus of Mr Zuber as Group Treasurer to build Westpac's liquidity profile so that Westpac could withstand any future funding constraints. The ability to demonstrate a strong liquid asset position was important to ensure that funding could be obtained and at a price that kept Westpac's cost of funding reasonable. Keeping its cost of funding low was particularly important to ensure that it remained competitive.

361    In the period 2007 to 2009, Mr Zuber observed the following general market trends concerning the availability of external funding for Australian trading banks from offshore investors:

(a)    an aversion to risk and a general reluctance to purchase bank bonds;

(b)    a concern about exposure to sub-prime mortgages and, in particular, bank exposure to those assets;

(c)    a view that liquidity was a key indicator of a bank's strength; and

(d)    given all the challenges in the US and Europe, the difficulty in convincing offshore investors that Australian banks were in a strong or stronger financial position.

362    By the end of 2009, it was Mr Zuber's belief that the Government Guarantee would not be extended or renewed and that Westpac would need to be in a position to satisfy its investors that it was strong and could handle global funding difficulties if they arose without the benefit of the Government Guarantee. To that end, he considered it imperative to ensure that Westpac had a range of surplus assets to liquidate if overseas or domestic markets became distressed or were no longer available so that it could continue to fund its business operations. In those circumstances, he wanted Group Treasury to invest in assets that could be liquified to fund Westpac's cash needs should a short-fall ever arise.

363    In 2007 to 2009, and then up to and including the relevant period, the liquid securities available to Group Treasury were predominantly:

(a)    securities issued by government and semi-government bodies (Government assets);

(b)    "repo-eligible" short-dated securities issued by banks, including Prime Bank Bills; or

(c)    "repo-eligible" long-dated securities issued by banks and residential mortgage backed securities (RMBS).

364    The market in Australia for short-dated Government assets at that time was not very deep, i.e. there was frequently insufficient liquidity, unlike, for example, in the US. In those circumstances, short term securities issued by other Prime Banks were the best option as such assets were liquid and "repo-eligible" with the RBA.

365    To qualify as a liquid asset, a security needed to be controlled by Group Treasury, not be part of any Financial Markets portfolio holding, and be eligible for "repo" with the RBA. Mr Zuber believed that Group Treasury needed to hold a diverse mix of liquid assets, from short term to long term, and from a range of different counterparties, in order to minimise counterparty credit risk and build diversity into the portfolio. Given its substantial business in AUD, the majority of its liquid assets were held in AUD. However, Westpac also held liquid assets in USD and NZD.

(c)    Repo-eligible securities generally

366    In order to ensure that the assets it held were the most liquid, Westpac invested in "repo-eligible securities" which could be deposited on a short or long term basis with the RBA in return for cash under a repurchase agreement or "repo". Under a repurchase agreement, Westpac could sell a repo-eligible security to the RBA, and agree to repurchase that security at a future time and at an agreed price.

367    These types of repurchase agreements:

(a)    would be used from time to time but not on a daily basis. The RBA expected Westpac to manage its funds so as to not be constantly short of cash unless it was a time of crisis; and

(b)    were distinct from the intra-day standing facility between Westpac and the RBA, through which Westpac could borrow and lend funds to the RBA at the cash rate throughout the day.

368    In the relevant period, "repo-eligible" securities included Government assets, AAA rated RMBS, Prime Bank Bills, and other longer dated securities from Prime Banks (as defined eligible by the RBA). All of these securities could be deposited with the RBA in return for cash during a crisis.

369    Repo-eligible assets permitted Westpac almost immediately to raise funds in the event that funding markets closed or became temporarily constrained. However, it was almost always the case that Westpac would have to pay more than the cash rate for those funds. For that reason, Mr Zuber did not consider that repo arrangements were a viable mechanism to manage daily fluctuations in Westpac's cash position. However, holding a large amount of repo-eligible assets did provide it with a secure buffer that it could use to fund in the event of some need or a funding crisis.

370    I would note that there were several types of repo agreements available:

(a)    Intra-day agreements where Westpac would sell an eligible security to the RBA and receive the current market value of the security in cash, and later on the same day repurchase the same security and pay the same market value (plus applicable transaction costs set by the RBA).

(b)    Daily open market operations, where participants submitted blind bids to the RBA for cash that it then lent to successful bidders (i.e. those who had submitted the highest yields). If successful, Westpac would sell a repo-eligible security to the RBA and receive the current market value of the security, and upon maturity, repurchase the same security and pay an interest rate that reflected its successful bid.

(c)    Overnight facilities for when it was unable to unwind an intra-day repo transaction with the RBA or unable to fund its ESA that day. This was an emergency facility, as any assets repurchased carried an interest rate fixed 25 points above the cash rate.

(d)    Costs of holding liquid assets

371    Throughout the relevant period, Westpac's holdings of liquid assets were considered as an insurance policy for any future funding crisis. An important part of that insurance policy was that liquid assets were held across all maturities, from the short term to long term. There was a material cost to holding liquid assets in that assets with a short term that could be liquidated quickly generally had a less advantageous rate of return to longer dated and less liquid assets. Westpac could achieve better rates of return than in Prime Bank Bills in a range of ways. For instance, investing in a three year floating rate note, issued at 120 basis points above BBSW rather than in 1 and 3 month Prime Bank Bills, which had a return that was likely to be close to where BBSW set. Alternatively, the funds used to invest in Prime Bank Bills in order to build and maintain Westpac's portfolio of liquid assets could be deployed in its lending business. If funds were lent to customers, Westpac would generate, on average, a return of at least 200 basis points above BBSW, being the net interest margin on those funds. However, given Westpac's objective to build its buffer of liquid assets (including short dated liquid assets) it was prepared to pay a price for greater security and flexibility in the event of future funding difficulties.

372    While holding liquid assets was costly, those assets could be held and then sold to provide necessary cash at short notice if required. The cost of holding liquid assets was balanced against the key benefits of Westpac being able to manage its funding position flexibly in response to changing market conditions, and it being more likely to project an image of strength to the market. If the market perceived Westpac to be strong (due to its large holdings of liquid assets), it was likely to find it easier to obtain wholesale funding, which in turn made it less likely that it would need to use its liquid assets.

373    By holding a buffer of liquid assets, if another financial crisis occurred, Westpac could look at its funding profile and identify cash that would be returning to Westpac in the next one to three months as a result of upcoming maturities of those assets or cash that Westpac could raise through liquidating its assets, either through commercial counterparties or, if necessary, through a repurchase agreement with the RBA.

374    In the relevant period, following the period from 2007 to 2009, all members of Group Treasury were consistently told by Mr Zuber that funding Westpac and building its holdings of liquid assets were the paramount objectives of trading activity. This message was conveyed in conversations that occurred almost daily with members of Group Treasury and more formal meetings (which were held at least weekly) to discuss funding and risk management. At these funding meetings, Group Treasury officers discussed matters such as upcoming issuances (both long and short term), how the business' balance sheets were performing, and highlighting the forecasted net cash needs of various divisions as loans were typically growing faster than deposits.

(e)    The Bank Bill Market as a source of funding

375    Witnesses for Westpac gave evidence that from a funding and liquidity management perspective, the Bank Bill Market had a number of characteristics that made it an attractive source of short term funding. In relation to the relevant period, Ms Johnston gave evidence of the following matters:

(a)    The domestic Bank Bill Market was generally very liquid and frequently (although not always) provided willing acquirers of Prime Bank Bills at attractive yields for short term issuance.

(b)    Westpac had a vested interest in ensuring the Bank Bill Market remained liquid both because it was a primary source of funding (often at a relatively low cost) and because it also used the market for part of its portfolio of liquid assets.

(c)    Raising some of its funds over shorter periods of time was a useful way to ensure that it was not paying excess amounts of interest for money that it did not yet need to employ in the business.

(d)    Raising short term funds in the Bank Bill Market was a useful way of funding a short cash position in the lead up to a planned long term funding deal. This would often give the Global Funding team, who were responsible for the long term funding of the Bank, flexibility around the timing of their issuance.

(e)    Raising short term funds was useful generally when looking to fund a short term funding gap, without unnecessarily resorting to long term funding which was more costly.

(f)    Given the regular large fluctuations in Westpac's balance sheet caused by cash flows unrelated to Group Treasury, such as corporate customers withdrawing or depositing cash, or an increase in mortgages and term deposits, trading in the Bank Bill Market was useful to manage these changes in the cash position. STIRR could issue NCDs relatively quickly if needed to raise cash, and if the cash position reversed for some unforeseen reason, STIRR could potentially buy Prime Bank Bills to avoid holding excess cash which was not required. In this way, the Bank Bill Market could be used to deal with daily known or unknown fluctuations in the cash position, as opposed to issuance in other markets which was often for a longer term, generally more costly and could not be reversed (such as a five year bond that could not be bought back after it had issued if the cash raised was no longer needed).

376    Ms Johnston considered that the Bank Bill Market had advantages over other potential sources of funding including:

(a)    funding from customer deposits with Westpac, which could not be controlled in a precise or timely manner (as compared to NCD issuance in the Bank Bill Market);

(b)    long term funding, such as bonds, which did not provide the same degree of short term flexibility as NCDs in the Bank Bill Market and took a longer period of time to plan and execute compared to trading each day in the Bank Bill Market; and

(c)    US short term funding (which usually had a one year term) which could not be reversed by buying back the securities issued; as such, Westpac could not move in and out of its US funding positions as it could in the Bank Bill Market.

(f)    Annual funding plans and formal funding documents

377    Westpac developed and circulated annual funding plans, which were subject to committee oversight, particularly by the ALCO, and endorsement by the Board. As Group Treasurer, Mr Zuber oversaw the preparation of Westpac's annual funding plans, which outlined its annual projected funding mix. The annual funding plan was commonly refined over time, as funding conditions onshore and offshore changed throughout the year. Westpac provided the following funding plans:

(a)    Westpac's funding plan for 2010, contained in a letter from Mr Zuber to Mr Michael Ryan of APRA dated 17 December 2009 (APRA letter);

(b)    Westpac's funding plan for 2011, contained in a memorandum to the Westpac Group Asset & Liability Committee dated 27 September 2010 (2011 Funding Plan); and

(c)    Westpac's funding plan for 2012, contained in a memorandum to the Westpac Group Asset & Liability Committee dated 22 September 2011.

378    These funding documents recorded Mr Zuber's concerns regarding the uncertain global economic conditions, and the priorities for Group Treasury in maintaining access to wholesale funding and building Westpac's liquid asset holdings in the relevant period.

379    For example, page 1 of the APRA letter stated:

Executive Summary

One of the most important learnings from the recent Global Financial Crisis is the importance of having a strong stable source of funding. This has necessitated a change in strategy where we have shifted our focus from asset growth as the key driver to now:

3. Increasing the quantum of our repo eligible liquid asset portfolio.

380    Page 3 stated:

Adverse credit and capital market conditions

If, because of adverse conditions, Westpac is unable to source appropriate funding, our first line of defence would be to liquidate a portion of our liquid asset portfolio and reduce our customer lending.

381    Page 4 stated:

A systemic shock in global financial systems

In the current uncertain global economic environment, there is a risk that another major systemic shock could occur that causes a further adverse impact on the Australian, New Zealand or global financial systems. Such an event could have a material adverse effect on financial institutions such as Westpac, including the undermining of confidence in the financial systems, reducing liquidity and impairing access to funding. The nature and consequences of any such event are difficult to predict, however, from our experience of the GFC, we expect our diverse sources of funding to provide access to wholesale funds, albeit at a much higher cost. In addition, our repo eligible liquidity buffer which has increased from $15bn to $80bn as insurance should access to wholesale markets become impaired.

382    Page 4 also referred to the impact of regulatory changes on holdings of liquid assets:

Further regulatory changes occur that necessitate changes to our funding and liquid asset positions APS 210

The proposed revisions to APS210 [an APRA prudential standard] are currently suggesting that we will need to significantly increase our holdings of liquid assets.

383    Page 2 of the 2011 Funding Plan stated:

The availability of offshore markets is one of the biggest risks we face and represents a critical supply constraint …

We rely on our liquid asset holdings to manage this risk. We hold sufficient liquid assets to absorb at least 6 months of offshore debt maturities.

384    The annual funding plans addressed all assets on Westpac's balance sheet. Prime Bank Bills were included in the category of domestic assets that were of less than twelve months duration. In addition to the annual funding plans, Mr Zuber also oversaw the preparation of:

(a)    monthly balance sheet reports that were sent to ALCO and discussed at meetings of that committee; and

(b)    the Weekly Maturity Forecast prepared by the Global Funding team in Group Treasury and circulated to the trading desks in Group Treasury.

385    During the relevant period Mr Zuber had conversations with members of Group Treasury responsible for executing Westpac's long and short term funding, as outlined in the various plans, reports, and forecasts. He also spoke with the traders on the STIRR Desk, who traded in the Bank Bill Market to fulfil Westpac's short term funding needs. Its short term funding needs did not need to be forecast as strictly as longer term deals as traders on the STIRR Desk could be flexible about when they raised funds by selling or issuing Prime Bank Bills.

(g)    The Cash Position Report

386    In addition to high level plans like the said funding documents, Westpac monitored its cash funding and liquidity on a day-to-day and intra-day basis through a variety of reports.

387    Each morning a member of the Cash Desk prepared the Cash Position Report and circulated it by email to Group Treasury at approximately 9.00 am. This Cash Position Report was a snapshot, as at 9.00 am, of Westpac's expected cash position at the end of the day on the basis that there would be no other unexpected movements in cash. As such, the Cash Position Report gave no indication of how the cash position might change by the end of the day. It was an estimate of the position.

388    The Cash Position Report was split into AUD and USD and included the following information:

(a)    "Starting position" summary, showing whether the net position for cash in each of AUD and USD was long or short. This was seen as the starting estimate for that day's cash position, on the basis that flows would occur during the day and change this position;

(b)    a cash forecast table made up of AUD and USD cash forecasts from the date of the report to approximately five weeks in the future, showing the cash forecast positions each weekday during that period;

(c)    the previous day's cash position forecasts for both AUD and USD; and

(d)    an upcoming maturities table which showed upcoming AUD-denominated maturities in various products for the following fortnight.

389    The email containing the Cash Position Report often included a summary of new and significant flows which had been identified or had occurred in the period since the previous report. From around September 2010, the Cash Position Report also included a small table recording the current and previous liquids positions which identified Westpac's total liquid asset holdings.

390    The Cash Position Report contained a number of assumptions and there were some items not accounted for in it. Relevant assumptions are set out below where I compare this report with the Weekly Maturity Forecast.

(h)    Weekly Maturity Forecast

391    The Weekly Maturity Forecast was circulated by email ahead of the weekly funding meeting and was a summary of estimated and actual cash flows from Group Treasury and the wider balance sheet. It identified forecast asset and deposit growth, the upcoming maturities of Westpac's different funding programs, as well as proposed new funding over the coming month. It showed (in AUD):

(a)    balance sheet cash flows, actual and forecast, which related to growth (positive or negative) in assets (mortgages/business) and deposits (retail/business); for example, an increase in asset growth from mortgages would cause a cash outflow;

(b)    cash outflows, which related to long and short term funding maturities;

(c)    cash inflows, which related to long and short term issuance both domestically and offshore;

(d)    the net cash flow (being a shortfall or excess), the cumulative shortfall or excess and a cumulative cash position on a monthly basis; and

(e)    a summary of the outstanding issuances for various short term funding programmes (such as NCDs), and a high level breakdown of the liquid assets position.

392    The Weekly Maturity Forecast was also based on a number of assumptions.

393    The key differences between the Weekly Maturity Forecast and the Cash Position Report were the following:

(a)    The Cash Position Report was a short term view of the daily cash balance, while the Weekly Maturity Forecast had a longer term focus and formed the basis of many discussions around longer term funding strategies.

(b)    The Cash Position Report was a very conservative daily estimate that:

(i)    assumed only NCDs issued by Westpac to a non-interbank customer would roll over on maturity;

(ii)    assumed none of Westpac's short term USD issuances would roll over on maturity, causing the USD position to become shorter over time; and

(iii)    did not include potential long term issuances.

(c)    The Weekly Maturity Forecast encompassed a broader set of cash flows than the Cash Position Report by:

(i)    identifying and assuming that all (or a large portion of) outstanding short term issuances in AUD and USD would remain on issue, which resulted in more cash being available than the Cash Position Report assumed;

(ii)    identifying upcoming long term offshore issuances; and

(iii)    more generally, forecasting wider balance sheet flows.

(i)    Liquidity risk management

394    During the relevant period Westpac operated under a "Liquidity Risk Management Framework", which was updated annually. This modelled Westpac's ability to fund under both normal conditions and during a crisis situation. At page 2 of that Framework dated February 2010, liquidity risk was defined as follows:

Liquidity risk is the risk that the bank will be unable to fund increases in assets and meet obligations as they come due, without incurring unacceptable losses. Two generally accepted types of liquidity risk are (1) funding liquidity risk the risk of not being able to meet efficiently both expected and unexpected current and future cash flow and collateral needs without affecting either daily operations or the financial condition of the bank and (2) market liquidity risk the risk that the bank cannot easily offset or eliminate a position at the market price because of inadequate market depth or market disruption.

395    At pages 3 and 4, Group Treasury's role in managing these risks was described as follows:

[Page 3] Group Treasury is responsible for the aggregate management of Westpac's balance sheet incorporating funding, liquidity, and capital management.

[Page 4] Group-wide liquidity management is the responsibility of the Group Treasurer under the oversight of the BRMC, ALCO and the Banking Book Risk Committee, with daily oversight from Market Risk Management … Group Treasury is responsible for monitoring the funding base of the Group and ensuring that this base is prudently maintained, adequately diversified and remains at all times within Group limits.

396    At page 6, liquid asset holdings and how Westpac used them to manage liquidity risk were described as follows:

Westpac holds a portfolio of liquid assets as a buffer against unforeseen funding requirements. These assets are held either in cash, government, semi-government or highly rated investment grade paper which are eligible for repurchase with Central Banks (repurchase eligibility as defined by the RBA). The large majority of these assets are held domestically in Australia and New Zealand. Determination of holding levels takes account of the liquidity requirements of our balance sheet under "normal" and "stress" conditions.

397    Group Treasury was expected to manage the liquidity risks outlined above by investing in liquid assets. Traders on the STIRR Desk in particular were responsible for Westpac's investments in short dated liquid assets, while traders on other desks invested in longer dated liquid assets.

(j)    Interest rate risk

398    It is convenient to deal with this topic comprehensively in a separate section of my reasons.

(k)    Profit

399    In carrying out its core objectives, Group Treasury also aimed to be a profitable part of Westpac. As noted above, the performance of the division and of individual officers was a relevant factor in their remuneration. Group Treasury would attempt to manage Westpac's liquidity and interest rate risk in a way that optimised Westpac's financial position. Traders on each desk were expected to conduct their activities to achieve Group Treasury's core objectives of funding and liquidity management, and risk management, in the most profitable manner possible but within Westpac's limits for risk and other parameters.

400    One of the principal ways in which Group Treasury made money was to position Westpac's balance sheet to profit from, or at least reduce any adverse effect of, potential future movements in interest rates. Traders would have views on matters that might affect interest rates and would hedge the interest rate risk in their individual portfolios with derivative products in a way that would seek to maximise the benefit of any potential movements in the cash rate. Some of the most profitable outcomes occurred when traders in Group Treasury took a view against the prevailing market view as to potential interest rate movements. That often occurred in periods of market volatility, when various financial markets were uncertain as to the way in which interest rates would move.

MANAGEMENT OF NON-INTEREST RATE RISKS

401    Group Treasury had in place risk management policies in relation to the following relevant matters: market risk, liquidity risk, credit risk, and operational risk. It also had other risks such as interest rate risk, which it managed and which I will discuss separately.

(a)    Market risk

402    The WIB RAS defined market risk as the "risk to earnings from changes in market factors such as foreign exchange rates, interest rates, commodity prices and equity prices". As part of managing market risks, the 2010 WIB RAS specified a Trading Book "Value at Risk" (VaR) limit of $65 million for all of the WIB, including Group Treasury. Additionally, there were two limits for the Banking Book managed by ALM: a VaR limit of $20 million and a Net Interest Income at Risk (NaR) limit of $160 million pre-tax. VaR is a widely used risk management tool for financial firms, and at the time it was accepted as a risk metric by regulators. During the relevant period Westpac also ran stress tests on a daily basis. The internal models used to quantify the VaR of market risk and any subsequent revisions were approved by APRA.

403    By April 2012, Westpac had developed market risk policies and specific risk limits that identified additional components of market risk and detailed "Trading versus Banking Book" position limits as well as distinct colour limits, with the "Red" limit being a "do not cross" limit and the "Amber" limit one that required attention if crossed. The 2012 WIB RAS contained procedures related to stress testing, profit and loss, and structured credit. For instance, the profit and loss provisions set for ALM and PRM detailed reporting procedures to senior management and the Board if single day profit/loss or cumulative 20-day profit or loss passed certain thresholds.

(b)    Liquidity risk

404    Management of liquidity and its associated risks was relevant for an institution like Westpac that relied heavily on wholesale funding markets over the relevant period. The WIB RAS set forth the key provisions governing liquidity risk management and Westpac's procedures and controls, including specific limits, to manage liquidity risk. The 2010 RAS stated that WIB's liquidity and funding risks was managed by Group Treasury and overseen by the ALCO. The WIB RAS detailed multiple liquidity risk metrics and noted that additional measures or processes could be instituted based on that testing and related liquidity risk factors, such as maturity mismatches. The liquidity risk section of the RAS was updated over time.

405    More broadly, Westpac had "Liquidity Risk Management Framework" documents that set out the methodology used to measure and control liquidity risk. It specified that Westpac had to hold a portfolio of liquid assets as a buffer against unexpected funding obligations. This buffer included cash, government, semi-government or highly rated investment grade paper that could be used for repo financing with central banks. During the relevant period Westpac was required to hold liquid assets worth at least $20 billion by APRA. The BRMC set a buffer over this limit, and established a global "Red" limit of $25 billion in liquid assets in 2010 and 2011 and $40 billion in 2012. This had to be met at all times. Further, in 2012 ALCO set additional targets for asset holdings, with a goal of holding at a minimum $80 billion in liquid assets, of which at least $15 billion should be "high-quality".

(c)    Credit risk

406    Westpac's internal policies considered credit risk a matter to be assessed and managed, being the risk of loss where a customer or counterparty failed to meet their obligations. These policies were outlined in Westpac documents.

407    Westpac actively monitored its credit portfolio to manage risk concentrations and assigned monthly risk limits per counterparty as part of its Issuer Risk Limits documents. In the context of Prime Bank Bills, there were limits to the exposure that Westpac could have to any one issuer. If such limits were reached, no additional Prime Bank Bills issued by that bank could be acquired by Westpac.

(d)    Operational risk

408    During the relevant period and as I have indicated, Westpac maintained OPCO that was comprised of both business and risk representatives. OPCO was part of the second line of defence and reported directly to the CEO and to the BRMC.

409    As an executive risk committee, OPCO oversaw and advised on the development of divisional RASs, and on risk management frameworks, policies and controls. OPCO reviewed the "Operational Risk Management Framework" and its effectiveness, the "Reputation Risk Management Framework", the operational risk management policies, profiles and thresholds, and the results from stress testing. Additional responsibilities included approving changes to operational risk policies, analysing emerging operational and compliance risks, as well as regulatory issues, and trends, and overseeing multiple subcommittees which handled risk, for instance divisional OPCOs. These monitored the effectiveness of management's implementation of the operational risk management framework within their division and escalated outstanding matters to Group OPCO.

410    OPCO reviewed various key policies and reports, such as the "Operational Risk & Compliance Report", "Assurance Report", and the "Reputation Risk Report". OPCO also reviewed audit results, the Basel Committee Operational Risk documents, the annual letter of assurance to APRA and processes to ensure compliance with APRA prudential standards. Further, it reviewed Westpac's risk management framework to confirm that it aligned with the "Three Lines of Defence" model that I have previously referred to. It also updated the "Operational Risk Assurance Model" from time to time.

411    I was taken to a report by OPCO, the "Operational Risk Report" for the quarter ending 30 June 2012. It recorded in relation to the June 2012 fine against Barclays Bank of £290 million for manipulating LIBOR and the Euro Inter-bank Offered Rate (EURIBOR) that "The LIBOR event is currently being reviewed for any similarities and potential exposures in Australia in relation to the Bank Bill Swap Rate (BBSW)". Of course, LIBOR and EURIBOR were set differently, a matter to which I will return later.

412    WIB's RAS listed the following five policies to ensure an organised and speedy identification, reporting, and resolution of operational risks:

(a)    All incidents were to be reported within five business days.

(b)    All risks had to have a "key indicator" and a corresponding tolerance level.

(c)    Key indicators for each business line were to be identified, monitored, and reported.

(d)    Key indicators that were out of tolerance were to be mitigated or managed after one month, escalated after two months, and fully reviewed after three months.

(e)    Risks were to be monitored quarterly through the "WIB Operational Risk & Compliance Report".

413    WIB's RAS during the relevant period stated that Westpac had a "very low appetite" for operational losses and "zero tolerance" for individuals who committed any intentional breaches of the Code of Conduct. Similarly, Westpac enforced a zero tolerance policy for any intentional regulatory breaches.

414    Moreover, Westpac's RAS confirmed the expectation that staff at all levels had to receive access to ongoing training and development opportunities to ensure that they had the necessary qualifications. Westpac also had a Code of Conduct during the relevant period that applied to all employees and listed the key principles of conduct expected. The Code of Conduct required those who were aware of any breaches of law or internal policies throughout the course of their employment to notify a manager or team leader immediately.

415    Further, Westpac committed to abide by AFMA's Code of Ethics, according to Westpac's "Dealing Room Policies for Risk in Financial Markets and Treasury". The AFMA Code of Ethics required Westpac's employees to, inter-alia, abstain from conducting any type of trading that could affect the normal supply and demand for a financial product, had the potential to "create artificial markets or prices," or did not reflect "genuine trading or commercial intention." The AFMA Code of Ethics also required AFMA members to have policies and procedures regarding potential conflicts of interest with clients. Westpac's policies to address conflicts of interest included four main steps: "identifying the conflict of interest, assessing and evaluating the conflict, implementing an appropriate response, and keeping documentation and records." Any conflicts of interest deemed to be of more than the usual potential importance or significance were to be escalated to the Business Unit Conflicts Committee or the Group Conflicts & Ethics Committee. More broadly, the "Conflicts of Interest Policy" stated that customers were to be "treated fairly, honestly and efficiently," in line with Westpac's policies, procedures, and the law. I will return later in these reasons to discuss the question of conflict of interest in more detail in the context of the s 912A question.

416    At this point, let me also note some additional matters about the "Dealing Room Policies for Risk in Financial Markets and Treasury":

(a)    In addition to the regulatory requirement to abide by the AFMA Code of Ethics, dealers were also required to be aware of and understand relevant Westpac and WIB compliance documents and policies;

(b)    Dealers were to ensure that customers understood the products being offered to them by Westpac and whether the product was appropriate and suitable for their needs; and

(c)    Dealers were to ensure that customers were provided with sufficient information on a transaction to allow them to make an "informed and independent" decision about the deal.

417    Again, I will return later in these reasons to discuss Westpac's policies and procedures when dealing with ASIC's s 912A case.

MANAGEMENT OF INTEREST RATE RISKS

418    During the relevant period, there were two types of interest rate risk that Group Treasury was responsible for managing:

(a)    outright or general interest rate risk; and

(b)    rate set risk or rate set exposure.

(a)    Outright or general interest rate risk

419    Outright interest rate risk refers to the exposure to movements in interest rates that change the value of Group Treasury's total assets and liabilities on a mark-to-market basis compared to previous valuations. This type of interest rate risk only arises when the interest rate of a debt instrument is fixed for a period of time.

420    Where an interest rate is fixed for a period of time, a change in the relevant market interest rate during that period will affect the mark-to-market value of the debt instrument. This has a profit and loss impact on the relevant trading book which is reflected in the Treasury Eagle report (Treasury Eagle). The sensitivity of those valuations to movements in interest rates constitutes outright or general interest rate risk.

421    By way of example, if Westpac makes a loan at 5% and interest rates increase, that loan is less valuable because the money lent at 5% could now potentially be lent at a higher return. This would result in a mark-to-market loss for Westpac. Similarly, if interest rates decrease, the loan is more valuable, resulting in a mark-to-market gain for Westpac.

422    The same principle may be applied across all transactions which have an exposure to interest rates, so that it is possible to calculate Westpac's overall sensitivity to changes in interest rates. Group Treasury undertook this task on a daily basis in the Treasury Eagle, which was produced around 5.00 pm each day and presented Group Treasury's interest rate risk position.

423    The Treasury Eagle grouped the interest rate risk by desk and according to different tenors or "maturity buckets" (from one day to 40 years). The tenors from 1 day to 3 months were referred to as "cash buckets". The interest rate risk was prepared so as to show the notional gain or loss in dollar value should interest rates increase by one basis point (BpV or PV01, which is the same as "DV01" referred to by Professor Stulz).

424    The Treasury Eagle showed the sensitivity of Westpac's transfer priced assets, liabilities, and other interest rate sensitive instruments, across all tenors to a one basis point increase in interest rates. The interest rate risk derived from Group Treasury's BBSW Rate Set Exposure was generally a component that, once crystallised, fed into the entire interest rate risk profile managed by Group Treasury. It was generated and circulated by a member of Group Treasury at around 5.00 pm each day. The Treasury Eagle captured the risk position from each of the individual portfolios in Group Treasury.

425    On the Treasury Eagle, a positive number in black represented that Westpac was "short", which meant the position was a net payer of interest and its profit would increase if rates increased. And a negative number in red represented that Westpac was "long", which meant the position was a net receiver of interest and its profit would decrease if rates increased. The report showed the notional gain or loss in dollar value per basis point if interest rates increased.

426    The basis point sensitivities were grouped into maturity buckets from one day out to 40 years, with the majority of Group Treasury's interest rate risk sitting in the period up to two years or less. The short term risk position was divided into tenor buckets starting from one day, to one month, two months, three months up to the first future (being the next expiring BAB Futures contract). Those tenors, up to three months, were often described as "cash buckets" being a shorthand way to refer to short dated risk.

427    Ms Johnston gave evidence about her use of and opinions about the Treasury Eagle. When she was assigned to STIRR in the relevant period, she observed that the STIRR book's risk position (as reflected in the Treasury Eagle) was generally always short. That is, the net position of its assets, liabilities and derivative transactions resulted in STIRR being a net payer of interest.

428    The short risk position in STIRR would make money per basis point if interest rates increased above the current level because STIRR would have previously locked in to pay interest at a lower level. This would be reflected in STIRR's mark-to-market revaluation on its net assets, liabilities and derivative transactions. If Mr Roden and Ms Johnston expected the RBA to increase the cash rate, the strategy of STIRR would often be to allow the book to "run with the position" because it might be profitable for the book. That is, it may not have been necessary to actively enter into derivative transactions or take other positions to reduce additional short interest rate risk. If Mr Roden or Ms Johnston thought that the cash rate was going to decrease then STIRR would be more active in seeking to minimise the short interest rate risk.

429    As I have said, outright interest rate risk was Group Treasury's exposure to movement in interest rates that changed the value of Group Treasury's total assets and liabilities compared to previous valuations. It arose due to the mismatch between its obligations to pay interest at differing rates and tenors on its customer deposits and wholesale funding (being liabilities) and its rights to receive interest at differing rates and tenors on its loans and other receivables (being assets).

430    Interest rate risk was centralised within Group Treasury. By centralising interest rate risk management in this way, Mr Zuber could work to manage Westpac's overall balance sheet risk position to ensure that Group Treasury was within risk appetite limits, whilst also minimising Westpac's funding costs.

431    Typically, Group Treasury's net position, as a result of the centralisation of Westpac's interest rate risk, was a short ("paid") position. This was largely due to the interest rate sensitivity generated by the wholesale funding portfolio.

432    Derivative transactions were a mechanism by which Group Treasury sought to manage interest rate risk. The primary derivative products traded by Group Treasury for that purpose were:

(a)    interest rate swaps;

(b)    BAB Futures;

(c)    forward rate agreements;

(d)    cross-currency swaps;

(e)    asset swaps;

(f)    interest rate options;

(g)    swaptions; and

(h)    inflation swaps.

433    STIRR was required to manage the interest rate risk in its portfolio within Westpac's overall appetite for interest rate risk, as set out in policies and frameworks that I have already noted, including:

(a)    the Dealing Room Policies for Traded Risk Market Risk Management;

(b)    the Dealing Room Policies for Risk in Financial Markets and Treasury;

(c)    the Market Risk Management Framework, Liquidity Risk Management Framework and Market and Liquidity Risk Management Framework; and

(d)    the risk limits as set by Market Risk.

434    In practice, the requirement to comply with Westpac's overall appetite for interest rate risk meant that STIRR would often have to take active steps to reduce and manage the short term interest rate risk that naturally arose in its book, to keep within the mandated parameters.

435    Ms Johnston said that she considered that what the RBA was going to do regarding the cash rate in the next six to twelve months was an important factor in STIRR's management of interest rate risk. Due to the large short term interest rate risk positions in STIRR, the portfolio could generate profit from taking a view on, and positioning the book to take advantage of, potential cash rate movements by the RBA. One of Ms Johnston's roles on the STIRR Desk was assisting Mr Roden to manage its short term interest rate risk. The two discussed expectations of future movements to the RBA cash rate, developments in other markets and financial and economic data that might inform the RBA's views. They would discuss different trading ideas and options to manage STIRR's risk. Mr Roden set the overriding strategy, and Ms Johnston assisted in its execution.

436    Westpac's AUD short term interest rate risk was centralised in STIRR's book. That included short term AUD interest rate risk arising from activities in Westpac's balance sheet, other portfolios in Group Treasury and STIRR's own activities.

437    The other portfolios that primarily contributed to STIRR's interest rate risk position were:

(a)    ALM the derivative transactions entered into between ALM and STIRR resulted in STIRR being a net payer of interest;

(b)    Liquids it bought and held long term liquid assets using funds lent to that book from STIRR at BBSW. Those loans resulted in STIRR being a net receiver of interest;

(c)    IMM the money raised in offshore markets converted into AUD using FX forwards and cross-currency swaps and the AUD proceeds were lent to STIRR, generally at BBSW. This resulted in STIRR being a net payer of interest; and

(d)    the Global Funding team, which raised long term funding that would be swapped back to a shorter term AUD liability (such as 3 month BBSW) using interest rate swaps. Internal Group Treasury trades were booked between the portfolios which resulted in STIRR being a net payer of interest.

438    Despite STIRR's rights to receive interest arising from its internal lending activities (mainly to the Liquids book), it almost always had a much greater exposure to pay interest, as a result of the significant amount of funding that was executed through IMM and the Global Funding team. In addition, other activities of the STIRR Desk gave rise to short term interest rate risk, such as trading in the Bank Bill Market, issuing NCDs, and entering into derivatives such as forward rate agreements, overnight indexed swaps and other types of swaps.

439    STIRR generally had a significant outright interest rate risk position. That interest rate risk position would be added to each day when the net position of STIRR's assets, liabilities and derivative transactions reset by reference to BBSW.

440    On any given day when STIRR had a BBSW Rate Set Exposure in a particular tenor, after BBSW had set in that tenor the rate STIRR would pay or receive on its exposure was fixed for that period. STIRR would then be exposed to the risk that interest rates in the future would move adversely to the rate at which it had already paid or received interest, which had been set by reference to BBSW that day. Essentially, STIRR obtained outright interest rate risk whenever it was locked in to pay or receive interest at a fixed rate, as the market valuation and re-valuation for that interest rate position could at any point move which altered the relative value of the rights and obligations.

(b)    Rate set risk or rate set exposure

441    The second type of interest rate risk managed by STIRR was rate set risk or rate set exposure. This is the risk arising from payment obligations that reset by reference to BBSW on each trading day. This type of interest rate risk also only arose in respect of debt instruments with interest rates which were fixed for a period of time.

442    Transactions entered into by Westpac gave rise to rights to receive or obligations to pay interest. Those rights and obligations were regularly re-priced or reset by reference to a benchmark. In the case of STIRR, the most common benchmark for repricing was BBSW.

443    On any given day, STIRR had either a net short or "paid" position, meaning that its obligations to pay interest resetting on that day exceeded its rights to receive interest resetting on that day, or a net long or "received" position, meaning that its rights to receive interest resetting on that day exceeded its obligations to pay interest re-setting on that day. This was referred to as STIRR's "rate set exposure", "rate set position" or "rate reset position".

444    Let me elaborate further.

445    On each trading day, Westpac on its own account was a party to a number of instruments which had floating leg payment obligations that reset by reference to the BBSW in various tenors. Westpac's net exposure to either pay or receive interest on a given day was referred to as the "rate set exposure".

446    Group Treasury's BBSW Rate Set Exposure varied on a daily basis. Sometimes the result of the sum of its rights and obligations to pay and receive interest was that it would have a net exposure to pay interest (referred to as a "short" exposure or a "paid set"). But on other days Westpac would have a net right to receive interest (referred to as a "long" exposure or a "received set") for a particular tenor. Group Treasury's BBSW Rate Set Exposure reset at or around 10.15 am each day when BBSW in each tenor was published by AFMA.

447    The rate set exposure was divided into one month, two month, three month, four month, five month, and six month buckets, and was managed in the STIRR book. The STIRR desk only managed Group Treasury's BBSW Rate Set Exposure and not Westpac's overall rate set position, which included the BBSW Rate Set Exposure of Westpac's Financial Markets division. Financial Markets had its own separate BBSW Rate Set Exposure that arose out of the trading between Financial Markets and its customers in derivative products.

448    As a result of the structural separation between Financial Markets and Group Treasury, Group Treasury did not manage the BBSW Rate Set Exposure arising out of Financial Market's customer transactions. Further, from 1 December 2008, Group Treasury's BBSW Rate Set Exposure included St George Treasury's BBSW Rate Set Exposure, but it did not include the BBSW Rate Set Exposure of St George's Financial Markets division. As noted above, until August 2010, St George's data was combined with Group Treasury's data manually.

449    STIRR officers had access to the systems through which Group Treasury's BBSW Rate Set Exposure could be obtained.

450    I should also note at this point that the parties before me agreed that exposures under the following instruments were included in Westpac's calculation of its rate set exposure if the product referenced BBSW:

(a)    interest rate swaps;

(b)    forward rate agreements;

(c)    cross-currency swaps where one leg of the swap used AUD;

(d)    interest rate options;

(e)    inflation swaps;

(f)    floating rate notes;

(g)    commercial loans;

(h)    deposit-taking facilities in respect of which Westpac had agreed to pay interest by reference to BBSW; and

(i)    Bank Bills held as part of Westpac's corporate bill lending activities.

(c)    Relationship between general interest rate risk and rate set risk

451    Outright or general interest rate risk and rate set risk are separate risks, although those risks are interrelated. The risks are interrelated in that, after a payment obligation is reset, the interest rate exposure arising from that payment obligation then forms part of the outright or general interest rate risk.

452    There is a difference between outright interest rate risk and rate set risk. In the course of giving evidence, Mr Roden described the difference as follows:

So in relation to rate set risk, once the set is concluded, we do have that risk. But that's a much easier risk to handle. Once it goes into our outright risk, it, effectively, goes into a washing machine, effectively, and for want of a better word, where we have multitudes of risks all come in together … Rate set risk is very different … Rate set risk is the is the point where a whole range of assets and liabilities have a reprice effectively, they have repricing there's a repricing event. That's what the rate set risk is.

453    According to Westpac, although STIRR would manage these types of interest rate risk in different ways, trading in Prime Bank Bills was particularly effective because it was a way in which STIRR could manage both outright interest rate risk and rate set risk. I will come back to this issue as it is quite contentious.

(d)    Trading in Prime Bank Bills to manage short term interest risk

454    If STIRR or Group Treasury had a "short" outright interest rate position (that is, was a net payer of interest), acquiring Prime Bank Bills had the effect of offsetting and reducing the total face value of the liabilities and other instruments pursuant to which Westpac paid interest. This is because a Prime Bank Bill is an asset pursuant to which Westpac receives interest. Similarly, an overall "long" position could be offset or reduced by selling or issuing Prime Bank Bills. In this way, trading in Prime Bank Bills could reduce Group Treasury's general or outright interest rate risk. ASIC did not challenge this position.

455    ASIC did not challenge any Westpac witnesses on the basis that trading in Prime Bank Bills was not an appropriate, effective or common place means of reducing outright interest rate risk. Rather, ASIC challenged Mr Zuber and Mr Roden on that basis that trading in Prime Bank Bills on any given day was only able to reduce Group Treasury's outright interest rate risk by a small amount. But I agree with Westpac that notwithstanding that a reduction on any given day may only be modest, there was nevertheless a reduction in the relevant exposure and the fact that it may have been a relatively modest impact ignores the magnitude of the outright interest rate risk which could only ever be managed in smaller increments.

456    Westpac also contended that trading in Prime Bank Bills was also a way in which STIRR could manage Group Treasury's rate set risk. Now STIRR was only responsible for managing Group Treasury's BBSW Rate Set Exposure. As I have said, STIRR did not manage Westpac's overall BBSW Rate Set Exposure risk, which included the rate set exposure arising out of trading between Westpac's Financial Markets division and its customers.

457    Westpac gave an example of how BBSW Rate Set Exposure could be managed in this way. If Group Treasury had a short BBSW Rate Set Exposure of $1 billion in the 1 month tenor, after BBSW set that day Group Treasury would be required to pay interest on that $1 billion at the 1 month BBSW rate from that day. If STIRR acquired $500 million worth of 1 month Prime Bank Bills during the BBSW Rate Set Window, those were assets that would receive interest at the yield traded that day (which would be at BBSW or close to it). As a result, Westpac would in effect pay interest at 1 month BBSW on only $500 million rather than $1 billion, thereby reducing its BBSW Rate Set Exposure. A similar reduction in a long rate set exposure would be achieved by selling or issuing Prime Bank Bills during the BBSW Rate Set Window.

458    Westpac contended that trading in Prime Bank Bills was an economically efficient way to hedge or reduce rate set risk for a number of reasons, including the following:

(a)    First, there is no additional cost for entering into the hedge over and above the cost of buying or selling the Prime Bank Bills.

(b)    Second, trading in Prime Bank Bills would reduce the BBSW Rate Set Exposure by the amount of the Prime Bank Bills bought or sold.

(c)    Third, trading in Prime Bank Bills could reduce or offset Group Treasury's rate set risk in circumstances where STIRR would otherwise be participating in the Prime Bank Bill market for funding or liquidity reasons.

(d)    Fourth, there were very frequently a range of other significant benefits and purposes of buying or selling Prime Bank Bills on a given day.

459    Westpac contended that the STIRR traders were also conscious of the potential (but unpredictable) effect that trading in Prime Bank Bills could have on the level at which BBSW set on a particular day. It is said that for this reason, all other things being equal, STIRR would seek to acquire or sell Prime Bank Bills on days when, if trading did have any effect on the rate at which BBSW set, that trading would not adversely affect Group Treasury's BBSW Rate Set Exposure position.

460    Now ASIC contended that buying or selling Prime Bank Bills at around 10.00 am on a particular business day does not hedge, offset or reduce BBSW rate set risk in the sense of the amount of net assets or liabilities that are resetting on that day. It says that no matter how many bills are bought or sold at around 10.00 am, Group Treasury will still have the same amount of exposure. Buying or selling Prime Bank Bills does not change that in any way. For example, if Group Treasury has a $1 billion short 1 month BBSW Rate Set Exposure on Tuesday, it will have to pay the 1 month BBSW on $1 billion. It says that buying any number of Prime Bank Bills at 10.00 am on that Tuesday only affects the obligation to pay 1 month BBSW on $1 billion by affecting where the 1 month BBSW rate sets. And that is what ASIC contends that Group Treasury was doing during the relevant period. It says that trading in that manner is the antithesis of hedging BBSW rate set risk. ASIC says that such trading proceeds on the footing that the risk is and will remain unhedged and that Westpac will profit from that unhedged risk by influencing BBSW to set favourably to the risk.

461    ASIC contended that BBSW rate set risk can be hedged, offset or reduced only by entering into instruments that directly reference the BBSW. For example, Group Treasury could enter into an interest rate swap or a FRA in an opposite direction to their BBSW rate set risk.

462    Further, ASIC contended that the witnesses who suggested that buying or selling Prime Bank Bills at around 10.00 am on a particular business day hedges BBSW rate set risk explained their use of the term "hedging" in a manner which ASIC says elided BBSW rate set risk with outright interest rate risk. ASIC contended that it was said by various witnesses that buying or selling Prime Bank Bills at around 10.00 am hedges BBSW rate set risk because, (a) going forward or (b) over the life of the Prime Bank Bills or (c) the net position is that, Group Treasury will receive interest (if buying Prime Bank Bills) or pay interest (if selling Prime Bank Bills) opposite to whatever its interest rate position is that is being hedged. But ASIC says that so to explain the concept in that fashion is to explain the hedging effect by reference to outright interest rate risk.

463    ASIC also contended that because buying or selling Prime Bank Bills was an imperfect hedge of outright interest rate risk and not a hedge of BBSW rate set risk at all, it is unsurprising that Financial Markets did not have the ability to buy or sell Prime Bank Bills. Only STIRR could do so. Every other desk and division had to manage its risk profile through derivative products. Such products could be used to manage outright interest rate risk and BBSW rate set risk. Now Mr Masnick accepted that Financial Markets used "proxy instruments" to hedge or reduce BBSW Rate Set Exposures, including BAB Futures, interest rate swaps and FRAs. Mr Sharma said that he traded derivatives, including interest rate swaps, Eurodollar futures, BAB Futures, cross-currency swaps, FX forwards, FX swaps, OIS and other futures to manage outright interest rate risk and BBSW rate set risk.

464    Further, ASIC contended that BAB Futures were particularly useful for hedging BBSW rate set risk and outright interest rate risk, particularly towards expiry and regardless of the tenor of the interest rate risk being managed. ASIC said that BAB Futures had certain advantageous features. It pointed to the following features:

(a)    BAB Futures were traded in far greater volumes than Prime Bank Bills were traded on the Bank Bill Market.

(b)    There was generally a lot of liquidity in the BAB Futures market during the relevant period.

(c)    There was liquidity in BAB Futures trading throughout the day.

(d)    BAB Futures were not required to be fully funded up front.

465    Moreover, Financial Markets used BAB Futures to hedge its BBSW rate set risk exposure during the relevant period.

466    ASIC also contended that FRAs were also useful hedging tools. FRAs could be used to hedge or reduce BBSW Rate Set Exposure without any basis risk because they reference the BBSW. FRAs were also leveraged, which meant that they were a lower cost means of hedging exposure from a funding point of view than trading in Prime Bank Bills directly. And they were traded throughout the day.

467    ASIC also says that OIS could also be used as hedging instruments. The market for OIS was generally quite liquid and enabled trades in large volumes. They were traded throughout the day.

468    ASIC also says that it is noteworthy that the word "hedging" appears irregularly in the communications which are in evidence. Mr Roden suggested, on at least one occasion, that hedging should be taken to be implied. But ASIC says that Mr Roden referred to hedging when he meant it. In his draft internal memorandum dated 29 August 2013 (after the relevant period) titled "Group Treasury Income Attribution and Analysis", Mr Roden makes no mention of Prime Bank Bills. But in my view Mr Roden was specifically discussing purpose built hedging instruments such as OIS, swaps and futures, rather than addressing all the ways that by conduct (i.e. trading in Bank Bills) one could produce a hedging effect.

469    Now I also accept that there are imponderables concerning Westpac's hedging thesis concerning rate set exposure. For example, if the purpose was to hedge, then it might be said that that should produce indifference to where the rate set. Further, it might be said that one should not be referring to the rate itself as Ms Johnston's 16 May 2011 email does. Further, the thesis does not seem to be expressly referred to in terms in any written or verbal communication. I have weighed such matters with the other evidence.

470    At this point it is also appropriate to refer to various s 19 transcripts that were tendered by ASIC.

471    ASIC says that notwithstanding the "ritual reference" to hedging by Westpac's witnesses at trial, both Ms Johnston and Mr Stokes said in their s 19 examinations that Prime Bank Bills were not or not regularly used to hedge BBSW Rate Set Exposures.

472    In Ms Johnston's s 19 examination, the following exchange took place:

Q.    And was the practice of the STIRR desk to use bank bills to hedge their rate set risk?

A.    Bank bill futures? Privilege.

Q.    No bank bills?

A.    No, not really.

473    In evidence before me, Ms Johnston sought to qualify what she then said, on the basis that "this was a very long section 19. I think I was quite confused with all the discussion around bank and different terminology".

474    In Mr Stokes' s 19 examination, the following exchange took place:

Q.    Okay. Would you ever use, when you were on the bills desk, would you ever trade bank bills on the day of your rate set risk to try to reduce that risk as well?

A.    Privilege. You would always trade you would always trade bank bills on the day of rate sets because you have a rate set every day and you trade bank bills just every day, so not really. Most of our bank bill trading was done for the two purposes of issuing NCDs to create cash or buying NCDs for liquid assets, yes.

Q.    Okay. So it wasn't really one of the reasons why, if there was no funding requirement, you wouldn't be trading bank bills to hedge a rate set?

A.    No, privilege, it's very ineffective because if you had a $5bn rate set, you can't trade $5bn worth of physical securities to hedge that rate set. It's just the rate set that you have is way too big to be nullified by any bill trading.

475    Mr Stokes sought to walk back from what he then said during cross-examination, on the basis that "I was giving an honest answer to what I thought the question was", and that trading Bank Bills was "very ineffective" in the sense that it could not be a total hedge.

476    Now I admitted these s 19 examination transcripts as prior inconsistent statements, but ASIC has also submitted that they should be admitted against Westpac as admissions pursuant to ss 81 and 87 of the Evidence Act. I disagree and have refused to do so.

477    Section 81 provides:

(1)    The hearsay rule and the opinion rule do not apply to evidence of an admission.

(2)    The hearsay rule and the opinion rule do not apply to evidence of a previous representation:

(a)    that was made in relation to an admission at the time the admission was made, or shortly before or after that time; and

(b)    to which it is reasonably necessary to refer in order to understand the admission.

478    Section 87(1) provides:

For the purpose of determining whether a previous representation made by a person is also taken to be an admission by a party, the court is to admit the representation if it is reasonably open to find that:

(a)    when the representation was made, the person had authority to make statements on behalf of the party in relation to the matter with respect to which the representation was made; or

(b)    when the representation was made, the person was an employee of the party, or had authority otherwise to act for the party, and the representation related to a matter within the scope of the person's employment or authority; or

(c)    the representation was made by the person in furtherance of a common purpose (whether lawful or not) that the person had with the party or one or more persons including the party.

479    Section 87(1) is engaged when a finding about one of the matters in paragraphs (a) to (c) is "reasonably open". The paragraphs "do not require a concluded finding of the relevant authority or common purpose. They merely require a conclusion that such a finding be one that 'is reasonably open'" (Ingot Capital Investments Pty Ltd v Macquarie Equity Capital Markets Ltd [No 4] [2006] NSWSC 90 at [11] per McDougall J). The relevant time upon which attention must focus is the time when the previous representation was made.

480    For the purpose of s 87(1)(a), it is apparent that the authority to be considered is the authority "to make statements … in relation to the matter with respect to which the representation was made". As McDougall J explained in Ingot v Macquarie [No 4] (at [20]):

the legislation is looking at the general authority of the person whose previous representation is sought to be tendered to make statements of the kind embodied in the particular representation, and not at the authority to make the particular representation (including in the circumstances in which it was made and having regard to the means by which it was made).

481    The "authority" referred to in s 87(1)(a) includes express or implied authority.

482    But s 87(1)(b) is a true alternative to s 87(1)(a). For s 87(1)(b) to apply, it is "sufficient … to show that the person was an employee of the party and the representation related to a matter within the scope of the person's employment" (Wright v Optus Administration & Anor (No 5) [2013] NSWSC 1717 at [14] per Campbell J). Indeed, as Campbell J then observed (at [18]):

It seems to me the scope of employment limb is much broader. Once one is satisfied that the person making the representation is an employee of the party then the second aspect is only that the representation related to a matter within the scope of the person's employment. It is not necessary to prove that the employee had actual or ostensible authority to make representations on behalf of the party.

483    ASIC says that s 19 examination transcripts have been admitted into evidence as admissions against defendant companies in previous cases, and it is well established that they may be so admitted (see generally Australian Competition and Consumer Commission v Leahy Petroleum Pty Ltd (2004) 141 FCR 183 at [97] and [98] per Merkel J (albeit in the context of a s 155 examination); Australian Securities and Investments Commission v Axis International Management Pty Ltd (No 5) (2011) 81 ACSR 631; [2011] FCA 60 at [60] per Gilmour J). For example in Australian Securities and Investments Commission v Fortescue Metals Group Ltd (No 5) (2009) 264 ALR 201; [2009] FCA 1586 at [122], Gilmour J admitted s 19 examination transcripts of a director and a company executive under s 87. In Australian Securities and Investments Commission v Astra Resources plc (2015) 107 ACSR 232; [2015] FCA 759 at [126], White J admitted s 19 examination transcripts of two directors and an "executive manager" under s 87(1)(b). In Australian Securities and Investments Commission v Sino Australia Oil and Gas Ltd (in liq) (2016) 115 ACSR 437; [2016] FCA 934 at [28], Davies J admitted a s 19 examination transcript of the managing director under s 87.

484    ASIC submits that Ms Johnston's statement in her s 19 examination transcript is an admission because it is adverse to Westpac's interests, it having set up its defence on the basis that Prime Bank Bills were being bought or sold to, among other things, hedge BBSW rate set risk. It says that the statement satisfies both s 87(1)(a) and (b) and points to the following matters. Ms Johnston's s 19 examination took place on 13 October 2015. At the time, she was a Westpac employee (albeit she had recently returned from a career break from June 2014 to August 2015), working either in IMM or the London office overseeing Group Treasury's London and European operations and executing offshore wholesale issuance for Westpac. She was an Associate Director. ASIC says that she had actual or implied authority on behalf of Westpac to make statements in relation to the subject of what she did while working on the STIRR Desk to hedge BBSW rate set risk and that s 87(1)(a) applies. Alternatively, ASIC says that at the time of the examination, Ms Johnston was an employee of Westpac and the statement related to how she carried out her duties while working on the STIRR Desk. It says that that is a subject related to a matter within the scope of her employment and s 87(1)(b) applies. Ms Johnston traded on the STIRR Desk and was one of the central traders in this proceeding. Moreover, Ms Johnston has worked extensively in Group Treasury, and STIRR was ultimately responsible for managing the short end interest rate risk for the whole of the division. She therefore had and has occasion to know how STIRR managed that risk.

485    ASIC says that Mr Stokes' s 19 examination took place in 2015. At the time, he was a Director of Group Treasury. He had been an employee of Group Treasury since 2002. ASIC says that he had authority on behalf of Westpac to make statements about how BBSW rate set risk was managed while he worked on STIRR and indeed afterwards. After leaving STIRR, Mr Stokes worked on the ALM desk. ASIC points out that STIRR was responsible for managing the short end interest rate risk of the ALM desk and the rest of Group Treasury, and so it may readily be inferred that the management of that risk (including BBSW rate set risk) is a subject relevant to Mr Stokes' role. It says that s 87(1)(a) applies and for essentially the same reasons so too does s 87(1)(b).

486    I would reject the statements made by Ms Johnston and Mr Stokes as admissions against Westpac. Let me explain my reasons.

487    As Westpac correctly points out, s 87(1)(a) applies where, at the time the representation was made, the person had authority to make statements on behalf of the party in relation to the relevant matter. Section 87(1)(b) applies where, at the time the representation was made, the person was an employee and the representation related to a matter within the scope of that person's employment or authority. So the relevant time at which to consider the authority is the time at which the representation was made.

488    Now at the time Ms Johnston participated in her s 19 examination and which is not disputed by ASIC, Ms Johnston was employed as an Associate Director either in IMM or in Westpac's London office, had recently completed a one year career break and importantly for present purposes had not worked on the STIRR Desk for over two years.

489    ASIC have not identified any material other than the fact of Ms Johnston's employment and that she had previously traded on the STIRR Desk, from which I can reasonably infer that Ms Johnston had express authority to make statements on behalf of Westpac in respect of the relevant matters at the time she was examined. Further, I agree with Westpac that it is not possible to infer from the nature of Ms Johnston's employment at the later time of making the relevant statement, being an Associate Director working on a different desk, that she had implied authority in respect of the relevant matters. In my view, ASIC cannot rely on s 87(1)(a) of the Evidence Act in respect of Ms Johnston's statements, even at the lower threshold of "reasonably open to find".

490    Further, at the time of her examination, the matters the subject of the statements did not form part of Ms Johnston's then current role. I agree with Westpac that such matters could not be said to fall within the scope of her employment or authority. Accordingly, ASIC cannot rely on s 87(1)(b) of the Evidence Act in respect of Ms Johnston's statements, even at the lower threshold of "reasonably open to find".

491    As for Mr Stokes' s 19 examination statements, at the time they were made he was employed in ALM. Mr Stokes had not worked on the STIRR Desk for more than ten years. I agree with Westpac that there is no basis to find that Mr Stokes had authority to make statements on behalf of Westpac in respect of the relevant matters at the time those statements were made or that the matters the subject of the statements fell within the scope of his employment or authority.

492    Further, I do not consider that the cases cited by ASIC by way of example are particularly helpful. They are relevantly distinguishable for the following reasons. In Ingot v Macquarie [No 4], the relevant admissions were made by a partner of Phillips Fox. McDougall J accepted that the partner was the "unlimited agent" of Phillips Fox, and held that the statements made by the partner were admissible against Phillips Fox. In ASIC v Fortescue Metals Group Ltd (No 5), the relevant admissions were made by a director and a senior member of management. In ASIC v Astra Resources plc, the relevant admissions were made by directors and an "executive manager". And in ASIC v Sino Australia Oil and Gas Ltd (in liq), the relevant admissions were made by the executive chairman and director. None of those cases are of assistance. They concern directors and members of senior management who clearly had actual or apparent authority to make the relevant statements on behalf of the relevant company. Those positions are distinguishable from the position of Ms Johnston and Mr Stokes. Perhaps closer to the mark is Wright v Optus Administration & Anor (No 5), which concerned an employee responsible for training making a contemporaneous statement about the training and supervision of trainees. But again this is not comparable to the position of Ms Johnston and Mr Stokes who had not been responsible for trading on the STIRR Desk for over two years and ten years respectively at the time of their statements.

493    But even if I were to accept these statements by way of admission, they are not of great assistance, and in any event have to be considered with the totality of the evidence.

494    To the extent there is any inconsistency between Ms Johnston's evidence and the statements made in her s 19 examination, I now have more recent statements of Ms Johnston's views in relation to trading in Prime Bank Bills. Ms Johnston has been cross-examined in respect of those matters. Further, her evidence was clear that one of STIRR's reasons for trading in Prime Bank Bills was to hedge or reduce rate set risk. Although the discrepancy with her s 19 examination hardly instilled me with confidence, on balance I am inclined to accept her evidence before me, which is consistent with other Westpac evidence.

495    Further, to the extent there is any inconsistency between Mr Stokes' evidence and the statement in his s 19 examination, again Mr Stokes' more recent statement is of greater probative value. Further, in circumstances in which Mr Stokes' trading on the STIRR desks was historic only, and occurred well before the relevant period, I agree with Westpac that the statements made by Mr Stokes in his s 19 examination are of less assistance in resolving the relevant hedging issue although they do raise cause for doubt.

496    Let me return to the principal question at this point. I would reject ASIC's contention that the buying or selling of Prime Bank Bills could not operate as some form of hedge (albeit imperfect) against BBSW Rate Set Exposure.

497    Westpac's witnesses were challenged on the basis that trading in Prime Bank Bills was not a method to hedge rate set risk at all. The reasons put to Westpac's witnesses variously were that:

(a)    Prime Bank Bills bought or sold on a particular day would not reference where BBSW set on that day, so would not hedge a BBSW Rate Set Exposure;

(b)    there was insufficient liquidity in the Bank Bill Market to hedge STIRR's entire rate set exposure;

(c)    there was no "leverage" when trading in Prime Bank Bills because STIRR was required to fund the purchase of any Prime Bank Bills bought; and

(d)    trading in Prime Bank Bills to manage rate set risk exposure could only be an effective hedge if the instrument was held for the entire period of the exposure, which did not occur from time to time.

498    Those witnesses were also challenged on the basis that trading in derivative products, in particular FRAs or BAB Futures, was a more effective way to hedge rate set exposure.

499    But each relevant Westpac witness gave evidence to the effect that trading in Prime Bank Bills for purposes which included hedging would likely occur at a level very similar to where BBSW set on that day. Notwithstanding the possibility of some divergence, Westpac's witnesses explained that trading in Prime Bank Bills to hedge a rate set risk exposure was "as close to perfect as we can get", "by far and away, the most perfect hedge that you can acquire", "it's a very perfect hedge, because… you're getting the exact opposite exposure", "a very, very close hedge… effectively, you're getting an asset and a liability at the exact same time and effectively neutralising that exposure and risk", "the closest instrument that would replicate a hedge for the rate set risk" and "the best hedge you could have done".

500    And even if Prime Bank Bills were acquired to hedge a short rate set exposure, and those Prime Bank Bills were acquired at a rate other than BBSW, there would still be an offsetting effect. This is because the short rate set exposure is offset by the opposite exposure generated by the Prime Bank Bills. In the course of responding to a hypothetical example where Prime Bank Bills were acquired at a rate other than BBSW, Mr Zuber explained the position as follows:

Well, it's a perfect hedge from a risk perspective. I have eliminated my risk. I may not have done it as profitably or as it's cost me. Hedging sometimes isn't free. In this instance, in your case, which, let's be clear, is quite unusual, it wasn't a clear hedge … I have eliminated my risk. My risk is smaller because my paid rate set and what I have bought … had net … A perfect hedge doesn't have to be at the same rate, a perfect hedge has reduced my risk.

501    Some Westpac witnesses were also challenged on the basis that Prime Bank Bills could not be acquired for genuine hedging purposes if the Prime Bank Bills acquired were sold shortly after the acquisition. But I agree with Westpac that this ignores the obvious fact that trading in Prime Bank Bills is dynamic. Circumstances change day-to-day.

502    Ms Johnston gave evidence regarding the circumstances in which Prime Bank Bills acquired for the purpose of hedging could be sold in the following period:

[I]f you buy bank bills at 5.04, around BBSW, and then if the rate that has offset your exposure potentially, if the rate then moves back down to five per cent, it's hedged you. You have made four basis points on that and you can sell them out and lock in that hedge if you would like.

503    Mr Park also gave the following evidence:

So with a large rate set exposure, buying the bills obviously reduces that amount of risk that he has got setting at that particular point in time. So it reduces that particular risk. Then you have got 24 hours that goes by after that, and then you have got the opportunity to change a hedge, so things might have happened like, you know, you might have interest rates move significantly, you might have a change in the path of interest rates in terms of what you expect interest rates to do. And so what you can then do is effectively unwind that hedge by selling those bank bills in the market, in the following days, and then you can kind of readjust your position going forward … [I]t offsets during the rate set, and then the following day, the world moves on, you have got 24 hours and then you can work out how you want to adjust your hedges going forward after that.

504    I found such explanations, in context, to be commercially plausible.

505    ASIC challenged the evidence of Westpac's witnesses on the basis that there was only limited liquidity in the Bank Bill Market which meant that on many days it was not possible to offset the entirety of Westpac's rate set risk by trading in Prime Bank Bills. But I agree with Westpac that this contention is based on a misconception that trading in Prime Bank Bills to reduce rate set risk is only worthwhile if it is possible to reduce the exposure entirely. Commercially, reducing part only of a rate set exposure is still a worthwhile strategy. Now Westpac's witnesses accepted that it would not be possible, for example, to fully hedge a $15 billion rate set exposure at reasonable prices by trading in Prime Bank Bills. But none accepted ASIC's case thesis that seeking to reduce part only of that exposure would not be worthwhile.

506    Further, it was put to Westpac's witnesses that a genuine hedging strategy would involve a trader continuing to acquire Prime Bank Bills until the liquidity in the market had been exhausted. But this ignores the other reasons why STIRR would trade in Prime Bank Bills and the costs involved in trading. Further, Mr Roden said:

Well, there's a limit to how much I physically may wish to buy for a range of reasons. I could have credit limit issues. I could have cash issues that all I'm prepared to take at this point in time. The nature of this market on some days is extremely fast and the volumes that you buy or sell can be uncertain.

507    In summary, I accept Westpac's position that trading Prime Bank Bills could be for a purpose of hedging BBSW Rate Set Exposure. But whether it was a purpose or a dominant purpose for trading on a particular occasion is a separate question. I reject ASIC's case that Westpac and its employees have concocted and confected such a hedging strategy or purpose. I will return to this topic again in the next main section.

(e)    BAB Futures

508    BAB Futures were traded contracts by which parties could take a position on where three month interest rates would be at the date of the BAB Futures expiry. During most of the relevant period, they traded through the ASX24 clearing house; for a time early in the relevant period this was the SFE as I have already explained. There is an issue between the parties as to what was the traded price of BAB Futures.

509    BAB Futures were the most liquid short dated product for participants hedging their risk or speculators taking a view on potential movements in the RBA cash rate. If participants in the BAB Futures market had the view that the RBA would move rates in the next three months, it was expected that the BAB Futures contracts expiring in the next three or six months would be actively traded by participants speculating on interest rate movements. The fact that BAB Futures were actively traded and liquid made them a useful product to hedge short term interest rate risk in STIRR, with the liquidity allowing large volumes to be traded if required.

510    BAB Futures represented where three month interest rates would be on expiry, with contracts expiring on a quarterly basis at midday on the Thursday before the second Friday in each of March, June, September and December. It was possible to trade BAB Futures expiring in any of those months for up to two years in advance. For example, if the upcoming quarter expiry was September 2017, that was called the "first future" and the "eighth future" would be BAB Futures expiring in June 2019. Once the September 2017 expiry has passed, the December 2017 contract would be considered the first BAB Future. Traders would often use the first and second BAB Future to hedge interest rate risk in the STIRR book, but it was also possible to use later BAB Futures to hedge more long dated risk.

511    I have already explained that an important feature of the BAB Futures contract was that any BAB Futures held on expiry were deliverable, that is, physically settled, which meant that a party that was "short" BAB Futures would be obliged to sell eligible 3 month Prime Bank Bills to the contractual counterparties at the closing contract price. Those counterparties would be "long" BAB Futures and obliged to pay the closing contract price for those 3 month Prime Bank Bills that are delivered. Delivery took place on the day following the expiry date. One BAB Futures contract was equivalent to a $1 million parcel of 3 month Prime Bank Bills. In practice, the vast majority of participants in the BAB Futures market did not hold BAB Futures to expiry but would "close out" their positions (that is, reduce the position to zero) by entering into an equivalent BAB Futures contract but taking the opposite side. That is, agreeing to sell rather than buy or vice versa.

512    Ms Johnston said that these BAB Futures trades had the effect of hedging movement in interest rates as follows:

(a)    a decrease in short term interest rates would result in a reduction in the STIRR book's mark-to-market value of its existing short term liabilities;

(b)    however, if STIRR was long BAB Futures, a decrease in short term interest rates would likely result in BAB Futures increasing in price and would result in an increase to the STIRR book's mark-to-market value of its BAB Futures position; and

(c)    in this way, the two exposures to movements in interest rates offset one another.

513    However, BAB Futures were not considered by Ms Johnston to be a "perfect" hedge for movements in current market interest rates because the implied yield of a BAB Future was a view on where three month interest rates would be out of a future date, being the expiry date, rather than where three month interest rates were on the trading date. This introduced basis risk, that is, the risk that three month interest rates on the trading date would not move in the same way as three month interest rates out of a future date. This was particularly the case when the first BAB Future expiry was two to three months away.

514    Whilst BAB Futures in theory reflected where three month BBSW would be on the day of expiry, the contract closing price did not set by reference to BBSW, and in practice the BAB Futures closing price often differed to BBSW at maturity. As BAB Futures were deliverable, a trader like Ms Johnston observed that in the last two weeks of trading before expiry, the price of BAB Futures was partially driven by market participants trying to close out of their open positions which meant that the BAB Futures price could sometimes be volatile and not reflective of the market's view of where 3 month BBSW would be at the expiry date.

515    If STIRR maintained a long position in the first BAB Future at the expiry date, STIRR would have to take delivery of the corresponding amount in Prime Bank Bills the day after the expiry date. Alternatively, if STIRR maintained a short position in the first BAB Future at the expiry date, STIRR would have to deliver the corresponding amount in Prime Bank Bills the day after the expiry date. Ms Johnston said that STIRR's usual practice was to try to reduce the BAB Futures position to zero (or close to zero) in the period leading up to the expiry date. By reducing any long position in the first BAB Future, STIRR would be unwinding its hedge of the short position in the STIRR book. That is, STIRR would have greater exposure to the risk of interest rates decreasing as a BAB Futures expiry date approached.

516    Ms Johnston said that on some occasions, STIRR would not completely close out its position and would take delivery or deliver Prime Bank Bills for one or more of the following reasons:

(a)    there would be a benefit in receiving Prime Bank Bills if STIRR wanted to build its holdings of those assets;

(b)    there would be a benefit in delivering Prime Bank Bills if STIRR wanted to raise money to meet its funding needs through the BAB Futures market or if STIRR wanted to get rid of particular Prime Bank Bills that it was holding;

(c)    there was an arbitrage opportunity between 3 month BBSW and the implied yield in the closing price of BAB Futures, such that the two rates were not the same and it would be possible to profit from the mismatch and either take delivery of, or deliver, Prime Bank Bills at a cheaper price than available in the Bank Bill Market; or

(d)    it was not possible to completely close out the BAB Futures position before expiry due to a lack of liquidity in the futures market.

517    Any decision to deliver or take delivery was generally discussed more broadly within Group Treasury and potentially at the weekly funding meetings, particularly in relation to the reasons set out at subparagraphs (a) and (b) above.

518    STIRR traded in BAB Futures as a way of managing outright interest rate risk. But the Westpac witnesses generally rejected the proposition that BAB Futures were an appropriate or effective way to manage rate set risk exposure. I must say that I found their evidence to be commercially compelling on this aspect.

519    The reason why BAB Futures were not an effective way to hedge a rate set risk (particularly in comparison to trading in Prime Bank Bills) was that the price for Prime Bank Bills was the market rate on the day the trade was undertaken, whereas the price for BAB Futures generally reflected the market participants' views as to what interest rates would be in the future. Further, even as the BAB Futures closeout approached, the price for BAB Futures would not necessarily align with BBSW. During the course of giving evidence, Ms Johnston explained the positon approaching the BAB Futures close as follows:

My understanding is that in the last couple of weeks, the price of bank bill futures is primarily driven by supply and demand because it's a deliverable contract. There might be people who have bought or sold those contracts, either hedge positions, take positions, and as they potentially looked to unwind those positions in anticipation of expiry, sometimes it's those supply and demand forces that can actually drive where the BAB futures trade, so it doesn't necessarily align completely with BBSW or move necessarily in line, and on the last day on expiry it doesn't set a BBSW… and sometimes there was dispersion which could present arbitrage opportunities.

520    During the course of giving evidence, Ms Johnston explained why she did not consider BAB Futures to be an effective way to hedge rate set risk:

I didn't think the bank bill futures were a very good hedge. If you have just gone past an expiry, the next BAB future is potentially three months away. It's a very inaccurate hedge to be hedging a movement in BBSW today using an instrument that is estimating where BBSW is in three months' time, and that kind of is the case probably for two of the months, and then… when you get close to expiry, my understanding is that it's driven more by supply and demand of people offloading their positions and unwinding, heading into expiry. So no, I didn't think it was a very good hedge.

521    Mr Sharma gave similar evidence:

So the issue with using BAB futures as a as a way to hedge interest rate risk in general is that there's a significant time variance between the interest rate period on the BAB future, versus the interest rate risk on the rate set, so you expose yourself to significant curve risk in relation to interest rate risk in general.

522    In response to a hypothetical example of acquiring BAB Futures on 9 June, to hedge a rate set risk exposure on 10 June, where the futures close-out is on 11 June, Mr Sharma explained the position as follows:

[A]round close-out there can be significant volatility in the bank bill futures market… So when you enter into a bank bill futures contract, all you're doing is moving your point in time for when you enter into interest rate risk exposure forward or backwards. So on in your example… All you have done is actually entered into more risk on the 9th and you're still exposed to whatever the volatility is on the 11th, and so around futures close-out… there can be anomalies as well.

523    Mr Masnick's evidence that he used BAB Futures as a "proxy instrument" to hedge rate set risk in Financial Markets is of no assistance in assessing the comparative advantages of trading in Prime Bank Bills to hedge BBSW Rate Set Exposure risk in circumstances in which Mr Masnick was not able to trade in Prime Bank Bills. Similarly, Mr Stokes' evidence that he traded derivative products to manage both outright and rate set risk on the ALM desk is of no value in circumstances where he was not asked in respect of which of those two risks he traded BAB Futures and, in any event, the ALM desk was not able to trade in Prime Bank Bills.

(f)    OIS

524    OISs were another product that STIRR would often use to hedge interest rate risk. OIS was a type of interest rate swap in which a fixed rate was exchanged for a floating rate. The floating rate reset daily based on the cash rate (rather than BBSW) and was compounded over a specified term (usually the term of the trade). For example, three month OIS was the market's view of where the cash rate would set over the next three months, compounded over three months. In this way, the rate at which three month OIS traded reflects the market's view of potential movements to the cash rate over the next three months.

525    ASIC notes that the market for OIS was generally quite liquid, enabled trade in large volumes, and that they were traded throughout the day.

526    STIRR would usually use OIS in the one to three month tenors to hedge the interest rate risk in STIRR's cash buckets (being the risk that the cash rate decreased and adversely affected STIRR's risk in tenors from one day to three months, as reflected in the Treasury Eagle). If Westpac received OIS, that meant Westpac received the fixed rate and paid the floating rate, which created a long interest rate risk position that would hedge STIRR's short position in its cash buckets. The hedge operated as follows:

(a)    a decrease in the cash rate would result in a reduction to the STIRR book's marked-to-market value of its existing cash bucket liabilities;

(b)    however, if Westpac received OIS, that decrease in the cash rate would result in an increase in the mark-to-market value of the OIS trade; and

(c)    in this way, the two exposures to movements in interest rates above offset one another.

527    The market for OIS was generally very liquid, and Ms Johnston observed that STIRR could often trade in very large volumes which made it useful when hedging was required. However, using OIS brought about basis risk, that is, the risk that STIRR's underlying risk (primarily based on BBSW) moved differently to OIS (based on the cash rate).

528    From time to time, STIRR would trade other contracts with OIS to hedge interest rate risk, such as FRA/OIS or BAB/OIS pairs. Ms Johnston understood that these types of trades to be longer term balance sheet hedges which offset the risk of potential future movements in interest rates adversely affecting Westpac's balance sheet.

(g)    FRAs

529    FRAs were agreements to exchange fixed and floating interest payments and could be used to manage interest rate risk or hedge BBSW Rate Set Exposure on a future date. A FRA was for one period and involved one payment by one party, that is the payment obligations of each party to a FRA were netted off so that only a single payment was required. Unlike BAB Futures, FRAs were OTC products that were not traded on any centralised exchange.

530    Ms Johnston said that she did not use FRAs as frequently as BAB Futures to hedge interest rate risk, as FRAs were quite illiquid and it was not a deep market so it was not possible to trade in a similar size as BAB Futures. Given that FRAs were illiquid from time to time, Westpac would often have to pay a premium to enter into one.

531    Contrastingly, ASIC has submitted that FRAs were useful hedging tools, that they could reduce rate set exposure without basis risk because they reference BBSW, and that being leveraged they were a lower cost means of hedging exposure from a funding point than trading Bank Bills directly. I must say that ASIC's submission at the general and theoretical level that it was put is true so far as it goes, but in reality I found Westpac's evidence that using FRAs to reduce rate set risk exposure had its limitations to be more compelling. Let me elaborate.

532    FRAs could manage interest rate risk if a trader held a view on the direction that interest rates would move. For example, if six weeks out from an RBA meeting they held the view that the RBA would increase the cash rate at that meeting, but FRAs were trading at a level that did not factor in any rate rise, they could enter into a three month FRA, resetting in seven weeks' time, in which:

(a)    Westpac paid the fixed rate and received the floating rate (i.e. BBSW);

(b)    the counterparty paid the floating rate and received the fixed rate; and

(c)    if the RBA then increased the cash rate, BBSW in each tenor would likely rise, and money would be made on the FRA when the counterparty paid Westpac the difference between the fixed rate and BBSW.

533    FRAs could also be used to hedge Group Treasury's BBSW Rate Set Exposure. FRAs impacted Group Treasury's rate set exposure, as the Bank would have to either pay or receive BBSW on the reset date. To hedge a rate set exposure, if on a future date STIRR had an obligation to pay based on where BBSW sets, Westpac could enter into a FRA locking in to pay a fixed rate and receive the floating rate BBSW on the same future date. This received BBSW position from the FRA offset the current paid BBSW position and reduced the rate set exposure.

534    Whilst not as common, STIRR would also use interest rate swaps to hedge interest rate risk, although interest rate swaps were generally only used to hedge longer dated risk. Ms Johnston said that she would often prefer to use a BAB Future rather than an interest rate swap, given the greater liquidity and alignment with the short term nature of the interest rate risk she was managing in STIRR.

535    Ms Johnston said that she was not aware of any electronic platform operated by Bloomberg Tradebook Australia Pty Ltd named "BETSY" or an electronic platform operated by BGC Partners (Australia) Pty Ltd named "BGC Trader". She had never traded interest rate swaps, or any other financial product, on those platforms. All interest rate swap transactions she entered into were in the OTC markets, in which trades were negotiated directly between counterparties.

536    A FRA, or forward rate agreement, is an over the counter agreement that involves the counterparties to the agreement exchanging fixed and floating interest payments on a notional amount at a time in the future.

537    A FRA can potentially reduce rate set exposure. For example, if STIRR was likely to have a short rate set exposure on some date in the future, Westpac could enter into a FRA where on that date it paid the fixed rate and received the floating rate (i.e. BBSW), and the counterparty paid the floating rate (i.e. BBSW) and received the fixed rate. Westpac's entitlement to receive BBSW on the future date would offset its short rate set exposure on that date.

538    But a limitation on using a FRA to hedge rate set risk exposure, compared to trading in Prime Bank Bills, was that the fixed leg of the FRA introduced additional outright or general interest rate risk. In the course of giving evidence, Mr Park identified this as a reason why a FRA was often not an appropriate way to hedge rate set risk:

[A] FRA is a swap, so it's a fixed for floating. So if you're trying to hedge a floating leg, you're taking on fixed interest rate risk at a forward space, so you're actually adding risk into your portfolio at that point in time… the floating leg will offset the floating leg, but you're taking on risk today by fixing that interest rate.

539    Ms Johnston also identified this shortcoming in the course of giving her evidence:

The other thing about when you do a FRA a number of weeks out, you are taking interest rate risk, and if there was an RBA meeting in there somewhere … you are taking then significant risk.

540    In addition to creating additional risk, other shortcomings with using FRAs to hedge rate set risk exposure during the relevant period were that FRAs were expensive and had limited liquidity.

541    Trading in FRAs to reduce rate set risk exposure had the following limitations:

(a)    the FRA market had limited liquidity;

(b)    there was a price to be paid for entering into a FRA;

(c)    the closer to the day of the rate set exposure, the less available liquidity in the FRA market and the more expensive the price;

(d)    the price for the FRA may be too high for it to be worth hedging exposure;

(e)    the liquidity in the FRA market may be too limited to meaningfully reduce the risk; and

(f)    as the price went up and the liquidity went down the closer to the day of the reset risk, using a FRA to hedge rate set risk was best done days in advance of the reset risk, which involved predicting the level of the reset risk which can be uncertain.

542    Further, liquidity and pricing were limitations to using FRAs to hedge rate set risk:

(a)    Mr Zuber said that "the liquidity during the relevant period was ordinary at best".

(b)    Mr Roden said that "you can [get a FRA one day in advance] but it tends to be very expensive … It might be a point or two, two basis points", and that "these markets were very illiquid in some of these periods. In fact, extremely illiquid". He also said in his second affidavit that "[o]ften FRAs at an appropriate price and volume are hard to find. Often counterparties in the market will be uncertain about the same possible market moves which can make it hard to find a counterparty willing to take on the risk at a competitive rate. As a result, the shorter the time period before the start date of the FRA, the more illiquid the market becomes and the higher the transaction costs".

(c)    Ms Johnston said that "pricing and liquidity in the FRA market was often less than ideal, and so there might have been times where we wanted to but you might not be able to get the liquidity that you want, the pricing might not seem very good", and that "there's issues with FRAs around liquidity and price". She also said in her affidavit that "FRAs are quite illiquid and it is not a deep market … Given that FRAs were illiquid from time to time, Westpac would often have to pay a premium to enter into one".

(d)    Mr Conway said that "[p]ost sort of 2007, 2008, the liquidity in that market [FRAs] sort of dried up".

543    One seeming advantage of using a FRA, rather than trading in Prime Bank Bills, to hedge a rate set exposure is that FRAs could be entered into in advance of the day of the exposure, whereas Prime Bank Bills could only be traded on the day of the exposure. But this advantage was more apparent than real. Mr Roden gave evidence that, given the volatility in Group Treasury's BBSW Rate Set Exposures, projected exposures would often change over time, which meant there were limits to how much exposure could be managed in advance of the day. Closer to the date of the relevant exposure (when the exposure could be predicted with more certainty), the FRA market was less liquid and more expensive.

544    I also agree with Westpac that the evidence of Mr Masnick that he used FRAs to hedge rate set risk in Westpac's Financial Markets division is of no assistance in determining whether FRAs were a better mechanism to hedge rate set risk exposure than trading in in Prime Bank Bills, given that Financial Markets was unable to trade in Prime Bank Bills. Similarly, Mr Rotcer Brizuela's evidence concerning NAB may likewise be disposed of; its Banking Book Risk Management (BBRM) desk used FRAs to hedge rate set risk exposure, but it is important to remember that that desk within NAB was unable to trade in Prime Bank Bills to manage rate set risk exposure.

RATIONALES FOR TRADING PRIME BANK BILLS

545    Mr Zuber, Mr Roden and Ms Johnston each gave evidence regarding what they considered to be possible commercial rationales for Westpac's trading of Prime Bank Bills. Much of this general material was not subjected to challenge by ASIC, notwithstanding their contention that Westpac personnel traded Prime Bank Bills for the purpose of moving the BBSW rate both on specific contravention dates and as a practice.

546    Without expressing any opinion on how likely any rationale might have been on any particular day or generally, what follows is a summary of the possible commercial rationales as they relate to Group Treasury's trading of Prime Bank Bills.

547    I have no reason to doubt any of this evidence in the generality with which it was expressed in terms of setting out for me the possible commercial rationales for trading Prime Bank Bills.

(a)    Purposes for buying Prime Bank Bills (non-interest rate risk)

Buying and holding Prime Bank Bills for liquidity management

548    From 2009 to 2012, Mr Zuber oversaw the building of Westpac's liquid asset holdings from around $20 billion to around $110 billion. Prime Bank Bills were part of those holdings. In the relevant period, his desire for Group Treasury to hold a reasonable amount of Prime Bank Bills meant that it usually held around $6 billion to $10 billion of Prime Bank Bills.

549    Because of the importance Mr Zuber gave to the objective of building liquid assets, he was willing (within limits) for Westpac to pay a price premium to hold short-dated liquid assets such as Prime Bank Bills, that is, rather than buying longer-dated securities which afforded a higher yield but would not have been as liquid. He said that Westpac would pay more for the flexibility of holding short-dated Prime Bank Bills which could be liquidated easily.

550    Further, whilst the STIRR Desk held a range of other short-dated securities in addition to Prime Bank Bills, Ms Johnston perceived that those other securities had certain limitations:

(a)    Government assets which, in her experience were not very liquid, and were very expensive when compared to Prime Bank Bills, often trading five to fifteen basis points below BBSW;

(b)    Bank Bills issued by non- Prime Banks (Non-Prime NCDs) which:

(i)    were not liquid at all, as Non-Prime NCDs could not be traded in the BBSW Rate Set Window or the Bank Bill Market more generally;

(ii)    were only available in relatively small amounts of $50 million to $200 million, as the issuers of those NCDs were quite small; and

(iii)    while Non-Prime NCDs were cheaper than Prime Bank Bills, that was because they were generally purchased from banks with a lower credit rating than the Prime Banks and as such their paper was priced at a margin above BBSW to reflect the increased credit risk. These securities would always be held to maturity and may have been rolled (that is, re-issued to Westpac) by the issuer at that time.

551    While Mr Zuber had a general idea for the level at which he wanted Westpac to hold Prime Bank Bills, he also desired flexibility so that Prime Bank Bills could be acquired at appropriate prices and when it best suited other aspects of Group Treasury's activities.

552    While Prime Bank Bills formed a key part of Group Treasury's liquid asset position, they also performed other functions. For example, Group Treasury maintained a buffer of short-dated securities, such as Prime Bank Bills, to fill gaps in the cash position as a result of unexpected cash outflows. It was not unusual for the Bank's cash position to change significantly in a very short period of time as a result of unexpected funds flow, such as movements of funds from large corporate customers, and in those instances, Prime Bank Bills could be sold to rebuild the Bank's cash position.

553    Mr Zuber said that he valued having and holding 1 month Prime Bank Bills in particular because they were very short-dated securities, which matured in a very short period of time and, as such, allowed a degree of flexibility that assisted in managing the liquidity and funding profile given the potential swings in the cash flow position. Three month Prime Bank Bills were attractive for similar reasons.

554    The risk of not having a liquid asset buffer is a crisis occurred was that the Bank would not be ready or able to fund itself as a going concern. Building up its short-dated liquid asset buffer outside of a crisis situation gave the Bank the ability to speak to investors from a position of balance sheet strength. Liquid asset holdings made Westpac a safe investment and therefore assisted the Bank to raise funds at a lower cost.

555    Ms Johnston also considered that Prime Bank Bills had additional strengths as a liquidity-related asset. She held the view that 1 month Prime Bank Bills in particular were often offered for sale at what she regarded to be attractive yields given their short term and low credit risk, which made 1 month Prime Bank Bills cheap to buy. In addition, 1 month Prime Bank Bills matured quickly or could be sold or repo'd with the RBA if Westpac's funding position or liquidity needs changed. Buying and holding 1 month Prime Bank Bills was a flexible way of managing short term liquidity and reducing surplus cash that would otherwise be left with the RBA.

556    In relation to this rationale, I note that ASIC submits, relying upon the evidence of Mr Zuber, that generally speaking, there was no pressing need to buy or sell Prime Bank Bills on any given day from a liquidity perspective.

557    Each of Mr Zuber and Mr Roden gave evidence that during the relevant period it was a practice of Group Treasury to hold a certain quantity of Prime Bank Bills as a liquidity buffer. Mr Zuber gave evidence of the average amount of holdings in each year of the relevant period. But I accept that in assessing this alternative explanation for trading, it is important to take into account that Mr Zuber in his first affidavit said:

[82]    While I had a general feel for the level at which I wanted Westpac to hold Prime Bank Bills, it was also necessary for there to be flexibility so that Prime Bank Bills could be acquired at appropriate prices and when it best suited other aspects of Group Treasury's activities …

[124]    While I always wanted Group Treasury to hold a buffer of Prime Bank Bills, I placed no constraints on the STIRR desk around holding any minimum or maximum amounts and did not direct the timing of when Prime Bank Bills should be bought or sold. Those decisions were left to the trader on the STIRR desk who participated in the Bank Bill Market each day, and was therefore were best placed to determine if prices and other market conditions were attractive.

[133]    Westpac could choose to hold onto Prime Bank Bills that it wished to sell or wait to issue its own paper until the price was attractive to Westpac. Equally, Westpac could wait to buy Prime Bank Bills until a desirable price was offered. Westpac would not buy or sell Prime Bank Bills at any price.

558    ASIC says that Mr Zuber acknowledged that generally speaking there was no pressing need to buy or sell Prime Bank Bills on any given day from a liquidity perspective. Accordingly, so ASIC says, this consideration was unlikely to rise above a matter that could be seen as consistent with a course of trading rather than the predominant cause of the trading.

559    I agree with ASIC, generally speaking.

Inventory management purposes

560    Because Prime Bank Bills were of relatively short duration, it was necessary to acquire new Prime Bank Bills constantly to replace them as they matured and redeploy the funds that had been returned to Westpac following their maturity. In order to ensure that Group Treasury was maintaining its inventory of Prime Bank Bills as efficiently as possible, it needed to look for opportunities to buy either at a good price or, in any event, where liquidity in the market could be found at a reasonable price.

561    There was a pattern of Westpac's inventory of 1 month and 3 month Prime Bank Bills decreasing over time (due to constant upcoming maturities) and then being built back up again by purchasing new Prime Bank Bills. There was also a general disposition towards buying and holding Prime Bank Bills in the one month tenor compared to other tenors.

562    Over the relevant period, it was usual for STIRR to hold more Prime Bank Bills with a duration of one month or less than Prime Bank Bills in any other tenor, and Westpac held a fairly steady inventory of around $6.3 billion of Prime Bank Bills in the one month or less tenors.

563    Westpac held on average per month:

(a)    $2.137 billion of 1 month Prime Bank Bills in 2010, and a further $4.158 billion of less than 1 month Prime Bank Bills; and

(b)    $1.882 billion of 1 month Prime Bank Bills in 2011, and a further $3.742 billion of less than 1 month Prime Bank Bills.

564    In addition, Westpac's daily average purchases of 1 month Prime Bank Bills in the Bank Bill Market during the relevant period were:

(a)    approximately $370 million in 2010;

(b)    approximately $344 million in 2011; and

(c)    approximately $305 million in 2012.

565    It was necessary for Westpac to be in the market almost every day to maintain the level of holdings of Prime Bank Bills in the one month tenor. For that reason, STIRR would try to acquire the asset consistently over time but would not necessarily acquire them every day. Finding liquidity in the Bank Bill Market (that is, volume) was an important part of ensuring that the buying of Prime Bank Bills did not become overly expensive.

566    During the relevant period, Mr Zuber regarded 1 month Prime Bank Bills as being reasonably good value, given their short term liquidity.

567    Ms Johnston also gave evidence regarding how the purpose of inventory management influenced her activities as a trader. She and Mr Roden would regularly discuss STIRR's inventory and how best to manage it. They would also discuss considerations around current market pricing of Prime Bank Bills. That is, what they considered to be fair value and whether Prime Bank Bills were trading at an attractive yield.

568    In order to maintain the level of inventory, she tried to consistently purchase Prime Bank Bills (primarily in the one month tenor) in the Bank Bill Market. There was no minimum or maximum requirement for holding Prime Bank Bills (I understand this to be a reference to Prime Bank Bills generally, excluding any counterparty-specific credit limit that Westpac may have had) and she said that she would not buy Prime Bank Bills at any price solely for the purposes of inventory management.

569    Ms Johnston said that she would capitalise on the opportunity to purchase Prime Bank Bills if the yields were high (and therefore the price was low) and could buy in volume at that price. However, she would not know what level other market participants were willing to sell Prime Bank Bills until trading commenced in the BBSW Rate Set Window.

Managing excess cash

570    Westpac's general approach to its Prime Bank Bill portfolio was to issue Prime Bank Bills in the six or three month tenors and to buy in the one, and to a lesser extent, the three month tenors. That allowed Westpac to obtain funds for a longer duration and deploy them more flexibly as short term investments. One month Prime Bank Bills produced a more favourable return than other short term investments and were highly liquid (both in trading and in their maturity profile). Accordingly, where Group Treasury had excess cash, the acquisition of short dated Prime Bank Bills might be a desirable mechanism to utilise those funds. Mr Zuber said that although theoretically Westpac could have placed its surplus funds in its exchange settlement account with the RBA, practically, this was not considered a feasible option for the approximately $6 billion to $10 billion dollars that was invested in the Prime Bank Bill portfolio. The RBA would have queried Mr Zuber about Westpac leaving such amounts in their accounts overnight. In addition, the RBA offered 25 basis points below the cash rate for amounts left with the RBA overnight, which was much less attractive than BBSW.

571    While managing excess cash was one rationale for purchasing Prime Bank Bills, building and maintaining Westpac's holdings of Prime Bank Bills was not necessarily constrained by its cash position on any given day. If Westpac's opening position for the day was short cash, Mr Zuber would not expect that to limit trading in Prime Bank Bills.

572    If Westpac was short on cash it was still possible to purchase Prime Bank Bills (and other assets). That is because it was possible to obtain cash:

(a)    at the cash rate in intra-day repo's with the RBA;

(b)    at the cash rate from other banks in the interbank overnight cash market;

(c)    pursuant to longer term repo arrangements with the RBA;

(d)    using FX forwards which convert Westpac's USD issuance to AUD; or

(e)    at the prevailing levels of BBSW by issuing NCDs to customers or in the Bank Bill Market.

573    In addition, cash could also be added to Westpac's cash position as a result of its general activities or movements in foreign currency, which impacted the amount of collateral posted for foreign currency derivatives.

574    Traders on the STIRR Desk were expected to be aware of Westpac's daily Cash Position Report, which I have previously described. It relevantly comprised customer deposits and upcoming maturities of securities held. All Group Treasury staff sat within close proximity to each other and as a result, communicated with each other throughout the day in relation to the changes in the cash position.

575    Ms Johnston said that purchasing Prime Bank Bills was an effective way for Westpac to invest excess cash at low risk and that holding Prime Bank Bills generated a return for any excess cash at a spread above the cash rate. At the same time, the cash invested remained reasonably accessible if funding conditions deteriorated and the Prime Bank Bills needed to be liquidated or could be held to maturity when the cash would be returned.

576    Now ASIC said that it is improbable that the cash position was a significant consideration for Group Treasury in its trading in Prime Bank Bills during the BBSW Rate Set Window.

577    In cross-examination, Mr Zuber acknowledged that "the cash was important but not necessarily overly critical". But ASIC says that in fact the precise composition of Group Treasury's cash position appears not to have constrained Group Treasury's trading at all or the witnesses' assessment of that trading. ASIC says that Mr Zuber's evidence offers a good example of this. He justified Group Treasury's buying of Prime Bank Bills as consistent with a desire to deploy excess cash on 12 of the 16 pleaded occasions. But of those 12 occasions, ASIC says that Group Treasury was short cash when combining the AUD and USD position on four occasions and Group Treasury was short AUD on three occasions.

578    Moreover, ASIC says that both Mr Zuber and Mr Roden said that being short cash would not prevent Group Treasury buying Prime Bank Bills. Mr Zuber also acknowledged that "it was not unusual for [Westpac's] cash position to change significantly in a very short period of time". And Mr Roden said that "the early morning position was a useful guide" only.

579    I must say that I tend to agree with ASIC that buying Prime Bank Bills may be consistent with the management of excess cash, but it does not appear to be the sole or dominant reason for the purchases on most occasions.

(b)    Purposes for selling Prime Bank Bills (non-interest rate risk)

Selling and issuing Prime Bank Bills to raise funds

580    The main business of Westpac was and remains to write loans to and gather deposits from corporate, business and retail customers. The net position of its loans and customer deposits results in a cash shortfall. One of the mechanisms by which Westpac raised the funds necessary to offset the net shortfall was to issue and sell Prime Bank Bills. During the relevant period, it was common for Westpac to sell or issue Prime Bank Bills in the three and six month tenors for funding purposes. Selling Prime Bank Bills would also support Westpac's banking operations in circumstances where there was a current or anticipated short term change in its cash flow.

581    From time to time, Westpac also issued or delivered 3 month Prime Bank Bills into the BAB Futures close as a means of both funding Westpac and seeking to take advantage of any price differential between the BAB Futures market and the Bank Bill Market. If the implied yield in the BAB Futures contract closed at a level that was lower than the 3 month BBSW, Westpac could take advantage of the arbitrage opportunity between the price of the BAB Futures and the price of the physical Prime Bank Bill. By delivering or issuing 3 month Prime Bank Bills into the BAB Futures close at a lower yield, Westpac could raise funds more cheaply than on the Bank Bill Market.

582    Ms Johnston provided some insight regarding fund raising on the Bank Bill Market. She received information regarding Westpac's funding position and strategy from funding meetings and Weekly Maturity Forecast, as well as the daily Cash Position Report and frequent conversations with members of Group Treasury who worked on the Cash Desk, IMM and in offshore offices. If other sources of funding were less attractive or unavailable to Westpac (such as offshore or long term markets), STIRR could sell or issue additional Prime Bank Bills.

583    She said that she would use the Bank Bill Market to sell Prime Bank Bills or issue NCDs of typically three to six months in duration to raise cash. Before selling or issuing she would consider:

(a)    the yield at which Prime Bank Bills were trading in the BBSW Rate Set Window on the day and on prior days;

(b)    the levels at which OISs were trading;

(c)    for funding in the six month tenor, the six month "strip" level, being a synthetic six month rate calculated using BAB Futures;

and would seek to issue at the level at which she could fund Westpac most cheaply, that is, to minimise its liability to pay interest on the Prime Bank Bills issued.

584    Westpac might also purchase 3 month Prime Bank Bills in the week leading up to a BAB Futures expiry in the event that an arbitrage opportunity arose between the two products. Even if there was no arbitrage opportunity, Westpac might decide to deliver Prime Bank Bills through the BAB Futures expiry as an alternative source of funding.

Selling Prime Bank Bills when they were no longer suitable

585    If STIRR acquired Prime Bank Bills for a particular purpose (e.g. delivery into the BAB Futures close), and the Prime Bank Bills were not suitable for that purpose, those Prime Bank Bills could be sold. Similarly, STIRR tended to sell early 3 and 6 month Prime Bank Bills before the end of the earlies, and late 3 and 6 month Prime Bank Bills before the end of the lates, to ensure that its inventory remained as liquid as possible.

(c)    Buying and selling Prime Bank Bills to manage interest rate risk

586    As I have previously discussed, this particular commercial rationale, its commercial effectiveness and its influence as a factor for trading in the minds of bank officers, was the focus of substantial submissions by ASIC. As I have already said, in my view the trading of Prime Bank Bills could manage both outright interest rate risk and rate set exposure risk. As set out above, one of Group Treasury's key functions was to manage interest rate risk, including the short term interest rate risk of Westpac which was centralised in the STIRR book. Trading Prime Bank Bills was one of the ways in which that short term interest rate risk could be managed. It could be used to manage outright interest risk as well as rate set exposure risk.

587    For example, buying Prime Bank Bills in circumstances where STIRR or Group Treasury was a net payer of interest (having a "short" position) had the effect of offsetting and reducing the total face value of liabilities and other interest rate sensitive instruments pursuant to which Westpac paid interest. That was because a Prime Bank Bill was an asset pursuant to which Westpac received interest, as such, any interest received on the Prime Bank Bill offset and reduced the total face value of the liabilities and other interest rate sensitive instruments pursuant to which Westpac paid interest. Such purchases reduced the "short" position and that effect was captured in the Treasury Eagle.

588    Moreover, as I have already indicated, trading Prime Bank Bills could affect Group Treasury's BBSW Rate Set Exposure, a process described by Westpac witnesses as "hedging". Buying Prime Bank Bills when Group Treasury had a short rate set exposure or selling Prime Bank Bills when Group Treasury had a long exposure, was what was described by Westpac witnesses as an almost "perfect hedge" for that exposure, which would be reduced by the amount of Prime Bank Bills bought or sold. For example, as Westpac contended, if there was a short Group Treasury rate set exposure of $1 billion in the one month tenor, after BBSW set that day Westpac would be paying 1 month BBSW on the face value of the rate set amount (i.e. $1 billion). If, during the BBSW Rate Set Window, Westpac acquired $500 million worth of 1 month Prime Bank Bills, those Prime Bank Bills would be assets receiving interest at the yield traded that day (i.e. BBSW or close to it) and, as a result, Westpac would effectively on a net basis pay 1 month BBSW on $500 million instead. The rate set exposure to pay interest would be reduced and offset by the Prime Bank Bills purchased under which Westpac had a right to receive interest. In this way, buying Prime Bank Bills reduced Westpac's exposure to paying interest by the amount of interest received on the Prime Bank Bills acquired. Similarly, as Westpac contended, if there was a long Group Treasury rate set exposure of $2 billion in the three month tenor, after BBSW set that day Westpac would be receiving 3 month BBSW on the face value of the rate set amount (i.e. $2 billion). If, during the BBSW Rate Set Window, Westpac sold $500 million worth of 3 month Prime Bank Bills, those Prime Bank Bills would be liabilities paying interest at the yield traded that day (i.e. BBSW or close to it) and, as a result Westpac would on a net basis effectively receive 3 month BBSW on $1.5 billion instead. The rate set exposure to receive interest would be reduced and offset by the Prime Bank Bills sold or issued pursuant to which Westpac was obliged to pay interest. In this way, selling Prime Bank Bills reduced Westpac's exposure to receiving interest by the amount of interest paid on the Prime Bank Bills issued or sold.

589    But of course and notably, the economic or arithmetic effectiveness of any "hedge" was affected by any discrepancy between the yield of the Prime Bank Bills that were purchased/sold and the BBSW that was actually set that day. Further, I agree with ASIC that the effectiveness of any hedging purpose was substantially less than Westpac suggests.

590    Let me also deal with another point.

591    The Bank Bill Market was not always highly liquid and Westpac may have sought to deal at a considerable volume in order to find either a willing seller or willing buyer of Prime Bank Bills. In such circumstances, trading may or may not have had a consequential effect on the yield at which Prime Bank Bills trade.

592    Accordingly, all other things being equal, Mr Zuber said that Westpac would seek to acquire or sell Prime Bank Bills on days where, if the level of the Bank's trading affected the yield at which Prime Bank Bills were trading, any effect on that yield would not also adversely affect Group Treasury's rate set exposure.

593    It was considered by Mr Zuber prudent not to take steps which could potentially put upward pressure on rates if Group Treasury had a short exposure or a paid set, as Westpac could ultimately pay more interest on its net liabilities that reset that day. As such, days where Group Treasury had a net short position were more favourable days upon which to acquire Prime Bank Bills and less attractive days on which to sell or issue Prime Bank Bills from a risk management perspective.

594    Similarly, according to Mr Zuber, it was considered prudent not to take any step which could potentially put downward pressure on rates if Group Treasury had a long exposure or a received set, as Westpac could ultimately receive less interest on its net assets that reset that day. As such, days where Group Treasury had a net long position were more favourable days upon which to sell or issue Prime Bank Bills and less attractive days on which to acquire Prime Bank Bills from a risk management perspective.

595    Ms Johnston as a trader made similar comments about her own activities. She was conscious that trading in Prime Bank Bills during the BBSW Rate Set Window could potentially affect the rate at which BBSW set. She would not know what price or volume other market participants were willing to buy or sell Prime Bank Bills and how trading and price action would unfold until trading commenced in the BBSW Rate Set Window. All other things being equal, if she had a degree of flexibility as to when she had to conduct a particular trade (such as knowing that she needed to issue Prime Bank Bills but not necessarily immediately), she might consider not trading on a day where if Westpac's trading did happen to affect the level of BBSW it might have an adverse effect on Group Treasury's BBSW Rate Set Exposure. But this was said to be just one of a number of considerations she would have regard to and it would not have interfered with buying or selling where more urgent needs arose.

596    Now ASIC has sought to leverage off such evidence to support its own case thesis.

597    It submits that it is "illusory and economically irrational" to distinguish between Westpac choosing to buy or sell on a particular day to avoid any detrimental effect on BBSW, and choosing to buy or sell on a particular day to positively seek a beneficial effect on BBSW, i.e. manipulate BBSW, the latter being its case thesis. This point was put by Mr Michael Borsky QC for ASIC. Now I acknowledge the intellectual force and economic rationality of his submission. Nevertheless I do not consider that if one traded for the former purpose, one would also necessarily be prepared to trade for the latter purpose. After all, the former is legal conduct, the latter (depending upon the satisfaction of various statutory elements or conditions) is not.

(d)    Trading in Prime Bank Bills to make a profit

598    In addition to the funding and liquidity purposes and risk management rationales outlined above, trading in Prime Bank Bills was a way in which Westpac could optimise its financial position. The traders on the STIRR desk had views of Prime Bank Bill yields and whether those yields represented fair value. If a STIRR trader saw an opportunity in the Bank Bill Market due to mispricing, it was expected that they either buy or sell Prime Bank Bills based on their view of fair value of Prime Bank Bill yields as long as Westpac's funding or liquidity position was not jeopardised. Mr Zuber said that it was not expected that other market participants would act to correct any anomalies in pricing as such participants had significant resources to take advantage of market anomalies and sophisticated views of interest rate movements.

599    That said, Westpac's Prime Bank Bill holdings and activities in the Bank Bill Market were small compared to Westpac's other activities. On average, during the relevant period, Westpac raised funds by issuing amounts approximately five times as much as its Prime Bank Bill holdings. The Prime Bank Bill portfolio was part of Group Treasury's broader approach to funding and liquidity, which was not focused purely on Prime Bank Bills.

(e)    Trading in Prime Bank Bills for the purpose of price discovery

600    When Westpac sought to buy or sell Prime Bank Bills, the trader on the STIRR Desk may have had to make an offer to sell or a bid to buy Prime Bank Bills to find out the level at which other participants in the Bank Bill Market were willing to trade. Transactions of that kind would primarily be for the purpose of price discovery in order to determine where other participants were willing to trade at large volumes.

601    For example, if Westpac was willing to accept a yield for Prime Bank Bills on a given day of around 4.63 to 4.64, a trader might initially submit a bid at 4.64. If that bid was the best bid in the market and was hit by a seller for $20 million worth of volume, he might then submit another bid at 4.64. If no further sellers were hitting the bid at 4.64, he or another market participant might submit a bid at 4.63. Being the best bid in the market, that new bid of 4.63 would display on broker screens. If that bid of 4.63 was hit by a seller, the trader might choose to bid again at 4.63 as it would appear that there was liquidity at the yield of 4.63. If that bid was not hit, he (or other market participants) might continue to decrease bids until further liquidity was found. However, if volume was being traded at 4.63, he would usually continue to bid at the same level.

(f)    Trading in Prime Bank Bills to support the operation of the Bank Bill Market

602    As a Prime Bank, Westpac had a responsibility under the AFMA Negotiable/Transferable Instruments Conventions (NTI Conventions) to support the operation of the Bank Bill Market. During the relevant period that involved providing consistent price support for all Prime Bank Bills in the Bank Bill Market, and endeavouring to maintain a minimum issuance level of $5 billion.

603    In addition to these AFMA responsibilities, Westpac had a genuine interest in the proper operation of the Bank Bill Market. As I have said, Westpac viewed Prime Bank Bills as important to building its liquid asset buffer and meeting short term funding needs. As such, the availability of investors and issuers or sellers in the Bank Bill Market was important. If market participants lost confidence in the operation of the Bank Bill Market, due to extreme volatility or otherwise, the market would cease to be liquid and Westpac would lose an important means of managing its funding and liquidity.

604    As to what this commitment meant at the level of individual traders, Ms Johnston said that providing price support and liquidity for Prime Bank Bills did not oblige Westpac to engage in two-way pricing (that is, providing both a bid and an offer). While she would show two-way prices if necessary, she would often enter a bid or offer through the brokers in the BBSW Rate Set Window if she could see that there were one-way offers or bids that did not have a counterparty on the other side (that is, a one-sided market). If her bid or offer was taken by another market participant, her trade would be supporting liquidity in the Bank Bill Market. She could then choose to show another bid or offer at the same or a different level.

(g)    Opportunistic trading for price

605    Ms Johnston said that she would trade in Prime Bank Bills for funding and liquidity purposes if she could find liquidity during the BBSW Rate Set Window at an attractive price. A trader would purchase Prime Bank Bills if there were sellers willing to offer at a high yield and therefore low price. Alternatively, they would sell or issue Prime Bank Bills if there were buyers willing to bid at a low yield and therefore high price. They would be opportunistic when trading to benefit from low prices (higher yields) if a buyer and high prices (lower yields) if a seller. As Westpac needed to both buy liquid assets and raise funds on a regular basis, it made sense to buy or sell Prime Bank Bills to take advantage of attractive prices on any given day.

606    Ms Johnston said that the view on whether or not Prime Bank Bills in the BBSW Rate Set Window were cheap or expensive on any given day was affected by a range of factors, and in particular, the longer term assessment of the RBA cash rate level, and the timing and direction of any potential moves. This assessment was informed by many factors, including:

(a)    Australian economic data, such as unemployment figures and the CPI;

(b)    trends in the trading of other financial products, such as BAB Futures and OISs;

(c)    events in other economies and financial markets generally;

(d)    the global outlook for interest rates in the US, UK and Europe; and

(e)    economic strategy papers and research providing market commentary on future movements in interest rates.

607    By being active and trading in the Bank Bill Market each day, a trader would also gain information regarding funding market sentiment which would help develop an assessment of the fair value for the spread between the cash rate and BBSW, which would influence the assessment of value for Prime Bank Bills. In order to determine if the yield at which Prime Bank Bills were trading was attractive, a trader would also consider where Prime Bank Bills were trading relative to other financial products.

608    If Prime Bank Bills were trading at a price that a trader thought was cheap (because the yield was high) then depending on other factors, they may buy Prime Bank Bills. Conversely, if Prime Bank Bills were trading at a yield that was thought to be low (making the price high), then depending on other factors, they may sell or issue Prime Bank Bills. While a trader would trade opportunistically to take advantage of attractive prices in the BBSW Rate Set Window, Ms Johnston said that she would also consider the needs of Group Treasury on any given day (such as executing short term funding or building short term asset holdings) which would also affect her view on fair value for Prime Bank Bills.

(h)    Compliance framework

609    Let me deal with one other matter at this point relevant to trading.

610    During the relevant period, Westpac did not have any policies or procedures that specifically addressed trading in the Bank Bill Market or making submissions to AFMA as to Westpac's view of the mid-rate of the yield for Prime Bank Bills.

611    But from July 2012, Westpac had in place a policy document entitled "Monitoring of WBC Rates Contribution to the AFMA 10am BBSW Rateset", but this policy focused entirely on the submission process as opposed to the underlying trading that informed the submissions. But in December 2013, Westpac, recognising that financial benchmarks played an important role in the financial systems' core functions of pricing and allocation of risk and capital, adopted a "WIB Benchmarks Policy". Further, from February 2014, Westpac had an updated version of the 2012 policy which dealt with additional requirements for monitoring trading during the BBSW Rate Set Window. Let me elaborate.

612    The July 2012 "Monitoring of WBC Rates Contribution to the AFMA 10am BBSW Rateset" policy required that the Dealing Room Risk Manager, Financial Markets & Treasury Risk monitor Westpac's BBSW contribution and ensure that they were consistent with market rates, by checking Westpac's contributed rates against those of other contributors each day.

613    The December 2013 "WIB Benchmarks Policy" had broad application to all rate submissions by WIB. In one sense it was considered a base line policy upon which was to be superimposed the additional best practices or rules/regulations in the relevant jurisdiction of operation. Although primarily concerned with submissions for benchmark rates, I note that the policy stated in relation to conflicts of interest:

Conflict of interest situations should be avoided as far as possible to ensure there is adequate separation between the Contributing Officer and the Supervisor, and other persons and/or businesses who actively engage in trading the asset class or any financial instrument that references the Benchmark.

614    The February 2014 "Monitoring of WBC Trading Activity AFMA 10am BBSW Rateset" was an updated version of the 2012 policy and required Treasury traders participating in the Bank Bill Market to complete a daily template setting out their trading activity and providing the rationale behind their actions. Potential rationales included issuance, credit purposes, liquid asset requirements, hedging interest rate risk and market making. These templates were to be reviewed by the risk manager.

615    Finally on this aspect, in terms of risk, I have referred earlier to Westpac's "Dealing Room Policies for Trade Risk: Market Risk Management" and its "Dealing Room Policies for Risk in Financial Markets and Treasury".

616    I will return to a discussion concerning Westpac's policies towards the end of my reasons when I address ASIC's s 912A case.

DYNAMICS OF THE BANK BILL MARKET AND TRADING STRATEGIES

617    Mr Zuber, Mr Roden and Ms Johnston all gave evidence regarding their general experiences with the Bank Bill Market. Most of this evidence was not challenged by ASIC. This material provides a useful understanding of how individuals at different levels of Westpac viewed the Bank Bill Market and how they approached it in their day-to-day activities.

618    In the relevant period, the Bank Bill Market operated very differently to other, more well-known, markets such as the market for listed equities or a commodities market. Mr Zuber cogently explained that this was so for a range of reasons.

619    First, there were a limited number of large participants that held sophisticated views on current and future interest rates, and therefore held views on the fair value of Prime Bank Bills.

620    Second, Prime Bank Bills were short term instruments used to raise funds at short notice and with little administrative burden by the four major Australian retail banks (including Westpac) and other Prime Banks over the relevant period. As such, for the Prime Banks, Prime Bank Bills were funding instruments with few practical constraints as to how much could be issued (within regulatory limits). There were no inherent characteristics that compelled a Prime Bank to sell or issue them.

621    Third and similarly, Prime Bank Bills were short term instruments that could typically be used to invest cash in return for a spread above the cash rate. As such, for the Prime Banks, Prime Bank Bills were investment instruments with no inherent characteristics that compelled a Prime Bank to buy them.

622    Fourth, many participants in the Bank Bill Market, in particular the Prime Banks, did not generally have to trade on any particular day. It was Westpac's practice to trade when factors such as liquidity, volume, and price were aligned. Those factors were not guaranteed every day.

623    Fifth, Prime Bank Bills generally traded in lines of $20 million. A bid or offer for Prime Bank Bills of a particular tenor usually did not indicate the volume of Prime Bank Bills ultimately for sale or being sought, other than the $20 million minimum amount.

624    Sixth, it was not possible to know the counterparty with which any participant was dealing or the issuer/acceptor of the Prime Bank Bills bought until after the trade was complete.

625    Seventh, Prime Bank Bill volumes were extremely elastic and liquidity in the market changed on a daily basis. This was primarily due to new economic and financial information from local and global markets, as well as vicissitudes in the funding and liquidity positions of Bank Bill Market participants. On some days, trading of Prime Bank Bills in a particular tenor would be negligible or such bills would not trade at all.

626    Eighth and as I have previously described, Prime Bank Bills traded according to two maturity periods of "earlies" and "lates". That is, in the first half of the month, Prime Bank Bills that mature from the 1st to the 15th of the month are liquid and trade on the Bank Bill Market. This period as known as "earlies". During this period, Prime Bank Bills which matured in the second half of the month were largely illiquid and were not regularly traded on the Bank Bill Market. In the second half of the month, Prime Bank Bills that matured from the 16th to the last day of the month are liquid and trade on the Bank Bill Market. This period was known as "lates". During this period, Prime Bank Bills which matured in the first half of the month were largely illiquid and are not regularly traded on the Bank Bill Market. The earlies/lates split could cause spikes in trading activity at the beginning and end of each maturity period, as market participants sought to sell Prime Bank Bills before they become illiquid towards the end of each period and to buy Prime Bank Bills at the beginning of each period.

627    Ninth, the vast majority of trading in the Bank Bill Market occurred during a 10 minute window around 10.00 am each day. The high speed of trading and the manual nature of the market (conducted through voice brokering and broker screens that were updated manually) meant that market participants had only a very small window of opportunity in which to trade.

628    Tenth, during the BBSW Rate Set Window a trader had to be prepared to make value decisions and react to information very quickly, in circumstances where there was limited information available in relation to the depth of the market and the market participants that day. As such, participants traded in the BBSW Rate Set Window with imperfect information.

629    Finally, market participants could not be distinctly characterised as either buyers or sellers, as participants could be both in the same or different tenors of Prime Bank Bills (although Mr Zuber said in cross-examination that Westpac was typically a buyer in the relevant period). Market participants, in particular Prime Banks, could trade in order to make a profit from temporary price impacts. That is, Prime Banks could trade to take advantage of arbitrage opportunities that arose, given the factors described above.

630    As I have already indicated, the earlies/lates dynamic was a particularly important feature of the Bank Bill Market. For example, trading on 1 June meant that:

(a)    1 month Prime Bank Bills being traded would mature from 1 July to 15 July;

(b)    3 month Prime Bank Bills being traded would mature from 1 September to 15 September; and

(c)    6 month Prime Bank Bills being traded would mature from 1 December to 15 December.

631    This dynamic also had the effect of reducing liquidity in the BBSW Rate Set Window when stock became "off the run". That is, 3 month Prime Bank Bills maturing on 10 November (which would be liquid and readily traded on 15 August) would become relatively illiquid from 16 August onwards. If those 3 month Prime Bank Bills maturing on 10 November were not sold on or before 15 August, STIRR might need to hold those Prime Bank Bills for a further month and a half until 1 October, when they would become 1 month Prime Bank Bills which could be more easily traded in the BBSW Rate Set Window.

632    For that reason, STIRR would often sell 3 or 6 month Prime Bank Bills at the end of the two week "earlies" or "lates" period in order to avoid holding too many assets that could not be readily traded in the Bank Bill Market in the new maturity period. However, it was Ms Johnston's usual practice to buy and hold 1 month Prime Bank Bills.

633    STIRR would often buy Prime Bank Bills at the beginning of the "earlies" or "lates" period in order to re-build STIRR's holdings of liquid assets. This pattern of buying stock at the beginning of a maturity period and selling towards the end of a maturity period was common among market participants who wanted to ensure the Prime Bank Bills they held were of the utmost liquidity.

634    It was also the case that at the end of each maturity period when most other market participants were selling, those days were highly liquid (from a buyer's perspective) and potentially good days to purchase Prime Bank Bills. Despite being at the end of a maturity band, a trader like Ms Johnston could potentially buy those Prime Bank Bills at an attractive price and keep them for repo with the RBA if necessary.

635    Overall, in the relevant period Westpac viewed the Bank Bill Market as a primary source of short-dated liquid assets and short term funding. As such, Westpac wanted other market participants to continue to trade in the BBSW Rate Set Window as those participants would not invest in Prime Bank Bills (including those issued by Westpac) or sell Prime Bank Bills if yields were highly volatile or there was a lack of liquidity.

(a)    Trading strategy Mr Zuber's evidence

636    Mr Zuber made the following comments concerning his understanding of trading strategy in the Bank Bill Market. Given his position, his understanding and knowledge has some probative force, although the general evidence given on this topic by Mr Roden and Ms Johnston had greater weight. Moreover, the vast majority of such evidence was not seriously challenged by ASIC.

637    First, at the beginning of each day, a trader would start trading with a view of where they thought the reasonable price for Prime Bank Bills would be on that day. That view would usually be informed by matters such as where BBSW set the day before plus any new information that had occurred since then that might affect market sentiment or the supply of, or demand for, Prime Bank Bills. If buying or selling pressure moved the yield at which Prime Bank Bills were trading away from a price that the market regards as reasonable, it was likely that other market participants would trade in a countervailing manner in line with their views of fair value. The nature of the participants in the BBSW Rate Set Window (very large financial institutions with sophisticated and well-researched views on the fair levels of interest rates and the resources necessary to profit from any pricing anomalies in the market) meant that Prime Bank Bill yields in the Bank Bill Market self-corrected and rarely deviated very far from fair levels.

638    Second, Westpac could choose to hold onto Prime Bank Bills that it wished to sell or wait to issue its own paper until the price was attractive to Westpac. Equally, Westpac could wait to buy Prime Bank Bills until a desirable price was offered.

639    Third, each market participant would have a slightly different view of the fair value of Prime Bank Bills, even if participants' underlying views of movements in interest rates were aligned. That is because each participant would have different drivers for trading each day that were to a large degree unknown to other parties. For example, a participant's immediate funding needs might influence their view of what was "fair value" for Prime Bank Bills on that day. This made it more difficult to predict how other participants would behave in the market.

640    During the relevant period, Mr Zuber regularly met with Group Treasury staff on a formal and informal basis to discuss Westpac's balance sheet, the funding position and upcoming deals, liquidity in various funding markets, views on domestic and global interest rates, external market conditions, activities at the other major trading banks and other financial market trends. All of this information fed into Westpac's trading strategy at or around 10.00 am on a particular day.

(b)    Mechanics and trading strategy Mr Roden's evidence

641    Mr Roden provided some insight into his views around the mechanics of trading Prime Bank Bills, most of which was not seriously contested by ASIC. It was Mr Roden's practice, and the practice he instructed other traders on STIRR to adopt, to remain in the market and continue trading after 10.05 am if other participants were still willing to trade.

Mr Roden's views on yields

642    Prime Bank Bills usually trade at a yield of some margin above the cash rate. When forming a view as to what the fair market value of a Prime Bank Bill and whether the yield for a Prime Bank Bill was reasonable (or was too high or too low given market conditions), Mr Roden took into account a number of factors including:

(a)    the current cash rate;

(b)    the market's and his expectations of possible cash rate movements;

(c)    Westpac's cash flows and his sense of other banks' cash flows, for example, expected collateral flows;

(d)    any premium that should be applied for credit risk;

(e)    the term to maturity of the Prime Bank Bill;

(f)    the levels of supply and demand for Prime Bank Bills (for example, at the end of the early or late period demand for Prime Bank Bills would likely be lower);

(g)    whether alternative funding or investment options presented better value;

(h)    the possible impact of offshore interest rates on domestic interest rates;

(i)    credit spreads; and

(j)    his assessment of risk in the market generally.

643    Other market participants might have different views about the appropriate yield for Prime Bank Bills based on their opinions on current or future market conditions or their individual organisation's cash flows. If this was the case, Westpac might be willing to bid several basis points higher or lower than other market participants in a particular tenor on a given day.

644    A number of different factors affected whether Mr Roden was willing to bid or offer a particular yield for a Prime Bank Bill. For example, on days that Westpac needed to purchase a large volume of 1 month Prime Bank Bills to maintain its liquid asset holdings, he might be willing to accept a slightly less favourable yield in order to find liquidity. Similarly, if the liquidity point appeared to be at a yield which was particularly attractive, he might adjust trading to take advantage of that perceived value and buy more Prime Bank Bills than initially intended on a particular day. Westpac's size and flexibility in terms of how and where it obtained its funding provided scope to take advantage of these opportunities as and when they arose.

645    Moreover, he gave evidence that the face value and trading amounts of Prime Bank Bills could provide a misleading impression of the actual amounts in question. While frequently billions of dollars in Prime Bank Bills could trade on a given day, in reality, the underlying funds flow, while significant, was not in that magnitude. The borrower of a Prime Bank Bill (the "seller" or "issuer") only paid the yield on the Prime Bank Bill and the lender (or purchaser) only received the yield on the Prime Bank Bill. For example, where the yield on a 3 month Prime Bank Bill with a face value of $100 million is 2%, ultimately, the interest component that the borrower pays and the lender receives would only be around $493,150 (using a simple interest calculation). As a result, small changes in the yield of a Prime Bank Bill would not significantly affect the cash flow between the lender and the borrower. If the yield on the Prime Bank Bill of $100 million dropped to 1.95%, this would mean that the borrower would pay and the lender would receive around $487,500, a difference of only around $5,650.

Mr Roden's views on the earlies/lates dynamic

646    To Mr Roden's observation, the earlies/lates convention and the related liquidity issues it caused had an impact on the yields at which Prime Bank Bills traded at particular times in the month. Assuming that market conditions were normal, as the 15th of the month approached, market participants holding "early" Prime Bank Bills would seek to sell those "early" Prime Bank Bills before they became illiquid on the 16th. Similarly, as the end of the month approached, market participants holding "late" Prime Bank Bills would seek to sell those Prime Bank Bills before they became illiquid on the 1st of the following month. All other things being equal, the increased supply available at these times in the month often had the effect of increasing the yields at which Prime Bank Bills were trading, and consequently, increasing the level at which BBSW set. Conversely, at the start of each early or late trading period, market participants would seek to buy Prime Bank Bills to build up their stock of liquid Prime Bank Bills. All other things being equal, this often had the effect of decreasing the yields at which Prime Bank Bills were trading, and consequently, decreasing the level at which BBSW set.

647    In addition, Westpac and other banks had a number of corporate customers with Bank Bill lines. These products were effectively loans made to corporate customers which were supported by an underlying Prime Bank Bill which Westpac (and other banks) might choose to hold or sell in the Bank Bill Market. Because these customers generally preferred 1 month loans, at the maturity date they would request to roll the loan onto another 1 month Bank Bill loan. Although there was no particular convention that applied to this product, if the next 1 month period fell on a non-working day, the new loan's maturity date would roll to the next working day unless that working day was in the subsequent month, in which case it would roll to the previous working day. As a consequence, this meant that customers with Bank Bill lines gradually rolled to the end of the month, leading to a build-up of selling or issuance of Prime Bank Bills around this time. This often added to the selling pressure caused by the conclusion of the late period at the end of the month, and in normal market conditions might have the effect of increasing Prime Bank Bill yields, and consequently increasing the level at which BBSW set.

648    If the market was expecting interest rates to go up, the usual pattern would be distorted. For example, if the market was expecting rates to increase, the drop in yields at the transition of a trading period might be less significant, or possibly, might reverse such that yields actually increased. Conversely, if the market was expecting rates to decrease, the drop in yields from the late to early period might be more significant.

Interacting with brokers

649    Most trading in the BBSW Rate Set Window was completed using a telephone speaker commonly referred to as a "squawk box". I have elaborated on this aspect in an earlier section of my reasons (The Bank Bill Market and BBSW, particularly section (c)). To communicate with the broker, the Westpac trader pressed a button on the speaker and spoke into it.

650    The Westpac trader on a particular day for each tenor would communicate to the broker Westpac's bids and offers for that tenor. Although other words may have been used, this was commonly communicated to the broker by stating the numbers which followed the decimal point in the yield that was being bid or offered for, followed by the month in which the Prime Bank Bills in that tenor matured. Other traders in the Bank Bill Market also traded in this manner.

651    Similarly, the broker would usually communicate the spread between the best bid and the best offer to the traders for each tenor over the squawk box by stating:

(a)    the numbers following the decimal point in the yield being bid;

(b)    the numbers following the decimal point in the yield being offered; and

(c)    the month in which the Prime Bank Bills in that tenor matured.

For example, if the best bid on 1 month Prime Bank Bills maturing in May was 4.67, and the best offer was 4.62, the broker might say "67 62 May". The broker might also only communicate the best bid, or only communicate the best offer, depending on the nature of trading on a given day. If this was the case, the broker might say, for example, "67 bid in May" or "62 May offered". In some instances, the broker might also omit the relevant month if, for example, the numbers after the decimal point in different tenors were substantially different.

652    Offers appeared on the broker screens in parcels or "lines" of $20 million worth of Prime Bank Bills (that is, a minimum number of $20 million of Prime Bank Bills could be purchased in a broker transaction). However, it was possible to offer more than $20 million. Mr Roden stated that generally speaking:

(a)    Brokers would sometimes, but not always, communicate to the relevant Westpac trader when trades occurred in their tenor between other participants or when the Westpac trader completed a trade. Although other words may have been used, this was commonly communicated using the words "done", or to indicate that an offer had been accepted, "taken" or "lifted", or to indicate that a bid had been accepted, "given" or "hit". Brokers also regularly used the word "yours" to indicate that a Westpac trader had taken an offer or been given stock in response to a bid.

(b)    Sometimes, but not always, the broker would also communicate the volume of stock purchased by the Westpac trader by referring to how many million dollars worth of stock they had purchased. For example, if the Westpac trader had purchased $100 million worth of Prime Bank Bills at a yield of 4.62, the broker might say "100 done at 62".

Finding the liquidity point

653    During the BBSW Rate Set Window, it was common for market participants, including Westpac, to start trading by making offers in relatively small amounts, for example the minimum line of $20 million, in order to find the liquidity point in the market in a process of price discovery. In this way, participants traded in smaller volumes to find the yield at which sellers or buyers on the other side were willing to trade in volume. This was necessary because only the best bid and offer was disclosed on the broker screens, and usually only for the minimum amount of $20 million, so participants did not have visibility of market depth, that is, the volume of available demand and supply away from the current best bid and offer (or even at the best bid and offer beyond volumes quoted in the screens).

654    Because bids and offers were usually, but not always, displayed in $20 million parcels and trading was done anonymously through brokers, participants were not usually aware of the counterparties with whom they were trading. Mr Roden stated that when trading in the BBSW Rate Set Window, a participant could never be sure exactly how much volume might be available at a particular yield, that is, how deep the market was. Although there might be an offer on screen at a yield of 4.63, that could mean that there was $20 million of volume available at that yield, or it could mean that there was $500 million of volume available at that yield.

655    Bids and offers could appear and disappear within seconds as trades were completed and quite often, the broker screens did not immediately show bids and offers. This at times made it difficult to keep track of the flow of trading. Because of the speed at which trading occurred during the BBSW Rate Set Window, it was not uncommon for a trader to have bought or sold more than he/she expected. If that occurred when Mr Roden, or another member of his team, was trading it was his practice to always honour and never cancel trades that were completed during the BBSW Rate Set Window. Similarly, individual banks, including Westpac, had no certainty as to how much trading volume occurred in the whole market on any particular day as not all trades were shown on screens, trading flows were difficult to follow and this information was not provided by brokers.

Issuer/acceptor credit limits and associated credit risk

656    Another feature of the Bank Bill Market that Mr Roden commented upon was that when acquiring Prime Bank Bills during the BBSW Rate Set Window, traders did not know which bank's Prime Bank Bills they were purchasing (i.e. the "issuer" or "acceptor" of the Prime Bank Bill) or usually which market participant they were purchasing the Prime Bank Bills from (i.e. the "counterparty" to the trade). Sometimes, if asked, brokers might communicate the identity of the counterparty to disclosed bids and offers. But more commonly traders did not find out the identity of the counterparty to their trades or which bank's Prime Bank Bills they had purchased until trade confirmations were received from the brokers later in the day.

657    Brokers would provide the details of trades completed during the BBSW Rate Set Window (i.e. volume, identity of issuer or acceptor of Prime Bank Bills traded and counterparties to trades) by around 12.00 pm each day. Brokers would usually do this by sending a confirmation email containing trade details, but sometimes Westpac traders might have telephone calls with brokers shortly after the BBSW Rate Set Window to get a summary of that morning's trading.

658    As Westpac could never be sure of the identity of the issuer or acceptor of purchased Prime Bank Bills in the Bank Bill Market, credit limits on the issuers/acceptors of Prime Bank Bills was an important consideration when trading in the BBSW Rate Set Window. Because a credit limit breach was viewed by Westpac's management to be very serious, it was Mr Roden's usual practice, and the practice he instructed other traders in STIRR to adopt, to keep a reasonable buffer on issuer/acceptor credit limits to ensure that any acquisition of Prime Bank Bills issued or accepted by a particular bank did not result in Group Treasury exceeding the credit limit for that bank.

659    If, for example, Westpac was close to its credit limit on CBA prior to the BBSW Rate Set Window, he might choose not to acquire Prime Bank Bills in the BBSW Rate Set Window on the chance that he might unknowingly purchase CBA Prime Bank Bills and breach the credit limit on CBA. Alternatively, if Prime Bank Bills purchased in the BBSW Rate Set Window turned out to be CBA Prime Bank Bills and Westpac was close to its CBA credit limit, he might then try to sell the CBA Prime Bank Bills the next day so as to reduce Westpac's credit exposure to CBA and bring the CBA credit limit back to a more acceptable level.

Potential effect of trading on the rate

660    Mr Roden's experience of the Bank Bill Market was that it was very difficult to predict with any confidence precisely what would occur. One of the complicating factors was that many of the participants would often trade in very large volumes that could, depending upon the countervailing appetite in the market, have an impact on the price. However, it was also frequently the case that very significant volumes of Prime Bank Bills would change hands with little or no apparent impact on the price because there was a corresponding appetite. Further, predicting when there would be genuine liquidity was very difficult.

661    Another feature of the market that he noted was that most of the participants seemed to have a very good understanding of the interest rate environment which meant that deviations from a reasonable level rarely lasted for very long. Where his team saw prices that were considered to be outside a sensible range they would regularly seek to take advantage of that bid or offer and their balance sheet and resources allowed them to buy and sell for value where prices looked wrong. Mr Roden assumed that other market participants operated in a similar way.

662    That said, Mr Roden was also aware that there would be instances where someone trading in large volumes (or even small volumes) could affect the rate. He was also aware that given the level of trading Westpac needed to do from time to time (either to fund itself or to maintain its Prime Bank Bill inventory), the trading could have an impact on the market. However, given the very regular occurrence that other participants in the market might have a corresponding or greater opposite desire to STIRR, Mr Roden considered that most of the time he could trade in the Bank Bill Market with little impact on price. Indeed, the rate could move contrary to the direction of his trading, that is, the rate could move up despite Westpac's buying or down despite Westpac's selling.

663    Those matters meant that Mr Roden was usually reluctant to speculate on the likely direction of BBSW and very rarely took anything for granted about the possible impact of Westpac's trading. His view was that it was preferable, where possible, not to trade in a way which, if trading did happen to have an effect on the level of BBSW, would adversely affect the rate reset position. He would generally seek to avoid large selling activity on days where Westpac had a paid rate set and generally seek to avoid buying significant volumes of bills on the day when he had a received rate set. He stated that this was not a fixed position and there would have been many days in which he would have ignored that position if Westpac's funding or liquidity management needs required trading on the given day. I have discussed earlier how ASIC has sought to leverage off such evidence in order to support its case thesis.

664    That particular consideration did not arise that frequently because in the overwhelming majority of cases, the team had short rate set positions and needed to buy Prime Bank Bills, and on days when they had larger rate set positions or where there was likely to be particular market volatility, they would often be seeking (as Mr Roden explained it) to hedge that position which also would mean that they were either buying on days when they had a paid rate set or vice versa. I have already discussed ASIC's challenge to that hedging rationale in relation to rate set risk as distinct from outright interest rate risk.

Direct trades

665    On most days during the relevant period, Westpac would also buy and sell Prime Bank Bills with institutional customers and other banks outside of the broker market. These trades were often, although not always, conducted at BBSW flat, that is, the parties would agree in advance that the trade would be at the rate that BBSW set that day.

How Mr Roden traded

666    Mr Roden made the following comments regarding his trading behaviour. First, during the relevant period he often, but not always, conducted Westpac's trading in Prime Bank Bills during the BBSW Rate Set Window for some or all tenors. Other STIRR traders also conducted Prime Bank Bill trading in the BBSW Rate Set Window during that period and it was very common (and was the practice on most days) for different STIRR traders to take responsibility for trading in different Prime Bank Bill tenors. For example, he might trade in the three and six month tenors and another trader might trade in the one month tenor. In circumstances where he was away or where he was for some reason unavailable, other traders within Group Treasury would conduct the trading. Sometimes that would be the more junior STIRR trader and sometimes that would be just another Group Treasury trader. In circumstances where he was on leave, a range of people could have operated the desk.

667    Second, it was his typical practice when managing STIRR's Prime Bank Bill trading to purchase Prime Bank Bills in the one month tenor to hold as liquid assets and to raise funds for Westpac by selling or issuing Prime Bank Bills in the three and six month tenors. Mr Roden considered that Westpac was generally a market maker in the three month tenor and would provide liquidity on both sides of the market, consistent with its obligations as a Prime Bank. Group Treasury also regularly used the three month tenor to raise funds by selling or issuing Prime Bank Bills and would purchase three month Prime Bank Bills to support a BAB Futures delivery, or from time to time to hold as part of its liquid assets portfolio.

668    Third, ensuring that Westpac had sufficient funds to meet its obligations was a priority during the relevant period. Mr Roden therefore favoured obtaining funding through the slightly longer tenors of three and six months as this enabled Westpac to lock in funding for a longer period of time which he and Westpac considered prudent. He could then invest the cash received from that funding by purchasing Prime Bank Bills in the shorter one month tenor where the cash would earn a return but would also be readily accessible should Westpac require it, either because the Prime Bank Bill would mature within 30 days or because he could sell the Prime Bank Bill or "repo" it with the RBA.

669    Fourth, another reason Mr Roden often purchased one month Prime Bank Bills was that during the relevant period he was of the view that one month Prime Bank Bills represented a good value proposition as a result of the shape of the yield curve at that time. The further out on the yield curve (i.e. the further away in time the relevant product matures), the greater the credit, market and liquidity risks to which the lender is exposed. For example, he stated that if the market expected the cash rate to be 2% for the next five years, on the yield curve the one year rate would be something like 2.25%, the two year rate 2.5%, the three year rate 2.75% and the four year rate 3% because there was more uncertainty and risk in lending funds over a longer period of time. Similarly, in the front-end of the yield curve, that is, for maturities of less than one year, there would be a steepness to the curve because there was more risk in lending for three months than there was lending for one month. However, for a time during the relevant period, the front end of the yield curve was flatter than Mr Roden would have expected, meaning that the three month rate was not materially higher than the one month rate. It was his view that the 1 month BBSW rate was higher than it should have been for much of the relevant period based on market conditions, thus representing value for the purchaser of the Prime Bank Bill. As a result, Westpac was usually a buyer most days in the one month tenor.

670    Mr Roden stated that although generally his approach was to purchase in the one month tenor and to raise funds in the six month and sometimes three month tenors, this was a general approach only and Westpac regularly participated on both sides of the market for each tenor where, for example, there was value in doing so or to satisfy its obligations as a Prime Bank to provide liquidity to the market.

671    In summary, throughout the relevant period there were a range of factors that influenced Mr Roden's approach to trading Prime Bank Bills on a given day including:

(a)    Westpac's funding and liquidity requirements on the day including potentially:

(i)    its cash position;

(ii)    the need to maintain Group Treasury's liquid asset holdings; and

(iii)    managing Group Treasury's exposure within applicable credit limits to holdings of Prime Bank Bills from particular issuers/acceptors.

(b)    The size and nature of Westpac's interest rate risk exposure in a particular tenor and the desirability of hedging this exposure (a proposition, as I have said, challenged by ASIC concerning rate set exposure risk but not outright interest rate risk).

(c)    His view on the likely or potential trading strategies of other participants in the Bank Bill Market and the manner in which their trading may impact the price of, or the demand and supply for, Prime Bank Bills.

(d)    His view on whether there was particular value in the price of Prime Bank Bills.

(e)    Whether there appeared to be liquidity in the market, and the extent of that liquidity.

(f)    The date on which the Prime Bank Bills in Westpac's inventory were issued and matured and, in particular, whether they would mature in the early or late part of any given month. For example, if Westpac was holding significant "earlies" in 3 month Prime Bank Bills, he might seek to sell them on or before the 15th of the month at which point they become illiquid.

(g)    His view and expectation of funding market conditions including:

(i)    Westpac's ability to raise funding offshore; and

(ii)    unexpected cash flows which might affect domestic funding conditions including Westpac's and other banks' collateral flows.

(h)    His view and expectation of general market conditions including:

(i)    likely movements in official interest rates;

(ii)    likely movements in future BBSW and other interest rates, including in response to events in domestic and global markets.

(i)    The availability and economic value of alternative trading options.

(j)    Westpac's obligations as a Prime Bank to make a market and provide liquidity in the Bank Bill Market.

(k)    Other limitations imposed by Westpac, including credit limits, short-term offshore limits, liquidity limits and/or market risk limits.

Relevant information sources used by Mr Roden

672    When developing a trading strategy, Mr Roden made use of the Treasury Holdings Report (described in more detail below as the TS Holdings Report), the Treasury Eagle and Cash Position Reports.

673    In addition, during the relevant period STIRR had access to a report which showed Westpac's daily rate set risk position for approximately the next month. This report was accessible electronically on Westpac's system, but it was not Mr Roden's practice to access it. More often, another member of the STIRR team would print a copy which he could review. He gave evidence that he looked at the report fairly regularly but not daily. The rate set risk report showed whether Treasury had a long or short exposure to BBSW for each tenor.

674    But Mr Roden was aware that the report was often not accurate as it did not include up to date information regarding retail and institutional assets and liabilities including corporate bill lending transferred from ALM. Westpac's corporate bill lending exposure was made up of its exposure from Bank Bill loans with customers which were managed by WIB and other corporate loans. Although the report did not include up to date information on this figure, he was usually aware of large corporate bill lending positions, particularly around the end of the month as customers with Bank Bill lines gradually rolled to those dates and he would take potentially large corporate bill lending positions into account when considering the rate set risk report.

675    Mr Roden also took into account external views regarding market conditions. In his experience, domestic and global market conditions affect the market for trading Prime Bank Bills. In particular, the market's views on whether the RBA was likely to move or hold the cash rate would usually be priced into the yield at which Prime Bank Bills traded. Similarly, his own views of domestic and global market conditions and possible movements in the cash rate affected his trading in Prime Bank Bills. In particular, he would consider the conditions in markets which Westpac accessed for alternative funding and possible adverse movements in the cash rate.

676    When considering whether the RBA was likely to move the cash rate, he had regard to a number of different data sources, market circulars and publications from the preceding weeks related to:

(a)    domestic market conditions which could affect the RBA's cash rate decision, including, for example, unemployment data or CPI information released by the Australian Bureau of Statistics;

(b)    global market conditions which could affect the RBA's cash rate decision, including, for example, the instability in Europe and the economic downturn experienced by the United States during the relevant period;

(c)    market expectations as to what the RBA's cash rate decision was likely to be, including, for example:

(i)    OIS rates;

(ii)    Westpac's economics team's expectations as reflected in its publications;

(iii)    Group Treasury's economics team's expectations as reflected in its publications;

(iv)    other banks' expectations published in market updates and circulars;

(v)    the views of influential finance journalists whose prognostications were often adopted (so Mr Roden believed) by large parts of the market and therefore sometimes created opportunities for Westpac to bet against the market by taking an opposite view; and

(vi)    public statements made by senior officials at the RBA including speeches, quarterly statements and minutes from RBA meetings.

He also had a general sense of the market view of what the RBA was likely to do from speaking to market analysts or other traders within Group Treasury.

677    Mr Roden stated that it was not unusual for him to form a different view to the market on the RBA's cash rate decision or market conditions generally.

(c)    Trading strategy Ms Johnston's evidence

678    Ms Johnston also gave evidence that provided a detailed view of her trading strategy during the relevant period. Again, most of this evidence about her general practices and strategies was not challenged by ASIC, save for the question of hedging rate set risk exposure and, of course, specific communications dealing with contravention dates that I will discuss later in these reasons.

679    On a typical day Ms Johnston would prepare for trading in the BBSW Rate Set Window by:

(a)    reviewing overnight market activity and considering current market trends;

(b)    reviewing and considering Westpac's cash position, funding and liquidity requirements that day and more broadly; and

(c)    reviewing internal reports and data relevant to STIRR and Group Treasury.

Overnight market activity and current market trends

680    Each morning, she would check what happened in global markets overnight, obtaining this information through:

(a)    emails circulated internally within Westpac;

(b)    summaries on market activity (including funding markets) provided by other banks; and

(c)    scanning trading screens which tracked market pricing information (such as trends in interest rates, FX, and equities).

681    She would look for potential insights from market commentary, particularly as her role on the STIRR Desk was more than just trading in the BBSW Rate Set Window but rather was to assist in managing the interest rate risk in STIRR. Market commentary would often indicate expectations about whether the RBA would increase or decrease interest rates.

682    In addition, she would have a more general sense of market activity from informal conversations with other members of Group Treasury about what trends were occurring in the markets in which they traded (for example, FX and offshore funding markets), whether Westpac had issued in the US market overnight and the rates at which that issuance occurred, and how that compared to BBSW. She would consider and compare the relative pricing of raising short term funds domestically and offshore to form a view on the kind of prices that might represent good value in the BBSW Rate Set Window.

BBSW/OIS spread

683    Ms Johnston would also consider the spread between BBSW and OIS, which was a reflection of funding market conditions in the Bank Bill Market and even more broadly, including general market perceptions of credit risk of Prime Banks. She would consider trends in the spread between BBSW and OIS over time rather than daily fluctuations. For instance, an increasing spread between BBSW and OIS could imply that the supply of funding was increasing because more banks were issuing in the Bank Bill Market and as a result the BBSW level may have increased. This would mean that the cost of issuance for each of the Prime Banks was also increasing. By contrast, if Prime Banks significantly reduced their issuance in the BBSW Rate Set Window, she would expect the spread between BBSW and OIS to contract. Ms Johnston closely followed movements in the spread between BBSW and OIS in Australia (and Euro and USD LIBOR/OIS offshore) which provided information on contemporaneous market dynamics. When she was in STIRR, her regular trading activity in the Bank Bill Market provided a sense of the usual BBSW/OIS spread and the dynamics of the market, which helped to determine good value in buying or selling Prime Bank Bills.

Foreign currency and interest rate data

684    Ms Johnston would be aware of movements in the exchange rate for the Australian dollar from computer screens and other monitors which provided real-time information on the exchange rate through the course of the day. These screens and monitors would also show different global and domestic interest rates, pricing for a variety of swaps, as well as currency movements for major currencies and other information. She also considered market commentary in relation to foreign exchange movements.

685    Westpac also had large bilateral collateral cash flows in connection with FX derivatives that were largely driven by movements in the AUD/USD exchange rate. If Ms Johnston saw a large movement in the exchange rate on a given day, she would expect a large cash flow impacting Westpac's cash position in about two days time.

686    She would also consider where BBSW had set in all tenors the previous day, as well as the general trend in movements in BBSW in the preceding week.

Cash Position Report

687    Ms Johnston would receive the Cash Position Report for the day shortly after 9.00 am and would consider it prior to trading in the BBSW Rate Set Window. She usually treated the Cash Position Report with some caution due to what she considered were the conservative assumptions it contained. She would have regard to the daily position and consider the Cash Forecast Table but would not rely heavily on the forecast beyond more than two weeks as it was likely to change significantly.

688    She interpreted the Cash Position Report with other contextual information, including:

(a)    information discussed in Group Treasury's weekly funding meeting;

(b)    the Weekly Maturity Forecast;

(c)    informal conversations with the Cash Desk as to expected flows for that day; and

(d)    informal conversations with other members of Group Treasury, including the IMM team responsible for short term offshore issuance.

689    Now the Cash Position Report did not constrain the amounts that a trader like Ms Johnston could spend in buying Prime Bank Bills in the BBSW Rate Set Window. It was not a formal or informal limit on trading activity, as the Cash Desk had resources available to manage Westpac's daily cash position. She was not required to ensure or responsible for ensuring that Westpac's cash position was positive every day.

Weekly Maturity Forecast

690    Ms Johnston would also consider the Weekly Maturity Forecast and Westpac's upcoming funding needs and planned issuance prior to trading in the BBSW Rate Set Window. This report also detailed how many Westpac NCDs were on issue.

Treasury Eagle

691    The risk in the Treasury Eagle identified the risk in STIRR at the end of each business day. That STIRR risk position was then updated throughout the course of the day in the STIRR blotter, a spreadsheet that Ms Johnston maintained with Mr Roden. The risk position for STIRR as captured in the Treasury Eagle was considered by Ms Johnston to be a fair substitute for the STIRR blotter. She would look at STIRR's risk position prior to trading in the BBSW Rate Set Window.

TS Holdings Report

692    Ms Johnston also had regard to a document called the TS Holdings Report which showed:

(a)    the inventory of short dated assets held by Group Treasury, both as total holdings and divided into:

(i)    assets that were maturing in one day, two days, three days, and less than one month;

(ii)    assets that were maturing in one month, two months, three months, four months, five months, six months and greater than six months; and

(iii)    within each of those tenors (except for maturities of one to three days), assets that were maturing from the first to the fifteenth day of the month and assets that were maturing from the sixteenth to the last day of the month. The TS Holdings Report indicated which assets were liquid and therefore could be readily traded in the Bank Bill Market being, of course, Prime Bank Bills;

(b)    available credit limits under "Credient SMT Limits" for all of Group Treasury; and

(c)    Government assets held by Group Treasury and if any assets had been "repurchased" to the RBA in exchange for cash.

693    But the TS Holdings Report did not include:

(a)    any reference to short term assets which were maturing on that day (although this figure could be obtained by looking at the previous day's TS Holdings Report);

(b)    any Westpac issued Prime Bank Bills; and

(c)    any reference to the yield or price at which Prime Bank Bills had been purchased.

694    Ms Johnston was responsible for producing the TS Holdings Report for each day during her time on the STIRR Desk. It was usually prepared around 9.00 am. A new TS Holdings Report was generated by running macros which linked the TS Holdings Report spreadsheet to data regarding Group Treasury's holdings of Prime Bank Bills and other short-dated assets, and Westpac's "Credient" credit limit monitoring.

695    The Credient system limits set out in the TS Holdings Report identified Westpac's available headroom against credit limits for various counterparties, which were not just applicable to Prime Bank Bills and other assets contained in the TS Holdings Report but also long term assets purchased by Group Treasury globally, for example, long-dated securities held in the Liquids book or short-dated assets purchased by New Zealand Treasury. These credit limits referred to the acceptor or issuer of the assets purchased (often referred to as the name of the paper). That is, Westpac might purchase Prime Bank Bills from a counterparty but the Prime Bank Bills might be accepted or issued, in the case of NCDs, by a range of Prime Banks including Westpac.

696    Prior to trading in the BBSW Rate Set Window, Ms Johnston would look at the TS Holdings Report to:

(a)    identify assets that were maturing in the next few days;

(b)    make sure it reconciled with the assets bought and sold the previous day;

(c)    check whether the limits looked correct against the previous day's trading and consider if the level of available limits would require the sale of assets such as Prime Bank Bills;

(d)    check what assets were available in the liquid tenors of the spreadsheet, being the tenors that could be readily traded if they were "earlies" in the first half of the month or "lates" in the second half of the month;

(e)    if they were coming to the end of an "earlies" period or a "lates" period (that is, towards the fifteenth of the month or the last day of the month), consider in conjunction with the available limits, if any, assets needed to be sold;

(f)    alternatively, decide whether any assets should be held, which would depend on current interest rates, any potential future changes to interest rates, credit limits and Westpac's cash position; and

(g)    identify any Government paper held by STIRR and which securities had been repo'd with the RBA.

697    Ms Johnston had regard to the credit limits reported in the TS Holdings Reports when deciding whether or not to sell any of Westpac's available inventory of Prime Bank Bills or buy Prime Bank Bills. Those decisions were also informed by her knowledge of what other Prime Bank Bills might be trading in the market, based upon her daily experience in trading in the BBSW Rate Set Window.

698    If the TS Holdings Report indicated that Group Treasury did not have large amounts of available limits for certain Prime Banks, Ms Johnston would consider:

(a)    selling the Prime Bank Bills accepted or issued by those banks to reduce Westpac's holdings and increase headroom against limits; or

(b)    buying Prime Bank Bills of an amount less than the available credit limit, so that any limit would not be breached.

699    Unsurprisingly, Westpac had very large credit limits for Australian Prime Banks, and relatively smaller limits for the foreign Prime Banks (being BNP Paribas (BNP) and JP Morgan). Despite the smaller limits for those foreign Prime Banks, Ms Johnston was usually not concerned by the size of those limits as JP Morgan and BNP were very small issuers in the BBSW Rate Set Window. The foreign Prime Banks did not have a great need for AUDs to fund their operations in contrast to the four major Australian retail banks. And if she did receive Prime Bank Bills from JP Morgan or BNP in the BBSW Rate Set Window, it was easy to sell those assets very quickly if necessary.

700    It was STIRR's usual practice to aim to keep a reasonable amount of headroom available against the credit limits for other Prime Banks, to ensure that even on days when STIRR wanted to buy a significant amount of assets, STIRR would still be within credit limits.

701    On one or two occasions when Westpac received more Prime Bank Bills than expected that were accepted or issued by a particular bank, Westpac could sell those Prime Bank Bills that day generally through a broker but outside of the BBSW Rate Set Window, or on a subsequent day. But it may have been necessary to offer a higher yield than the yield at which Westpac had purchased the Prime Bank Bills in order to sell them.

702    I will return to this question in a moment.

Run sheet

703    A run sheet was used by traders. The run sheet was printed with information detailing Westpac-issued paper that was maturing that day. Mr Roden and Ms Johnston would keep track of what trades were entered into over the course of the morning by writing them down by hand. This included customer trades dealt through Financial Markets and interbank trades generally dealt through brokers. From the run sheet, STIRR could summarise the cash impact of the day's trading.

704    In the course of the morning between 8.30 am and 10.30 am, members of the Financial Markets division's sales team would relay customer requests to buy or sell Westpac issued NCDs. STIRR would confirm if Westpac was willing to issue or buy back those NCDs and, if necessary, at what rate (that is, if the NCDs were not dealt at BBSW). It was common for customers to roll over their Westpac NCDs. From experience, Ms Johnston expected most customers to roll over their Westpac NCDs which would usually occur at the relevant rate of BBSW that day. She would update the run sheet by hand with this information, as any customer roll overs increased the cash position on the same day. Conversely, if the NCD was not rolled over by the customer, that paper would mature and Westpac would have to pay the face value to the customer, resulting in a decrease to the cash position.

705    Following the BBSW Rate Set Window, Ms Johnston would fill in the run sheet with the broker trades that had occurred. It was her usual practice to speak to the relevant broker who would verbally confirm the trades that had occurred in the BBSW Rate Set Window. She would record those trades by hand on the run sheet.

706    Around 10.30 am, after the run sheet had been updated with:

(a)    customer bill rolls and other transactions executed through Financial Markets;

(b)    trades through the brokers, primarily in the BBSW Rate Set Window; and

(c)    direct trades with smaller bank counterparties,

Mr Roden or Ms Johnston would provide the net cash flow to the Cash Desk to update the cash position for the day.

Group Treasury's BBSW Rate Set Exposure

707    Prior to trading in the BBSW Rate Set Window, Ms Johnston was aware of Group Treasury's BBSW Rate Set Exposure each day in the one, three and six month tenors. Those exposures were expressed as the face value dollar amounts of the net assets, liabilities and derivative transactions resetting by reference to BBSW. She had regard to Group Treasury's BBSW Rate Set Exposure in order to understand the total interest rate risk arising from that exposure as Group Treasury would need to be aware if this risk would cause Group Treasury to breach its risk limits.

708    Apparently Ms Johnston applied the following rules of thumb:

(a)    every $1 billion of rate set exposure in the one month tenor gave rise to approximately $8,000 of basis point risk;

(b)    every $1 billion of rate set exposure in the three month tenor gave rise to approximately $24,000 of basis point risk; and

(c)    every $1 billion of rate set exposure in the six month tenor gave rise to approximately $48,000 of basis point risk.

709    That is, if Group Treasury's rate set exposure was not hedged and crystallised after BBSW was published in a particular tenor, those basis point values would be added to the risk positions in the Treasury Eagle for the corresponding tenor.

Positions in other instruments

710    Ms Johnston would also consider STIRR's position in BAB Futures if it was a week or less from the BAB Futures expiry. It was possible to fund Westpac or take delivery of 3 month Prime Bank Bills through the BAB Futures close out rather than through trading in the Bank Bill Market. For this reason, she would consider the size and direction of the BAB Futures position and the price at which BAB Futures had traded, prior to trading in the BBSW Rate Set Window.

711    Further, Ms Johnston would consider the rates offered in Bills/LIBOR (being the rate at which a party can convert USD into AUD, expressed as BBSW plus a spread) and FX forwards. She would consider whether that rate indicated that it was more or less attractive to fund Westpac in offshore markets, or to lend excess cash via these products (in exchange for USD), rather than investing in assets or issuing in the Bank Bill Market.

(d)    Credit limits

712    As I have already discussed above, as part of their trading strategy the traders on the STIRR Desk were required to consider the credit risk associated with buying Prime Bank Bills. Westpac managed credit risk by setting various limits on the types of securities that it could hold that were issued by third parties. Secondary market trading limits (SMT limits) were a type of credit limit that applied to investments in liquid debt securities and credit derivatives held in Westpac's trading book portfolios, for both Group Treasury and Financial Markets. Prime Bank Bills fell within the SMT limits regime. As such, SMT limits placed a limit on the amount of Prime Bank Bills Group Treasury could purchase from any single bank, and different credit limits applied for different banks.

713    During the relevant period, traders in Group Treasury did not have accurate real-time data in relation to credit limits but were able to approximate numbers from Westpac's traded credit limit monitoring system. On average, Westpac had available headroom for the three other major Australian banks as follows:

(a)    in 2010: $2.077 billion for ANZ, $2 billion for CBA, and $2.301 billion for NAB;

(b)    in 2011: $1.596 billion for ANZ, $1.5 billion for CBA, and $2.33 billion for NAB; and

(c)    in 2012 up to 6 June 2012: $2.974 billion for ANZ, $2.252 billion for CBA, and $2.225 billion for NAB.

Given the average size of these available headroom amounts, Westpac's trading in the Bank Bill Market during the relevant period would very rarely, if ever, have been constrained by a concern that SMT limits would be breached.

714    In relation to the two foreign Prime Banks during the relevant period, JP Morgan and BNP, these banks were not large issuers in the Bank Bill Market as they had very small operations in Australia and did not require a large amount of AUDs to fund their local activities. For this reason, the JP Morgan and BNP SMT limits which applied during the relevant period would not have significantly restricted Westpac's trading in Prime Bank Bills. In any event as I have already said, if Westpac bought Prime Bank Bills from a foreign bank during the BBSW Rate Set Window, that paper could be sold again very quickly to avoid any issue with SMT limits, or could be included in a stock switch of repo-eligible assets with another bank.

715    Generally speaking, SMT limits may have been a reason for selling Prime Bank Bills, to stay within Westpac's credit limits.

OPPORTUNITY AND INCENTIVES TO MANIPULATE

716    ASIC contends that Westpac's structure provided incentives to manipulate.

717    Now as I have said, only Group Treasury traded in Prime Bank Bills and Group Treasury was separate from Financial Markets. But ASIC says that having the trading function in Group Treasury did nothing to quell the incentive for Westpac to seek to influence BBSW to its advantage. And it did not preclude the flexibility of the relevant Westpac STIRR trader attending to funding or other responsibilities on a day which also enabled exploitation of the opportunity to influence the rate set.

718    Further, ASIC says that various financial incentives motivated Westpac's STIRR traders to trade to influence BBSW. I will elaborate in detail on this in a moment.

719    Further, ASIC also says that Westpac's consolidated book gave Group Treasury a particular interest in keeping the 1 month BBSW low to benefit its very large short 1 month exposures. As ASIC points out, Group Treasury had a financial interest when the exposure was short or paying in keeping the BBSW down as its liabilities were set off it. In an email to Mr Murray Regan of AFMA, Mr Roden stated that Westpac's exposures "to the 10am rate setting process" were:

extremely broad, diverse and significant. They include in excess of $50 Billion in liquid assets, $50 Bill in Corp loans, $120 Billion in wholesale funding and , $310 Billion in mortgage loans and deposits of $300 Billion. All of these are impacted by the rate setting mechanism [Email, Mr Roden to Mr Murray Regan (AFMA), 3 August 2012].

720    Further, ASIC correctly points out that Mr Zuber and other witnesses conceded that the rate set exposure would be taken into account by Westpac, as a rational market participant and all else being equal, to the extent that Westpac would avoid trading to adversely affect Group Treasury's rate set exposure. I have previously discussed ASIC's argument concerning the economic rationale of the flipside.

721    Let me elaborate on some of these themes.

(a)    Opportunity

722    ASIC says that Westpac had the opportunity to carry out the Rate Set Trading Practice at the bank or divisional level, and also at the STIRR level.

723    In relation to the bank or divisional level, and as Mr Conway said, Westpac had:

the second biggest balance sheet in Australia, so you know, when those two balance sheets enter the market to trade [that is, CBA and Westpac], to hedge, their activities are quite likely to be the ones that are moving around the rate when they're hedging.

As Mr Conway also acknowledged, "depending on what else is on the other side, trading in the market, could move the rate".

724    Further, STIRR was the only division of Westpac to trade Prime Bank Bills. No other division of Westpac traded Prime Bank Bills on the Bank Bill Market.

725    Further, Westpac's traders had wide discretion as to the days on which they chose to sell or buy Prime Bank Bills. Mr Roden agreed that he had the discretion to time the days on which he sold or issued bills in particular tenors "to days when market conditions were favourable or when, if my trading were to have any impact on the level at which BBSW set, this would not increase Westpac's risk position". And as Mr Zuber explained:

Westpac could choose to hold onto Prime Bank Bills that it wished to sell or wait to issue its own paper until the price was attractive to Westpac. Equally, Westpac could wait to buy Prime Bank Bills until a desirable price was offered. Westpac would not buy or sell Prime Bank Bills at any price.

Further, Mr Zuber placed no constraints on the STIRR Desk around holding any minimum or maximum amounts and did not direct the timing of when Prime Bank Bills should be bought or sold. Those decisions were left to the trader on the STIRR Desk who participated in the Bank Bill Market each day, and was therefore best placed to determine if prices and other market conditions were attractive.

726    Further, during the relevant period Westpac's trading was unconstrained by any policy dealing specifically with trading in the Bank Bill Market or the setting of BBSW.

727    I accept that because of inter-alia these objective structural elements, Westpac had the opportunity by its trading to influence where BBSW set on certain occasions. The question of opportunity and capacity are contestable propositions given how the Bank Bill Market operated, the conduct and reactions of other participants, and also how BBSW was set. I will elaborate on this later.

(b)    Westpac's traders acknowledge that they had regard to the BBSW Rate Set Exposure when trading

728    It is not in doubt that STIRR employees monitored Group Treasury's BBSW Rate Set Exposure on each trading day prior to trading Prime Bank Bills in the BBSW Rate Set Window.

729    Further, Mr Zuber gave evidence that "on many occasions we traded bank bills with the knowledge of what our rate set position was, with the knowledge that potentially that could have an impact on the rate".

730    Moreover, Mr Roden deposed that:

I was of the view that it was preferable, where possible, not to trade in a way which, if our trading did happen to have an effect on the level of BBSW, would adversely affect our rate reset position. That meant, in essence, I would generally seek to avoid large selling activity on days where we had a paid rate set and generally seek to avoid buying significant volumes of bills on the day when I had a received rate set.

[T]here was discretion available to me to time the days on which I sold or issued Bank Bills in particular tenors to days when market conditions were favourable or when, if my trading were to have any impact on the level at which BBSW set, this would not increase Westpac's risk position.

731    And in cross-examination, he agreed that he had discretion to time the days on which he sold or issued Prime Bank Bills or purchased Prime Bank Bills, subject to limits.

732    Further, Ms Johnston deposed that:

[I]f I had a degree of flexibility as to when I had to conduct a particular trade (such as knowing I needed to issue Prime Bank Bills but not necessarily immediately), I might consider not trading on a day where if Westpac's trading did happen to affect the level of BBSW it might have an adverse effect on Group Treasury's Rate Set Exposure;

[I]t sometimes made sense to time any purchasing activity on days when Group Treasury's Rate Set Exposure was paid … [as it was, for example, on 20 May 2010, 20 September 2010, 22 September 2010] … in order to avoid potentially adversely affecting the Bank's position in the event that Westpac's trading affected the rate at which BBSW set.

733    Further, Mr Hosie understood that Westpac's trading could affect the yield at which Prime Bank Bills traded and therefore the submissions that were made by panellists and where BBSW set. From time to time this understanding impacted upon his decisions about when to trade in the Bank Bill Market. The prospect that his trading could have this effect impacted his decisions about when to trade. Mr Hosie said that "we would certainly be conscious of what our rate sets were" and "a larger rate set would, you know, introduce more risk, so we were conscious of those". Mr Hosie agreed that if he wanted to buy Prime Bank Bills he might pick a paid set, but there were other factors to consider. He agreed that the larger the paid set, the more of a consideration it would have been.

734    Further, Westpac admitted in its further amended defence that its trading in the BBSW Rate Set Window was typically influenced by considerations which included an assessment of the likely or potential trading strategies of other Bank Bill Market participants and the manner in which their trading may impact the price and availability of Prime Bank Bills or the Bank Bill Market's appetite to acquire Prime Bank Bills.

735    Further, when trading Prime Bank Bills in the Bank Bill Market, Mr Roden took into account his view of the likely or potential strategies of other participants. Further, Westpac's traders believed or suspected that the trading strategies adopted by other banks included trading to influence BBSW. Mr Roden suspected that UBS was engaging in manipulation of the Bank Bill Market by trading to move the rate at which BBSW set. He also suspected and complained to AFMA about other banks who he suspected were doing the same thing. Ms Johnston was also aware that people in the industry occasionally speculated that participants in the BBSW Rate Set Window were trading to affect the rate at which BBSW set and may have thought other banks' trading was suspicious.

736    Further, Mr Conway speculated that CBA used a cover to achieve its objective of trading with the purpose of trying to move the rate. Mr Park was also aware of the possibility that UBS was trying to influence the rate. Mr Sharma had heard rumours about Prime Banks trading Prime Bank Bills to manipulate the rate set. And Mr Duignan was aware of speculation in the market that Prime Banks were trading in the BBSW Rate Set Window for the purpose of moving the rate to suit their rate set exposures.

(c)    Westpac's BBSW Rate Set Exposure

737    Westpac's traders were conscious during the relevant period of the importance of the rate at which BBSW set. Ms Johnston told Mr Greg Ryan, another employee of Westpac, that BBSW was "an international benchmark which a lot of different products get set off" (and that her team "manage BBSW and the rate set … actually … kind of manage and monitor where BBSW gets set") [Phone call transcript, Ms Johnston to Mr Greg Ryan, 24 May 2010].

738    Further, on 3 August 2012, Mr Roden told AFMA that Westpac's exposures "in excess of $50 Billion in liquid assets, $50 Bill in Corp loans, $120 Billion in wholesale funding and , $310 Billion in mortgage loans and deposits of $300 Billion" were all "impacted by the rate setting mechanism" [Email, Mr Roden to Mr Regan (AFMA), 3 August 2012].

739    Further, as I have already indicated, employees and officers of STIRR, including Mr Roden and Ms Johnston, were able to access Group Treasury's exposure to the BBSW rate set each day. On each day during the relevant period:

(a)    the Group Treasury BBSW Rate Set Exposure was identifiable through Westpac's internal systems;

(b)    the Group Treasury BBSW Rate Set Exposure on any given day included the exposure as at that point in time for each day into the future; and

(c)    employees and officers of STIRR had access to the systems through which the Group Treasury BBSW Rate Set Exposure could be obtained.

740    Let me say something at this point about the quantification and direction of the BBSW Rate Set Exposure relevant to Group Treasury on the specific contravention dates, although I will return to this later. ASIC submits that I should use column F (and not column J) of schedules 1 and 2 to the Statement of Agreed Facts (SOAF) tendered before me (s 191 of the Evidence Act) when determining the direction and quantity of the Group Treasury BBSW Rate Set Exposure to which Westpac had regard when trading Prime Bank Bills in the Bank Bill Market for each contravention date.

741    ASIC contends that column F (and not J) accords with the BBSW Rate Set Exposure to which Westpac's traders had regard when trading on behalf of Westpac on each of these dates.

742    Now Westpac traders had regard to a daily rate set email circulated prior to trading in the BBSW Rate Set Window. The daily rate set email set out the ALM and PRM rate set exposure data on Westpac's Murex system, which is the BBSW Rate Set Exposure in column F (and not column J) of schedules 1 and 2 to the SOAF.

743    ASIC's contention that Westpac's traders had regard to the BBSW Rate Set Exposure recorded in the daily rate set email (and therefore the BBSW Rate Set Exposure in column F, and not column J, of schedules 1 and 2 to the SOAF) is said to be supported by the following evidence:

(a)    Mr Roden's statement to Ms Johnston that he had "14 billion of one month" on 6 April 2010, which is best represented by column F (and not column J) of schedule 2 to the SOAF [Phone call transcript, Mr Roden to Ms Johnston, 6 April 2010].

(b)    Mr Roden's stated belief that he was a "receiver" on 10 June 2010 (as he told others) [Phone call transcript, Mr Roden to Mr Sharma, 8 June 2010] and therefore his belief that he had a net long rate set exposure to the 3 month BBSW. Column F of Schedule 1 to the SOAF for 10 June 2010 contains a net long rate set exposure position, whereas column J does not.

(c)    Ms Johnston's statement to Mr Sharma on 21 September 2010 that she was a "$2 billion payer tomorrow", which is best represented by column F (and not column J) of schedule 2 to the SOAF [Phone call transcript, Ms Johnston to Mr Sharma, 21 September 2010].

(d)    The BBSW Rate Set Exposures referred to in Ms Johnston's email of 16 May 2011 for the contravention dates 1 December 2010, 6 December 2010, 1 March 2011, 4 March 2011, 1 June 2011 and 6 June 2011, which are each best represented by column F (and not column J) of schedule 2 to the SOAF.

744    Now column J of schedules 1 and 2 to the SOAF represents a calculation of Group Treasury exposures which also incorporates exposure data from St George, Corporate Bill Lending exposure data and Transfer Priced Balance Sheet rate set risk exposure for St George and Westpac. But the daily rate set emails circulated to Westpac's traders prior to trading in the BBSW Rate Set Window did not include:

(a)    Corporate bill lending exposure (represented in column H of schedules 1 and 2 to the SOAF); or

(b)    Transfer Priced Balance Sheet rate set risk exposure (represented in column I of schedules 1 and 2 to the SOAF).

745    Corporate bill lending rate risk was not substantially recorded on Westpac's Murex system. Westpac corporate bill lending rate risk was recorded on Westpac's AS400 system, and St George corporate bill lending rate risk was stored on the St George Citadel system. Ms Johnston's evidence was that this type of risk was estimated or based on an assumption of how many bills would roll over on a particular day, and the details of which bills had in fact rolled over would not be known until potentially the next day. Ms Johnston and Mr Stokes received access to the corporate bill lending rate risk system (the commercial bills report) on 7 May 2010. Mr Johnston agreed in cross-examination that it appeared that she received access to this system in May 2010 and she could not recall requesting access to it before then but she went on to say that she may still have seen the relevant data in other sources, even before she received access to such a report.

746    As for the ALM (Transfer Priced Balance Sheet) risk, this was recorded in a spreadsheet prepared by another team in Group Treasury. On 20 May 2010, Ms Johnston reported that such a spreadsheet had not been received for some time. Ms Johnston discussed the spreadsheet with Mr Greg Ryan on 24 May 2010, and her evidence was that he did not consider it to be complete (or "all inclusive"), and she wanted to investigate this [Email, Mr Greg Ryan to Ms Johnston, 24 May 2010].

747    This spreadsheet estimated ALM (Transfer Priced Balance Sheet) risk as at a particular date, and for a number of days following (including non-business days).

748    Now it was the general practice at the time under both the Following Business Day Convention and the Modified Following Business Day Convention in the ISDA definitions for risk falling on a non-business day to roll forward to the next business day. Ms Johnston explained in cross-examination that the usual convention used by Group Treasury was the Modified Following Business Day Convention, which is consistent with what is set out in the BBSW Procedures that I described earlier in these reasons. Under the Following Business Day Convention, the risk would roll forward from a non-business day to the next business day in every case. Under the Modified Following Business Day Convention, the risk would be rolled back if rolling forward would mean crossing into a new month. Under both Conventions, the risk for example for Saturday 1 May 2010 and Sunday 2 May 2010 would have rolled forward to Monday 3 May 2010, as this would not require rolling a date forward to a different month. A trader who had access to the spreadsheet could see the estimated risk for each day separately, and would not roll back risk from the weekend of 1 and 2 May 2010 to Friday 30 April 2010. So if a trader had seen the spreadsheet prior to trading on 30 April 2010, they would have seen the ALM (Transfer Priced Balance Sheet) risk for that day of $2.139 billion in the 1 month (and not $3.492 billion as shown in column I, and included in column J, of schedule 2 to the SOAF).

749    ASIC submitted that the consolidated data represented in column I of schedule 2 to the SOAF was prepared on the basis that rate set risk for weekend days were rolled back to the previous business day, which is contrary to both Conventions. This is relevant for the contravention date on 30 April 2010, where column F shows a short exposure and column J shows a long exposure, and I will address the matter further later in these reasons. But I would note now that even if I accept ASIC's submission above that ALM (Transfer Priced Balance Sheet) risk for that day of $2.139 billion (and not $3.492 billion), I would still be left a column F showing a short exposure and column J showing a long exposure.

750    Prior to August 2010, the rate set risk recorded on Westpac's Murex system and included in the daily rate set emails (and represented in column F to the schedules to the SOAF) did not include BBSW rate set risk from St George's Citadel and Kondor systems, which were not at that time stored on Westpac's Murex system. Mr Stokes' evidence was that prior to the migration of St George data onto the Murex system in August 2010, he manually added St George risk exposure to a report that he saved on Westpac's hard drive. He does not recall having a practice of printing out hard copies of this report. Further, he did not know whether other Treasury employees looked at the report on Westpac's hard drive on any given day during the relevant period.

751    I will return to the question of the recording and quantification of the exposures (and those to which the traders had access) in the "Contravention Dates and Alleged Practice" section of my reasons. But let me say now that I am not able to say that traders only ever had access to and relied upon the column F figures (as ASIC contends) or conversely at all times had access to and relied upon the column J figures (as Westpac contends). Each contravention date needs to be considered separately. But I would note that the issue only becomes significant on dates where say column F shows a short exposure and column J shows a long exposure or vice versa, or where the columns show the exposure in the same direction but the quantum of the exposure in one column as compared with the other column is significantly less, say by one or more orders of magnitude. Accordingly, when I discuss the contraventions dates I will only elaborate on this issue when it is necessary to do so for a particular date as having significance to the relevant trader's knowledge which then may inform the question of purpose.

(d)    Financial incentives

752    Westpac not unreasonably expected Group Treasury to make a profit. Indeed, Group Treasury had profit targets. And it may be accepted that during the relevant period Group Treasury had a financial interest in where BBSW set.

753    Group Treasury had an interest in keeping the 1 month BBSW low to benefit its usually very large short 1 month BBSW exposures, which Mr Zuber explained resulted from or were at least exacerbated by basis swaps that it entered into to convert its 3 month interest rate exposure to 1 month interest rate exposure, with rate sets every 90 days. Further, on a significant number of trading days during the relevant period, Westpac purchased 100% of the 1 month Prime Bank Bills sold during the BBSW Rate Set Window. Mr Zuber agreed that Group Treasury generally was a significant buyer in 1 month Prime Bank Bills.

754    Mr Steven Carritt was an expert witness called by Westpac, who had been employed by CBA's Treasury for 20 years and had been responsible for its Asset and Liability Management desk. He was retained by Westpac to provide expert evidence on topics relating to the management of a bank's treasury function and the uses of Prime Bank Bills to assist the Treasury function. His evidence was that he would see it as rational for a market participant who has the appetite and discretion to execute transactions, to do so on days when there may be "offsetting positions", the offsetting position being any movement in the BBSW in the direction of the participant's BBSW Rate Set Exposure that day, all other things being equal. The ceteris paribus assumption is of course of central importance given the other influences, factors and reasons which might motivate a trader's strategy or course of dealing on a particular day. At a more general level I would note at this point that in some of ASIC's case presentation and cross-examination of Westpac's witnesses, ASIC tended to downplay or even ignore that significant assumption.

755    Let me deal with another aspect of financial incentives.

756    Traders such as Mr Roden were awarded not insubstantial bonuses during this period, and Group Treasury's profit and loss was at least a factor in the awarding of those bonuses. Mr Zuber agreed that the traders had a financial incentive to trade in a way that was profitable for Westpac. But Mr Zuber said that "doing this type of activity, I just didn't see it as significant enough that it would be a motivation for someone to do that", that is, engage in the relevant manipulation.

757    For example, in his 2011/2012 performance review, Mr Roden was given an "Outstanding" rating in relation to the "Financial" heading. The measurement for this heading was "STIRR P&L" and the target and stretch target was "STIRR above $200m". This heading was given a 70 weighting. The performance review noted:

Treasury's total revenue for 2012 was $1.1billion. Colin's direct management of the short-end domestic risk generated $350 million of revenue. Colin has led the day to day risk business during one of the most challenging environments to a truly outstanding result. His management of the macro risk profile has resulted in Treasury generating superior returns both in absolute and on a relative basis.

758    In his previous 2010/2011 performance review, it was noted that:

This year Colin Roden has contributed and estimated $500m to Treasury p&l which is more than his prior year (09/10) contribution. Colin is the primary shortend book runner across all bank positions and chief trader of the Treasury desk Colin has made $500m this year through managing the short end through periods of significant interest rate volatility, where the curve has priced rate hikes as well as substantial emergency easings into the curve.

759    And in the 2009/2010 performance review, it was noted that "Colin has produced a robust well above plan 1st half financial performance, he has capitalised on opportunities which the market has presented in the shortend of the Australian market".

760    Generally, Westpac's bonus structure allowed for Group Treasury staff to earn many multiples of their base salary as a result of their financial performance. Mr Roden was well remunerated for his financial performance.

761    Mr Masnick confirmed that Financial Markets employees, including Mr Parker (who entered into BBSW Referenced Products with Westpac's customers), were also remunerated based on financial performance. Mr Masnick confirmed that Mr Parker was very well compensated during the relevant period as a result of his financial performance for Westpac.

762    Now I accept that these incentives, at the divisional level and at the employee level, created incentives for Group Treasury to manipulate where BBSW set.

763    ASIC has relied on the following communications to suggest that Mr Roden considered it important to the profitability of the division to be able to influence the BBSW. I should note two matters at this point. First, I have not corrected typographical errors in the contemporaneous documents. Second, I will say something later about communications involving Mr Garfield Lee of CBA.

764    In one communication with Mr Lee of the CBA, Mr Roden said that by having the sole ability to buy and sell Prime Bank Bills for Westpac, Group Treasury was able to make "a couple of 100 million dollars … very esasily" [Chat log, Mr Roden to Mr Lee, 12 March 2010], though the statement seems jocular in context.

765    In another communication with Mr Lee, Mr Roden referred to the need to "have control over BBSW … to protect yourself" [Chat log, Mr Roden to Mr Lee, 10 February 2011]. Mr Roden here was responding to the concerns of Mr Lee, who it was proposed would run a "single currency basis book" of "cash v 1m v 3m v 6m curves". Mr Roden said that such a book would be loaded with illiquid risk and one would not want such a book unless one had control over BBSW. Mr Roden explained this as a reference to access to Prime Bank Bill trading to mitigate risk. I accept this. Moreover, I do not agree with ASIC that this is only plausible if one understands his reference to mitigating risk as trading to affect the rate at which BBSW sets.

766    ASIC says that Mr Roden's statement "i wouldnt go near an aussie basis book without bbsw control" should similarly be understood as a reference to seeking to control the rate at which BBSW set [Chat log, Mr Roden to Mr Lee, 10 February 2011]. Mr Roden said that an Aussie basis book is "going to give you floating rate exposures in Aussie dollars, pay and receive sets". Group Treasury used basis swaps to switch its risk from 3 month to 1 month, which increased the size of its 1 month BBSW Rate Set Exposures. Mr Roden said that his reference to "BBSW control" was to "some methodology to mitigate some of my risk, because bank bills is really the only efficient mechanism for doing that". But ASIC says that the suggestion that it was most efficient to fully fund the purchase of Prime Bank Bills to offset the risk that Group Treasury had created by entering into basis swaps, particularly when it could reduce or offset its risk with other derivative transactions that would not affect its profit and loss, is implausible. ASIC says that at the very least, Mr Roden was referring to having "control" over all of the bank's BBSW transactions so as to have sufficient market power, funding and ability to trade in a way that could affect the rate at which BBSW set. I disagree with ASIC's analysis which significantly over-states what can be gleaned from this informal chat.

767    Further, ASIC made reference to the fact that Mr Roden explained to Mr Lee in another communication that Group Treasury had "control" of BBSW, it bought "all teh bank bills … and we fund as we choose" [Chat log, Mr Roden to Mr Lee, 25 March 2010]. But Mr Roden said that this was a reference to "within Westpac, Westpac Treasury bought all the client bank bills from Institutional Bank … and hold them within Treasury and fund how Treasury needs to fund". Now ASIC says that this is implausible and submits that Mr Roden was referring to Westpac's participation in the Bank Bill Market. But I am not prepared to reject Mr Roden's interpretation.

768    Generally speaking, it is one thing to say that incentives existed. But it is another thing to say that the traders engaged in manipulative trading, let alone that there was the Rate Set Trading Practice of the type contended for.

769    But I do agree that the financial incentives (albeit indirect as they are not directly referable to specific trading activities) are supportive of ASIC's case.

(e)    The structures of other banks

770    Westpac has contended that structural differences between Westpac on the one hand and NAB and CBA on the other hand gave rise to different incentives and motivations when trading in the Bank Bill Market. On this aspect, I do not consider that the contended for structural differences takes Westpac far. I propose to be circumspect about my discussion of the other banks as the evidence was necessarily incomplete and they were not parties before me.

771    Westpac says that NAB's and CBA's Prime Bank Bill traders, who sat within their respective Markets divisions, were incentivised to trade in a way to maximise the profit for their own book without regard to the interests of Group Treasury more generally. But contrastingly Westpac says that the performance of Westpac's traders, who sat within Group Treasury, was measured on the basis of overall group performance rather than individual portfolio performance. It also says that matters such as ensuring that Westpac had the funding it needed and was protected from interest rate risk, market risk and credit risk as much as possible were the primary considerations.

772    Westpac says that that by reason of these suggested primary considerations, Group Treasury's relatively minor gains from manipulating BBSW to benefit rate set exposure would have been irrelevant and difficult to discern. It also says that the conflicts between the interests of Financial Markets and Group Treasury (or their equivalents) found at NAB and CBA were not present at Westpac. But ASIC says that Westpac's submission is not supported by the evidence. It says that the conflicts which arose at the other banks did not derive from manipulation by STIRR (or STIRR equivalent) traders, but from the opposing rate set exposures of the Markets book and the Treasury book on particular days, a circumstance which was avoided under Westpac's structure.

773    ASIC says that in the case of NAB, the structural differences between NAB and Westpac were relatively insignificant.

774    Further, ASIC says that in the case of CBA, the structural differences between CBA and Westpac were initially relatively insignificant and later minimised by CBA to increase returns on profits from managing rate set risks. It also says that CBA's Treasury desks traded to influence BBSW from time to time to suit their own rate set exposure, again inconsistently with the thrust of Westpac's argument.

775    In elaboration ASIC says that the organisational structure of NAB and CBA meant that the rate set exposure within the equivalent STIRR desks on a particular day could be the opposite of the rate set exposure of a Treasury desk. It says that occasional complaints from Treasury about manipulation by the equivalent STIRR desks were attributable to a lack of alignment of the rate set exposures. But it says that that conflict was avoided under the Westpac model which placed the STIRR function within Group Treasury.

776    In relation to CBA, ASIC has also pointed to the White Paper authored by Mr Carritt in February 2011 which stated:

One basis point reduction in wholesale funding costs, as a result of contrary defensive rate set management by Markets Funding Desk/swap book compared to a focus on defensive activity on the funding rate set by Treasury under an integrated domestic funding model NPAT $13m pa, NIM 1bp. This benefit will be hard to track owing to the fact that Treasury's activity will be aligned to managing the benchmark BBSW rate set.

777    ASIC says that the White Paper envisaged that if CBA Treasury took charge of trading in the Bank Bill Market then contrary defensive rate set management, that is, trading by Markets to influence the rate set adversely to the interests of CBA Treasury, would be avoided and replaced by the same activity by CBA Treasury, but "aligned to managing the benchmark BBSW rate set" to Treasury's advantage. And Mr Carritt conceded that "managing the benchmark BBSW rate set" meant "moving the rate to your advantage in the Window". Thus, so ASIC says, the White Paper recommended moving CBA to a Westpac style structure (control by Treasury of trading in the Bank Bill Market) with an attendant advantage that trading to move the rate set could bring about a profitable reduction in wholesale funding costs.

778    Let me deal with these questions in more detail.

NAB structure

779    Westpac has contended that its structure during the relevant period may be contrasted with the structure of the NAB. It says that in particular, NAB's BBRM desk, within NAB Group Treasury was responsible for managing NAB's short term interest rate risk. In that respect, BBRM at NAB was the structural equivalent of STIRR at Westpac, save that BBRM was not responsible for NAB's short term funding and liquidity. But it says that an important difference was that BBRM was not able to trade in Prime Bank Bills.

780    It says that at the NAB, the short term interest rate trading desk (STIRT) was responsible for trading in short-term interest rate products, including Prime Bank Bills. STIRT was part of NAB's Products and Markets division, which was responsible for trading financial products with NAB's customers. NAB's Products and Markets division was the structural equivalent of Financial Markets at Westpac.

781    Westpac says that the consequence of NAB's structure was that STIRT was incentivised to trade in a way to maximise the profit made by its book, without regard to the interests of BBRM or NAB Group Treasury more generally.

782    It says that Mr David Page, a former employee of NAB, gave evidence that it was in the interests of NAB Group Treasury that the Bank Bill Market be free of manipulation given its importance to NAB as a source of short-term funding and liquid assets. But it says that STIRT did not have the same appreciation for the importance of the Bank Bill Market, and treated it as a market to be manipulated for the benefit of its own book. Mr Page identified this as at least a partial explanation for the way in which STIRT traded in the Bank Bill Market.

783    Westpac says that the consequence of those differences is that NAB's STIRT desk had different incentives and motivations when trading in the Bank Bill Market compared to the incentives and motivations for Westpac's traders in Group Treasury, who were not managing a swap book but managing the funding, liquidity and interest rate risk of the whole Westpac balance sheet. Westpac's Group Treasury traders' performance was measured on the basis of the overall group performance rather than individual portfolio performance, and matters such as ensuring that Westpac had the funding it needed and was protected from interest rate risk, market risk and credit risk as much as possible were the primary considerations. Relatively minor gains from interest rate exposure it is said would have been irrelevant and difficult to discern.

784    Now I accept that during the relevant period, NAB's organisational structure consisted relevantly of two separate divisions: Group Treasury and Products & Markets (formerly known as Global Markets). Group Treasury comprised Global Liquidity, the Short Term Funding Desk and BBRM. Group Treasury's BBSW rate set exposures were concentrated in the books of BBRM, which managed the net interest rate exposure of the bank. Within Products and Markets, STIRR relevantly also had BBSW rate set exposures.

785    But I agree with ASIC that Westpac's submission, that NAB's STIRT desk was incentivised to maximise the profits of its own book without regard to the interests of BBRM or Group Treasury, overstates the evidence. STIRT traded to ensure NAB was adequately funded and liquid and was the conduit for issuing NCDs into the interbank market for the Short Term Funding Desk. NAB's STIRT desk was incentivised to engage in trading for the purpose of influencing the yields of Prime Bank Bills and BBSW, because that produced profits for its book.

786    Further, both NAB and Westpac had "one face" to the Bank Bill Market. In both instances it was STIRT (or STIRR). And STIRT did not trade solely in respect of its own exposure without regard to Group Treasury. NAB's STIRT desk traded on instructions and in coordination with other desks and traded to create liquidity and ensure NAB was adequately funded.

787    Further, as ASIC points out, NAB's Liquidity and BBRM desks also directed trading to influence BBSW, which is inconsistent with Westpac's submission that a Treasury function will lack an incentive to engage in that kind of activity. The evidence of Mr Rotcer Brizuela, an employee of NAB, was that both Liquidity and BBRM, in coordination with Mr Page (Short Term Funding), directed NAB's Prime Bank Bill traders to influence BBSW for the benefit of their respective books.

CBA structure

788    Westpac also contends that its structure may be contrasted with the structure of the CBA.

789    Westpac says that like Westpac (and NAB), CBA had a structural separation between its Treasury and its Global Markets divisions. Unlike Westpac, however, for at least part of the relevant period, traders within CBA's Global Markets division were responsible for trading in Prime Bank Bills. As a consequence, from time to time conflicts arose between the interests of CBA Global Markets and CBA Treasury.

790    Westpac says that during the relevant period, Mr Carritt was involved in a process which sought to transfer responsibility for trading in Prime Bank Bills from CBA Global Markets to CBA Treasury. The purpose of that transfer was to avoid those conflicts and improve the ability of CBA to engage in Prime Bank Bill trading for liquidity, funding and risk management purposes, without the effect of that trading being neutralised by Global Markets' own trading. This can be seen on the part of CBA as a recognition of the advantages of Westpac's structure which involved Treasury undertaking Prime Bank Bill trading. Westpac says that that structure avoided conflicts of the types to which Mr Carritt referred in his evidence, and permitted trading in Prime Bank Bills to focus on funding, liquidity and management of interest rate risk.

791    Now in relation to CBA, I will keep my observations to a minimum given that ASIC has brought separate proceedings against it making similar allegations to those made against Westpac, and in respect of which I am the docket judge with those proceedings not finally disposed of as at the date of the publication of these reasons.

792    For a time, trading occurred in the Global Markets division, before a restructure transferred Prime Bank Bills trading to the CBA Treasury. CBA sought to have one face to the market, that being Mr Mark Hulme, to avoid competing desks trading against each other in the BBSW Rate Set Window.

793    Now Westpac submits that the earlier structure gave rise to conflicts between the interests of Global Markets and Treasury, and the revised structure permitted trading in Prime Bank Bills to focus on funding, liquidity and management of interest rate risk. But ASIC says that submission ignores the following matters. First, at CBA, trading to influence the rate was not confined to Global Markets. Traders within CBA Treasury also traded to influence the rate, although I accept that reliance was placed for such a proposition mostly on rumours that came to Mr Carritt's attention about Treasury traders acting to influence the rate. Second, conflicts were apparently resolved at CBA by the separate divisions co-ordinating with each other to resolve opposing positions on terms which enabled trading to the advantage of the division with a remaining exposure. For these and other matters I am inclined to agree with ASIC that there is little in the different structure of CBA whether before or after the restructure to assist Westpac.

SUSCEPTIBILITY OF THE BANK BILL MARKET AND BBSW TO MANIPULATION

794    Considerable evidence was led before me concerning whether and to what extent the Bank Bill Market was susceptible to manipulation, and accordingly whether and to what extent BBSW was susceptible to manipulation. It is convenient to deal with some relevant matters going to these questions at this point in my reasons. I would note though that in the parties' submissions from time to time the concepts were blended. But this is understandable given that the vast majority of trading of Prime Bank Bills occurred in the BBSW Rate Set Window. But where it is necessary to distinguish the two concepts I have done so, particularly where the question arises as to the effect or likely effect of Westpac's actual trading behaviour on BBSW itself.

795    Now it is relevant to note at the outset that Westpac has accepted that trading in the BBSW Rate Set Window could affect the BBSW.

(a)    The views of participants

796    As ASIC points out, the contemporaneous communications show that Westpac's traders thought that their trading may have affected BBSW. They included:

(a)    "yeah it's been really low. Thanks thanks to um Col" (this is part of a conversation that does not relate to a contravention date, but as the parties made broader submissions about it, I have addressed the conversation further later in these reasons) [Phone call transcript, Ms Johnston to Mr Sharma, 8 March 2010].

(b)    "your fingerprints all over it" [Chat log, Mr Conway to Mr Hosie, 1 July 2011].

(c)    "we got it back down" [Phone call transcript, Mr Hosie to Mr Roden, 1 July 2011].

(d)    "thanks to yours truly" (this is part of a conversation that does not relate to a contravention date, but as the parties made broader submissions about it, I have addressed the conversation further later in these reasons) [Phone call transcript, Ms Johnston to Mr Sharma, 10 August 2010].

(e)    "I pushed the one month down …" [Phone call transcript, Mr Roden to Ms Johnston, 6 April 2010].

(f)    "we got the one month down to 4.23 'cause we had a 13 or 14 yard rate set" [Phone call transcript, Mr Duignan to Ms Johnston, 7 April 2010].

(g)    "looks like you're keeping the one month down" and "you're a force to be reckoned with" [Phone call transcript, Ms Johnston to Mr Roden, 8 April 2010].

(h)    "I believe it could have got to the figure if we hadn't been on the bid" [Email, Ms Johnston to PRM traders, 4 March 2011].

797    ASIC has contended that where it is believed a market cannot in fact be manipulated, that market is less susceptible to manipulation by reason of that fact. But ASIC says that if participants in the industry believed Prime Banks could trade in the BBSW Rate Set Window to influence BBSW, it made such trading more likely and thus the market more susceptible to such manipulation. Generally speaking I would agree with ASIC's submission at the high level at which that submission is expressed.

798    It is apparent that Westpac's employees from Group Treasury and Financial Markets were of the opinion that trading during the BBSW Rate Set Window could affect the rate at which the BBSW set, that trading by Group Treasury during the BBSW Rate Set Window might have affected the rate at which the BBSW set and that the BBSW was susceptible to manipulation.

799    Mr Roden said that he was aware that there would be instances where someone trading in large volumes (or even small volumes) could affect the rate. He knew that Westpac's trading could have an impact on the market. He was conscious that STIRR's trading could have an impact on the level of BBSW. He traded Prime Bank Bills on Westpac's behalf with the knowledge that his trading could affect the BBSW. Indeed, he sought to trade and timed his trading so as to trade in a way that would not be to the disadvantage of Westpac.

800    Ms Johnston was aware that trading in Prime Bank Bills during the BBSW Rate Set Window could potentially affect the rate at which BBSW set. In cross-examination, Ms Johnston's evidence was that by her remark in a phone call with Mr Roden on 8 April 2010 to the effect that it looked like he had been keeping the 1 month [BBSW] down, she was probably observing the level at which 1 month bills were trading and, knowing that Westpac was generally a buyer in that tenor, enquiring whether Mr Roden had been buying.

801    Mr Hosie understood that it was possible that his trading during the BBSW Rate Set Window might impact the price at which Prime Bank Bills traded at 10.00 am, and therefore the submissions made by BBSW panellists and ultimately where BBSW set.

802    Mr Park acknowledged that he knew that buying a large volume of Prime Bank Bills in the Bank Bill Market could have a price impact, and that trading which had a price impact could affect that rate set position or rate set risk of Westpac.

803    Further, as ASIC has correctly pointed out, the documentary evidence reveals the following examples consistent with industry (and Westpac) participant perceptions that trading in the BBSW Rate Set Window could affect where BBSW set:

(a)    Mr Roden told Mr Parker (as I have said, then a Senior Dealer on the Short AUD Swaps Desk in Financial Markets, who did not give evidence for Westpac) on 5 June 2008 that "IF WE WERE NOT THERE IT [the BBSW] WOULD BE FFLAT TO PL US 20".

(b)    On 23 November 2009, Mr Joshua Masters (then a credit trader in Financial Markets) directed Mr Stokes to engage in "no funny business" because he had "a big rateset todasy".

(c)    On 8 March 2010, Ms Johnston said to Mr Sharma in a phone call that "- yeah it's been really low. Thanks thanks to um Col".

(d)    On 6 April 2010, Mr Roden said to Ms Johnston: "I pushed the 1 month down. It was gonna set at 30, right, and then I got it down to 23".

(e)    On 7 April 2010, Mr Duignan said in a phone call with Ms Johnston: "yesterday was really good 'cause we got the one month down to 4.23 'cause we had a 13 or 14 yard rate set".

(f)    On 8 April 2010, Mr Roden said in a phone call with Ms Johnston: "Yeah that's it, that was a cracking result for the 32 today I can't believe it" and "I could have got it to 30 but that's completely pointless doing that cause I would have had to spend money and what's the point of spending money right it's just ridiculous so … that's the story Johnston".

(g)    On 20 May 2010, Ms Johnston left Mr Roden a voice mail recording that "…and 1 month ticked up a point, um I tried to keep that down because we had a pretty decent rate set but there was quite a bit of resistance, ummm ok."

(h)    On 24 May 2010, Ms Johnston had the following telephone conversation with Mr Greg Ryan:

Ms Johnston:    Well, we just have to man… we um, so have you… do you k    now about BBSW?

Greg:        Yeah, yeah

Ms Johnston:    So that's obviously an international benchmark which a lot of different products get reset off so, um, so for example in the St George swaps tab um you can see like today St George has 13.8 billion resetting off the 1 month rate today…

Greg:        Yep

Ms Johnston:     so, um, in our team here we manage BBSW and the rate set, so we actually, um, [pause] we kind of manage and monitor where BBSW gets set

Greg:        Yep

Ms Johnston:    and if we've got a large rate set and it's interest, in the bank's interests to, umm, try and decrease that rate or something,

Greg:        Oh OK, yeah

Ms Johnston:    we kind of just oversee how that gets set …

Greg:        Yeah

Ms Johnston:     as well as a number of other banks.

(i)    Ms Johnston took credit for the 1 month rate being "really low" by responding to Mr Sharma in a phone call on 10 August 2010: "thanks to yours truly".

(j)    On 1 July 2011, Mr Roden told Mr Hosie that Mr Michael Dodds (a trader from ANZ) that "Doddy got run over which is a good thing."

(k)    On 1 July 2011, in reference to how low the 1 month was, Mr Hosie said to Mr Roden in a phone call that, "Yeah mate, we got it back down".

(l)    On 30 January 2012, Mr Roden provided Mr Whitfield with a "market update" by email stating that they were "[s]pending money trying to get the BBSW rate set as low as possible in an attempt to reduce our cash bucket negative carry."

804    Mr Zuber agreed that the BBSW was susceptible to manipulation. He gave evidence that someone could set out to trade Prime Bank Bills in the Bank Bill Market during the BBSW Rate Set Window to manipulate the BBSW and could potentially be successful in doing so.

805    Further, Mr Roden suspected UBS of possibly trading with the purpose to move the rate, and complained to AFMA about its conduct and other banks who were "doing the same thing", although in his cross-examination he sought to attribute his comment to a different context concerning short-selling although I am not sure that this diminishes the force of ASIC's point in this context. Mr Roden explained:

And those complaints you made to AFMA [regarding UBS] were in relation to buying and selling for the sole or the dominant purpose of moving the rate; that's right, isn't it? No. It was about the process of short-selling.

You mean UBS short-selling, for example; buying at the end of one day and then selling into the set the next day? Yes.

But the purpose of that selling being to move the rate? Possibly, but again, I don't know their it struck me as an unusual it was it is difficult to find a proper commercial rationale for someone doing that.

806    Further, as I have said, Ms Johnston was aware that people in the industry occasionally speculated that participants in the BBSW Rate Set Window were trading to affect the rate at which BBSW set and from time to time she may have thought other banks' trading was suspicious.

807    Further, Mr Conway speculated that CBA used a "cover" to achieve its objective of moving the rate up.

808    Further, as ASIC pointed out, other communications show that Westpac knew that the BBSW was capable of being manipulated:

(a)    Mr Roden told Mr Zuber in an email on 16 February 2009 that "having a derivative contract linked to BBSW" was "likely to exacerbate" the problem of "blatant manipulation".

(b)    Mr Conway in an email to other Westpac employees on 21 September 2009 said "BBSW rate set was jammed lower today".

(c)    Mr James Land of Financial Markets told Mr Bill Evans (WIB) that the "BBSW function is broken and has been for sometime. This view is strengthened by the random volatility that occurs each day as players try and jam the rateset" [Email, Mr Land to Mr Evans, 17 November 2009].

(d)    Mr Park referred to the market for betting on the State of Origin as "so small, it's even smaller than the bloody bill market... everyone knew what was going on", to which Mr Sharma responded: "it was like a 3 month BBSW rate set" [Phone call transcript, Mr Duignan, Mr Park and Mr Sharma, 29 September 2010].

(e)    Mr Parker observed that "fears of a ramp into bbsw today" had been "confirmed" [Chat log, Mr Parker and other Westpac officers, 8/9 December 2010].

(f)    Mr Parker wrote on another occasion in an email on 7 July 2011 sent to Mr Masnick (and Mr Greg Burke): "3mth bills set high yesterday @4.98% someone ramping the set … sellers in 10.03am pusshing though 4.96 …".

(g)    Mr Conway, when asked in a conversation with Mr Lee on 24 November 2011 if he had "even done col/sophie's gig" said: "nope … it is open for ramping". Mr Zuber agreed in cross-examination that "ramping" was language which Westpac employees should not use.

809    Further, as ASIC also correctly points out, the evidence of other experienced industry participants was also to the effect that the Bank Bill Market and BBSW were susceptible to manipulation.

810    The evidence of NAB former and current employees, Mr Page, NAB's Head of Cash Desk, Mr Brizuela of NAB's BBRM and Mr Brock Johnson, former Bank Bill Market trader, was that NAB's Bank Bill Market traders sometimes traded for the purpose of influencing the level at which the BBSW was set in a way that was favourable to their BBSW rate set exposure. NAB's business records similarly record that NAB's Bank Bill Market traders sometimes traded for that purpose and with success.

811    Mr Page commenced employment with NAB in 1994 on NAB's STIRT desk. In 1999 he moved to the Short Term Funding Desk as its Senior Dealer and was the Head of the Short Term Funding Desk. Based on his market experience of over 20 years, his evidence was that it was possible for the BBSW to be deliberately manipulated by market participants. He believed that NAB traders attempted to deliberately manipulate the BBSW and that traders from the STIRT desk at NAB viewed the BBSW rate set as a "game" of who had the biggest "bullets", by which he meant stock to buy or sell to get the BBSW rate set to where they wanted it to be. To Mr Page's knowledge, the rate setting practice persisted after 2008 and he continued to facilitate it.

812    Mr Brizuela joined NAB in 1998. During the relevant period he was a Manager in BBRM, had received a dealer's accreditation from AFMA. He knew that STIRT at NAB sometimes sought to manage its rate set exposure by trading to influence the BBSW, based on his observations of its trading and on his conversations with its traders and Mr McVicar.

813    Mr Johnson started at NAB in 2004 as an analyst in Market Risk and moved to STIRT where he worked as a trader. In 2009 he moved to the Cross Currency desk. In 2013 he left the bank to join a corporate debt consultancy firm. Whilst with STIRT, he observed the practice of STIRT traders seeking to move the BBSW by either selling or buying significant volumes of Prime Bank Bills in the BBSW Rate Set Window.

814    Further, the evidence of Mr Carritt concerning CBA's practices including the documents put to him in cross-examination, tended to suggest that there were perceptions among some traders and employees at CBA that trading with a purpose of affecting BBSW might be achieved. Reference was made to opinions contained inter-alia in the following CBA's business records. Mr Roman Groblicki expressed the opinion that "buying and selling bills/CD's by them [CBA's Swap's desk] is purely to affect BBSW rate settings" [Email, Mr Groblicki to Mr John Pilkington (both CBA), 11 May 2012]. The White Paper (p 17) referred to CBA's Markets division having the ability to influence the BBSW higher to suit their position. It was stated that: "Markets is able to influence a higher BBSW rate set to suit their position while this is often contrary to the Group's balance sheet exposure".

815    Further, in cross-examination Mr Carritt also appeared to accept the following:

(a)    People more senior than Mr Carritt within CBA told him that traders in the Markets division of CBA traded Prime Bank Bills with the purpose of moving the rate.

(b)    Mr Carritt's statement in an email to Ms Lyn Cobley (CBA Treasurer) that "[a]ll you can say is that GM will influence to the extent it suits GM" was his description of the behaviour of Global Markets, from time to time, to influence the rate sets and was a statement of purpose [Email, Mr Carritt to Ms Cobley, 24 February 2011].

(c)    Mr Groblicki in his email of 11 May 2012 to Mr Pilkington of CBA in which Mr Groblicki expresses the opinion that "buying and selling bills/CD's by them [CBA's Swap's desk] is purely to affect BBSW rate settings" appears to have assumed that CBA's swaps desk had an interest in trading to affect the BBSW rate.

(d)    He thought potentially that the swaps team could move the rate.

(e)    In his view, jamming, ramming or ramping is a reference to perceived conduct of trading for the purpose of moving the rate set up or down.

(f)    It was possible that traders were seeking to affect BBSW.

816    Further, on the question of industry perceptions, the RBA published a report in September 2011 which recognised that to manipulate the BBSW as the LIBOR had potentially been manipulated, this would have to be achieved by "manipulat[ing] the traded market directly" [RBA Bulletin paper, September Quarter 2011].

(b)    Features of the Bank Bill Market rendering it susceptible to manipulation

817    I agree with ASIC that there were a number of objective characteristics of the Bank Bill Market which made it susceptible to manipulation and potentially BBSW being susceptible to manipulation. On this aspect I accept the evidence of Professor Talis Putnins, an expert for ASIC (I will address his background later), who identified the following five principal features that caused the Bank Bill Market to be susceptible to manipulation:

(a)    incentives by derivative exposures based on BBSW;

(b)    BBSW being determined from a single non-random point in time;

(c)    concentration of participants in the Bank Bill Market;

(d)    barriers to entry; and

(e)    position limits.

Incentives from exposure to BBSW Referenced Products

818    Professor Putnins observed that the phenomenon of a considerable volume of products and contracts that derive their value from the BBSW (or the yields of Prime Bank Bills) constituted a significant incentive to market participants to manipulate the Bank Bill Market and the BBSW. The evidence led before me confirmed that the BBSW Rate Set Exposure of the traders' book was significant to the traders' management of their portfolio of assets and liabilities.

819    Further, Mr Simon Bishop, an expert for Westpac (I will address his background later), did not contest that profit might be gained from trading which had the effect of improving the BBSW set. Now he sought to demonstrate that costs would outweigh the gains, but he accepted that the gains may be greater than his calculation if the effect of the trading was to move the BBSW by more than 3 basis points. Moreover, after being subjected to a close and effective cross-examination on this aspect by Mr Peter Collinson QC for ASIC, it was well apparent that each of the costs (other than the more expensive trading in the BBSW Rate Set Window) Mr Bishop referred to was unsubstantiated or insupportable.

820    Further, Mr Zuber accepted that, all else being equal (a critical assumption), Westpac as a rational market participant would seek to acquire or sell Prime Bank Bills on days where, if the level of Westpac's trading affected the yield at which the Prime Bank Bills were trading, any effect on that yield would not also adversely affect Group Treasury's BBSW Rate Set Exposure. Accordingly, as a matter of (as he described it) "prudent interest rate risk management", if Group Treasury had a short exposure or a paid set, Westpac would not take a step which could potentially put upward pressure on rates. Now Mr Bishop accepted that an economically rational actor with a short rate set exposure would not merely undertake the passive activity of avoiding issuing or selling Prime Bank Bills, but would go further and trade to move the rate set downwards assuming capacity to do so and an available profit from such activity. But this of course assumes that all else is equal and further that it was lawful to do so.

821    Further, Westpac accepted that it would benefit from any movement in the BBSW in the direction that suited its BBSW Rate Set Exposure on that day. That is, it would benefit from the BBSW setting lower when it had a short exposure or payset, and would benefit from a higher BBSW when it had a long exposure or receive set.

822    Further, AFMA also recognised that "BBSW panellists must manage a conflict of interest in situations where they … have a derivatives position that may be dependent on BBSW" [AFMA Board briefing paper, 24 April 2012]. This may provide a motivation that might induce a person to attempt to manipulate the price at which securities trade. Manipulation of the physical market can facilitate profiting in the futures market. Ashley Black, "Regulating Market Manipulation: sections 997-999 of the Corporations Law" (1996) 70 Australian Law Journal 987 cites the decision of US v Regan, 937 F 2d 823 (2d Cir, 1991) where underwriters to the issue of convertible bonds (convertible to common stock at a fixed price) arranged for the sale of shares for the purpose of reducing the market price of those shares, which in turn required bonds to be issued with a higher interest yield in order to place them with investors.

823    Further, Westpac took its BBSW Rate Set Exposure into account when trading in the Bank Bill Market.

824    Now I accept that all of these matters are probative of the ability of trading in the Bank Bill Market to affect BBSW and its susceptibility to manipulation.

Fixed time for trading to determine reference rate

825    Professor Putnins identified that a specific non-random time for specifying the trading which is used as the basis for determining a key reference rate is a feature rendering a market susceptible to manipulation. He compared the use of trading at 10.00 am in the Bank Bill Market to stock exchange prices at the close of trading, and the notoriety of such circumstances creating a likelihood of manipulation. I tend to agree.

826    Now it was suggested by Mr Bishop, and put to Professor Putnins in cross-examination, that the force of the general point was diminished due to uncertainties in the effect of trading at the critical time, but that was rejected by Professor Putnins. Moreover, Mr Bishop accepted that he did not have sufficient knowledge on the point.

827    Now I agree that there were uncertainties in the effect of trading at the critical time, although I do not consider that such uncertainties diminish the general force of Professor Putnins' point that to fix a time for trading to determine the BBSW made the BBSW more susceptible to manipulation.

828    I, of course, am not here commenting on the effect or likely effect of specific bids, offers or trades.

Concentration of market participants, barriers to entry and liquidity

829    The Bank Bill Market had a small number of participants, of which only Prime Banks had the capacity to issue NCDs, which also in my view made the Bank Bill Market more susceptible to manipulation.

830    AFMA recognised the "lack of diversity in the Prime market," [NTI Committee meeting of 24 May 2010, draft minutes] the post-GFC reduction of the amount of "Prime Bank Paper on Issue" and the reduced "level of trading activity" in 2010-2011 [AFMA Board briefing paper, 24 April 2012].

831    Further, from early 2010, the Bank Bill Market suffered from reduced liquidity. Other traders in the market observed that in about 2010, there was reduced liquidity in the market and that this made it easier as Mr Brizuela said for "banks to push the rate".

832    Further, Westpac's witnesses also recognised that the Bank Bill Market had limited liquidity, particularly outside of the BBSW Rate Set Window. In mid-2010, Ms Johnston knew that the Bank Bill Market barely started trading before 9.58 am. Even within the few minutes around 10.00 am, Westpac's traders recognised that there was no certainty as to what volumes would be traded on a particular day. Outside of the BBSW Rate Set Window, there was only some very limited liquidity at around 4.20 pm to 4.30 pm.

833    Further, Professor Putnins identified a key feature of the Bank Bill Market as its far greater concentration of participants. I agree with ASIC that the participant concentration where only a few buyers and sellers dominate the trading of Prime Bank Bills meant that those few traders may have an outsized influence over BBSW.

834    Now I accept that other market participants had an ability to engage in competitive responses which would temper a participant's efforts to affect prices. But as Professor Putnins explained, notwithstanding the resources or sophistication of other participants, they may potentially encounter problems with identifying and reacting to trading behaviours in the limited window before the determination of BBSW. The nature of the marketplace may not have enabled or may not have been conducive to self-correcting behaviour in time to avoid a participant trading to influence BBSW.

835    Further, Mr Conway's evidence was that "CBA is the biggest balance sheet in Australia, Westpac's the second biggest balance sheet in Australia" and that "when those two balance sheets enter the market to trade … their activities are quite likely to be the ones that are moving around the rate".

836    Further, as Ms Johnston confirmed in a telephone conversation with a representative of the RBA on 9 June 2010, the Bank Bill Market was "much smaller … than the US", and that "BBSW here, it gets affected very much … here if the banks go and do, you know, a couple of billion dollars of issuance into these short term markets it just clogs up straight away".

837    Further, Mr Paul Howarth, a NAB trader, referred to other participants being "suffocated by Westpac stock" [Email, Mr Howarth to NAB officers, 9 March 2012]. Mr Rob Collins of NAB observed that, in respect of 1 month bills, "Westpac bal[ance] sheet take[s] it all" [Chat log, Mr Collins to Mr Corbett (Credit Suisse), 18 July 2011].

838    Further, traders in the market also knew that the Prime Banks had sufficiently large balance sheets to influence BBSW should they wish to do so.

839    Generally I accept that the Bank Bill Market is a highly concentrated market by financial market standards, and that such concentration is itself a form of illiquidity. The lack of breadth and depth in the market may have resulted in the trading decisions of each participant being more likely to affect prices than in a less concentrated market. But I do not consider this observation to be anything other than self-evident.

Position limits

840    Professor Putnins identified certain position limits (such as credit and interest rate risk limits) which could, if a participant was unprepared, inhibit the ability of market participants to react to manipulative conduct and temporary price impacts.

841    Mr Bishop accepted the proposition in principle, but considered (for the purpose of ascertaining the circumstances on respective contravention dates) that it would be necessary to ascertain empirically whether in fact other market participants were constrained by their limit positions from acting. I tend to agree.

General

842    Let me deal with some other matters.

843    First, as Professor Putnins also explained, the mechanism by which BBSW was set during the relevant period was a relatively crude manual system, which made the Bank Bill Market more susceptible to manipulation. An electronic market would have allowed participants to respond immediately.

844    Second, there is the reputational question. A bank's reputation for trading in a particular way could affect the way in which other participants in the market traded, making it more likely that trading could influence or manipulate.

845    Indeed, ASIC has submitted that there is relevant evidence of Westpac's reputation for trading in the Bank Bill Market. Some examples from the evidence include the following:

(a)    Prior to the relevant period, Mr Stephen Reid of NAB wrote "in a chat with David Herriot from Citibank ages ago, he has showed he has had rate setting to the tune of 2 bio, that went against him when WBC pushed the rate down & he couldnt issue to offset it" [Email Mr Reid to Mr Mike Krohn (NAB), 9 March 2005].

(b)    Mr Greg Betts (ABN AMRO) told Mr Roden: "okay BBSW …let you go do some ramping …" [Chat log, Mr Betts to Mr Roden, 23 September 2008].

(c)    Mr Webster of Bank of Scotland saying "if he [Mr Roden] wants them lower all i can do is try to stop [the] wave [of] buying …. [I] cannot push it up as hes there for 10 y[ards] if needed then after 10 goes to acceptors and seells stock based on over limits" [Chat log, Mr Webster to Mr Paul Woodward (ANZ), 29 October 2009].

(d)    Mr Howarth of NAB wrote: "Bills/OIS is out quite significantly in the 3 month. At Mondays set it was in to 12 (!) on WBC ramping" [Email, Mr Howarth to NAB bank officers, 30 October 2009].

(e)    Mr Harper of Credit Suisse in a chat said "[I] have an axe to stop seeing bbsw getting manipulated" and "pass it on [to] the WBC" [Chat log, Credit Suisse and NAB officers, 4 March 2010].

(f)    Mr Hermeet Najjhur of NAB said, in a chat "someone has driven [d]own the rateset but think we will end up ghigher over next couple of days … west[p]ac being buying" [Chat Log, Mr Najjhur to Mr Imran Ismail (JP Morgan), 18 March 2010].

(g)    Mr Collins of NAB said, in response to a comment from Mr Sam Ellis of ANZ about "col": "its nasty when he has big ones against you", "he is the biggest by miles" and "so no one to keep him honest" [Chat log, Mr Collins to Mr Ellis, 25 June 2010].

(h)    On 4 August 2010, Mr Howarth in a chat said "there seems to be a fair bit of interest in the set tomorrow … interesting", to which Mr Corbett of Credit Suisse said "col still the main manipulator" [Chat log, Mr Howarth and Mr Corbett, 4 August 2010].

(i)    Mr Michael Tsakiris of NAB telling Mr McVicar (also of NAB) that "the biggest thing's Westpac, right. If they want to get involved then they can move it the most" and "There's a lot of stock out there but it's all about Westpac, if they want to step up … and do something" [Phone call transcript, Mr Tsakiris to Mr McVicar, 12 July 2011].

(j)    Mr McBride of NAB saying "looked like nab pushing it higher" and "[W]estpac and a few other cronies trying to downset it" [Chat log, NAB bank officers, 9 September 2011].

(k)    Mr Najjhur of NAB wrote on another chat "he reckons west pac going to drive 1 month lower … so to confirm we have 2 yards of one month", "sold 240 at 39 to Westpac 1 month through tullets" and "220 1 month to Westpac through icap" [Chat log, Mr Najjhur to Mr Tsakiris, 19 December 2011].

(l)    Mr Hall of NAB: "wbc selling today. Puck it makes me laugh" to which Mr Page responded "yup they have a lot of ammo" [Chat log, Mr Hall and Mr Page, 7 June 2012].

846    Mr Lee of CBA said in a chat after trading on 9 December 2010: "the rateset today was amazing…. Roden just played that beautifully" [Chat log, Mr Lee with NAB bank officers, 9 December 2010].

847    Further, when other banks believed that Westpac "can move it the most" [Phone call transcript, Mr Tsakiris and Mr McVicar (both NAB), 12 July 2011], and that there was little they could do if Mr Roden "wants them [the 1m] lower" [Chat log, Mr Webster and Mr Woodward, 29 October 2009], the consequence may have been that they got out of Westpac's way.

848    I accept that reputation could facilitate trading that could influence or manipulate.

849    Third, ASIC has submitted that Westpac, as an AFMA panellist, made submissions on each trading day which, together with the submissions of other panellists, determined the rate at which BBSW set on that day. In this way, so ASIC submits, Westpac had the ability to "deliberately contribute (or omit) data to the sample, gaining strategic influence over the result" by making or omitting to make "additional trades of whatever sort the benchmark tabulates" (Verstein A, "Benchmark Manipulation" (2015) 56 Boston College Law Review 215 at 231 and 232).

850    Further, ASIC says that the potential for manipulation of the BBSW in respect of the submissions process was recognised and recorded by AFMA as a concern as to whether the submissions accurately reflected the rate at which Prime Bank Bills were trading at 10.00 am on any given day ("the potential for panellists to be able to influence rate-sets to their advantage") [NTI Committee meeting of 26 May 2010, draft minutes]. AFMA also recognised "a separate question about the reliability of those traded rates as a fair reflection of underlying market conditions, such that they are not impacted by attempts by any trader to influence rate outcomes" [AFMA Board briefing paper, 24 April 2012].

851    Further, ASIC says that on the pleaded contravention dates Westpac generally made the most favourable submissions to AFMA based on its BBSW Rate Set Exposure for that day. It has submitted the following

(a)    For each of the sale contravention dates:

(i)    Westpac submitted the highest rate within the range submitted on that day in that tenor; and

(ii)    Westpac's submission was higher than the rate at which BBSW set on each day.

(b)    For eight of the purchase contravention dates (6 April 2010 (in the case of the 1 month), 20 May 2010, 6 December 2010, 1 March 2011, 4 March 2011, 1 June 2011, 6 June 2011 and 6 June 2012) Westpac submitted the lowest rate in the range submitted on that day in that tenor;

(c)    For seven of the purchase contravention dates (30 April 2010, 20 May 2010, 6 December 2010, 1 March 2011, 4 March 2011, 1 June 2011 and 6 June 2012) Westpac's submission was lower than the rate at which BBSW set on each day; and

(d)    For only one purchase contravention date did Westpac submit the higher of the two submitted rates (22 September 2010).

852    I must say at this point that even accepting these statistics, ASIC never ran or pleaded a case questioning Westpac's behaviour concerning its submissions to AFMA as a panellist or any suggestion that such submissions were distorted or were not relevantly Westpac's genuine best estimates as required. Accordingly I have put to one side ASIC's "most favourable submissions" thesis. But of course Westpac's conduct in the BBSW Rate Set Window could affect the pool of potential inputs used to form the estimates upon which BBSW was then calculated.

853    Finally, I would note the following. The minutes of AFMA's BBSW Committee meeting on 26 May 2010 attended by Ms Johnston for Westpac notes the following:

NTI Committee consensus was that there are many inter-related and contributing factors, including:

    The Early/Late periods in which Bills are traded plays a large part in the volatility, ie: early purchased stock becomes illiquid after mid-month, prompting selling prior to mid and end-month to avoid holding illiquid stock;

    A lack of diversity in the Prime market, which has five participants, down from eight in prior years;

    The contraction in issuance of Australian denominated short term securities by the financial services industry relative to issuance levels pre-GFC, as evidenced by RBA statistics;

    The decrease in credit limits and increased restrictions on balance sheets since the GFC this has offset the impact of the decrease in issuance;

    The materially larger size of the derivatives markets dependent on BBSW rate-sets within their portfolios as compared to the size of the physical Bills market;

    The strong appetite for stock from the funds/investment industry, historical predisposition of the Australian government/semi government and corporate entities to use mid and end-month dates relative to their funding and market / interest rate sensitivity profiles;

    The potential for panellists to be able to influence rate-sets to their advantage.

854    But it may be queried whether the last point relates to the making of submissions or the anterior conduct in the BBSW Rate Set Window or both.

855    Further, the minutes of AFMA's NTI Committee meeting on 8 February 2011 (at which Ms Johnston was chairperson) records the following under the heading "Market liquidity and conditions Market Integrity":

Members noted that Bank Bills liquidity remains less than ideal, with only a couple of prime banks regularly issuing stock into the 10:00am rate-set. Members agreed that this is a reality of the diversified funding profile of the majors, and their reduced reliance on the Australian short term debt markets.

Members noted there are still the occasional instances of delays in broker screen price updates around the 10:00am rate-set, recognising that the concentrated activity is still overwhelming manual screen input and impacting the flow of information.

Members noted that overselling into the 10:00am rate-set has diminished, however has not been fully eliminated.

The committee will continue to monitor these issues.

(c)    Conclusion

856    In summary, and based upon the foregoing, in my view the Bank Bill Market and the BBSW were susceptible to manipulation from conduct during the BBSW Rate Set Window. But the degree of susceptibility depended upon the conditions and activity of the participants trading on a particular day, including the volume of Prime Bank Bills being traded in each tenor, the number of participants and their interest. It is not feasible to engage in generalisations concerning the degree of susceptibility save to say that I would reject the suggestion that the Bank Bill Market and BBSW were very susceptible to manipulation. I would also note that the variability of the degree of susceptibility was not only trading day dependent, but also a matter that may not have been able to be predicted by participants in advance of the BBSW Rate Set Window on many if not most of the trading days.

RELIABILITY OF WESTPAC'S WITNESSES

857    Before turning to the specific contravention dates, it is appropriate to say something concerning the reliability of Westpac's witnesses and ASIC's submissions in that regard. ASIC exemplified its points by reference to specific communications and excerpts of cross-examination from which I was encouraged to draw more general conclusions on reliability. But I have refrained at this point from reaching any conclusions about the evidence of a witness on a specific communication relating to a contravention date. I have instead at this point restricted myself to assessing the reliability of a witness in more general terms, although informed by the totality of the evidence including that led in relation to the contravention dates.

(a)    Mr Zuber

858    I found Mr Zuber to be a very experienced and talented banker, whose evidence was helpful and reliable. Nevertheless, ASIC has made a number of criticisms of his evidence that it is necessary to address.

859    First, it says that his analysis of each alleged contravention date was carefully constructed and amounted to nothing more than a reconstruction of possible reasons for trading on those days. But Mr Zuber did not suggest otherwise. Indeed, he accepted that looking at trading on a particular day after the event, you could only "back-solve legitimate commercial reasons" as to why the trader traded as he did. Mr Zuber described his task in the following terms:

I came up with what I thought to be some reasons of what he was doing and how he conducted himself and I saw nothing in any of those trades that gave me any implications that there was even an instance where his activity was to manipulate the rate. It was all commercial.

860    Second and relatedly, ASIC says that Mr Zuber's evidence was also formulaic so far as the contravention dates were concerned. But Mr Zuber explained that approach in the following justifiable terms: "in a number of the cases, we were buying bank bills, so it's not unusual that as I constructed my position together, I'm going to the same five or six things that we would be doing on various days in how we operated". Now ASIC says that the formulaic language which has been deployed tends to make the point that on any occasion of buying or selling Prime Bank Bills it could invariably be characterised as consistent with one or more Group Treasury functions, but that what motivated trading on a particular day as opposed to any other day is what was important. Now so much is true, but that is hardly a criticism of the reliability of Mr Zuber's evidence.

861    Third, ASIC says that Mr Zuber's evidence is of little assistance in considering the contravention dates because he did not refer to the communications upon which ASIC relies. ASIC says that his failure to do so suggests that he never properly investigated the purposes for trading of the traders. ASIC says that he proceeded on the assumption that STIRR was not manipulating the BBSW.

862    Fourth, ASIC says that Mr Zuber had limited experience in trading Prime Bank Bills. He only traded Prime Bank Bills in the late 1990s into the early 2000s through the then markets desk. He left day-to-day decision-making about trading to the individual traders.

863    But generally speaking, ASIC had to accept that save in one respect, Mr Zuber otherwise gave straightforward evidence doing the best that he could with the task he had taken upon himself. ASIC also accepted that Mr Zuber made appropriate concessions. But it says that the one exception was his evidence about Group Treasury's practice of trading on days when any impact of its trading on yields and the BBSW would not be disadvantageous having regard to its BBSW Rate Set Exposure. ASIC says that Mr Zuber revealed a reluctance to accept the corollary, namely, that Group Treasury traded on days when any impact of its trading on yields and the BBSW would be advantageous having regard to its BBSW Rate Set Exposure. ASIC says that he begrudgingly accepted this, albeit in qualified terms. ASIC says that this demonstrated a consciousness on his part to protect the interests of Westpac.

864    Now as to these criticisms I would observe the following.

865    First, it is too simplistic to suggest that Mr Zuber's evidence concerning the contravention dates should be discounted because it is no more than reconstruction. Mr Zuber is and was throughout the relevant period the Group Treasurer of Westpac with overall responsibility and oversight for Group Treasury's operations. He personally was the manager of the STIRR desk prior to Mr Roden. He gave cogent evidence of the objectives of Group Treasury, how buying and selling Prime Bank Bills furthered those objectives and, applying that framework to the particular contravention dates, identified commercial reasons for which Group Treasury bought or sold Prime Bank Bills which were consistent with the trading on each contravention date. I agree with Westpac that such evidence is not pure speculation. It is evidence of corporate practices which probatively informed one of the principal issues for determination.

866    Second, as to the complaint made that Mr Zuber's evidence was formulaic because for many contravention dates he gave the same potential explanations for the trading, as Westpac correctly submits, that Westpac's trading on the majority of contravention dates may be largely for the same reasons for which Group Treasury usually trades Prime Bank Bills is hardly surprising or a basis for criticism in terms of Mr Zuber's reliability. It merely indicates that the trading may be consistent with the pursuit of legitimate trading strategies.

867    Third, as to the suggestion that Mr Zuber's evidence is of little assistance because he did not refer to the contemporaneous communications on which ASIC relies, I agree with ASIC that this lessens the force of Mr Zuber's evidence. But in fairness, it was not Mr Zuber's role to opine on the actual subjective purposes of Mr Roden and Ms Johnston. It would have been inappropriate for Mr Zuber to express opinions on subjective purpose based upon his interpretation of communications to which he was not a party. His evidence was properly limited to identifying the reasons for which Group Treasury normally traded Prime Bank Bills and whether the objective circumstances of the impugned trading were consistent with such reasons. Such evidence is relevant to my task of determining whether Westpac traded on the contravention dates for legitimate commercial reasons or for the purpose of manipulating the rate or a combination of both.

868    Fourth, as to the suggestion that Mr Zuber had limited experience in trading Prime Bank Bills, Mr Zuber did have a prior role as Manager of the STIRR desk. But in any event the criticism could not detract from the strength of the evidence he gave in relation to the way in which buying and selling Prime Bank Bills furthers the principal or significant objectives of Group Treasury.

869    Fifth, as to the suggestion that Mr Zuber only begrudgingly accepted in qualified terms that Westpac traded on days when any price impact of its trading would have a positive effect on its rate set position, the transcript references referred to by ASIC do not fully support such a characterisation, although there is some force in ASIC's point.

(b)    Mr Roden

870    ASIC has challenged the reliability of Mr Roden's evidence. But I must say that for the most part I found his evidence to be reliable. He is a highly skilled individual who for the most part gave helpful and informative evidence. But there were aspects of his evidence that I found to be problematic in relation to some of the contravention dates which I will discuss in more detail later. But the fact that I have not accepted some aspects of his evidence is no basis for a wholesale rejection of his evidence.

871    Let me now deal with ASIC's criticisms.

872    First, it is said that aspects of his first affidavit dealing with the contravention dates were detailed, but formulaic. It says that he accepted in his first affidavit that he could not remember the trading on each date, or the communications upon which ASIC relies, and accordingly his opinions were expressed "at no higher a level than supposed consistency of the trading with particular objectives".

873    Second, ASIC says that it is significant that he omitted to note that buying or selling Prime Bank Bills on a particular day when Group Treasury had a short or long BBSW Rate Set Exposure respectively would have a potentially beneficial effect on the movement of the BBSW, which movement would be consistent with one of Group Treasury's core objectives, namely profit-making.

874    Third, ASIC says that under sustained cross-examination by Mr Philip Crutchfield QC over a number of days, the quality of his evidence fluctuated. It is said that he was at times unresponsive and evasive. For example, ASIC says that Mr Roden became much more firm about his reasons for trading (in particular on 6 April 2010) during cross-examination and as it progressed, compared to the uncertainty expressed in his first affidavit. ASIC says that the true position must be that he could not remember and therefore did not know why it is that he traded on the relevant contravention dates, including 6 April 2010. ASIC says that the fact that his evidence became increasingly firmer was a tell-tale sign of Mr Roden taking up a combative position in defence of his and his employer's interests.

875    Fourth, ASIC says that Mr Roden often sought to inappropriately downplay contemporaneous communications which were harmful to his and Westpac's position by fobbing them off as mere banter, humour or colour.

876    Fifth, ASIC says that Mr Roden also sought to downplay other communications by referring to people that he was communicating with as more junior or as a mere irritant. It says that the course of communications between Mr Roden and Mr Lee do not bear out Mr Roden's attempt to distance himself from Mr Lee. And as for Mr Roden's regard for Ms Johnston, ASIC says that the calls between them do not support his downplaying of her status.

877    Sixth, ASIC says that Mr Roden could not offer credible explanations for communications harmful to his and his employer's interests. It is said that his refusal to accept the natural and ordinary meaning of some of the communications is revealing and damaging to his reliability. Further, ASIC says that Mr Roden had to resort, at times, to the notion that an apparently benign explanation such as hedging was implied or that it went without saying because it was understood by those involved in the communication. It says that this stance was not credible.

878    Now some of ASIC's criticisms have force but others do not.

879    As to the weight of Mr Roden's evidence being reduced for being no more than reconstruction that rose no higher than consistency with Group Treasury's objectives because he could not recall the trading on particular contravention dates, there is some force to ASIC's point. But as the head trader on the STIRR desk he gave detailed and probative evidence of the reasons for which STIRR generally traded in Prime Bank Bills and then analysed the available data of the objective circumstances of STIRR's trading on each contravention date by reference to those reasons. His evidence directly bears upon the primary issue, but of course it has its limitations in relation to his reconstruction of the purposes for trading on the contravention dates.

880    Further, as to the suggestion that Mr Roden became firmer about his reasons for trading, particularly on 6 April 2010, as his cross-examination progressed, Mr Roden did not claim to have a specific recollection of the trading on each day, but he did say that he had a general recollection of market conditions from around the time and the way in which the market was operating. That position did not change during the course of his evidence and he did not claim in cross-examination for the first time to recall particular trading. I would also note that if Mr Roden did become firmer in terms of his answers or demeanour, it was more a reflexive stance to the length and combative nature of the cross-examination; sustained and occasionally repetitive cross-examination is not unknown to produce such a reaction from a witness.

881    Further, as to Mr Roden being criticised for downplaying some communications as banter, humour or colour, I agree with Westpac that various of the communications on which ASIC has relied are of that character. Westpac says that ASIC's criticism in truth is that Mr Roden did not accept that what made them humorous was the existence of the Rate Set Trading Practice. But it says that it can hardly be a valid criticism of Mr Roden's evidence, which denied the existence of the Rate Set Trading Practice, that he sought to downplay communications as humorous without accepting that they were joking about the existence of such a practice. I agree with Westpac that the criticism proceeds from the assumption that Westpac was manipulating BBSW when that is the matter in issue to which Mr Roden's evidence was directed.

882    Further, as to the suggestion that Mr Roden sought to downplay his communications with Ms Johnston because of her more junior status, and with Mr Lee of the CBA because he was simply a "serial pest", several points can be made. To the extent that Mr Roden made reference to Ms Johnston as a junior employee, it was in response to being asked to speculate about her state of mind or knowledge. As for Mr Roden's characterisation of the nature of his relationship with Mr Lee, I agree with Westpac that to some extent it was borne out by the content of the communications themselves.

883    Further, as to Mr Roden being criticised for refusing to accept the natural and ordinary meaning of the communications on which ASIC has relied and that his explanations "beggar belief", as Westpac says, this amounts to a criticism that Mr Roden disagreed with ASIC's interpretation of the communications and his rejection of ASIC's case theory that he was trading to manipulate BBSW. Now I have not accepted some aspects of Mr Roden's evidence as I will later explain. But in my opinion, for the most part Mr Roden gave credible and responsive answers. He also made concessions where called for, including that he probably should not have said what he said to Mr Parker in certain chats having regard to the information barrier between Group Treasury and Financial Markets; see for example his telephone conversation with Mr Parker on 24 May 2010.

(c)    Ms Johnston

884    ASIC has also challenged the reliability of Ms Johnston's evidence, but again I found her for the most part to be a reliable witness. She is a very proficient and intelligent individual. However, like Mr Roden, there were aspects of her evidence that I found to be problematic in relation to some of the contravention dates which I will discuss in more detail later.

885    ASIC made the following submissions.

886    First, ASIC makes the point, which I accept, that Ms Johnston did not recall any of the events or communications the subject of these proceedings. She did not recall trading on any of the relevant contravention dates or her purpose for trading on any of those dates.

887    Second, ASIC says that the evidence bears out that Ms Johnston followed the directions of Mr Roden. And it is said that accordingly Mr Roden must accept the responsibility for the way he trained and directed "young and impressionable" people. But that description frankly underplays Ms Johnston's considerable skill. In any event, none of that goes to the question of reliability

888    Third, ASIC says that it is unsurprising that Ms Johnston's trading style was not generally as aggressive as Mr Roden's. She was more often focused on trying to keep BBSW from moving against Westpac's interests rather than "dominating" the market. That may be so, but for the moment I am still dealing with the reliability of her evidence.

889    Fourth, ASIC says that Ms Johnston was seeking to give the most helpful evidence for both herself and Mr Roden that she could. It is said that she was well prepared to give her evidence. I might interpolate here that one would hope so. The fact that she was well prepared in giving her evidence should be a reason to compliment rather than to condemn. ASIC also says that it was apparent from her answers that she was at times trying to anticipate where a line of questioning might be heading. For example, when she was asked if she had received a black eye because she had spent too much on Prime Bank Bills without succeeding in keeping BBSW down, her answer was "I think the communication you're referring to I haven't seen in a very long time given I'm not a party to it". And when it was put to her that Financial Markets and Group Treasury should not have been sharing details of their expectations about trading in the Bank Bill Market, her response was "I'm not sure if you're going to take me to a specific communication."

890    Fifth, ASIC says that Ms Johnston was cautious in agreeing to propositions put to her which ought to have been accepted without qualification. It says that she qualified more than 100 of her answers with the word "potentially" or the expression "not necessarily".

891    Sixth, ASIC says that Ms Johnston failed to make appropriate concessions when, for example, inappropriate communications between her and Mr Parker were put to her.

892    Seventh, ASIC says that Ms Johnston also sought to distance herself from a more candid prior statement about Westpac's practices in relation to hedging, which was also made under affirmation and which was inconsistent with the evidence given in this proceeding. ASIC has referred to part of her s 19 examination where Ms Johnston said that she did not use Prime Bank Bills to hedge BBSW Rate Set Exposure risk, but did use them to hedge outright interest rate risk.

893    Generally, ASIC says that Ms Johnston's evidence must be considered in circumstances where she had no recollection of the facts in issue in these proceedings. It says that her evidence adds little to the contemporaneous records which are "clear on their face". Further, ASIC says that what was "most telling" about her evidence was that she could not proffer any reading, other than the natural reading contended for by ASIC, of some of the clearest statements describing her purpose for trading Prime Bank Bills.

894    Let me make the following points.

895    As to Ms Johnston being criticised for being unduly cautious and careful in her evidence, and for being well prepared to give her evidence, I agree with Westpac that this is not a balanced criticism. ASIC's allegations called into question her integrity and probity. She was entitled to be considered, careful and prepared in giving her evidence. Moreover, in some respects her reticence in cross-examination to agree in unqualified terms with propositions put to her reflected more the general terms in which such propositions were framed or, from her perspective, the implicit assumptions embedded in those propositions which she was not prepared to accept.

896    Further, as to the criticism that Ms Johnston failed to make appropriate concessions about the inappropriateness of her communications with Mr Parker, I agree with Westpac that this assumes those communications were inappropriate. But the communications for the most part between Ms Johnston and Mr Parker did not involve Ms Johnston disclosing specific Group Treasury positions or trading strategies.

897    As I have said, I found her evidence for the most part to be reliable, but there are some aspects that I have not accepted that I will elaborate on later. But I would say at this point that her evidence given as part of an earlier s 19 examination hardly instilled me with confidence as to her evidence before me concerning the question of using Prime Bank Bills to hedge BBSW Rate Set Exposure risk (in addition to their use in hedging outright interest rate risk).

(d)    Mr Sharma

898    ASIC has submitted that Mr Sharma did not give balanced evidence and that his evidence should be treated with caution. It is said that he would not accept the plain meaning of the communications put to him. And that more tellingly, he strove to give answers to questions that "toed the company line". Thus, ASIC points to the following examples:

(a)    He gave evidence that Ms Johnston's nickname was the "perfumed steamroller" only because she was a woman, but this did not explain the "steamroller" portion of the nickname.

(b)    He gave evidence of his discussion with Mr Roden on 8 June 2010 about f***ing the rate set on 10 June 2010, but said that it was a discussion about volatility and about different rate sets.

(c)    He gave evidence about a discussion on 17 November 2010 about avoiding opposite positions to Mr Roden in December 2010, which he said was a discussion about volatility only.

899    But as Westpac points out, the criticisms of Mr Sharma's evidence, which is said not to have been balanced and should be treated with caution, were simply based upon his unwillingness to accept ASIC's case as to the meaning of certain communications, namely, that the "steamroller" in Ms Johnston's nickname referred to her trading for an improper purpose, that Mr Roden was referring in their telephone call on 8 June 2010 to manipulating BBSW, and that the warning from Mr Park on 17 November 2010 to avoid the market on days in early December was because Mr Roden would be manipulating the rate. I agree with Westpac that the fact that Mr Sharma did not agree with ASIC's characterisation of these communications and the necessary consequences flowing from ASIC's interpretation, being that Mr Roden and Ms Johnston were manipulating BBSW, is no basis of itself to discount his evidence.

900    Generally I found Mr Sharma to be a reliable witness save for some aspects of communications dealing with specific contravention dates that I will discuss later, including in relation to the 17 November 2010 phone call.

(e)    Mr Park

901    ASIC has made various criticisms of Mr Park's evidence.

902    First, it is said that in cross-examination Mr Park was careful to disclaim any ability to venture a view on the reasons for UBS trading on a particular day, yet he had no such qualms venturing a firm view that Mr Roden was hedging rate set exposure by his trading. It is said that there was a stark contrast in his evidence in this respect. In relation to the call which took place on 7 April 2010, Mr Park initially offered firm evidence that Mr Roden was trading to hedge his exposure. He claimed to be confident of that, even though he admitted he did not remember the conversation. But he ultimately conceded that he had no better knowledge of Mr Roden's planned trading than to speculate on what might be possible or, as Mr Park ultimately accepted, probable. ASIC says that what Mr Park was avoiding accepting was that Mr Roden and STIRR had concrete plans on how to trade based on their BBSW Rate Set Exposures. It is said that so much had to be drawn out of him under cross-examination. Further, ASIC says that the 17 November 2010 call also provides an example. It is said that to begin with, he explained what he said on the call as "what I thought might happen". This then became what "was likely" to happen. This then became what was "possible that he was going to" do. Mr Park then landed on "[a] probability. Okay".

903    Second, ASIC says that Mr Park suggested that ramping meant something anodyne. ASIC says that what Mr Park was evidently trying to avoid accepting but without persuasive force was that ramping meant deliberately trading to influence the BBSW.

904    Third, ASIC also says that some of Mr Park's answers diverged from the weight of the other evidence of Westpac's witnesses in a manner which undermined his reliability. It is said that he would not accept that FRAs, BAB Futures and interest rate swaps could hedge BBSW Rate Set Exposure, when Westpac's other evidence was to the contrary (or so ASIC asserted).

905    In my view, Mr Park was for the most part a reliable witness, but there were some problematic aspects that I will discuss later rather than now when dealing with the contravention dates. Let me make some other observations at this point.

906    In part, ASIC has criticised Mr Park's evidence because he did not accept ASIC's case theory. And so his refusal to accept that ramping necessarily meant deliberately trading to influence the rate. But his evidence as an experienced market participant is relevant to the meaning of that term. And I agree with Westpac that such evidence is not in and of itself to be dismissed or discounted by assuming that the meaning must be something different.

907    Further, as to ASIC's submission that the reliability of his evidence is undermined because he did not accept that BAB Futures and FRAs can hedge BBSW Rate Set Exposure, the evidence does support the proposition that on one view BAB Futures are not an effective hedge of rate set risk. This is due to the fact that they reflect the value of 3 month Prime Bank Bills at the expiry date rather than the day on which the rate set risk arises. In other words the party with the rate set risk is still exposed to changes in BBSW between those two dates. And so Mr Park's evidence that BAB Futures are not a direct hedge of rate set risk. Further, Mr Park's explanation about the downside to using FRAs (or any interest rate swap) to hedge rate set risk was persuasive.

(f)    Mr Duignan

908    ASIC has also criticised Mr Duignan's evidence. It says that he did not give balanced evidence and that his evidence should be treated with caution. It says that on the one hand he was eager to distance himself from STIRR's trading when it suited: "Well, Mr Roden wouldn't discuss with me what his strategies were at any time" and "I think I was stating that Mr Roden bought paper and I don't really know what the reason was for". But on the other hand, it is said that he was eager to point to RBA cash decisions, even when doing so was not relevant to the question being put, and even when it led Mr Duignan to give evidence as if he did know what Mr Roden's purposes for trading were. Further, ASIC says that Mr Duignan also offered an implausible explanation of the call with Mr Park and Mr Sharma on 17 November 2010.

909    Now as to ASIC's primary complaint that Mr Duignan was eager to distance himself from Mr Roden's reasons for trading when it suited him but happy to speculate about those reasons when he thought it would assist Westpac's case, I agree with Westpac that this is an unfair criticism having regard to the questions that were put to him.

910    He was asked about his conversation with Ms Johnston on 7 April 2010 and it was put to him that in that conversation he and Ms Johnston were not discussing Mr Roden's purpose for hedging interest rate risk on 6 April 2010. His answer was that Mr Roden did not discuss trading strategies with him. It was then put to him that he would not know the reasons for which Mr Roden and Ms Johnston traded on any given day, to which he answered that he would not be told the reasons for trading although he may sometimes know what they were doing if there were credit issues etc. There is nothing inconsistent between that evidence and the evidence he gave in relation to the conversation with Mr Park and Mr Sharma on 17 November 2010. He did not venture reasons for Mr Roden's trading. It was put to him that Mr Park was telling him in the call to avoid the days discussed because Mr Roden would be buying and selling, to which he said he thought Mr Park was telling him to avoid those days because it was going to be volatile, with an RBA day and futures delivery also in the mix. He was then asked if Mr Park was assuming in the call that Mr Roden would be buying on 6 December 2010. Mr Duignan's answer that the day before an RBA interest rate decision Mr Roden would be managing his risk as he saw fit was not purporting to give evidence of Mr Roden's reasons for trading but was an answer to a question which asked him what he understood Mr Park to be assuming as to whether Mr Roden would be buying on 6 December 2010. I think that there is nothing in ASIC's point at least in relation to its inconsistency criticism about commenting about Mr Roden's reasons for trading.

911    As to the other complaint about Mr Duignan's evidence, that his evidence in relation to the 17 November 2010 call with Mr Park and Mr Sharma was implausible, Westpac says that this assumes the correctness of ASIC's interpretation of the conversation. I agree that it so assumes. I will deal with this call later.

912    Generally speaking I found Mr Duignan's evidence to be reliable save as to some problematic aspects dealing with some contravention dates that I will discuss later, including in relation to the 17 November 2010 phone call.

(g)    Mr Hosie

913    ASIC says that Mr Hosie's evidence in cross-examination was very guarded. ASIC accepts that whilst that is understandable when new documents were shown to him, it is a trait that manifested throughout his cross-examination. ASIC says that it appeared that Mr Hosie strove to give honest answers, but they were answers "skewed by confirmation bias to toe the company line". In my view, the criticism is no more than that he did not accept ASIC's case theory. I agree with Westpac that that is not a justification in and of itself for discounting his evidence.

(h)    Mr Conway

914    ASIC has also criticised Mr Conway's evidence.

915    It submits that Mr Conway was not always candid in giving evidence about what he understood certain communications to mean. It is said that he showed a reluctance to acknowledge any potential meaning unfavourable to Westpac's forensic interests, and at times gave almost flippant answers. Thus, so ASIC submits, he initially purported to be unsure about the meaning of the "first rule of [F]ight [C]lub". And he noted that the word "tasty" had to do with eating. Further, on the one hand he would disclaim any knowledge of STIRR's reasons for trading, whilst on the other hand readily proffering that what STIRR was doing was hedging.

916    Now the claim that he initially purported to be unsure about the meaning of the "first rule of [F]ight [C]lub" is incorrect. Further, the dismissal of his evidence of what he understood Mr Hosie meant by "tasty" in their 1 July 2011 chat is also unjustified. He denied that he understood Mr Hosie to be saying that he traded to move the rate lower, which was the meaning put to him, and said that he thought potentially Mr Hosie was just saying that he was buying, as in devouring the available supply. Given the obscurity of Mr Hosie's comment, Mr Conway's evidence was not flippant. Generally speaking, Mr Conway was a reliable witness who gave responsive answers.

(i)    Mr Stokes

917    Generally, ASIC accepts that Mr Stokes' evidence was candid and responsive. But it says that the one exception was that in giving his evidence, it was apparent that Mr Stokes was "singing from the same hymn [book]" as other Westpac witnesses when it came to trading Prime Bank Bills to hedge, and trading only so as to avoid detrimentally affecting the BBSW but not so as to affect it beneficially. ASIC says that he pressed this evidence even in the face of his admission in his s 19 examination that he did not trade Prime Bank Bills to hedge BBSW rate set risk. ASIC says that given that he otherwise gave straightforward evidence, his willingness to change position is revealing of his disposition to favour Westpac's forensic interests. I found Mr Stokes to be generally reliable, save that the s 19 examination transcript did throw doubt on one aspect of his evidence.

(j)    Mr Masnick

918    In my view Mr Masnick was a reliable witness.

919    ASIC correctly accepted that Mr Masnick's evidence about derivative products and trading platforms appeared to reflect the views of a market participant which were honestly held, although ASIC said that those views are not determinative.

920    Further, ASIC accepted that Mr Masnick gave candid and probative evidence about the ways in which Financial Markets could and did hedge both BBSW Rate Set Exposure and outright interest rate risk, without trading in Prime Bank Bills.

921    But ASIC says that when Mr Masnick was asked about the confidentiality barrier between Group Treasury and Financial Markets, he was unwilling to make any concession that a conversation which took place was improper. ASIC says that his reluctance to do so revealed a concern on his part not to harm the forensic interests of Westpac. I must say that given that he was not a party to the relevant communication, which itself was general, this is hardly a substantial criticism.

(k)    General

922    Although I have made some general observations as to reliability at this point, I should say that in doing so I have considered all of the written and oral evidence led. But it has been convenient at this point to set out in general terms my perspective of some of the witnesses in order to facilitate the discussion in the next section. But the fact that my reasons are expressed in the linear order that I identified at the outset of my reasons does not indicate that a later section has not informed an earlier section (and of course vice versa). Further, my views on the reliability of any one witness concerning their evidence of subjective purpose(s) for trading is necessarily informed by, inter-alia:

(a)    objective features of the Bank Bill Market and BBSW and their actual and perceived susceptibility to manipulation;

(b)    the structure of Westpac, its trading strategies, risks, relevant incentives, and perceived gains and losses of trading;

(c)    all contemporaneous documentation; and

(d)    evidence of other witnesses and their reliability.

923    Further, as will become apparent from the next section I have concluded that some of Westpac's witnesses gave problematic evidence concerning the purpose for trading on 6 April 2010, 20 May 2010, and 1 and 6 December 2010. But I would say now that I do not consider that such a finding warrants any wholesale rejection of their other evidence, whether concerning other contravention dates or more generally.

924    Further, I should also note that in analysing Westpac's strategies for trading Prime Bank Bills I have accepted that a potential purpose for the trading may have been to hedge BBSW Rate Set Exposure. I have rejected ASIC's case that such a purpose generally was a confected afterthought by Westpac's witnesses. But whether it was in fact a purpose or the sole or predominant purpose for trading on a particular contravention date is, of course, another matter. Let me now turn to the contravention dates.

CONTRAVENTION DATES AND ALLEGED PRACTICE

925    Before discussing the specific contravention dates and ASIC's more general case concerning the Rate Set Trading Practice, it is necessary to make some preliminary observations concerning the evidence of key conversations relied upon by ASIC, their mode of proof and other evidentiary themes both as to the quality of the evidence and its subject matter.

(a)    Preliminary matters

926    First, a substantial number of audio files recording telephone conversations were tendered in evidence. Accompanying those audio files were competing transcripts submitted by each of the parties purporting to more faithfully record each conversation. In some cases, over the course of the trial the parties agreed on the transcripts of a number of audio files. In such a case, a third version was tendered. But in other cases, the parties not only failed to agree on a transcript, but instead tendered updated versions of their own transcripts. Further, different witnesses often had different transcripts put to them of the same conversation. Moreover, a single witness may have been shown one transcript in preparing their affidavit evidence but had other different transcripts put to them during the course of their oral evidence. Such a state of affairs has not been that desirable from a forensic perspective, but it has been understandable in a trial of this complexity. In order to make these reasons both consistent and comprehensible, I have taken the following steps:

(a)    First, where an agreed transcript was tendered, I have based my decision on that transcript and where necessary reproduced relevant aspects. And even where a witness has had a different version put to them in cross-examination, I have used the agreed transcript in these reasons except where there has been a material difference between the transcript that the witness was commenting upon and the agreed transcript, in which case I have taken into account the difference.

(b)    Second, where there has been no agreed transcript, I have considered and reproduced the competing versions but relied upon the most "advanced" version of each party's transcript. And even where the witness has had a different version put to them in cross-examination I have used the latest version in these reasons except where there has been a material difference between the transcript that they were commenting upon and the latest version, in which case I have taken into account the difference.

927    By taking these steps, I have balanced the interests of consistency in these reasons with the necessary degree of forensic discrimination that needs to be given in some cases to different versions of the transcripts for the same conversation.

928    Second, as to form, several other points should be made. Any spelling or other errors in those parts of the transcripts which I have reproduced appear in the versions tendered by the parties without further correction by me unless I indicate otherwise; that comment also applies to extracts from other contemporaneous documents that I have reproduced here or elsewhere in these reasons. Further, in some cases I have given electronic code references to some of the transcripts so that it is clear what I have referred to. I should note that for some of the transcripts I have given a composite reference. For example, if I have used, say, WBC.0003.0005.0070 (_WBC), that is a shorthand reference to the ASIC version (WBC.0003.0005.0070) and the Westpac version (WBC.0003.0005.0070_WBC). If the reference is to, say, WBC.0003.0005.0062_AGR, then that is a reference to an agreed transcript; but not all agreed transcripts carry such a tag as in some cases one party will have agreed with the other's version without any new composite being prepared. I should also say, perhaps counter-intuitively, that the letters "WBC" at the start of a document reference indicate that it was originally produced by ASIC; but if such letters are reproduced at the end of a document reference then this indicates a Westpac modification. And if a document reference commences with "WIB" then this indicates that it was originally produced by Westpac. Further, if the letters "ASIC" appear at the end of a document, then this indicates an ASIC modification. Further, occasionally a party may have amended its own version, in which case it may be marked as, say, WIB.705.001.0021_WBC, i.e. a Westpac produced transcript updated by Westpac. Finally in terms of form, the times for conversations discussed in these reasons have, where necessary, been converted from GMT to AEST even though the primary source may have used GMT.

929    Third, I have listened to many of the relevant audio files particularly those relating to the key communications. I have listened to these audio files for a number of reasons. Where transcripts have not been agreed, I have listened to the particular audio files in an endeavour to resolve any difference in the competing transcripts. But sometimes that has been a fruitless exercise. Further, whether transcripts have been agreed or not, I have listened to the audio files in order to assess from tone and tempo the mood of the conversation. Was what was communicated said in jest? Were conspiratorial tones used? Was there optimism, confidence or just matter of factness? Was there big-noting by the speaker or admiration by the recipient? One could go on with identifying the possibilities. I do not propose to do so. But what is well apparent is that an analysis of the face value of the words used has its limitations. Context is all important. So too is an appreciation of human nature. A confident statement may be inversely proportional to the speaker's knowledge. A statement made by a person of a benefit or result achieved may overstate their casual role, whether through ignorance or ego. A statement made by a person may involve game playing. Perhaps it may involve the new millennium vice of virtue signalling, although it seems that this affliction is less prevalent in the finance industry than elsewhere. There are all of these possibilities and more. I would also note that on many occasions during the trial, Mr Crutchfield QC for ASIC took every available opportunity to play various of the audio tapes for my benefit and that of the witnesses. Accordingly, for most of the key conversations I have heard their audio tapes on more than one occasion.

930    Fourth, all of the trading relevant to this proceeding and the communications relating thereto took place more than five years ago. Unsurprisingly, none of Westpac's witnesses remembered the communications or the trading which took place. Various Westpac witnesses proffered their reconstruction of the communications and trading. But ASIC says that very little weight should be given to such evidence.

931    Now I accept that where trading was sought to be analysed and legitimised, the witnesses based their analysis on limited information provided to them to undertake the exercise, and sometimes without regard to all of the contemporaneous communications. Further, some witnesses did not examine trading, or a failure to trade, on days close to the alleged contravention dates. For example, Mr Zuber said that on 1 and 6 December 2010, Mr Roden may have been buying Prime Bank Bills for the purposes of potentially being able to deliver into the futures close-out, and he did not think it was the case that Mr Roden was buying on 1 and 6 December 2010 to sell out on 9 December 2010 to benefit Group Treasury's BBSW Rate Set Exposure. But Mr Zuber did not consider Westpac's trading information or activities on 9 December 2010 and he could not say why Mr Roden would sell all of the relevant Prime Bank Bills on 9 December 2010. Further, I accept that even where witnesses dealt with both contemporaneous communications and the trading that occurred on a relevant contravention date, some of those witnesses may have assessed the potential commercial rationales for the trading without factoring in the communications.

932    Further, I also accept that I should be cautious in the value and weight that I give to direct evidence of purpose. It is trite to observe that such evidence is given ex post facto and is usually self-serving.

933    Further, I also accept that in any contest between contemporaneous documents and the recollections of persons with an interest in the outcome of the litigation, greater weight is usually accorded to such documents. Generally, they provide a "safer repository of reliable fact, particularly when it is clear that they have been prepared by a person with no reason to misstate those facts in the documents and where there is no suggestion that the documents are other than genuine" (Evans v Braddock [2015] NSWSC 249 at [74] per Hallen J).

934    Further, receding memories are usually attendant with a tendency to give formulaic evidence reflecting an attempt to reconstruct that which the witness does not recall, which formulaic evidence is to be received with caution (see e.g. Bathurst Regional Council v Local Government Financial Services Pty Ltd (No 5) [2012] FCA 1200 at [1248] per Jagot J). More generally, undoubtedly human memory is fallible and a witness at the time of giving evidence may not be giving reliable evidence that reflects the content or context of a previous communication or the reason for an act, particularly where the act was repetitive and there was no particular reason why it should have been noteworthy then or remembered now.

935    Further, where a witness shows a mystifying reluctance to accept the plain meaning of language used in a contemporaneous document, their evidence should be given less weight, but not necessarily because the witness is untruthful (The Bell Group Ltd (in liq) v Westpac Banking Corporation (No 9) (2008) 70 ACSR 1; [2008] WASC 239 at [1051] and [1052] per Owen J).

936    Further, where the clear meaning of a contemporaneous communication diverges from the evidence now given by a witness, it may be sufficient for me to find that the witness was doing his or her best to reconstruct a possible explanation for doing or saying what they did on a particular occasion, but that having regard to all of the evidence, the purpose that the witness held on the particular day was the purpose disclosed by the plain words of the relevant communication. I would adopt that approach concerning some aspects of Mr Roden's and Ms Johnston's evidence concerning the trading on 6 April 2010, 20 May 2010 and 1 and 6 December 2010.

937    Fifth, let me now say something concerning the vernacular of the traders. And it concerns the use of the "f***" word and its various derivatives. There is little doubt that linguistic Darwinism has favoured the English language. And part of its natural advantage springs not only from its capacity to either create vocabulary or unashamedly appropriate it from elsewhere, but its subsequent diverse and rich deployment. The "f***" word and its use by the traders in the present context is a classic example. It has been used as both a transitive and intransitive verb. It has been used in an active sense and a passive sense. It has been used in the past tense and the future tense. It has been used as an adjective. It has been used as a noun including as a verbal noun. Someone even tried to use it as an adverb. Occasionally it has been deployed not in any context that a formal grammarian would encourage, but simply to reflect an emotional response. Sometimes disappointment or exasperation, sometimes pleasant surprise or even admiration. Sometimes criticism, sometimes positive reinforcement. Even more occasionally, it has been used to indelicately communicate the thought that caution was being thrown to the wind. Clearly, the "f***" word and its derivatives are not terms of art in the finance industry. Nevertheless, their use in otherwise polite conversation appears to have been well understood by the colourful interlocutors.

938    More generally, the language games used by the participants, albeit not quite as attractive as the philosophical description given by Ludwig Wittgenstein, should be seen for what they are. They are frank discussions. They are designed to communicate information quickly and without the usual social niceties. They proceed on an unstated foundation of concepts and strategies well understood. To state the obvious is to be avoided. Robustness and directness are prized. That is the context in which the participants' communications must be analysed.

939    Sixth, during the trial much reference was made to the catchy nicknames of the traders and other participants. This was entertaining stuff in the theatre of trial. But as a trial judge mining facts from the coal face of forensic inquiry, such distractions must be put to one side. Nevertheless, there is one nickname that resonates with anyone who lived through the golden age of television journalism that I will say something further about later.

940    Seventh, ASIC has relied on communications that contain language such as "jamming" or "jam", "ramp" or "ramping", "bullets", "ammo" or "ammunition", or "shenanigans". ASIC has also relied on communications that contain observations from other participants in the Bank Bill Market. To the extent that those communications contain speculation about the behaviour of market participants, they have little probative value in assessing whether Westpac engaged in the Rate Set Trading Practice, although they may be relevant to the question of the susceptibility of the Bank Bill Market or the BBSW to being manipulated. Let me say something further on terminology.

941    ASIC has sought to attach a particular trade meaning to such terms. But there is a difficulty. The way in which such terms were used by some participants does not necessarily inform the way in which they were used by other participants. Moreover ASIC has not established that the expressions or particular words were terms of art with an industry-wide meaning. Accordingly, the context of the particular communication in which the particular term has been used is all important. The evidence establishes the following:

(a)    The term "ramp" or "ramping" could just be referring to trading or the effect of trading in large volumes. But sometimes the term was used to refer to trading in a way which might influence yield or which may involve BBSW setting at an unexpected level.

(b)    The term "jam" or "jamming" could refer to a range of things, including being a reference to trading or volatility as a result of active trading including manipulative trading.

(c)    The term "shenanigans" was often used as a reference to large amounts of volume being available in the Bank Bill Market or the market being volatile. But sometimes it could refer to inexplicable or improper trading.

(d)    The terms "ammo", "bullets" and other similar terms are in most contexts properly understood as a reference to a portfolio of securities or stock that a trader may wish to sell. But the use of such terms may say little about the purpose for which those securities were held or traded, although in some contexts such use may resonate with perceived weaponisation, including potentially for a manipulative purpose.

942    Eighth, I raised with ASIC's counsel whether the actual movement in BBSW on the day of trading for a particular contravention date as compared with the preceding day or the following day is probative evidence of a manipulative purpose. In other words, take a day where Group Treasury's BBSW Rate Set Exposure was short and it was said that Westpac was buying Prime Bank Bills to drive down BBSW. How should I treat evidence that BBSW moved down on that day as compared with the previous day? Does this confirm manipulative purpose? What if BBSW stayed the same? Should I treat this as neutral on the question? And what if BBSW moved up? Should I treat this as negating evidence of a manipulative purpose or positive evidence of a non-manipulative purpose because BBSW moved in the opposite direction to what ASIC's case theory would suggest?

943    To most of these possibilities, ASIC said that such movement was equivocal evidence in confirming or demonstrating a manipulative purpose. If BBSW stayed the same, then ASIC said that this just established that BBSW had been maintained by the manipulative trading. If BBSW went in the opposite direction to ASIC's case thesis, then it said that it would have moved even further in the opposite direction but for the manipulative trading. But where BBSW moved in the same direction down (i.e. on the buying scenario conjoined with a short BBSW Rate Set Exposure), then it was happy to bank such a result as confirming its thesis. In other words ASIC was seeking to have the best of all possible worlds.

944    This all demonstrated to me that to use changes in BBSW on a particular contravention date to confirm or falsify manipulative purpose was largely an arid forensic exercise. Accordingly, although for each of the contravention dates I have set out the changes in BBSW, drawing conclusions therefrom has not been of assistance.

945    Ninth, if one looks at the trading on each of the contravention dates, to my mind there is no evidence to show that:

(a)    each of the trades was not genuine in terms of the objectively ascertainable indicia; or

(b)    Westpac did not obtain the best value that it could either on the buy side or the sell side.

946    In other words, from what I could tell the trades were genuine and at market value if one considers the objectively ascertainable indicia, putting to one side questions of subjective purpose for the moment.

947    Now intuitively one might think that to make such an observation was against ASIC's case thesis. But ASIC disagreed. Its point was that a manipulative trade, the only "hidden" indicia of which was Westpac's alleged manipulative purpose, could for all relevant purposes look like and masquerade as a genuine trade. So again, it was seeking to have the best of all worlds. If a particular trade looked dodgy or non-genuine, then it sought to bank this as support for its case thesis. But if the trade looked and tasted like a genuine trade, then in any event ASIC said that it could still be masquerading as a manipulative trade and therefore was not inconsistent with its case thesis.

948    As I have said, to my mind when looking at the trading on each of the contravention dates, each of the trades was genuine when considering objectively ascertained indicia. Moreover, Westpac obtained or appears to have obtained the best value that it could either on the buy side or the sell side. I will return to these questions at a conceptual level much later in my reasons when discussing ss 1041A and 1041B in terms of their construction and application.

949    Tenth, for each of the contravention dates I have set out some estimates made by Mr Zuber concerning the financial consequences of one basis point changes in BBSW. Westpac sought to make much of the very modest sums that could potentially be earned from manipulation, seeking to suggest that there was little incentive, with ASIC suggesting the opposite particularly if there were three basis point changes or more. On this aspect, in my view ASIC's case on financial incentives to manipulate, if the circumstances were right, is the more compelling one.

950    Eleventh, the discussion that follows concerns trading in the Bank Bill Market and the purpose for such trading on the contravention dates. But there is no discussion of the trading in or pricing of interest rate swaps, cross-currency swaps and BAB Futures on those dates, except in a secondary sense insofar as it may be relevant as to why Prime Bank Bills may have been traded. ASIC does not contend that there was direct manipulation of the trading in the markets for such derivative products or their pricing. Rather, ASIC contends that the effect or likely effect in those markets or on their pricing was indirectly caused by and through the manipulation of BBSW through trading in the Bank Bill Market.

951    Let me now descend into the detail concerning the contravention dates. I will discuss later the question of the existence of the Rate Set Trading Practice. I will also say something later concerning alternative hypotheses and their interaction with the onus of proof when dealing with questions of dominant purpose relating to the purpose(s) for trading on the contravention dates. I will also analyse later what is meant by dominant purpose, but would say at this point that I have taken dominant purpose to be the prevailing or principal purpose. Now Westpac has said that I should adopt a "but for" test for determining dominant purpose. But that is a test for causation, not a test for dominant purpose. Nevertheless, such a test is a useful forensic tool to assist in determining the question of dominant purpose, to the extent that it throws light at least on whether a purpose is a necessary purpose as distinct from a sufficient purpose. Let me say now that in terms of my findings of dominant manipulative purpose on 6 April 2010, 20 May 2010, 1 and 6 December 2010, that if such a manipulative purpose had not existed for the trading on those dates, Westpac would in all likelihood have not traded to the extent that it did.

952    One final preliminary note. My findings concerning each specific contravention date are also informed by the following matters. First, and generally, the totality of the evidence. Second, for a particular contravention date, what I have analysed and said about all other contravention dates. Third, the setting of more general evidence concerning the alleged Rate Set Trading Practice even though I have discussed this later in my reasons. In that respect the linear sequence of my reasons cannot fully reflect all of the iterative analysis that I have had to undertake. Fourth, the incentives to manipulate and the susceptibility to manipulation of the Bank Bill Market and BBSW that I have already discussed, and the effect or likely effect of trading for such a dominant purpose that I discuss later. Finally, to the extent that any part of ASIC's case is circumstantial, I have considered all evidentiary facts to the extent that they might afford proof or an indication that was prospectant, concomitant or retrospectant, adopting Wigmore's conceptual categorisation, and have assessed the probative force of their combined weight.

(b)    6 April 2010

953    ASIC's case is that Westpac bought 1 month and 3 month Prime Bank Bills on 6 April 2010 for the sole or dominant purpose of lowering or maintaining the level at which BBSW set in both the 1 month and 3 month tenors. In summary, I have accepted ASIC's case concerning this contravention date in terms of the existence of the relevant dominant purpose.

1 month tenor BBSW and Westpac rate set exposure

954    On 6 April 2010, Westpac bought $1.622 billion of 1 month Prime Bank Bills, which was 100% of the trading volume in the Bank Bill Market on the day. The purchases were made at yields of between 4.23 and 4.28, although approximately 93% of that volume was acquired at yields between 4.23 and 4.25 being the liquidity point in the Bank Bill Market at which there was substantial volume on offer.

955    On 6 April 2010, the 1 month BBSW set at 4.2300. AFMA received submissions for the 1 month BBSW of between 4.23 and 4.25. Westpac's submission was 4.23.

956    The 1 month BBSW in the days before and after 6 April 2010 was as follows:

30 March 2010

4.1750

31 March 2010

4.2067

1 April 2010

4.2433

6 April 2010

4.2300

7 April 2010

4.3500

8 April 2010

4.3200

9 April 2010

4.3100

On the following day, 7 April 2010, the 1 month BBSW increased to 4.3500, apparently due to the RBA announcement on the afternoon of 6 April 2010 of a 25 basis point increase in the cash rate.

957    On Tuesday 6 April 2010, Group Treasury had the following component BBSW Rate Set Exposures in the 1 month tenor:

(a)    a short exposure in PRM of $13,196,390,749.19;

(b)    a short exposure in ALM of $1,466,000,000.00;

(c)    a short exposure in St George of $1,840,000,000.00;

(d)    a long exposure in Corporate Bill Lending of $1,132,984,211.21; and

(e)    a short exposure in ALM (transfer priced balance sheet) of $14,729,865.83.

958    ASIC contends that Group Treasury had a short BBSW Rate Set Exposure in the 1 month tenor of $14,662,390,749.19. But Group Treasury's aggregate BBSW Rate Set Exposure that day in the 1 month tenor was a short exposure of $15,384,136,403.81. This discrepancy is due to the position taken by ASIC at trial that the BBSW Rate Set Exposure of a given day is to be obtained by netting out the component exposures in the PRM and ALM books. As I have previously explained, the table annexed to the parties' SOAF shows this "PRM+ALM" figure in column F for each contravention date. The table shows in additional columns component exposures in St George (column G), Corporate Bill Lending (column H) and ALM (transfer priced balance sheet) (column I) and then in column J the "Aggregate Group Treasury", which is the aggregate figure of the five component exposures. On most contravention dates, the discrepancy between PRM+ALM and aggregate Group Treasury does not affect my analysis. But on several contravention dates, the discrepancy is large enough to change the direction of the exposure. In other words, on some of the contravention dates ASIC submits that Westpac had a short exposure when the aggregate Group Treasury exposure was long and vice versa. That discrepancy gave rise to an argument on the part of Westpac for some contravention dates that ASIC's purpose case theory was inconsistent with the exposure that Westpac actually had. ASIC's response was to say that the traders may not have known the aggregate Group Treasury position (including St George), but may only have known of the PRM and ALM components. I will deal with this issue when I need to.

959    Now apparently, 6 April 2010 was Group Treasury's largest BBSW Rate Set Exposure of the year in the one month tenor.

3 month tenor BBSW and Westpac rate set exposure

960    Further, on 6 April 2010, Westpac bought $1.02 billion of 3 month Prime Bank Bills, which was 72.86% of the market volume on the day. The purchases were made at yields of 4.47 and 4.48.

961    On 6 April 2010, 3 month BBSW set at 4.4683. AFMA received submissions for the 3 month BBSW of between 4.46 and 4.48. Westpac's submission was 4.47.

962    In the days before and after 6 April 2010 the 3 month BBSW was as follows:

30 March 2010

4.5250

31 March 2010

4.4933

1 April 2010

4.4183

6 April 2010

4.4683

7 April 2010

4.5400

8 April 2010

4.5217

9 April 2010

4.5367

963    The 3 month BBSW on the following day, 7 April 2010, increased to 4.5400, again apparently due to the RBA announcement on the afternoon of 6 April 2010 of a 25 basis point increase in the cash rate.

964    On Tuesday 6 April 2010, Group Treasury had the following component BBSW Rate Set Exposures in the 3 month tenor:

(a)    a short exposure in PRM of $6,432,138,813.00;

(b)    a long exposure ALM of $3,432,000,000.02;

(c)    a short exposure in St George of $200,000,000.00;

(d)    a long exposure in Corporate Bill Lending of $364,851,195.04; and

(e)    a long exposure in ALM (transfer priced balance sheet) of $707,664,150.06.

965    ASIC contends that Group Treasury had a short BBSW Rate Set Exposure in the 3 month tenor of $3,000,138,812.98. But Group Treasury's aggregate BBSW Rate Set Exposure that day in the 3 month tenor was a short exposure of $2,127,623,467.88.

Knowledge of Westpac Group Treasury BBSW Rate Set Exposures on 6 April 2010

966    Now I accept that Westpac and in particular Mr Roden knew the BBSW Rate Set Exposure for 6 April 2010 prior to trading on 6 April 2010. Employees and officers of STIRR (including Mr Roden) had access to this information. Mr Roden conducted Westpac's trading on 6 April 2010, in both the 1 month and 3 month tenors. Prior to trading in the BBSW Rate Set Window on 6 April 2010, Mr Roden knew Group Treasury's BBSW Rate Set Exposure in the 1 month tenor at least exceeded $14 billion (short) and that Group Treasury's BBSW Rate Set Exposure in the 3 month tenor at least exceeded $2 billion (short).

967    It is also not in doubt that the profit/loss of the books in Group Treasury would have been potentially affected by movement in the BBSW on 6 April 2010. Mr Zuber gave evidence that Group Treasury's BBSW Rate Set Exposure would have benefited by about $118,000 and $46,000 from a single basis point move in the 1 month and 3 month BBSW rate set on that day respectively, but those figures are based on the aggregate Group Treasury exposure.

Mr Roden's trading on 6 April 2010

968    On 6 April 2010 Westpac purchased 100% of the 1 month Prime Bank Bills traded through ICAP and Tullett in the Bank Bill Market, with a face value of $1.622 billion. It purchased about 73% of the 3 month Prime Bank Bills traded through ICAP and Tullett in the Bank Bill Market, with a face value of $1.02 billion.

969    In my view, the relevant trader in both the 1 month tenor and the 3 month tenor was Mr Roden. Now although Mr Roden did "not have any specific recollection of Westpac's trading on 6 April 2010 or events on that day" and did "not recall … trading on that day", he gave evidence that he had listened to audio recordings of trading that day "which appear to be of a broker communicating with me".

970    In my view the natural and ordinary meaning of the contemporaneous communications support ASIC's case that Mr Roden traded with the sole or dominant purpose of influencing the BBSW to set lower. Let me go through each of these in turn.

Telephone call between Mr Roden and Ms Johnston on 6 April 2010 (shortly prior to RBA announcement) [Agreed transcript WIB.705.002.0054_AGR]

971    Let me begin with a communication between Mr Roden and Ms Johnston on 6 April 2010 before the RBA announcement. The time of this call is not known, but both parties accept in their submissions that it was prior to the RBA announcement at 2.30 pm that day. There is an agreed transcript of the conversation, and a version of the transcript was put to Mr Roden. The audio of the conversation was also played to Mr Roden. I have also listened to it.

972    It would appear that this conversation is a reference to the risk position which Group Treasury was taking on the RBA cash rate decision later that afternoon. Mr Roden said "we'll find out very shortly whether or not we win the Lottery because … I've got no f***in idea what they're gonna do", Ms Johnston noted that it "[s]hould be very interesting" and Mr Roden responded "I got no idea what they are going to do so we will find out shortly". It is readily apparent that they were discussing the imminent RBA announcement on the cash rate. It is also apparent that Mr Roden genuinely at this time had little idea of what the RBA was going to do.

973    In my view this conversation demonstrates that Mr Roden was uncertain on 6 April 2010 as to whether the RBA was likely to raise the cash rate or hold it steady, contrary to ASIC's suggestion that his view was that the RBA would raise the cash rate. I say this notwithstanding that Mr Roden was the author or source of the view in the Treasury Update dated 31 March 2010 that the RBA was likely to increase the cash rate. Further, I reject ASIC's suggestion, based on an analogy with a professional punter and a short priced favourite, that it was very unlikely that Mr Roden had a last minute change of view about the RBA decision and decided to hedge his bet by reducing his exposure to the RBA decision on the day.

974    In my view, on 6 April 2010 Mr Roden was uncertain whether the RBA would raise the cash rate. I accept Westpac's submission that his trading that day was influenced by that uncertainty.

Telephone call between Mr Roden and Ms Johnston at 5.21 pm on 6 April 2010 [Agreed transcript WBC.0003.0005.0055_ASIC]

975    Let me deal with the second communication between Mr Roden and Ms Johnston on that day.

976    At about 5:21:49 pm AEST on 6 April 2010, Mr Roden spoke on the phone with Ms Johnston, who was in London. This transcript is now agreed, although earlier versions of the transcript were put to Mr Roden and Ms Johnston, as well as Mr Bishop and Professor Putnins. The audio of this conversation was played to Mr Roden. I have also listened to it.

977    The first relevant exchange is as follows:

Ms Johnston:    You've been looking OK today

Mr Roden:    Yeah, no, we made about 12 million bucks today, right, so that's not that's what you call a good day, right?

978    In my view the reference to a profit of $12 million made on 6 April 2010 was a reference to the estimated profit made by Group Treasury due to revaluations on the whole of its book which was exposed to domestic interest rates due to the increase in the official RBA cash rate announced at around 2.30 pm on 6 April 2010 as well as other positions and trading outcomes across the ambit of Group Treasury's activities that day. In my view it was not a reference to the calculation of profit arising from the movement in the BBSW Rate Set Window that day. In Mr Roden's first affidavit he said "I believe I am likely referring to the profit that was made due to the revaluation across the whole of the STIRR book as a result of the RBA rate rise".

979    As to the calculation of this figure, Mr Roden gave evidence that Westpac looked at the difference between the closing interest rates from the previous day and the closing interest rate from the current day in each tenor (taking into account new positions that were entered into during the day). Although other factors may also have contributed to the end of day profit/loss, the difference between the two closing rates would be multiplied by the end of day risk position in each portfolio for that tenor and that result would form a substantial part of Treasury's end of day profit/loss result. In this instance, the Treasury Eagles for 1 April 2010 and 6 April 2010 record that at closing on 6 April 2010, the STIRR portfolio:

(a)    was short approximately $913,623 per basis point in the three month tenor. The closing rate in the three month tenor was 4.42 on 1 April 2010 and 4.54 on 6 April 2010, as recorded in the Treasury Eagle for each tenor, being an increase of approximately 12 basis points between the two days. As STIRR was short, for each basis point increase in the three month rate, it gained $913,623, meaning that on 6 April 2010, the STIRR portfolio revalued and gained approximately $10,963,476 in the three month tenor; and

(b)    was short approximately $170,126 per basis point in the one month tenor. The closing rate in the one month tenor was 4.24 on 1 April 2010 and 4.33 on 6 April 2010, as recorded in the Treasury Eagle for each tenor, being an increase of approximately 9 basis points between the two days. As STIRR was short, for each basis point increase in the one month rate, it gained $170,126, meaning that on 6 April 2010, the STIRR portfolio revalued and gained approximately $1,531,134 in the one month tenor.

980    Accordingly, the upward movements in interest rates between 1 April 2010 and the close of business on 6 April 2010 meant that in the one and three month tenors Group Treasury's mark to market position and therefore profit/loss result improved by around $12,494,610.

981    Mr Roden then goes on to refer to Group Treasury's 1 month and 3 month BBSW Rate Set Exposure:

Mr Roden:    we had a massive rate set today, like we had like …

Ms Johnston:    Oh really

Mr Roden:    f***in 14 billion of one month.

Ms Johnston:     Gee

Mr Roden:    I had to buy like I bought like fu… I had 14 billion 1 month, 'cause it's a long you know, you got the 5 day run

Ms Johnston:    Yeah [inaudible]

Mr Roden:    … and 3 yards of 3 month, probably about 4 yards of 3 months, so I ended up buying like 2½ billion dollars of [inaudible]

Ms Johnston:    Shivers…

982    This part of the discussion concerned Mr Roden's trading that morning in the Bank Bill Market, from where he says "I had to buy". He also accepted that when he said "we had a massive rate set today, like we had like … f***in 14 billion of one month" that he was referring to Group Treasury's 1 month BBSW Rate Set Exposure. His focus in this part of the conversation with Ms Johnston was his rate set exposure.

983    Now I would note that he stressed the word "massive" and his tone suggests that he was a reluctant buyer, at least of those volumes. Further, Ms Johnston's word "Shivers" more suggests a sympathetic ear rather than one of admiration.

984    Ms Johnston's evidence was that she understood Mr Roden's statement "14 billion 1 month" to be a reference to Group Treasury's BBSW Rate Set Exposure in the 1 month tenor, and "3 yards of 3 month, probably about 4 yards of 3 months" to be a reference to Group Treasury's BBSW Rate Set Exposure in the 3 month tenor. She says that her reactions to Mr Roden's statements reflected her view that these exposures were "relatively large exposures".

985    Mr Roden explained to Ms Johnston: "I had to buy like … so I ended up buying like 2 ½ billion dollars".

986    Now ASIC contends that the natural meaning of this statement is that Mr Roden had to spend a lot of money to buy Prime Bank Bills in the BBSW Rate Set Window to get the BBSW down. I am inclined to agree. But Mr Roden's evidence is that:

I believe I was likely expressing a view that it was necessary to buy Prime Bank Bills that day to reduce Treasury's one month BBSW rate set exposure, which was $14 billion, and also to reduce Treasury's net short interest rate risk position. As expressed in my call with Ms Johnston on 6 April 2010 … it appears that I was very uncertain as to whether the RBA would change the Cash Rate and my usual practice would have been to reduce the amount that Treasury stood to lose if the RBA held the Cash Rate by reducing Treasury's net short interest rate risk position if that was possible.

987    Mr Roden repeated this explanation in cross-examination:

the reason why I bought on this day, I believe, is because I had formed the view that the rate the probability of a rate hike was probably closer to fifty-fifty and I was trying to reduce my risk.

988    In my view, ASIC's interpretation should be preferred for the following reasons.

989    First, this part of the