Supreme Court New South Wales Medium Neutral Citation: UGL Rail Pty Ltd v Wilkinson Murray Pty Ltd [2014] NSWSC 1959 Hearing dates: 20 to 24, 27, 28, 30, 31 October 2014, 3 to 5, 7, 10 to 12, 20 and 21 November 2014 Decision date: 15 December 2014 Jurisdiction: Equity Division - Technology and Construction List Before: Ball J Decision: Judgment for the plaintiff in the sum of $5,059,026.17 Catchwords: TORTS – negligence – whether acoustic engineering advice provided by defendant negligent – whether plaintiff contractor relied on advice when installing insufficient material in breach of head contract – where reliance required contractor to prove what non-negligent advice would have been and what it would have done had it received non-negligent advice – whether claim apportionable under Civil Liability Act 2002 (NSW), Pt 4 – plaintiff who contributes to its own loss not a "concurrent wrongdoer" – whether plaintiff contributorily negligent TRADE PRACTICES – claim for damages under Trade Practices Act 1974 (Cth), s82 for breach of s 52 – whether advice given by defendant misleading or deceptive – whether plaintiff contractor relied on advice when installing insufficient material in breach of head contract – whether claim apportionable under Trade Practices Act 1974 (Cth), Pt VIA – plaintiff who contributes to its own loss not a "concurrent wrongdoer" – whether plaintiff contributorily negligent DAMAGES – where plaintiff settled claim brought against it by head contractor for breach of head contract – where plaintiff claims cost of settlement against defendant under contractual indemnities – causation – whether negligent or misleading or deceptive advice caused plaintiff to enter into settlement agreement – Civil Liability Act 2002 (NSW), s 5D – s 5D not applicable to claims under Trade Practices Act 1974 (Cth), s 82 but words "by conduct of another" require a causal connection – whether s 5D applies to claim under contractual indemnity not clear and depends on construction of indemnity provisions – causation in this case dependent on reasonableness of settlement whichever test of causation applied DAMAGES – where plaintiff settled claim brought against it by head contractor for breach of head contract – where plaintiff claims cost of settlement against defendant under contractual indemnities – whether settlement reasonable – whether amount paid pursuant to settlement reasonable in the circumstances – whether appropriate that defendant be liable for the full extent of the settlement where only causally responsible for part of it – Civil Liability Act 2002 (NSW), s 5D(1)(b) – court to determine what would be a reasonable settlement when actual settlement unreasonable CONTRACT – where negligent advice in breach of agreement between plaintiff and defendant – whether plaintiff estopped from asserting breach of contract based on common assumption – whether plaintiff breached implied term that parties cooperate – claim statute barred by Limitation Act 1969 (NSW), s 14 where breach occurred in 2003 and proceedings commenced in 2011 Legislation Cited: Civil Liability Act 2002 (NSW), Pt 1A, ss 5A, 5B, 5D, 5O, Pt 4, ss 34, 35 Law Reform (Miscellaneous Provisions) Act 1965 (NSW), s 9 Limitation Act 1969 (NSW), s 14 Trade Practices Act 1974 (Cth), ss 52, 82, Pt VIA Cases Cited: AI Mclean Pty Ltd v Hayson [2008] NSWSC 927 Adeels Palace v Moubarak [2009] HCA 48; (2009) 239 CLR 420 Allianz Australia Insurance Ltd v GSF Australia Pty Ltd [2005] HCA 26; (2005) 221 CLR 568 BHP Billiton (Olympic Dam) Corporation Pty Ltd v Steuler Industriewerke GmbH (No 2) [2011] VSC 659 Biggin & Co Ltd v Permanite Ltd [1951] 2 KB 314 BNP Paribas v Pacific Carriers Ltd [2005] NSWCA 72 Bovis Lend Lease Limited v RT Fire Protection Limited [2003] EWHC 939 (TCC); (2003) 89 ConLR 169 Commonwealth v Amann Aviation Pty Ltd [1991] HCA 54; (1991) 174 CLR 64 Commonwealth v Cornwell [2007] HCA 16; (2007) 229 CLR 519 Concrete Pty Ltd v Parramatta Design and Developments Pty Ltd [2006] HCA 55; (2006) 229 CLR 577 Drayton v Martin (1996) 67 FCR 1 Edwards v Insurance Office of Australia Ltd (1933) 34 SR (NSW) 88 General Feeds Inc Panama v Slobodna Plovidba Yugoslavia [1999] 1 Lloyd's Rep 688 Gibbs v Guild (1881) 8 QBD 296 Hadley v Baxendale (1854) 9 Ex 341; (1854) 156 ER 145 Hawkins v Clayton [1988] HCA 15; (1988) 164 CLR 539 Henville v Walker [2001] HCA 51; (2001) 206 CLR 459 Howe v Teefy (1927) 27 SR (NSW) 301 Hudson Investment Group Ltd v Atanaskovic [2014] NSWCA 255; (2014) 311 ALR 290 I & L Securities Pty Ltd v HTW Valuers (Brisbane) Pty Ltd [2002] HCA 41; (2002) 210 CLR 109 Jones v Dunkel [1959] HCA 8; (1959) 101 CLR 298 King v Western Sydney Local Health Network [2013] NSWCA 162 Lawfund Australia Pty Ltd v Lawfund Leasing Pty Ltd [2008] NSWSC 144; (2008) 66 ACSR 1 March v E & MH Stramare Pty Ltd [1991] HCA 12; (1991) 171 CLR 506 McRae v Commonwealth Disposals Commission [1951] HCA 79; (1951) 84 CLR 377 Monaghan Surveyors Pty Ltd v Stratford Glen-Avon Pty Ltd [2012] NSWCA 94 Nykredit Mortgage Bank Plc v Edward Erdman Group Ltd (No 2) [1998] 1 All ER 305 Paul v Cooke [2013] NSWCA 311; (2013) 85 NSWLR 167 QBE Insurance (Australia) Ltd v Wesfarmers General Insurance Ltd [2010] NSWSC 855 Rail Corporation NSW v Fluor Australia Pty Ltd [2009] NSWCA 344 Raphael Shin Enterprises Pty Ltd v Waterpoint Shepherds Bay Pty Ltd [2014] NSWSC 743 Rennie Golledge Pty Ltd v Ballard [2012] NSWCA 376; (2012) 82 NSWLR 231 Roe v Minister for Health [1954] 2 QB 66 Secured Income Real Estate (Aust) Ltd v St Martins Investments Pty Ltd [1979] HCA 51; (1979) 144 CLR 596 Strong v Woolworths [2012] HCA 5; (2012) 246 CLR 182 Tanna v Deutsche Bank (Asia) AG [1996] NSWSC 408 Tanna v Deutsche Bank (Asia) AG (Court of Appeal (NSW), 15 September 1998, unrep) Travel Compensation Fund v Tambree [2005] HCA 69; (2005) 224 CLR 627 Unity Insurance Brokers Pty Ltd v Rocco Pezzano Pty Ltd [1998] HCA 38; (1998) 192 CLR 603 Wallace v Kam [2013] HCA 19; (2013) 250 CLR 375 Wardley Australia Ltd v State of Western Australia (1992) 175 CLR 514 Zanner v Zanner [2010] NSWCA 343; (2010) 79 NSWLR 702 Texts Cited: DA Bies and CH Hansen, Engineering Noise Control: Theory and Practice (2nd ed, 1996, E & FN Spon) D Villa, Annotated Civil Liability Act 2002 (NSW) (2nd ed, 2013, Lawbook Co) JD Heydon, Cross on Evidence (8th Aust ed, 2010, LexisNexis) The Negligence Review Panel, Review of the Law of Negligence: Final Report (2002) W Courtney, "Settlements Following Breach of Contract" (2013) LMCLQ 157 Category: Principal judgment Parties: UGL Rail Pty Ltd (Plaintiff) Wilkinson Murray Pty Ltd (Defendant) Representation: Counsel: MA Jones SC with Ms JA Steele and Ms CO Gleeson (Plaintiff) RA Cavanagh SC with SJ Walsh (Defendant) Solicitors: DLA Piper (Plaintiff) Thompson Cooper Lawyers (Defendant) File Number(s): 2011/123652 Judgment Introduction 1. In these proceedings, the plaintiff, UGL Rail (formerly known as United Group Rail Pty Ltd), alleges that the defendant, Wilkinson Murray, gave negligent or misleading advice in relation to the steps that should be taken to comply with contractual specifications concerning reverberation control in the Epping to Chatswood Rail Link (ECRL) causing UGL Rail losses said to total $26,429,541.15 plus interest and costs. 2. The ECRL comprises two adjacent cylindrical rail tunnels each approximately 13 km long connecting Chatswood and Epping stations, together with three intermediate stations at Macquarie Park, Macquarie University and North Ryde. 3. Originally, on 6 July 2002, the Director General of the New South Wales Department of Transport entered into a contract (the CIVSYS Deed) with a joint venture between Thiess Pty Ltd and Hochtief Aktiengesellschaft (THJV) for the design and construction of what was then the Parramatta Rail Link. In 2003, the scope of the project was modified to become the ECRL and, on 1 January 2004, the CIVSYS Deed was assigned from the Director General to the Rail Infrastructure Development Corporation, later known as the Transport Infrastructure Development Corporation (TIDC). Under arrangements that are not in evidence, TIDC would make the ECRL available to RailCorp to operate train services between Epping and Chatswood once the link was complete. 4. On 20 December 2002, THJV entered into a subcontract with Alstom Australia Limited (Alstom) for the systems installation component of the ECRL (the Systems Subcontract) for a total price of approximately $179 million. The work subcontracted to Alstom included work associated with the design, construction and testing of sound absorbent panels to be installed in the tunnels to meet the reverberation times specified in the Works Brief, which set out the scope of work to be undertaken by THJV pursuant to the CIVSYS Deed. 5. On 10 June 2003, Alstom entered into a consultancy agreement with Wilkinson Murray (the Consultancy Agreement) under which Wilkinson Murray agreed to provide acoustic consulting services in connection with Alstom's obligations under the Systems Subcontract. It will be necessary to say more about the nature of those services shortly. For present purposes, it is sufficient to observe that the services included advice on the nature and extent of the sound absorbent panels that should be installed in the tunnels to meet the reverberation times specified in the Works Brief and compliance testing of the installed panels. 6. Wilkinson Murray provided advice to Alstom in May 2003. There are issues concerning whether the advice was intended to be final advice, whether it was qualified and whether Alstom relied on it. 7. On 2 June 2005, United Group Limited, UGL Rail's parent company, and UGL Rail entered into a sale and purchase agreement with Alstom by which UGL Rail acquired the transport business of Alstom and became responsible for the Systems Subcontract. The Systems Subcontract was novated to UGL Rail by a deed of novation dated 16 September 2005. On the same day, Alstom and UGL Rail entered into a deed of assignment of contracts by which Alstom assigned to UGL Rail various contracts, including the Consultancy Agreement. 8. The work under the Systems Subcontract was actually performed by UGL Infrastructure Pty Ltd, another subsidiary of United Group Limited. The arrangements by which UGL Infrastructure performed the work are not clear from the evidence. Frequently, the parties did not distinguish between UGL Rail and UGL Infrastructure, referring to them interchangeably as "United". This judgment adopts that practice where it is not important to distinguish between the two companies. 9. In about April or May 2007, United installed a single row of acoustic panels in a test section of the ECRL between Macquarie Park and Macquarie University stations and, on 21 May 2007, Wilkinson Murray undertook reverberation time testing in that section of the tunnel. It advised United on 25 May 2007 that the measurements significantly exceeded the specifications contained in the Works Brief. It will be necessary to say more about the nature of the advice given by Wilkinson Murray and the steps taken to solve the problem. It is sufficient by way of introduction to observe that an additional row of panels was installed in the test section of the tunnel and further testing was done, which revealed that the tunnels still did not satisfy the reverberation specification. As a result, United increased the size of the second row of panels and the two rows of panels were installed throughout the tunnels. Following their installation, the acoustic engineers engaged by TIDC, Heggies, conducted further tests and concluded that the tunnels did not meet the reverberation specification. Wilkinson Murray disputed Heggies's conclusions. However, on 10 October 2008, TIDC issued a defect notice to THJV in relation to the non-compliance with the reverberation specification and, on 30 October 2008, it issued a defect notice to THJV in relation to excessive in-car noise. It is accepted between the parties that, although the failure to meet the reverberation specification contributed to in-car noise, there were other causes of the failure to meet the contractual requirements relating to in-car noise. 10. Following service of the defect notices, TIDC and THJV formed a working group to investigate possible solutions to the problem of in-car noise, which was TIDC's principal concern. Various proposals were developed in conjunction with United. Between 11 March 2009 and 20 March 2009, there was a mediation between TIDC and THJV, which was attended by United, concerning the disputes that had arisen. At that mediation, the parties agreed in principle to undertake certain noise mitigation works and that completion of that work would be treated as discharging THJV's and UGL Rail's obligations to meet the contractual specifications concerning noise, including the contractual specifications concerned with in-tunnel reverberation. 11. Terms of settlement were agreed and, in connection with the settlement, UGL Rail and THJV entered into a deed on 23 June 2009 entitled "Deed Concerning Clause 8.2(b) Direction for Variation" (the Settlement Deed) under which UGL Rail agreed to perform certain work at no cost to THJV that TIDC had sought to be done in order to ameliorate the problems with in-car noise. A corresponding deed was entered into between TIDC and THJV. 12. UGL Rail was insured in respect of the claim that was made against it. Its insurance claim was settled for the sum of $25,027,000. That sum was paid directly by the insurers to UGL Infrastructure, which undertook the work in discharge of UGL Rail's obligations under the Settlement Deed. It is UGL Rail's position that the cost of undertaking the work in accordance with the Settlement Deed was less than the cost of undertaking the work necessary to meet the reverberation specification and that entry into the Settlement Deed had the added advantage of capping its liability, whereas there could be no guarantee that any rectification work it undertook would achieve the specification for in-tunnel reverberation. It contends, in those circumstances, that entry into the Settlement Deed was a reasonable basis on which to settle the claim against it in respect of the defect notice relating to reverberation. It also contends that it did not itself have the resources to undertake the work it had agreed to undertake under the Settlement Deed and that, in those circumstances, it was reasonable to engage UGL Infrastructure to do that work. It submits that the amount it paid UGL Infrastructure to perform that work (in effect by causing the insurers to pay UGL Infrastructure directly) was reasonable, even though it has emerged that UGL Infrastructure was able to do the necessary work for substantially less than UGL Rail recovered from its insurers and paid (by direction) to UGL Infrastructure. 13. UGL Rail's claim raises a number of issues. The principal ones are: * whether the advice given by Wilkinson Murray in 2003 and 2007 was negligent; * whether the advice given by Wilkinson Murray in 2003 and 2007 was misleading or deceptive and in contravention of what was then s 52 of the Trade Practices Act 1974 (Cth) (TPA); * whether UGL Rail relied on the advice; * whether the settlement embodied in the Settlement Deed was a reasonable settlement of UGL Rail's liability in respect of the failure to meet the reverberation specification; * whether the amount paid by UGL Rail (by direction) to UGL Infrastructure for the work that UGL Infrastructure did was reasonable. 1. Wilkinson Murray also raises a number of specific defences to UGL Rail's claim that need to be dealt with. Those defences raise the following additional issues: * whether UGL Rail is estopped from asserting that Wilkinson Murray failed to comply with its contractual duty to provide its services with due care and skill in 2003 because, following the provision of the May 2003 advice, it was a common assumption of the parties that Alstom would install half the level of recommended acoustic panels and conduct further testing to determine whether that was sufficient before proceeding to install further panelling if necessary; * whether the claim in respect of advice given in 2003 is statute barred; * whether it was an implied term of the Consultancy Agreement that the parties would cooperate in the performance of the agreement and UGL Rail breached that implied term in various respects; * whether UGL Rail's claim is an apportionable claim within the meaning of Pt 4 of the Civil Liability Act 2002 (NSW) (CLA) and Pt VIA of the TPA and, if so, whether UGL Rail is a concurrent wrongdoer by reference to whose conduct any liability of Wilkinson Murray must be reduced; * alternatively, whether UGL Rail was guilty of contributory negligence; * whether Wilkinson Murray acted in a manner that was widely accepted in Australia by peer professional opinion as competent professional practice and, for that reason, is not liable by reason of s 5O of the CLA. Some concepts of acoustics 1. Reverberation is caused by the reflection of sound waves off surfaces – most significantly, the surfaces that contain the space in which the original sound was produced. Reverberation affects both the intensity, measured in decibels (dB), and the quality of the sound in the space. 2. A distinction is drawn between early reflections and later ones. It is the later reflections that constitute reverberation and the measurement of reverberation is concerned with those later reflections. 3. Reverberation is measured by reference to the time taken for the level of sound to reduce a specified number of decibels from the time the sound source stops. The time taken for that reduction is called the reverberation time (RT). The standard measure of reverberation time is the time taken for a sound to decay 60 dB after the sound source is stopped. It is approximately the time taken for a relatively loud sound to become inaudible. That standard measure is referred to as RT60. It is possible to plot the reduction in decibels against time. The resulting graph is called a decay curve. The measurement of RT60 is usually taken from a particular time or decibel reading after the sound source stops – for example, from a point on the decay curve that is 5 dB below the initial level to a point that is 65 dB below that level. 4. Although RT60 is used as the standard measure of reverberation, the decline in decibels is usually measured over shorter intervals and the decay curve for the measured interval is extrapolated to obtain RT60. That is particularly so where it is not practical to measure the time taken for the sound to decay by 60 dB. Two measures, which took on particular significance in this case, are RT30 – the time taken for the sound to decline in intensity by 30 dB – and RT20 – the time for the sound to decline in intensity by 20 dB. 5. When the rate of decline of the intensity of a particular sound in a space after the source of the sound is stopped is linear, it is simple to derive RT60 from RT30 and RT20. RT60 is twice RT30 and three times RT20. However, the calculation of RT60 using a simple straight line extrapolation of the decay curve obtained for RT30 and RT20 can produce substantially different results where the rate of decay of the intensity of the relevant sound is non‑linear. 6. The reverberation time is affected by the frequency of the sound (perceived as pitch), which is measured in hertz (Hz), and the acoustic properties of the relevant surfaces and, in particular, the extent to which those surfaces transmit, absorb and reflect sound waves of particular frequencies. 7. Frequencies in the audible range of sound range from approximately 20 Hz to 20,000 Hz. Most single sources generate sound in a number of frequencies. Consequently, it is usual to measure the reverberation time at different frequencies. Frequencies can be divided into bands. Octaves (and fractional octaves) are the most commonly used bands. An octave is a band where the upper limiting frequency is twice the lower limiting frequency. The centre frequency is the geometric mean of the upper level limit and the lower level limit. The centre frequencies in the audible range are 31.5, 63, 125, 250, 500, 1k, 2k, 4k, 8k and 16k Hz. It is common to measure the reverberation time of each of those frequencies, or a subset of them, when determining the reverberation characteristics of a particular space. 8. Various formulas have been developed to calculate the reverberation time of a particular space depending on the acoustic properties of the surfaces that contain it. The classical formulas were originally developed early last century by Sabine, and improved by Eyring, to deal with spaces with relatively high absorption. Those formulas recognise that the reverberation time is a function of the volume of the space and the sound absorbent characteristics of the surfaces that contain it. Consequently, the formulas can be used to calculate the amount of sound absorbent material of particular characteristics that need to be introduced to the boundaries of a space of known volume to achieve a particular reverberation time. The formulas assume an idealised space in which the reverberant soundwaves are evenly diffused in the space and the sound propagation is uniform in all directions. However, the formulas provide reasonably accurate results for rooms of typical size and shape under usual conditions. The formulas have been improved in various ways by other acoustic engineers, including Fitzroy, who developed a formula designed to deal with non‑uniform distribution of absorption. However, those improvements have generally been concerned with regular shaped enclosures. Professor Kang, who gave expert evidence for UGL Rail, and Dr Burgemeister, who gave expert evidence for Wilkinson Murray, agree that the available formulas do not produce reliable results for tunnels, which are both unusual in shape and not completely enclosed. Significantly, the reverberation in a tunnel is not diffuse, so that the determination of the reverberation time depends on the point at which it is measured, and the decay curve is not linear, so that determination of RT60 by extrapolation from RT30 and RT20 may yield substantially different results. Similarly, there is no direct relationship between the absorption levels of particular surfaces and the reverberation time. The placement of the same amount of sound absorbent material in different positions may affect the reverberation time measured at particular points. 9. The placement of sound absorbent material may also affect reverberation where the space in question is not particularly large when compared to the wavelength of the sound being considered. The wavelength is the distance over which the wave's shape repeats. If a sound wave is moving with constant speed, the wavelength is inversely proportional to the frequency of the sound: the higher the frequency, the shorter the wavelength. In simple terms, in a space which is very large compared to the wavelength of the sound being considered, most sound waves which make up the reverberant sound will have been reflected off all bounding surfaces that contain the space. Consequently, the precise placement of the sound absorbent material has less effect on which sound waves are absorbed. On the other hand, where the size of the room is not large compared to the wavelength of the reverberant sound, the precise placement of the sound absorbent material may have a significant effect on reverberation. 10. There are limits on the extent to which the level of sound in a space can be reduced by the introduction of sound absorbent material. According to DA Bies and CH Hansen, Engineering Noise Control: Theory and Practice (2nd ed, 1996, E & FN Spon), a leading textbook in the field: Experience shows that it is extremely unusual to achieve noise reductions in excess of 3 or 4 dB(a) using this form of control [ie, the introduction of sound absorbent material to the space], which can be exorbitantly expensive when large spaces or factories are involved. (p 6) The reason for that is that the introduction of sound absorbent material can only affect the reverberant sound field. It cannot affect the intensity of direct sound waves. AS2460 1. At the time of the events giving rise to these proceedings, the measurement of reverberation time in rooms was governed by a joint Australian/New Zealand Standard known as AS2460. The standard is expressed to cover the measurement of reverberation time in all types of room. 2. Clause 5.1.2 of AS2460 relevantly provides: The test sound source shall produce a sound pressure level sufficient to provide decay curves with the required minimum dynamic range without contamination by background noise. Under normal circumstances the reverberation time is evaluated from the averaged slope of the decay curve beginning from a sound pressure level of 1 dB to 5 dB below the start of the reverberant decay, or beginning 0.1 seconds to 0.2 seconds after the sound source has been switched off, for a level interval of not less than 30 dB. In the event that insufficient signal is available, it may be necessary to evaluate the reverberation time over a smaller range of the decay of sound pressure level. It is permissible to evaluate the reverberation time for a level interval of less than 30 dB, but not less than 20 dB. However, in such cases the report shall note the level interval over which the reverberation time was determined. 1. Clause 6.2.1 of AS2460 provides: Measurements of reverberation time may be required for a variety of purposes, such as assessment of the amount of room absorption for noise control purposes, assessment of the reverberation time for sound system calculations or verification of building performance against a design brief. The number of measurement positions used to characterize the reverberation time shall be chosen to achieve an appropriate coverage in the room. Clause 6.2.2 deals with measurement spacing. It relevantly provides: No microphone position shall be too close to any source position in order to avoid excessive influence from the direct sound. 1. Section 8 sets out how the reverberation time is to be evaluated from decay curves. Clause 8.1(a) largely repeats cl 5.1.2. It states: The level interval is selected by commencing 2 dB to 5 dB below the start of the reverberant decay, and ending 6 dB to 9 dB above the noise floor. The resulting level interval shall be not less than 20 dB. If the level interval is less than 30 dB, but not less than 20 dB, an appropriate statement shall be included in the test report … 1. Clause 8.1(b) sets out various means of determining the reverberant decay rate of the level interval selected. Clause 8.2 deals with non-linear decay curves. It states: In cases where the decay curve is not a straight line, a unique reverberation time cannot be determined. The following procedure shall be followed: (a) Where the decay curve takes the form of two straight lines, establish a single break point appropriate to all traces at that frequency, in terms of level relative to the initial level. (b) Measure the slope of the upper section and the slope of the lower section of the curve and the appropriate level intervals reported. The minimum acceptable level interval for measurement of each slope shall be 10 dB. … The Consultancy Agreement 1. Wilkinson Murray was retained to provide consultancy services in relation to two issues. One was achieving the reverberation times specified in the Works Brief. The other concerned advice in relation to ventilation and train breakout noise control. This case only concerns Wilkinson Murray's advice in relation to tunnel reverberation times. The Consultancy Agreement provides a separate fee for those services, which was $16,000. As will become apparent, Wilkinson Murray did a substantial amount of work in relation to tunnel reverberation times before the Consultancy Agreement was signed. 2. Clause 2.1 of the Consultancy Agreement provides: The Consultant acknowledges that the Principal relies on the Consultant to perform the Services with all due skill, care and diligence and that the Principal has entered into this Agreement relying on the representations given by the Consultant set out in clause 3.2. 1. Clause 3.1 relevantly provides: The Consultant must: (1) commence and proceed to perform the Services promptly; … (6) perform the Services: (a) with all due skill, care and diligence; (b) in accordance with the requirements of the Principal's Preliminary Requirements, the Authorities, the Law, Standards and this Agreement; (c) so as to permit the Principal to comply with its obligations under the Contract; (d) by the completion date or dates and milestone dates referred to in Item 10 of the Annexure or such other dates as are directed by the Principal from time to time or contained in the program approved by the Principal under Clause 4.1; 1. Clause 3.2(1) states that Wilkinson Murray "represents and acknowledges that it has the skill, competence, resources and experienced personnel available to perform the Services". 2. "Services" is defined to mean: [A]ll services which the Consultant is required to perform in accordance with the Agreement that are further described or particularised in the Annexure and Schedule 1, together with such other services as would ordinarily be provided by a professional consultant exercising all due skill and expertise in the performance of services such as or similar to the Services in connection with a project such as or similar to the Project. 1. The Annexure states that the Services are those set out in Schedule 1. Schedule 1 states that the Services are those defined in Specification PRL-CSA207400 Rev A. Schedule 1 also states: The Specification PRL-CSA207400 Rev A refers the Project Works Brief [sic], which is included in the Agreement. The Consultant acknowledges that it has received a copy of the Project Works Brief electronically as a PDF file. 1. Clause 3.0 of the Specification provides: All works performed under this specification shall comply with the requirements of the Parramatta Rail Link Project Works Brief … All works performed and equipment supplied under this specification shall conform in material, construction, workmanship and performance to the latest edition of the appropriate Australian Standards. If there is no Australian Standard applicable, the works shall comply with the current relevant International Standards, British Standards and relevant Codes. … In the event of lack of clarity or conflict of requirements, the contractor shall seek the ALSTOM Engineers' clarifications and approvals in writing. 1. Clause 4.2 of the Specification sets out the scope of work in relation to the control of tunnel reverberation times. Clause 4.2.1 provides: The Contractor shall perform the following works associated with the design aspect for the control of tunnel reverberation times. a) Determine tunnel sound absorbent panel requirements for control of tunnel reverberation times to meet the requirements of the Project Works Brief. b) Co-ordination of design works with ALSTOM and THJV & their consultants as necessary; including attending meetings as necessary; to establish a cost effective means for control of the tunnel reverberation times. c) Provide report(s) (including sketches/drawings) detailing the technical requirements for the sound absorbent panels, that will enable compliance with the required Tunnel Reverberation Times. This report must satisfy the the [sic] MUAP requirements for the Project and be accepted by all stakeholders. d) Review of the ALSTOM preferred tenderer's technical tender submission to ensure compliance with the technical requirements of the Wilkinson Murray Report and Project Works Brief, prior to the placing of an order by ALSTOM. 1. Clause 4.2.2 provides that Wilkinson Murray would make two inspections of the sound absorbent panels during the process of manufacture to ensure that the panels were constructed to meet any technical requirements identified by Wilkinson Murray and would provide two inspections during the process of the installation. 2. Clause 4.2.3 deals with compliance testing. It provides: The Contractor shall perform the following works associated with the compliance testing aspect for the control of tunnel reverberation times. a) Provision of a testing proposal that meets the requirements of the Project Deed; developed in conjunction with ALSTOM's other performance testing requirements; for the compliance testing of the sound absorbent panels against the tunnel reverberation time requirements. b) Testing of the tunnel reverberation time; including provision of testing equipment; to verify sound absorbent panel's compliance. c) Provision of a report demonstrating the compliance of the sound absorbent panels against the tunnel reverberation time requirements. d) Attendance at meetings as required, to satisfy Parramatta Rail Link and stakeholder expectations that tunnel reverberation time requirements have been complied with. 1. Clause 3.14.10 of the Works Brief sets out the reverberation times that had to be met. It provides: Design Criteria The reverberation times (RT60) in the tunnels must not exceed the values listed in the following table, evaluated in the absence of trains. Frequency Band 250Hz 500Hz 1kHz 2kHz 4kHz Reverberation Time RT60 (seconds) 2.5 2.2 1.8 1.5 1.2 Reference must be made to AS2640 [sic] for definitions and measurement methods for reverberation times. The dimensions, materials and locations of the acoustic panels in tunnels must be determined by the Contractor. They must be durable during the design life, and must not be prone to damage during normal maintenance activities, or susceptible to corrosion and fatigue failure. Wind generated by train operations must not cause fibreglass particles, or the like, to be released from the panels. The table in cl 3.14.10 was repeated in cl 5.1 of the Specification. The reference to "AS2640" is clearly an error. The correct reference is to AS2460. 1. The Consultancy Agreement contains two indemnities in favour of Alstom. Clause 6 provides: 6.1 The Consultant indemnifies the Principal against any loss, damage, liability, cost or expense of any kind including (without limitation) any indirect or consequential loss or damage ("Loss") incurred, or suffered by the Principal in connection with the Consultant's performance of the Agreement, any negligence of the Consultant, its employees, servants, agents or sub‑contractors or any breach of the Agreement by the Consultant. 6.2 Without limiting Clause 6.1, if the Consultant's performance of the Agreement or any negligence of the Consultant, its servants, agents and sub‑contractors or any omission or default of the Consultant causes the Principal to, or contributes to the Principal becoming, liable for the allowance or payment of any Loss or other penalties or liquidated damages under the Contract, then the Consultant shall indemnify the Principal against any and all such Loss or other penalties or liquidated damages. 6.3 The Consultant acknowledges that: (1) the provisions of the Contract under which Loss or other penalties or liquidated damages may be imposed upon the Principal are available for inspection at the offices of the Principal during normal business hours; and (2) prior to entering into the Agreement, it was given an opportunity to inspect such provisions. 1. Clause 18.1 provides: The Consultant indemnifies the Principal against: (1) any costs and expenses incurred, and losses and damages suffered by the Principal; and (2) any legal costs (on a solicitor and own client or full indemnity basis, which ever is greater) and other costs and expenses incurred by the Principal in connection with a demand, action, arbitration or other proceeding (including mediation, compromise, out of court settlement or appeal); arising directly or indirectly as a result of or in connection with any breach or non‑performance of any of the obligations of the Consultant or any negligent act or omission of the Consultant under this Agreement whether express or implied. 1. Clause 21.2(2) of the agreement provides: The Principal may at any time transfer, assign or novate or otherwise deal with the whole or any part of this Agreement. The Consultant hereby irrevocably consents to any novation of this Agreement to the Principal or its nominee and agrees to do all things (including executing the Deed of Novation) necessary or desirable to give effect to a novation of this Agreement. 1. It is also noteworthy that cl 5.1 of the Systems Subcontract also contained an obligation to design and construct the works the subject of the subcontract in accordance with the Works Brief. Under cl 16.18 of the Systems Subcontract, Alstom agreed to indemnify THJV against: (a) any liability to or claim by any other person; and (b) all costs, losses and damages suffered or incurred by the THJV, arising out of, or in any way in connection with, the Subcontractor's breach of a term of this Deed. It follows that, if the tunnel failed to satisfy the reverberation times specified in the Works Brief, Alstom, and UGL Rail as its assignee, were liable to indemnify THJV for any costs, losses and damages THJV suffered as a consequence of that failure in accordance with cl 16.18. TIDC had similar rights against THJV under the CIVSYS Deed. Background facts Events prior to May 2003 1. Alstom was involved with THJV's tender for the Parramatta Rail Project (as the ECRL Project was then known) and, in connection with the tender, it and THJV obtained preliminary advice in March 2002 from Wilkinson Murray on various noise-related issues affecting the project, including in-tunnel reverberation. Wilkinson Murray's preliminary advice was prepared largely by Mr Gross, although, according to the report, it was prepared by Mr Murray, one of its directors, and checked by Mr Bridge, who at the time was a senior engineer at Wilkinson Murray. The report appears to have been based on some calculations performed by Mr Lawrence, which used the classic Sabine formula and assumed a tunnel length of 50 m. Both Mr Murray and Mr Bridge gave evidence. Mr Gross and Mr Lawrence did not. The preliminary report was addressed to Gutteridge Haskins & Davey Pty Ltd, consulting engineers, who had been engaged by THJV in connection with its tender. In that report, Wilkinson Murray made the following recommendation in relation to reverberation control: To meet the reverberation time limits in dual track tunnels, it is recommended that sound absorbent panels be fixed to the walls of the tunnel. A strip of depth 550mm is to be applied to each side of the tunnel at a height of approximately 1m and this strip is to be continuous along the tunnel. Each panel should be constructed using an infill of 50mm thick glasswool faced with a perforated metal of open area at least 20%. The overall panel may be fixed directly to the tunnel wall. … The report went on to attach a diagram of the proposed panels and to give specifications of the sound absorption coefficients of the glasswool to be included in the panels. It was against that background that Alstom approached Wilkinson Murray to provide acoustic consulting services following THJV's successful tender and Alstom's appointment as the subcontractor responsible for the systems installation component of what became the ECRL Project. 1. Mr Murray was approached by Mr Jeremy Mortier at the beginning of October 2002 about providing consulting services in connection with the project. Mr Mortier was the engineer at Alstom who had primary responsibility for the reverberation control aspect of the project. There was correspondence between Mr Murray and Mr Mortier on the terms on which Wilkinson Murray would provide its services and Wilkinson Murray commenced work on the project well before the Consultancy Agreement was signed, although at a time when the drafting of the Consultancy Agreement was well advanced. The parties agree that Wilkinson Murray performed that work on the terms contained in the signed Consultancy Agreement. Advice in May 2003 1. In May 2003, Wilkinson Murray provided Alstom with a report concerning the quantity and type of acoustic panels to be placed in the ECRL in order to meet the contractual specifications in respect of reverberation time set out in the Works Brief. The report was prepared by Mr Bridge and reviewed by Mr Murray. It was sent to Mr Mortier on 15 May 2003. 2. The report observed: A circular tunnel is not a simple environment in which to predict reverberation time. The mode of reflections within a cylinder are significantly different from those in a more conventional rectangular room. The prediction methodologies are therefore limited in their accuracy within such an environment. The report went on to explain that, in those circumstances, Wilkinson Murray had undertaken reverberation time measurements in a section of the Eastern Suburbs railway tunnel west of Kings Cross station, which had three bands of acoustic panelling each approximately 660 mm across. The report recorded Wilkinson Murray's findings and stated: It was also found that the spread of measurement results was relatively smooth for the range of noise source and measurement locations. These findings provide a high level of confidence that the reverberation time in the Parramatta Rail Link tunnel can be consistently and accurately measured. The results also allow validation of the reverberation time prediction model. That model was explained in these terms in the report: Various adaptations were made to the [Fitzroy/Eyring] model to establish the best model configuration to predict the measured results. The best fit appears to be the use of the model without consideration of the three modes [a reference to different types of relection], but combining the three modes in a way equivalent to the original Eyring model. Taking a representative tunnel length of 30m with tunnel ends being totally sound absorbent (or open to the atmosphere) also resulted in good agreement. 1. On that basis, Wilkinson Murray recommended various options based on different types of sound absorbent material. In the case of glasswool, it recommended 1.2 m per metre of tunnel length on the basis that the glasswool was enclosed in perforated metal sheeting but was not wrapped in mylar. Wilkinson Murray also stated: In order to comply with the reverberation time requirements in the Works Brief, the acoustic absorption can be located in a single band at any position in the tunnel. However, to best meet the intended purpose it should be located in two bands one on each side of the tunnel and at a height approximately midway between the wheels and the carriage windows. 1. In reaching the conclusions expressed in the report, Mr Bridge performed various calculations using computer programs that incorporated the classical formulas, including the Eyring and Sabine formulas or variants of them. His first set of calculations was based on the calculations performed by Mr Lawrence for the preliminary report prepared in 2002 with some minor changes to the assumptions that had been made by Mr Lawrence. Those calculations used a computer program that had been used by Mr Lawrence known as NoiseWorks. Mr Bridge then prepared a second set of calculations using an excel spreadsheet, which included various formulas based on different combinations of the classic formulas. He assumed rooms of various lengths (representing the length of the tunnel) and assumed that the end walls (corresponding to the beginning and end of the tunnel) were lined with fully absorbent material. The formulas based on the Fitzroy equation produced radically different results from those that were not. 2. Mr Bridge then arranged for Mr Lawrence to carry out tests in the Eastern Suburbs railway tunnel. Mr Bridge was not present when the tests were carried out. As I have said, Mr Lawrence did not give evidence, although he was available to do so. The report states that Mr Lawrence conducted his tests in that part of the tunnel which was west of Kings Cross station. That part of the tunnel was approximately 70 m in length. Mr Lawrence carried out tests with the sound receiver at various distances, ranging from 5 m to 20 m from the sound source, and included in his calculations a mean result. The results showed that reverberation time increased with the distance between the sound source and receiver, which is something that Mr Murray says that he noted at the time, although it appears that Mr Bridge did not. Mr Bridge says that he then compared the mean results with his predictions. He adjusted some of his assumptions to obtain results that most closely corresponded to the mean results obtained by Mr Lawrence and concluded that the best fit was the model that he had created based on the Eyring formula and assuming a room of 30 m in length. He made no adjustments to that model to take account of the differences between the Eastern Suburbs rail tunnel (which is not cylindrical) and the ECRL. He then used that model to calculate the amount of sound absorbent material to be included in the ECRL to achieve the specifications set out in the Works Brief. Those calculations showed that minimal sound absorbent panelling was required. He discussed his opinion and calculations with Mr Murray and did some further calculations before finalising the report. It is apparent that the recommendations made in the report were based on the calculations using the original NoiseWorks program and an assumed tunnel length of 50 m, not on the formula that Mr Bridge says he identified as the best fit for the Eastern Suburbs rail tunnel. Mr Bridge said in cross-examination that the final recommendations were based on the NoiseWorks model because Wilkinson Murray had "decided to be more conservative in our advice". 3. Up until the time when Mr Murray gave evidence, UGL Rail had assumed, on the basis of what was stated in the report, that Mr Lawrence conducted his tests in that part of the Kings Cross tunnel which was west of Kings Cross station. However, during cross‑examination, Mr Murray gave evidence that, in fact, Mr Lawrence had conducted his tests in that part of the tunnel which is east of Kings Cross station and that the report is mistaken when it says otherwise. There is no evidence of the length of that part of the tunnel. However, it is substantially longer than 70 m. 4. I am not prepared to accept Mr Murray's evidence. It is inconsistent with what is stated in the report. He could not give a satisfactory explanation for why he had not picked up the error in the report earlier even though he knew the length of the test tunnel was an important issue in the case. No explanation was given for why Mr Lawrence was not called to give evidence. He is someone who could be expected to know the true position and Wilkinson Murray's failure to call him supports the conclusion that the report is accurate when it says that testing occurred in the tunnel west of Kings Cross station: see Jones v Dunkel [1959] HCA 8; (1959) 101 CLR 298 and the discussion of the rule derived from that case in JD Heydon, Cross on Evidence (8th Aust ed, 2010, LexisNexis). 5. Mr Mortier discussed Wilkinson Murray's report with Mr Johnson, who at the time was the engineer at Alstom responsible for the management of the project. Following receipt of the report, Mr Mortier and Mr Bridge also had a number of conversations about it. There is a dispute about what was said. Mr Mortier says that, during one of the conversations, Mr Bridge said words to the effect that "we may be able to get away with a lesser amount of material". Mr Mortier first gave evidence to that effect orally. In his original affidavit, he said that, during one of the conversations he had with Mr Johnson, he suggested that they test one row of panels and see what result they got before installing two rows. Mr Johnson agreed and Mr Mortier then raised the matter with Mr Bridge who replied "Okay". Mr Bridge accepts the version of the conversation given in Mr Mortier's affidavit. Subsequently, on 28 May 2003, Mr Mortier sent Mr Bridge a fax raising the question of what sound absorbent material was best to use from a costs point of view and whether it would best to bag the material in mylar. The fax also said: Please note that in relation to qualifying/verifying the recommendations of the report, we will intend on fitting out a section of tunnel (say 50m) and performing reverberation time measurements to determine if the second row of tunnel acoustic panels fitted to the other side of the tunnel are required. 1. I accept Mr Mortier's evidence. It is apparent that Mr Mortier was the person to first raise the possibility of testing one row of panels. It is unlikely that he would have done so unless Mr Bridge had said something to cause him to believe that it may be possible to get away with a single row of panels. It is not surprising that Mr Bridge would have said something to that effect. Mr Bridge's own analysis suggested that the recommendation in the report was a conservative one and that one row of panelling was likely to be sufficient. I do not think Mr Bridge's response ("Okay") meant that he was resiling from the recommendations made in the report. However, I think that by saying "okay" Mr Bridge intended to convey the opinion that he thought that the approach proposed by Mr Mortier was a reasonable one. 2. There was further correspondence between Mr Mortier and Mr Bridge concerning the type of material that should be used. In that correspondence, Mr Bridge advised against bagging the material in mylar. No criticism is made of that advice or of Mr Bridge's advice in relation to the sound absorbent material that should be used. 3. As I have said, the Consultancy Agreement was signed on 10 June 2003. Events following May 2003 advice 1. Following receipt of Mr Bridge's advice, Mr Mortier prepared a number of drawings showing the dimensions of the acoustic panels and their proposed location in the tunnels. The drawings allowed for one row of acoustic panels, each 0.6 m2, on the right hand side of the tunnel as shown on the drawing (opposite the safety walkway). Mr Mortier says that he allowed sufficient space for a second row of panels immediately above the first. The drawing for the panels showed perforated panels with a minimum 15 per cent open area and a maximum of 30 per cent open area. That specification was consistent with Wilkinson Murray's advice; and again, there is no criticism of that aspect of the advice. 2. It is not clear from the evidence what happened then. Mr Mortier prepared a specification for the sound absorbent panels without further assistance from Wilkinson Murray. The final version of the specification was dated 5 October 2004 and Alstom used that specification and the drawings prepared by Mr Mortier to obtain quotes from material suppliers, although Mr Mortier was not involved in that process. Mr Mortier says that he ceased to work on the project in June 2005, although it appears that he continued to have a peripheral involvement after that time. Before Mr Mortier left the project, he prepared an email stating that the design work in relation to reverberation control to be performed by Wilkinson Murray was complete and that Wilkinson Murray had yet to provide test procedures for measuring reverberation times. The email also states: In terms of tunnel sound panels, we were to fit out a section of the tunnel, perform reverberation measurements and then determine final sound panel requirements with respect to the necessity or otherwise for panels on both sides of the tunnel. 1. Following United's acquisition of Alstom's transport business in June 2005 and the novation of the Systems Subcontract to UGL Rail in September 2005, UGL Rail appointed Sound Control Pty Ltd to manufacture the acoustic panels from components obtained from other suppliers. 2. In December 2005, Wilkinson Murray provided a report addressed to Alstom setting out the procedures that Wilkinson Murray proposed to adopt for measuring noise associated with the project, including tunnel reverberation times. In relation to reverberation times, the report stated that all measurements would be performed in accordance with AS2460. 3. In about March 2006, Mr Skantha Rajendra took over responsibility at United for the reverberation control aspect of the project. On 15 March 2006, Mr Yallamas of Sound Control sent to Mr Rajendra revised drawings for the acoustic panels. The revised drawings showed panels of the size specified by Mr Mortier but with significant unperforated borders. Mr Rajendra made some comments on those drawings and the final "as built" drawings were issued on 28 March 2006. Mr Rajendra approved sample panels on 18 July 2006 and, in an email dated 13 September 2006, confirmed that United would like Sound Control to proceed with the manufacture of the panels. Advice in 2007 1. Following installation of a sample row of panels, Mr Sebastian Giglio of Wilkinson Murray conducted reverberation time tests on 21 May 2007. By that stage, most of the services had been installed in the tunnel. There were approximately 200 m of test panelling. According to Mr Giglio, the panels were 1100 x 620 mm and installed with a 100 mm overlap. The faces of the panels were perforated but with a 50 mm border on the long edges (top and bottom) and 100 mm border on the short edges, giving an effective absorptive area for each panel of 0.47 m2. In performing the tests, Mr Giglio placed the microphone approximately 70 m in from the start of the panelling and conducted various tests with the sound source ranging from 5 m to 160 m from the microphone. Mr Giglio sent a summary of the results to Mr Murray, who was working in Hong Kong at the time. They had a number of discussions concerning the results of the tests. 2. On 25 May 2007, Mr Murray sent Mr Rajendra a fax reporting on the results of the test. After observing that the test revealed reverberation times significantly in excess of those specified, Mr Murray said: The existing amount of acoustic panel in the trial section is insufficient to meet the specified reverberation time limits. It will therefore be necessary to add additional acoustic panel, and at least the second panel on the opposite side of the tunnel will be required. However, our current modeling, validated by the latest measurements, indicates that even more acoustic panel will be required. A third acoustic panel identical to the existing one is likely to be required, and this panel would need to be located at a position which is not adjacent to one of the other two panels. For example, the third panel would best be located at a high location towards the top of the tunnel. Our modeling indicates that three panels along the full length of the tunnels would result in compliance with the reverberation time limits at most frequencies, but with a small exceedance of the limits at 250 Hz. The small exceedance may not eventuate, or may be small enough to be considered acceptable. To meet the limits, we therefore recommend three acoustic panels separated around the tunnel perimeter. It may be a reasonable strategy to fit two panels for further testing, before the decision is made to incorporate the third panel. 1. Although the fax refers to modelling, there is no evidence that Wilkinson Murray had done any additional modelling that formed the basis of this opinion. 2. By this stage, Mr Ben Stelzer had taken over from Mr Johnson as the project manager, although Mr Johnson continued to maintain an overview of the project and remained responsible for resolving any major problems that arose. 3. Mr Rajendra, after considering Mr Murray's fax and discussing it with Mr Stelzer, discussed the fax with Mr Giglio. It was agreed that, as a first step, a second row of test panels would be installed immediately above the first row because of the difficulties of installing panels on the opposite side of the tunnel. 4. There was other correspondence within UGL Rail concerning the results. In particular, on 28 May 2007, Mr Stelzer sent Mr Johnson an email in which he said: I've spoken to Jeremy [Mr Mortier] about this too and he recalls that there was significant unknowns regarding how much acoustic material was required and that the decision was made to proceed with one row of panels and to do testing later on to check. If this is so, then it looks like we forgot to plan for this to be done in terms of design, time and money. See below in Skantha's email that it looks like W+M determined that they thought two rows would be required for the panel size we have chosen. It seems to me now that we will need to work out a location in the tunnel section to install another row and do additional testing to see if one row is enough 1. Following the installation of the second row of test panels, Mr Giglio conducted further tests on 4 June 2007. This time, Mr Giglio performed tests where the distance between the sound source and receiver ranged from 0.6 m to 10 m. The test revealed that reverberation time increased as the distance between the sound source and receiver increased and that, for most frequencies at all distances, the measured results were greater than the contractual specification. 2. Mr Giglio sent Mr Rajendra an email setting out the results obtained where the sound source and receiver were 1 m apart. Commenting on those results, Mr Giglio said: These are still about 20% higher than the specification. (A tolerance of 5-10% would usually be considered acceptable for a reverberation time measurement). So, either this item is negotiated with the Client or a third row of absorption panels is added. Given the current configuration I don't think a third row of panels will fit above the existing two rows. However, for new work I think it would be possible to have three rows, one atop the other, all on the one side of the tunnel. 1. Mr Rajendra responded on 5 June 2007 saying: Thankyou for your prompt response. Obviously this isn't what we wanted to hear. I have spoken to Ben (our Project Manager) and at this stage we would like to explore the following: Have two rows of panels. As we have already purchased all the material for the first row of panels I would like to keep this the same and modify the second row of panels so we can get the reverberation time to what is required by the specification. With the second row, are there changes to the panel design we can do such that we meet the requirements, eg make the panel wider (increasing the face area), use a different absorbent material, change the thickness etc. As mentioned to yourself on Monday, a third row is not really an option for us as there is hardly anywhere in the tunnel that will let us have a continuous run of panels. Even with the second row we are having to move survey marks, we clash with signalling platforms, compressed air pipework etc. Could you please explore what changes need to be made to the second row of panels such that we can meet the reverberation requirements in the spec. Mr Rajendra said that the matter was urgent. 1. Mr Giglio replied on 6 June 2007. He referred to the possibility of increasing the size of the second panel. In that context, he said: Our original calculations for the tunnel were based on 1.2m2 but the recent testing has shown that even this may be insufficient to meet the specified reverberation times, for reasons outlined by Barry Murray in his correspondence of 25 May 2007. We are confident that with 1.5m2 of absorption per m of tunnel length that the specified reverberation times will be complied with or very nearly so. 1. There is internal correspondence within United at the time that suggests that a third row of panels would not fit into the tunnel. That is certainly what Mr Stelzer said in an email dated 6 July 2007 to Mr Garry Ewen, one of THJV's employees. 2. In accordance with Mr Rajendra's email, Mr Giglio prepared a rough diagram for a larger panel. Mr Rajendra modified that drawing in consultation with Sound Control and a row of the larger panels was installed in the test section of the tunnel. Those panels were manufactured with a non-perforated border contrary to the drawings submitted by Mr Rajendra. 3. On 9 July 2007, Mr Rajendra placed an order with Sound Control for the second row of panels for the whole tunnel because the delivery of the material had a long lead time. 4. Mr Giglio conducted further tests on 9 August 2007. On this occasion, Mr Giglio took readings at distances of 0.6 m, 1 m and 2 m from the noise source. He reported on those results to Mr Murray by email shortly afterwards. The email said in part: The border was a mistake by the fabricator and the actual final panels will be perforated edge-to-edge. I think that the reverb time will come down when the whole tunnel is lined because even at 100m length of lining (50m each side of the mircrophone [sic]), sound can travel 343m in one second (yes, I am brilliant I know). I think the last time I quoted the results at 0.6m. A tunnel is not a classical Sabine space and so where is the 'reverberant field' vis-à-vis the relative measurement location source-microphone? Average T20 results 250 500 1kHz 2kHz 4kHz 0.6m 2.6 2.1 1.8 1.4 1.1 1m 2.7 2.2 1.9 1.6 1.2 2m 3.0 2.2 2.1 1.7 1.2 Goal 2.5 2.2 1.8 1.5 1.2 The Client is sweating on the results as they must get started on the other 26km in order to finish on time. Obviosuly [sic] they are hoping for no more increases in absorption 1. Mr Murray replied on the same day saying "I say that we should go with this, but with perf to the edge". 2. On 13 August 2007, Mr Giglio reported the results for the 0.6 m test to Mr Rajendra. He said: We think the results will be acceptable, based on the following: * Currently there is only 50m of absoprtion [sic] panels each side of the test location along the tunnel. Having more absorption along a greater length of tunnel will improve the result; * The final panels will be perforated edge-to-edge whereas the prototype panels still have the non-perforated border; therefore in the final result there will be more effective absorption material in the tunnel; * There is usually a tolerance in compliance assessment for reverberation time. If some frequencies are 10% higher than spec then this would be considered acceptable 1. Mr Rajendra replied that there would still be a 20 mm border for structural reasons, but that the test panels had a 50 mm border top and bottom and a 100 mm border on the sides. He asked whether Mr Giglio still believed that that would be acceptable. 2. Mr Giglio replied on 14 August 2007 in the following terms: On 8 June our recommendation was for a minimum of 1.5m^2 of 'effective' perforated panel area per 1m length of tunnel, including having perforated sides on the wider panel. Based on your advice below I understand that now there will be approximately only 1.35m^2 of effective perforated panel per m length of tunnel. The test last week was with approximately 1.3m^2. I was hoping for the additional 15% perforated area, based on our June 8 advice. This, together with a greater length of lined tunnel, was expected to provide reverberation time compliance. Based on the testing carried out I cannot be satisfied that the current proposal as it stands (as described by you below) will unequivocally achieve compliance at project completion. I think we have no other choice, but I am not comfortable and feel I cannot guarantee compliance (even though the chances are good). It appears that, in preparing this email, Mr Giglio had miscalculated the effective area of the perforated panels and, in particular, the size of the panels tested the previous week. 1. Following receipt of that email, Mr Rajendra placed a variation order for the second row of panels with Sound Control. 2. Mr Rajendra replied to Mr Giglio's email on 15 August 2007. He set out his understanding of the test results that had been achieved to date and observed that he understood Wilkinson Murray would be comfortable that the required specifications would be met if the perforated area was increased by an additional 15 per cent. The email continued: As I mentioned to you on the phone, this is not possible as there are now spatial constraints in the tunnel that are prohibiting us from increasing the height of the panels. From the table above, you will notice that with a 20mm border an extra 9% of perforated area/metre will be gained. In light of the above, could you please advise UNITED of the reverberation times expected once the panels are installed with a 20mm border. 1. Mr Giglio responded on 12 September 2007 in the following terms: It is not really possible to calculate that accurately without doing a detailed acoustical model of the tunnel. All we can really do is try to maximise the effective acoustical absorption area within the constraints available, including the non-ideal location of the panels (distributed would be better than adjacent) and the non-perforated border. At the end of the day, to satisfy the Client I would think that if the targets are met at most frequencies and are within say 5-10% of the target at other frequencies then I think that that would be sufficient. WM's last test showed that the result was just about there, the final proposed construction will have slightly more absorption area as well as the full length of tunnel being treated. WM can't guarantee the final result but based on the evidence collected and calculations that have been carried out it will just scrape in. If you want a guaranteed result then that would probably require a row of absorption on the opposite side of the tunnel. 1. Following receipt of that email, Mr Rajendra finalised the design of the second row of panels with Sound Control and issued full construction drawings on 28 September 2007. It is not clear from the evidence precisely what area of acoustic panelling was actually installed after allowance is made for the non-perforated borders and the overlap in the panels. It was less than 1.5 m2 per metre of tunnel. Mr Challis, an expert subsequently retained by United, appears to have assumed that it was 1.445 m2 per metre of tunnel. In final written submissions, the parties agreed that the total effective area of sound absorbent material that was installed at that time was approximately 1.36 m2 per metre of tunnel. Test results in 2008 1. Installation of the two rows of panels was completed on 17 January 2008, following which Mr Giglio was asked by United to conduct further testing. Mr Giglio conducted those tests on 29 January 2008. Mr Giglio reported to United on those tests on 4 February 2008. It appears from the report that the tests were carried out with distances between the source and receiver ranging from 1 m to 5 m at five different locations along the tunnel. According to Mr Giglio, the test revealed that the reverberation times were within specification at two of the five test locations and exceeded the specification at certain frequencies at three of the five test locations. 2. On 3 July 2008, Mr Giglio was informed by United that it was TIDC's view that the reverberation times in the tunnel did not comply with the specifications. Mr Giglio carried out further tests on 9 July 2008 at 39 locations in the tunnel. He reported on those results on 10 July 2008 to Mr Ian Turnbull. Mr Turnbull's precise role is not clear from the evidence. However, he was one of the United engineers responsible for managing the project. Again, the tests were carried out at various locations in the tunnel and using various distances between source and receiver, ranging from 1 m to 5 m. According to Mr Giglio's results, the reverberation times exceeded specification in a number of locations at a number of frequencies. 3. Mr Johnson returned full time to the project in June or July 2008. 4. It appears that, at some stage, Wilkinson Murray became aware that TIDC had engaged Heggies to advise it on compliance with the reverberation specification and that Heggies had conducted tests showing that the specification had not been met. In response, Mr Giglio prepared a formal report to be sent to United dealing with the issue, which he discussed with Mr Murray. The report took some time to prepare. A version of it was sent by Mr Murray to Mr Turnbull on 2 September 2008 and the final report was sent by Mr Giglio to Mr Turnbull on 3 September 2008. The report summarised results obtained from the 29 January 2008 tests and further tests that Mr Giglio had carried out on 21 July 2008. Based on the results recorded in the report, the report asserted: Reverberation time tests are considered to have met the project goals. Variations between tests can be explained by considering the fact that the reverberation time reported by the instrument is very sensitive in this particular space to where the software determines the start of the sound level decay with time. The report also stated: In the ECRL tunnel, the average absorption coefficient is around 0.1 (β=0.9) as there is approximately 1.5m of the tunnel perimeter that is treated with acoustic absorption material. If the amount of absorption material was doubled to achieve β=0.8 then the reduction in sound level in the tunnel would be of the order of 2decibels. In general, a doubling of absorption represents almost a halving of the reverberation time. Therefore, it can be seen that small differences in the reverberation time of 10%, 20% or even 50% will not have a significant effect on the resultant sound level in the tunnel. 1. On 5 September 2008, Mr Heggie sent to Mr Murray a sample of the Heggies test results and, in a subsequent telephone conversation, told Mr Murray that he (Mr Heggie) was clearly of the view that the reverberation times did not comply with specification. It appears from a note of the conversation prepared by Mr Murray that Heggies had installed additional absorption panels on the opposite side of the tunnel, which had produced compliance, and that they intended to conduct further tests with reduced panelling to determine the amount of panelling necessary to ensure compliance. 2. By this time, TIDC had also identified a problem with the in-car noise. Commenting on the situation in an email dated 10 September 2008 to Mr Johnson and others, Mr Stelzer said: Clearly this could be a serious issue. Given the problems TIDC / RailCorp are having with in train noise, I doubt we are going to wriggle out of this without some work to do. Sounds like we are not going to be able to convince Richard Heggie our results are adequate without further testing and that could easily generate some areas of non‑compliance. I'd suggest that we hold our line with TIDC for the moment 1. Heggies conducted further tests on 2 October 2008, which were attended by Mr Giglio. Mr Giglio sent Mr Turnbull a summary of the results of those tests the following day. Heggies prepared a detailed report dated 10 October 2008 on the tests. A copy of that report was provided to Mr Johnson. It is apparent from Mr Giglio's summary and the Heggies report that Heggies conducted tests at 0 m, 5 m, 10 m and 20 m from the cross‑sectional plane containing the sound source. They derived RT60 from RT30 and performed various tests using different quantities of additional sound absorbent material, ranging from no additional panelling up to three additional polyester panels, each 1200 mm high. Their tests showed that the reverberation time exceeded the specification for all distances and frequencies by a substantial margin with the existing panelling and exceeded the specification for some frequencies and distances even with three additional polyester panels. TIDC serves defect notices 1. On 10 October 2008, TIDC sent THJV a letter giving notice that the principal's representative had identified the following defect: [T]he reverberation times in the tunnels exceed the values listed in the table in Section 3.14.10 – Acoustic Panels of the Works Brief. … The letter required that the defect be rectified by 1 May 2009. It also included a copy of a report prepared by Heggies. THJV passed a copy of the letter and report on to United. 1. Mr Johnson responded to that letter and report on 15 October 2008 taking issue with the report. 2. Wilkinson Murray prepared a detailed response dated 23 October 2008 to Heggies's report. Wilkinson Murray's report explained the difficulties in applying AS2460 to tunnels. Wilkinson Murray provided United with amended versions of that report on 27 October 2008 and 3 November 2008. 3. In the meantime, on 30 October 2008, TIDC served a notice on THJV alleging various breaches of the CIVSYS Deed arising from excessive in-car noise. 4. In November 2008, UGL Rail retained Challis Consulting Pty Limited to provide a second opinion in relation to compliance with the reverberation specification. 5. On 12 November 2008, THJV sent United a copy of TIDC's defect notice in relation to in-car noise. The covering letter stated that, to the extent that THJV had any liability for the defect, THJV would look to UGL Rail. 6. In November 2008, Heggies began to investigate the effect on in-car noise of adding additional volumes of absorptive material in the tunnel, its placement in different locations, the benefit of high speed polished rail grinding techniques and the use of rail dampers. It appears from an email dated 16 October 2008 sent by Mr Johnson that TIDC was seeking to reduce in-car noise by approximately 15 dB and that it was seeking to obtain a reduction of 4 to 7 dB through the introduction of additional panelling, a reduction of 4 to 5 dB through rail dampers and a reduction of 4 to 5 dB through additional rail grinding. Establishment of working group to resolve problems with in-car noise and Mr Challis's reports 1. On 2 December 2008, TIDC wrote to THJV accepting in principle a without prejudice proposal that they and RailCorp together form a working group to proceed expeditiously to seek to agree on a technical solution to the matters the subject of the defect notice served on 30 October 2008 relating to in-car noise and to refer the dispute to mediation. In the meantime, TIDC agreed to fund the direct cost of materials and 50 per cent of the direct costs of installation with the balance of the direct costs being funded by THJV. 2. Following that agreement, Mr Johnson attended meetings of the working group as a representative of United. The working group consisted of representatives from TIDC, including its technical adviser, Heggies; RailCorp, including its technical adviser, Sinclair Knight Merz; THJV, including its technical adviser, Mr Renzo Tonin; and United and its technical adviser, Mr Louis Challis. The focus of the working group was on the reduction of in-car noise. The target set by TIDC was to achieve a sound level of 80 dB for 80 per cent of the journey time in the vestibule of a K-set train. For the remaining 20 per cent of the journey, the in‑train noise levels were not to exceed 85 dB. The working group considered various options for achieving the goals set by TIDC and investigations were conducted by members of the working group into the merits of each option and an estimate of their costs. 3. Independently of that work, Mr Challis conducted reverberation testing in early December 2008 and provided United with a draft report on 8 December 2008 setting out the results of his tests. The draft reported the following results using RT30 to derive RT60: Specified Contract Average of 10 decays Specified Frequency Hz Requirement RT60 @ one position on 2/12/08 (125 Hz – not specified Not specified 8.13 (at 160Hz) 250 2.5 4.62 500 2.2 2.63 1k 1.8 2.25 2k 1.5 2.16 4k 1.2 1.68 8k Not specified 1.12 1. After observing that the measurements were particularly high for low frequencies, the report observed that "the achievement of the conforming reverberation time at 250Hz will dominate the scope of the amelioration program required to achieve conformance of the five specified frequencies". Mr Challis expressed doubt that that specification could be achieved using the existing panels and suggested utilising a supplementary panel configuration that was thicker than the one that had been installed. 2. On 23 December 2008, United received advice from Mallesons Stephen Jaques to the effect that, if there was a failure to meet the specified reverberation times, United would be liable for the cost of work to rectify that breach but that it was more likely than not that United would not be liable in relation to the allegations set out in the defect notice in respect of in-car noise. 3. Mr Challis prepared a further draft report for United dated 8 January 2009. That draft, like the previous one, concluded that the tunnels did not meet the reverberation specification. It also stated that it was not possible at that stage to express a view of how much additional acoustic panelling would need to be added to meet the specification. 4. Mr Challis produced his final report on 29 January 2009. That report included a section setting out Mr Challis's opinion on the amount of additional acoustic panelling that was required in order to meet the specification. Mr Challis's recommendation was summarised in the following table: Frequency Parameter 250Hz 500Hz 1kHz 2kHz 4kHz Height of unity absorption wall lining required for RT30 compliance 4750mm 2400mm 3600mm 3600mm 3600mm Height of unity absorption wall lining required for RT20 compliance 2500mm 1445mm 2645mm 3000mm 2800mm 1. In expressing that opinion, Mr Challis relied on the results of the tests conducted by Heggies, which were set out in their report dated 10 October 2008. It is not easy to follow what Mr Challis did. It appears that he obtained from tests conducted by RMIT the precise absorption coefficients for the panels that were already installed and the material used by Heggies at each test frequency so that he could calculate the effective height of the panels. So, to take a simple example, if the height of the panel was 1200 mm and its absorption coefficient for a particular frequency was 0.5, Mr Challis treated the effective height of the panel for that frequency as being 600 mm. Mr Challis calculated the effective heights for the existing panels and for the additional one, two and three panels tested by Heggies for each frequency. He then plotted the effective heights against the reverberation time to reach a conclusion concerning the effective height of panelling required for each frequency. In the case of the 250 Hz frequency, Mr Challis calculated the effective height of the existing panels and three tests panels as 3496 mm. With that amount of panelling, RT60 (based on RT30) was 3.3 seconds. In that case, Mr Challis appears to have projected his graph to conclude that, in order to reach a reverberation time of 2.5 seconds (the amount specified in the Works Brief), it would be necessary to install panelling with an effective height of 4750 mm, although Mr Challis's calculations appear to contain an arithmetic error. That error aside, assuming that the same panelling was used, that equated to additional panelling of 6765 mm. According to Mr Challis's calculations, substantially less panelling was required to meet the specifications for other frequencies, as is apparent from the table set out above. 2. On the same day Mr Challis produced his final report, Mr David Anderson of RailCorp distributed an email to members of the working group, including Mr Johnson, purporting to summarise the various options, their relative advantages and disadvantages and their estimated costs. The email included a table substantially in the following form: Option Scenario Total Pre- tender Cost Estimate ($M) Av % of Trip over Target Rank (in terms of cost & target compliance) A No Dampers – All Absorption 22.4 41% 6 B All Absorption w Dampers at Selected Locations 23.4 21% 5 C Dampers at Selected Locations – no Lower Walls Abs. 20.8 22% 3 D All Dampers with Optimised Absorption 20.1 20% 1 E All Absorption Except 4ft – Dampers at Selected Locations 20.8 20% 2 F Dampers – All 4ft, No Lower Wall & Min Under WW Abs (3.1km) 21.6 21% 4 1. At a meeting of the working group on 10 February 2009, it was agreed that the preferred technical solution was option D, which appears also to have included rail grinding. Following that meeting, tenders were issued to various suppliers for rail dampers and acoustic panels. The mediation 1. There was a mediation between TIDC and THJV that commenced on 11 March 2009 and ran for several days. United was not a party to the mediation but agreed to attend it. 2. Prior to the mediation, on 10 March 2009, Mr Johnson prepared a cost estimate comparing the cost of option D (referred to as the "TIDC Solution") and various options proposed by Mr Tonin, Heggies and Mr Challis to meet the reverberation specification. Mr Johnson said in his first affidavit that he concluded that the cost of the TIDC solution was approximately $21,241,000 (assuming that TIDC would be responsible for the supply of the dampers and making certain other assumptions), the cost of the Tonin solution was $22,417,700, the cost of the Heggies solution was substantially the same as the Tonin solution, and the Challis solution would be approximately $1,000,000 less. In performing those calculations, Mr Johnson assumed that the cost of installing acoustic panelling in accordance with the various solutions per metre of additional panelling was the same as the cost per metre of installing acoustic panelling in accordance with the TIDC solution. 3. Mr Johnson prepared a note of his calculations. The note relevantly was in the following terms: In Car Noise Solution Labour Material Vehicles/Misc Total Dampers $1,260,500 $5,922,000 $689,000 $7,871,500 Floor Panels $1,540,700 $1,860,000 $689,000 $4,089,700 Wall Panels $4,201,800 $3,800,000 $1,378,000 $9,379,800 TOTAL $7,003,000 $11,582,000 $2,756,000 $21,341,00 [sic] TONIN Solution 97,200 sq metres of Acoustisorb2 or 3. 21,600m long by 4.5m high. The In car noise solution has 33,900 lineal metres of panel. Approximating the Tonin solution to 3.75 rows of panels by 1200mm high, gives 3.75 x21,600m or 81,000 lineal metres. Prorata the "Wall Panels" above gives 81,000/33,900 =2.39 Wall Panels $10,042,300 $9,082,000 $3,293,400 $22,417,700 x2.39 The discrepancy between Mr Johnson's affidavit and his note is likely to be a typographical error. The note went on to explain that the estimate in relation to the Heggies solution was a "guestimate based on TIDC position paper stating 3 rows of additional panels bigger and thicker than those currently installed". The note also stated that Mr Challis had been "suitably vague" in defining a solution. However, in reaching a cost estimate, Mr Johnson assumed that there would be a total of 3500 mm of acoustic panelling per metre with additional material behind the existing panelling. 1. Mr Johnson says that his estimate was likely to be an under-estimate because he used a pro rata figure and had not made any allowance for the additional costs associated with the logistical difficulty of installing additional rows of panels. 2. Mr Johnson also prepared a spreadsheet setting out his calculations for the costs of the TIDC solution and a summary of those calculations. The summary was in the following terms: Acoustic Treatment Summary of Cost Estimate Item Manufacturer Total Supply Cost Dampers S&V $ 5,922,500 4 foot blocks Adbri $ 1,801,910 Cladding Advance Metal $ 2,892,176 Glasswool Fletcher Insulation $ 666,393 Bedding mortar for blocks (assumed) $ 54,894 Fastenings for wall panels (assumed) $ 240,999 Grand Total Material Supply $ 11,578,872 Total Labour and Plant Labour Costs $ 5,698,000 (tunnel, yard, excludes drivers) Supervisory Costs + Site Overheads $ 1,305,000 (supervisor, eng, safety) Hi-rail vehicle costs $ 2,732,000 (includes driver) Forklift Costs $ 24,000 Grand Total Installation Labour & Plant $ 9,759,000 TOTAL $ 21,337,872 Contingency 10% 2,133,787 Overhead & Profit 17.5% 4,107,540 GRAND TOTAL $ 27,579,199 1. Mr Gary Boyd, who was a quantity surveyor engaged by United's insurers, also undertook some preliminary check calculations of the TIDC solution. His draft calculations indicated that the costs would be approximately $33.5 million – higher than those estimated by Mr Johnson. According to Mr Boyd, the main differences were the assumptions he made concerning the number of acoustic panels required, labour efficiencies and the rate of overhead and profit. In calculating the amount that should be allowed for labour, Mr Boyd took into account the fact that reduced access meant shorter shifts since shuttle trains would be running in the tunnels. Mr Boyd had not, at that stage, prepared comparison costings of a solution designed simply to meet the reverberation specification. 2. It is not entirely clear what happened at the mediation. United was represented by Mr Tim Burt, its General Counsel, who did not give evidence, and Mr Johnson. Mr Boyd also attended the mediation. It is apparent that TIDC was pressing THJV to agree to the TIDC solution and to pay the costs of that solution. One issue that was discussed at the mediation was the comparative costs of the TIDC solution and the costs of meeting the reverberation specification. Reporting to Mr Andrew Summers on that issue, who at the time was the Chief Executive of UGL Infrastructure, Mr Burt said in an email dated 17 March 2009: The rectification works proposed by TIDC ('TIDC Works') are designed to address the in-car noise issue. These works will involve the installation of additional acoustic panels (approx 38,000 m2) on the walls, acoustic blocks between the tracks and dampers on the rails. The works that would be necessary to meet the specified reverberation times ('Reverberation Works') are different to the TIDC Works, and would involve the installation of additional acoustic panels (approx 97,000m), but no acoustic blocks or dampers. The TIDC Works have been costed at $22 - 25m, while the Reverberation Works have been costed at $25 - 27m (both exclusive of contingency, margin and overhead). 1. Other issues that were discussed included how the TIDC solution would be implemented. One issue was the particular suppliers of materials, such as rail dampers and the acoustic panels. Another important issue was the extent to which THJV would be given access to the tunnels to undertake rectification work. At the time, RailCorp was threatening to limit access to four days per week. 2. Although the dispute did not settle at the mediation, it was agreed in principle between TIDC and THJV to adopt the TIDC solution, with TIDC agreeing to pay for the costs of the dampers and THJV being responsible as between them for most of the other costs. Mr Johnson says that he concluded at the mediation that it was in United's interest to agree to the TIDC solution because: * it provided certainty, since TIDC has accepted the design risk; * there was no guarantee that any reverberation solution proposed by the experts would meet the specification; * the cost of any reverberation solution was uncertain and would have increased significantly if it was necessary to install panels at height; * it would be extremely commercially and practically difficult to undertake the works required to comply with the reverberation specification; * as part of the TIDC solution, TIDC agreed to waive any claim for liquidated damages and agreed to release the security. Events following the mediation 1. Following the mediation, there were further negotiations between the parties. Mr Simon Campbell took over the project management responsibilities from Mr Johnson and one of his responsibilities was to prepare a detailed costing of the TIDC solution. Mr Campbell prepared that costing in conjunction with Mr Johnson, who remained involved with the project until June 2009, when he moved to Brisbane. Mr Campbell says that the costing was prepared on the same basis as United would have used if it had been tendering for the work. It took into account quotes that had been received for materials and made a number of assumptions concerning access to the tunnel. The total cost of the estimate was $36,718,564. 2. At the same time, Mr Boyd also revised his own cost estimate and prepared cost estimates for solutions directed at meeting the reverberation specification. Mr Boyd's final report, or at least what appears to be a final draft of that report, is dated 2 June 2009. In the report, Mr Boyd says he compared the TIDC solution embodied in what was then a draft of the Settlement Deed with various proposals to meet the reverberation specification. Mr Boyd said that a number of those were too expensive to report on in detail. He concentrated on a solution which he described as the "Leo/Challis Option 2", which was based on a report prepared by Mr Leo dated 12 March 2009. Mr Leo was from Marshall Day, who had been engaged by United's insurers. The report referred to by Mr Boyd was not in evidence. Mr Boyd described the Leo/Challis Option 2 solution in these terms in his report: a. In the original solution proposed by Leo / Challis, Ross Leo proposed an additional 3.91m of 100mm thick acoustic material around the girth of the tunnel … b. The Challis 2 solution amended this advice by saying that the girth could be reduced by approximately 0.4m assuming that additional acoustic material could be placed behind the existing reverberation treatment originally fixed in the tunnel. … c. The revised solution proposed that 2 additional acoustic panels (1 x approx 500mm wide and 1 x approx 900mm wide) could be placed behind the existing panels with a new steel cover plate being used to re-fix the panels back to the tunnel wall. This would avoid the need for approximately 9,500m2 additional steel panels. 1. Mr Boyd estimated that the costs of the TIDC solution based on the terms of the draft Settlement Deed excluding any allowance for overheads and profit were $25,247,799. That figure only related to the costs of the rectification work. It did not include other payments to be made by UGL Rail under the Settlement Deed. Mr Boyd's estimate for implementing the Leo/Challis Option 2 solution was $34,116,831, although in an email dated 26 May 2009 he had estimated those costs as being $31,959,137. The explanation for the difference is not clear from the evidence. 2. On 3 June 2009, DLA Phillips Fox, on behalf of UGL's insurers, sent Mallesons Stephen Jaques an offer to settle UGL Rail's claim for $25,027,000. The offer was calculated as follows: United Claim Underwriter's Offer 1 Direct Costs $25,271,750 $24,836,129 2 An allowance for provision of bank guarantees for longer than required by the contract and an allowance for defect correction $76,043 $nil 3 Sub Total (1 +2) $25,347,793 $24,836,129 4 Clause 11.3 Payment $2,000,000 $see below 5 Clause 6.3 Delay Costs1 $2,916,000 $see below UGL Productivity $2,099,034 6 Direct Labour 77.7 hours @ $27,000 $211,758 $included in 1 Management 3/24 of $1,694,064 7 General Contingency – 2.5% of costs (excluding materials) $417,854 $411,670 8 Sub Total (4+5+6+7) $7,664,646 $411,670 9 Overheads 7.7% of Total Costs $1,951,780 $1,614,350 10 Margin 7% of total cost $1,774,346 $0 11 Sub Total (9+10) $3,726,793 $1,614,350 12 Total (3+8+11) $36,718,564 $26,826,149 13 Less Allowance for early receipt of funds $800,000 14 Total (13 + 14) say $26,027,000 15 Less Excess $1,000,000 15 Total $25,027,000 [sic] 1 There is no clause 6.3 of the Deed. We assume that this is actually intended to refer to clause 6(b). Please let us know if that is correct 1. DLA Phillips Fox went on to explain the circumstances in which the insurers would make the payments referred to in 4, 5 and 6. It also said they would consider paying THJV's and TIDC's costs but only to the extent that they "relate to RT60 defect". 2. Mr Johnson discussed the proposed settlement with Mr Summers. It appears that, some time after receiving the insurers' offer, Mr Summers spoke to Mr Richard Leupen, the Chief Executive Officer of United, and recommended acceptance of the insurers' offer and entry into the Settlement Deed. In making that recommendation, Mr Summers said he formed the view that the settlement was reasonable because: * the advice UGL Rail had received was that it was liable to THJV for works not meeting the reverberation specification; * UGL Rail was not required to conduct any tests or meet the requirements of any test or measurement in respect of the completion of the works the subject of the proposed settlement; * UGL Rail was not required to accept any design responsibility for the TIDC solution. It simply had to undertake the work referred to in the Settlement Deed. * the works to be undertaken in the Settlement Deed were less expensive than works required to comply with the reverberation specification. 1. Mr Leupen accepted Mr Summers's recommendation. 2. On 10 June 2009, UGL Rail accepted its insurers' offer and, on 23 June 2009, entered into the Settlement Deed. 3. Under the terms of the settlement, TIDC directed THJV to carry out certain work and, pursuant to cl 8.2(b) of the Systems Subcontract, THJV directed UGL Rail to carry out the variation works at no cost to THJV. It was specifically agreed that United did not assume or undertake any design responsibility or warrant any of the design characteristics in respect of the variation works except in relation to "Fixings" (that is, the means by which absorbent panels were to be fixed to the walls, absorbent blocks were to be fixed to the trackbed and glasswool batts were to be fixed to cable trays). The Deed specified various dates by which certain portions of the work was to be completed. 4. Under cl 11(a) of the Settlement Deed, United agreed to pay THJV the sum of $1,867,402.30 on account of TIDC's and THJV's costs of and incidental to the dispute and, under cl 11(b), a further $1,251,048.33 in respect of THJV's costs of attending to the variation works. United also agreed to pay a further $2,000,000 to THJV if United did not complete the works on time. 5. Work had in fact commenced on implementing the TIDC solution in late May 2009. The quantum experts retained by the parties (Mr Makin, who was retained by UGL Rail, and Mr Young, who was retained by Wilkinson Murray) agreed that the total costs of doing that work were $13,668,109, excluding any profit component and GST. However, during the course of the hearing, spreadsheets were produced by UGL Rail that indicate that the total costs incurred by UGL Infrastructure in completing the work (excluding any profit component and GST) were $14,940,268.95. UGL Rail's claim 1. UGL Rail's total claim is $37,361,892, calculated as follows: Item Amount Cost of initial two rows $5,641,585.61 Amount paid to UGL Infrastructure $25,027,000 TJHV and TIDC costs $3,430,295.69 UGL Rail's legal costs $358,717.85 Total $34,457,599.15 Less cost of installing current panelling $8,028,058 Total $26,429,541.15 Plus interest from 12 June 2009 to 21 November 2014 $10,932,350.86 Total $37,361,892 1. One observation to make about this claim is that UGL Rail has not deducted from the claim the additional costs it would have incurred if it had installed the correct amount of acoustic panelling originally. Rather, what it has done is to include as damages the costs of installing the original two rows of panelling on the basis, presumably, that those costs were wasted but to deduct the total costs of installing the panelling recommended by Professor Kang in order to meet the reverberation specification. There may be a question whether that was the correct approach. However, it was not suggested in this case that the two approaches lead to substantially different results. Did Wilkinson Murray breach a duty it owed UGL Rail? The allegations 1. The allegation that Wilkinson Murray breached a duty that it owed UGL Rail has two limbs. First, it is alleged that the services provided by Wilkinson Murray in 2003 were not performed with due care and skill. The relevant services are said to be the provision of the May 2003 report and the provision of advice on the question whether it would be reasonable to install a single test row of panels rather than installing the recommended quantity of panelling immediately. A long list of particulars is given for why the services were not performed with due care and skill. Essentially, though, what is alleged is that Wilkinson Murray did not understand the acoustic properties of the tunnels and, for that reason, made incorrect recommendations about the amount of sound absorbent material that it would be necessary to install in order to meet the reverberation specification. It is also alleged that Wilkinson Murray engaged in misleading or deceptive conduct. The relevant conduct is said to be the representations alleged to have been made in cls 3.2(2)(a), 3.2(2)(b) and 3.2(2)(c) of the Consultancy Agreement (that the services would be free from defects in design (cl 3.2(2)(a)), that they would be suitable for their intended purpose and of good quality (cl 3.2(2)(b)) and satisfy the requirements of the Works Brief (cl 3.2(2)(c))) and representations in the report that the acoustic absorption could be located in a single band in any position in the tunnel and that, to best meet the intended purpose, the acoustic absorption should be located in two bands, one on each side, at a height approximately midway between the wheels and the carriage windows. 2. The second limb relates to the services provided in 2007. Those services consist of the advice given concerning the extent to which the two rows of panels that were ultimately installed would meet the reverberation specification. It is alleged that those services were not performed with due care and skill. Again, a long list of particulars are given. Essentially, though, the complaint is that Wilkinson Murray used inappropriate testing procedures. In particular, it used RT20 to derive RT60 when it should have used RT30; and it used the distances between sound source and receiver that were close to zero when it should have used a range of distances up to 20 m. It is also alleged that Wilkinson Murray engaged in misleading or deceptive conduct. The allegations based on cls 3.2(2)(a), 3.2(2)(b) and 3.2(2)(c) are repeated, although it is difficult to make much sense of those allegations in this context. To the extent that those clauses contained representations, those representations were made years earlier. UGL Rail alleges that Wilkinson Murray made a number of other representations. Those representations, in substance, are: * three panels along the full length of the tunnels would result in compliance with the reverberation specification at most frequencies with a small exceedance at 250 Hz, which may not eventuate and which may be small enough to be considered acceptable; * the reverberation time measurements following the fitting of a second row of test panels exceeded the specification by about 20 per cent and that exceedance would need to be negotiated with the client or a third row of panels would need to be installed; * Wilkinson Murray was confident that, with 1.5 m2 of absorption material per metre of tunnel length, the specified reverberation times would be complied with or very nearly so; * whilst Wilkinson Murray could not guarantee the results with two panels of the dimensions that it was proposed to install, based on the evidence and calculations that had been carried, what was proposed would just scrape in and to guarantee the results would probably require an additional row of absorption on the opposite side of the tunnel. The 2003 services 1. There is a question of how it could be said that Wilkinson Murray breached a duty it owed to UGL Rail in respect of the 2003 services when those services were provided to Alstom. It is true that Alstom's rights under the Consultancy Agreement were assigned to UGL Rail. However, Alstom did not purport to assign any claim it may have had for negligence or under the TPA; and there is no evidence that Alstom had suffered any loss as a consequence of the breach of duty. This question was touched on by Wilkinson Murray in its final written submissions, although it was never seriously pursued. In my opinion, the answer to the question is to be found in the way in which UGL Rail puts its case. UGL Rail maintains that, as a consequence of the assignment of the rights under the Consultancy Agreement and Wilkinson Murray's consent to the assignment, UGL Rail was entitled to rely on advice given by Wilkinson Murray prior to the assignment; and that is what it did. I deal with the question of reliance below. However, it seems to me that, as a consequence of what happened, Wilkinson Murray made the same promises to UGL Rail as it did to Alstom in respect of the advice that it had given. In giving that advice in the circumstances it did, it owed a duty of care not only to Alstom but to any assignee of Alstom's rights; and if it engaged in misleading or deceptive conduct in 2003, the only question is whether UGL Rail relied on that conduct in 2007. 2. As to the question whether Wilkinson Murray breached a duty that it owed, in my opinion, any acoustic engineer acting with reasonable care and skill would have appreciated that a circular tunnel with a length of approximately 13 km was likely to have unusual acoustic properties. Appreciating that fact, such an engineer who was unfamiliar with the acoustic properties of tunnels would have undertaken some research on that topic before making recommendations on the sound absorbent material to be installed in a tunnel as a means of achieving specified reverberation times. Both Professor Kang and Dr Burgemeister agreed that it was important for a consultant acoustic engineer to be aware of current theory for design and that, if faced with an unfamiliar space, it was important for the consultant to undertake research or seek advice. Mr Murray conceded in cross-examination that that was the position. He gave the following evidence: Q. I take it that you agree that in order to start designing for such an unusual space your company would need to become familiar with the type of space you were dealing with? A. Yes. Q. And I take it the best way to become familiar would be to undertake some research so as to discover what fellow engineers in Australia and around the world designing for this type of space were doing? A. Yes, probably. 1. It was also apparent from literature that was readily available at the time that two features of a tunnel are that reverberation increases along its length and that the decay curve is not linear, with the result that the classic formulas used for predicting the effect on reverberation time of the installation of certain quantities of sound absorbent material are unreliable. That was also obvious from the preliminary calculations performed by Mr Bridge and the results of the testing Mr Lawrence did on the Eastern Suburbs railway tunnel. The various calculations Mr Bridge performed using the classic formulas produced such different results that that must have raised a question concerning their application. The results that Mr Lawrence obtained showed that reverberation time increased with the distance between receiver and sound source. Mr Murray appreciated that, although Mr Bridge did not. Mr Bridge purported to use the results of the tests to calibrate his "model" but, in fact, the recommendations made in Wilkinson Murray's report were not based on that calibrated model. That could only have been because Wilkinson Murray recognised that the calibrated model itself was unreliable. All those things would have indicated to an acoustic engineer acting with reasonable care and skill that the classic formulas could not be relied on. 2. Knowing that, in my opinion, an acoustic engineer acting with reasonable care and skill would have concluded that the only reliable ways of determining the amount of acoustic panelling required were to conduct a computer simulation, to build a scale model or to install sample panelling in the tunnel in question and determine the necessary amount of acoustic panelling through a process of trial and error – in effect, using the tunnel itself as a model. However, in my opinion, an acoustic engineer acting with reasonable care and skill would have appreciated that, because of the size of the tunnel and the fact that the reverberant sound field was not diffuse, the precise placement of the acoustic panels was likely to be important and that, consequently, a trial and error approach using the actual tunnel would make it more difficult to determine the optimal location of sound absorbent panels. 3. In making the recommendations that it did in its May 2003 report, Wilkinson Murray did not conduct any modelling or try any combination of panelling in the actual tunnel. Consequently, it was not in a position to make those recommendations. 4. Wilkinson Murray submits that the conclusions of the previous paragraphs are mistaken for two reasons. First, it submits that it was unusual at the time for acoustic engineers in Australia to build a scale model or to use computer modelling. It relies on evidence given by Dr Burgemeister that the use of physical acoustic scale modelling is very unusual and is not (and was not) in common use by Australian consultants. Although Wilkinson Murray accepts that programs were available at the time to permit computer modelling, such as Odeon, they were expensive and not in common use – a submission supported by evidence from Dr Burgemeister. Consequently, it submits that it was not negligent in failing to use computer modelling. In making these submissions, Wilkinson Murray relies on s 5O of the CLA, which relevantly provides: (1) A person practising a profession (a professional) does not incur a liability in negligence arising from the provision of a professional service if it is established that the professional acted in a manner that (at the time the service was provided) was widely accepted in Australia by peer professional opinion as competent professional practice. (2) However, peer professional opinion cannot be relied on for the purposes of this section if the court considers that the opinion is irrational. Second, Wilkinson Murray submits that the advice contained in the May 2003 report was only preliminary in nature and it was always intended by the parties that it would be refined at a later stage following testing of the results achieved in the actual tunnel. 1. I accept that computer modelling and scale modelling were not commonly used in Australia in 2003. However, computer modelling was available and it appears that a number of firms of acoustic engineers in Australia had the capacity to undertake computer modelling at that time. Moreover, it is not to the point to say that computer modelling was not common in Australia at the time. Nor was that the problem that had to be addressed. The evidence is that the ECRL was the first tunnel of its type for which acoustic panelling had to be designed to control reverberation; and the only other railway tunnel in Australia containing acoustic panelling was the Eastern Suburbs railway tunnel. To say that computer modelling was not commonly used to solve common problems says nothing about what an acoustic engineer acting with reasonable care and skill would do when faced with the unusual problem that confronted Wilkinson Murray. 2. In any event, neither of the points made by Wilkinson Murray addresses the essential difficulty with the May 2003 report. Contrary to Wilkinson Murray's submission, the report was not expressed to be preliminary in nature and it was not qualified in any way. Rather, it asserted that the approach it took provided "a high level of confidence that the reverberation time in the Parramatta Rail Link tunnel can be consistently and accurately measured". It may be that computer and scale modelling were not common at the time. But absent modelling of one of those types or actual testing in the tunnel, there was no basis for the assertion that Wilkinson Murray made in its report. That was obvious from the testing that Wilkinson Murray actually did, which showed that the classic formulas did not produce reliable results for tunnels and which, when used to select a formula, produced results that were so counterintuitive that Wilkinson Murray itself chose not to use the selected formula as a basis for the recommendation made in its report. Wilkinson Murray was negligent because it put forward recommendations concerning the acoustic panelling that should be installed in a way that suggested those recommendations were reliable and had a rational basis, whereas any acoustic engineer acting with reasonable care and skill would have appreciated they were not and did not. 3. In my opinion, it was also negligent for Wilkinson Murray to have advised that it was sensible to test one row of panels before installing a second row. For the reasons I have given, any acoustic engineer acting with reasonable care and skill would have appreciated that the test results could not validate the use of a variant of one of the classic formulas to predict reverberation times in tunnels generally. Rather, such an engineer would have concluded that any coincidence between the results obtained in the tunnel and the formula used by Mr Bridge was just that – coincidence. When Mr Bridge accepted that it was sensible to test one row of panels, he was conveying the impression that the advice given by Wilkinson Murray was conservative and that it may be possible to get away with less acoustic panelling. He had no rational basis for that view. 4. For similar reasons, Wilkinson Murray's recommendations were misleading and deceptive. Having regard to the context in which the representations were made, the representations concerning the amount of panelling that should be installed and its location carried with them a representation that those representations had a rational basis and that it would be reasonable to rely on them, whereas, for the reasons I have given, they did not and they could not be relied on. 5. On the other hand, I do not accept that clauses 3.2(2)(a), 3.2(2)(b) and 3.2(2)(c) of the Consultancy Agreement amounted to misleading or deceptive conduct. In my opinion, they were not conduct. Rather, they were simply contractual promises setting out what Wilkinson Murray agreed to do: Raphael Shin Enterprises Pty Ltd v Waterpoint Shepherds Bay Pty Ltd [2014] NSWSC 743 at [84]-[86] per Sackar J and cases there cited; see also Lawfund Australia Pty Ltd v Lawfund Leasing Pty Ltd [2008] NSWSC 144; (2008) 66 ACSR 1 at [59]-[61] per Brereton J. The 2007 services 1. The 2007 services have two components. The first is the testing of the various configurations of sample panelling. The second is the recommendations that Wilkinson Murray actually made. There is an obvious connection between the two since the first drove the second. However, it will be helpful to consider them separately. 2. Wilkinson Murray submits that the 2007 services fell outside the scope of the Consultancy Agreement, that it owed no duty of care in relation to the provision of those services and that it did not engage in misleading or deceptive conduct in connection with them. 3. In support of the first contention, Wilkinson Murray submits that there is nothing in the Consultancy Agreement that contemplated a "test and see" approach. In its submission, the agreement certainly did not contemplate a test and see approach where there were restrictions placed on what could be done because many of the services had already been installed in the tunnel with the result that there was limited space for additional acoustic panelling. What the Consultancy Agreement contemplated was the provision of a concept design and the testing of that design once it had been installed. 4. In support of the second contention, Wilkinson Murray submits that UGL Rail has not satisfied the requirements of s 5B of the CLA. That section provides: (1) A person is not negligent in failing to take precautions against a risk of harm unless: (a) the risk was foreseeable (that is, it is a risk of which the person knew or ought to have known), and (b) the risk was not insignificant, and (c) in the circumstances, a reasonable person in the person's position would have taken those precautions. (2) In determining whether a reasonable person would have taken precautions against a risk of harm, the court is to consider the following (amongst other relevant things): (a) the probability that the harm would occur if care were not taken, (b) the likely seriousness of the harm, (c) the burden of taking precautions to avoid the risk of harm, (d) the social utility of the activity that creates the risk of harm. 1. In my opinion, there is no merit in the suggestion that the work performed by Wilkinson Murray in 2007 did not fall within the scope of the Consultancy Agreement. It is true that what was actually done by Wilkinson Murray did not correspond precisely to the three stages set out in the Specification. However, the definition of "Services" is not limited to that work. It includes "such other services as would ordinarily be provided by a professional consultant exercising all due skill and expertise in the performance of services such as or similar to the Services in connection with a project such as or similar to the Project". The work done in 2007 by Wilkinson Murray falls within that description. It was part of the services provided by Wilkinson Murray to assist UGL Rail to determine what quantity of acoustic panelling should be installed in the tunnels so as to meet the reverberation specification. That work went to the heart of the work that Wilkinson Murray was engaged to perform. 2. It is unclear on what basis Wilkinson Murray says that the requirements of s 5B of the CLA were not satisfied. There was a clear risk of harm to UGL Rail if Wilkinson Murray did not take reasonable care both in conducting its testing and making recommendations concerning the amount of acoustic panelling that should be installed. That risk was foreseeable and not insignificant. It is obvious that the requirements of s 5B(1) are met. 3. It is desirable to return to the question whether Wilkinson Murray engaged in misleading or deceptive conduct in the provision of the 2007 services after considering the question of breach of contract and negligence. 4. As I have said, the question whether Wilkinson Murray exercised reasonable skill and care in carrying out the testing in 2007 centred on two issues. One was whether it was appropriate to use RT20. The other was whether it was appropriate to use only the results for tests where the receiver was in or close to the perpendicular plane containing the sound source. 5. Wilkinson Murray submits that the choice of RT20 or RT30 required an exercise of judgment. It was reasonable for Wilkinson Murray to choose RT20 because the choice of that measure was more consistent with the purpose behind the reverberation specification. TIDC was seeking to control reverberation as a means of controlling in-car noise. RT20 better reflected the effect of reverberation on in-car noise. In addition, it was apparent that UGL Rail was concerned about costs. The selection of RT20 meant that the reported reverberation times were less, with the result that less acoustic panelling needed to be installed. The selection of RT20 in those circumstances was supported by Dr Burgemeister, who expressed the view that: [S]ince the early part of the decay is most relevant to the resulting noise levels within the tunnel, it would seem most reasonable to adopt the T20. Dr Burgemeister's conclusion was itself supported by the international standard for measuring reverberation, ISO 3382-2:2008. That standard was published in 2008. However, according to Wilkinson Murray, it still provides a benchmark of what was reasonable at the time. It expressed a preference for RT20. In doing so, it gave the following reasons: a) The subjective evaluation of reverberation is related to the early part of the decay; b) For the estimation of the steady-state sound level in a room from its reverberation time, it is appropriate to use the early part of the decay; and c) The signal-to-noise ratio is often a problem in field measurements, and it is often difficult or impossible to get an evaluation range of more than 20 dB. This requires a signal-to-noise level of at least 35 dB. 1. In my opinion, these submissions misunderstand the nature of the task that Wilkinson Murray was to undertake. Wilkinson Murray was engaged to assist UGL Rail to comply with the specifications set out in UGL Rail's contract with THJV. Wilkinson Murray must have understood that, if UGL Rail failed to comply with those specifications, it would be exposed to a claim for rectification work which potentially could run into millions of dollars. The question was whether it was reasonably open to Wilkinson Murray to interpret the contractual specification as requiring measurements by reference to RT20 in those circumstances. In my opinion, it was not. The specification in the Works Brief, which was expressly incorporated into the Systems Subcontract, set out the requirements for RT60 for certain specified frequencies. The Works Brief specifically stated that "[r]eference must be made to AS2640 [sic] for definitions and measurement methods for reverberation times". Although it is accepted that AS2460 is more appropriate for rooms rather than tunnels, it was not suggested that it could not be applied in measuring reverberation in the ECRL. 2. In my opinion, cl 5.1.2 of AS2460 (quoted in para 26 above) is unambiguous. It means that RT30 should be used if it can be. If it cannot be used (because insufficient signal is available), it is permissible to use RT20 but, in that event, the report must note that that is what has been done. In my opinion, an acoustic engineer acting with reasonable care and skill would not have interpreted cl 5.1.2 in any other way and would have appreciated that RT30 had to be used unless it could not be because insufficient signal was available. There is no suggestion that that exception applies in this case. 3. Wilkinson Murray submits that non-linear decay curves are governed by cl 8.2(b) of AS2460 and that that clause indicates that it is appropriate to use a decay curve of at least 10 dB. It is not easy to understand cl 8.2(b) in the absence of expert assistance. However, whatever precisely it means, I do not see how it can be interpreted as permitting the measurement of reverberation time over a range of less than 30 dB, or 20 dB where a measurement over 30 dB is not possible. Rather, it is concerned with how the decay curve should be reported where it cannot be reported as a single straight line but can be reported as two straight lines. It has nothing to do with the range over which the decay curve should be measured. Neither of the experts nor any of the witnesses suggested that cl 8.2(b) had any application to this case. 4. There is a suggestion in Wilkinson Murray's submissions that it was necessary for it to choose RT20 because UGL Rail was constrained in the amount of additional panelling that could be installed in the tunnels and it would have been impossible for UGL Rail to meet the specification if RT30 had been used. That submission takes various forms and goes to reliance as well as negligence. It will be necessary to say more about the submission later. However, to the extent that it is said to go to the question whether Wilkinson Murray was negligent, in my view, the point is irrelevant. Wilkinson Murray's task was to report accurately on reverberation times so that UGL Rail knew whether it complied with the specification or not and could make sensible decisions about what it would do in the event of non-compliance. By using RT20 rather than RT30, Wilkinson Murray failed to do that. 5. A similar analysis applies to the choice of the distance between sound source and receiver. Wilkinson Murray submits that it was appropriate to choose a distance along the tunnel of zero (or something close to it) because that best captured the effect of reverberation on in-car noise. Professor Kang disputed that conclusion. He pointed out that, in the case of a train, the source generating the initial sound was not a point but a line; and, in those circumstances, it was his view that the reverberant sound field as a whole would have a substantial effect on in-car noise. 6. Professor Kang's evidence strikes me as plausible and I accept it, although there may be a question – not answered by the evidence – of how significant an effect the reverberant field at various distances has on in-car noise. Whatever the precise effect, the question is what was required by the contract and, in particular, whether the approach taken by Wilkinson Murray was reasonably open having regard to the terms of the specification and AS2460 and the fact that the reverberant sound field in the tunnel as a whole would have an effect on in-car noise. In my opinion, the approach taken by Wilkinson Murray was not reasonably open. 7. Clause 6.2.1 of AS2460 relevantly provides "[t]he number of measurement positions used to characterize the reverberation time shall be chosen to achieve an appropriate coverage in the room". The question is how this requirement is to be applied to a tunnel. In a normal room, the reverberant sound field is largely diffuse. Consequently, it is possible to take measurements at a number of points and to take the average of those measurements as providing a fair characterisation of the reverberant sound field in the room. In the case of a tunnel, reverberation time increases with the distance between the sound source and point of measurement. Consequently, both Professor Kang and Dr Burgemeister agreed that it was not appropriate to average the results obtained at different points along the tunnel from the sound source. On the other hand, it is not appropriate to choose a single point along the tunnel since the results will vary depending on the point chosen. To choose any single point will give a misleading picture of the sound field in the tunnel as a whole. In those circumstances, in my opinion, an acoustic engineer acting with reasonable care and skill would report on a number of points that, taken together, fairly described the acoustic characteristics of the space. It is noteworthy that that is what Wilkinson Murray did when it took its earlier readings. When the specification states that the reverberation time must not exceed specified amounts for certain frequencies, it must be understood as stating that the specification must be met for all points that, when taken together, fairly describe the acoustic characteristics of the space. There is no doubt room for differences of opinion about the range of distances over which measurements should be taken. However, I do not think that it was reasonably open for Wilkinson Murray to conclude that the acoustic properties of the tunnel were fairly described by reporting only on results obtained when the sound source and receiver were in the same perpendicular plane, or close to it. It is apparent that Wilkinson Murray chose those results not because they fairly described the characteristics of the tunnel but because they suggested that the specifications would be met. 8. As I have said, UGL Rail complains about a number of statements Wilkinson Murray made during the course of the testing in 2007. However, the critical statement was the last statement made by Wilkinson Murray before United installed panelling throughout the tunnel. If UGL Rail relied on any statement, it must have been that one. UGL Rail must have understood that Wilkinson Murray's final advice superseded any previous advice that it had given. 9. Wilkinson Murray's final advice was given in Mr Giglio's email dated 12 September 2007. In that email, Mr Giglio said that the last tests showed that the result was "just about there", that Wilkinson Murray could not guarantee the final result but that "based on the evidence collected and calculations that have been carried out it will just scrape in" and that, if United wanted a guaranteed result, "that would probably require a row of absorption on the opposite side of the tunnel". 10. In my opinion, Wilkinson Murray did not have a reasonable basis for making any of those statements and an acoustic engineer acting with reasonable care and skill would not have made them. The results were only "just about there" because Wilkinson Murray was reporting on the basis of inappropriate measurements. The opinion that United would "just scrape in" was based on the same flawed approach. It was clear that United would not just scape in if Wilkinson Murray used RT30 and appropriate measurement distances. Mr Giglio had no basis for saying that, if United wanted a guaranteed result, it would probably need to install a third row of panels on the opposite side of the tunnel. He had performed no calculations from which it could reasonably be concluded that a third row of panels would guarantee compliance. Indeed, it appears that he had performed no calculations at all. For the reasons I have given earlier, in my opinion, such a statement could only be made on the basis of computer modelling, a scale model or actual tests in the tunnels. 11. Wilkinson Murray suggests that this is not enough to establish that it was negligent. What UGL Rail had to do, and did not do, was prove that its advice that three panels were sufficient was wrong. I do not accept that submission. It was still negligent for Wilkinson Murray to suggest that there were good grounds for believing that two panels were likely to be sufficient. Moreover, as I will explain shortly, I am satisfied based on Professor Kang's modelling that three panels would not have been sufficient and that, in order to meet the reverberation specification, it was necessary to install a total of 3000 mm of acoustic panelling per metre of tunnel, which is well in excess of the three panels recommended by Wilkinson Murray. 12. Again, for similar reasons, in my opinion, the statements in the 12 September 2007 email were misleading. In the context in which they were made, they were opinions that carried with them an implied representation that Mr Giglio had a reasonable basis for making them, which he did not. Causation 1. Section 5D of the CLA provides: (1) A determination that negligence caused particular harm comprises the following elements: (a) that the negligence was a necessary condition of the occurrence of the harm (factual causation), and (b) that it is appropriate for the scope of the negligent person's liability to extend to the harm so caused (scope of liability). (2) In determining in an exceptional case, in accordance with established principles, whether negligence that cannot be established as a necessary condition of the occurrence of harm should be accepted as establishing factual causation, the court is to consider (amongst other relevant things) whether or not and why responsibility for the harm should be imposed on the negligent party. (3) If it is relevant to the determination of factual causation to determine what the person who suffered harm would have done if the negligent person had not been negligent: (a) the matter is to be determined subjectively in the light of all relevant circumstances, subject to paragraph (b), and (b) any statement made by the person after suffering the harm about what he or she would have done is inadmissible except to the extent (if any) that the statement is against his or her interest. (4) For the purpose of determining the scope of liability, the court is to consider (amongst other relevant things) whether or not and why responsibility for the harm should be imposed on the negligent party. 1. Section 5A(1) of the CLA provides: This Part applies to any claim for damages for harm resulting from negligence, regardless of whether the claim is brought in tort, in contract, under statute or otherwise. The Part to which s 5A(1) refers includes s 5D. Therefore, s 5D must be applied in any case to which Pt 1A of the CLA applies: Adeels Palace v Moubarak [2009] HCA 48; (2009) 239 CLR 420 at [48]. 1. Section 5D reflects a "structural approach" to the question of causation (Zanner v Zanner [2010] NSWCA 343; (2010) 79 NSWLR 702 at [4]) by separating causation into two elements: "factual causation" and "scope of liability": Adeels Palace v Moubarak at [42]; Wallace v Kam [2013] HCA 19; (2013) 250 CLR 375 at [12]; Paul v Cooke [2013] NSWCA 311; (2013) 85 NSWLR 167 at [83]. The distinction drawn in s 5D(1) between factual causation and scope of liability "should not be obscured by judicial gloss": Wallace v Kam at [14]. 2. Section 5D(1)(a) involves the determination whether the negligence complained of was "a necessary condition of the occurrence of harm", which involves nothing more or less than the application of the "but for" test of causation: Wallace v Kam at [14]; Strong v Woolworths [2012] HCA 5; (2012) 246 CLR 182 at [18]; Adeels Palace Pty Ltd v Moubarak at [45]. The effect of s 5(2) is that there are some "exceptional" cases where "legal causation is established even though the plaintiff cannot prove that the defendant's negligence was a necessary condition of the occurrence of the harm": Paul v Cooke at [84]. In those cases, the court is to consider (amongst other relevant things) whether or not and why responsibility for the harm should be imposed on the negligent party: s 5D(2). 3. In addition, s 5D(1)(b) requires the plaintiff to establish that it is appropriate for the negligent party's liability to extend to the harm so caused. A determination of the scope of liability under subs (1)(b) is, in contrast to the factual inquiry under subs (1)(a), "entirely normative": Wallace v Kam at [14]; Hudson Investment Group Ltd v Atanaskovic [2014] NSWCA 255; (2014) 311 ALR 290 at [105]. It is intended to "extend beyond what have traditionally been seen as elements of causation, to cover questions raised by intervening and successive causes, foreseeability and remoteness": Monaghan Surveyors Pty Ltd v Stratford Glen-Avon Pty Ltd [2012] NSWCA 94 at [69]; see Zanner v Zanner at [6] for further considerations that might apply to a determination under s 5(1)(b). It has been noted that, although the two questions have been delineated by s 5D(1), "questions of factual causation and scope of liability…do not readily fall into separate and independent watertight compartments": King v Western Sydney Local Health Network [2013] NSWCA 162 at [34]. Therefore, the strength of a causal connection (or any other factual matter relevant to factual causation) might be relevant to, and relied on in, the court's determination of what is "appropriate" under s 5(1)(b): Paul v Cooke at [110]. 4. The High Court held in Wallace v Kam that, in a case falling in an established class, the normative inquiry under s 5(1)(b) is to be answered through the application of precedent: id at [22]. In a novel case, s 5D(4) provides that, in determining the scope of liability, the court is explicitly to consider and explain in terms of legal policy whether or not, and if so why, responsibility for the harm should be imposed on the negligent party: id at [23]. The normative considerations discussed above in relation to s 5(1)(b) are also relevant to subss (2) and (4) of the CLA: Zanner v Zanner at [6]. 5. The test stated in s 5D makes explicit what was implicit in the common law's approach to questions of causation in negligence. As the High Court explained in Wallace v Kam (at [11]): The common law of negligence requires determination of causation for the purpose of attributing legal responsibility. Such a determination inevitably involves two questions: a question of historical fact as to how particular harm occurred; and a normative question as to whether legal responsibility for that particular harm occurring in that way should be attributed to a particular person. The distinct nature of those two questions has tended, by and large, to be overlooked in the articulation of the common law. In particular, the application of the first question, and the existence of the second, have been obscured by traditional expressions of causation for the purposes of the common law of negligence in the conclusory language of "directness", "reality", "effectiveness" and "proximity". (footnote omitted) 1. Although the Ipp Report intended s 5D to provide "helpful legislative guidance" to judges in explicitly articulating reasons for imposing liability or not (The Negligence Review Panel, Review of the Law of Negligence: Final Report (2002) at [7.49]), it did not intend generally to change the common law. Despite that intention, Allsop P in Zanner v Zanner noted that, to the extent that the structure of s 5D conforms more closely to the approach of McHugh J in March v E & MH Stramare Pty Ltd [1991] HCA 12; (1991) 171 CLR 506 rather than that of Mason CJ, it would tend to indicate that it did: at [5]. In March v E & MH Stramare, McHugh J adopted the "scope of risk" test stated by Denning LJ in Roe v Minister for Health [1954] 2 QB 66 at 85: Starting with the proposition that a negligent person should be liable, within reason, for the consequences of his conduct, the extent of his liability is to be found by asking the one question: Is the consequence fairly to be regarded as within the risk created by the negligence? If so, the negligent person is liable for it: but otherwise not. (McHugh J's emphasis.) 1. His Honour explained that this test "enables relevant policy factors to be articulated and justified in a way which is not possible when responsibility is limited by reference to commonsense notions of causation or to more specific criteria such as 'novus actus interveniens', 'sole cause' or 'real cause', all of which conceal unexpressed value judgments": March v E & MH Stramare at 536. 2. Leeming JA in Paul v Cooke recognised that s 5D, particularly subss (2) and (4), plainly enough reflects a substantive change to the way in which courts are to determine issues: at [41]. His Honour observed (at [105]) that the approach of McHugh J in March v EH Stramare Pty Ltd at 535 is now reflected in s 5D(1)(b). Whether or not the approach taken under s 5D will produce a different result to the approach under common law has not been decided: Adeels Palace v Moubarak at [44]; Strong v Woolworths Ltd at [19], [28]. In any case, whilst it would be incorrect to substitute the language of the statute with common law principles, the above statements indicate that the approach of McHugh J is "helpful in applying the open ended language in s 5(1)(b) and (4)" of the CLA: Hudson Investment Group Ltd v Atanaskovic at [106]. 3. It is clear that s 5D applies to claims in contract and negligence. It is not entirely clear that it applies to a contractual indemnity in respect of loss arising from negligent conduct since, strictly speaking, a claim under an indemnity is not a claim for damages; and the right to an indemnity depends on the form in which the indemnity is given rather than on the principles of common law relating to the recovery of damages as those principles have been stated or modified by the CLA. It seems clear that s 5D does not apply to claims for damages under s 82 of the TPA for misleading or deceptive conduct in contravention of s 52. Section 82 of the TPA provides that a person who suffers loss or damage "by conduct of another person" done in contravention of s 52 may recover the amount of that loss or damage. In using the expression "by", s 82 adopts its own test of causation: see Allianz Australia Insurance Ltd v GSF Australia Pty Ltd [2005] HCA 26; (2005) 221 CLR 568 at [99] per Gummow, Hayne and Heydon JJ; Travel Compensation Fund v Tambree [2005] HCA 69; (2005) 224 CLR 627 at [30] per Gleeson CJ, at [54] per Kirby J; Henville v Walker [2001] HCA 51; (2001) 206 CLR 459 at [18] per Gleeson CJ, at [96] per McHugh J; I & L Securities Pty Ltd v HTW Valuers (Brisbane) Pty Ltd [2002] HCA 41; (2002) 210 CLR 109 at [26] per Gleeson CJ, at [84] per McHugh J. That test is not displaced by s 5D of the CLA. As Basten JA pointed out in Monaghan Surveyors Pty Ltd v Stratford Glen-Avon Pty Ltd [2012] NSWCA 94 (at [79]): The claim under the Trade Practices Act will not be subject to s 5D, because s 82, which provided the remedy by way of damages for breach of s 52, was not constrained by State law, unless that law were picked up and applied by another Commonwealth law: Insight Vacations Pty Ltd v Young [2010] NSWCA 137; 241 FLR 125 at [85]–[91]. It was not suggested in this case that any relevant Commonwealth law had that operation in respect of s 52 of the Trade Practices Act (as then in force). Nor was it suggested in this case that any relevant Commonwealth law had that effect. 1. The application of s 5D to contractual indemnities and claims for damages for contravention of s 52 of the TPA is not purely theoretical since, in the present case, Wilkinson Murray asserts that UGL Rail has not satisfied the requirements of s 5D(1)(a) and UGL Rail seeks to rely on evidence given by Mr Johnson of what he would have done if Wilkinson Murray had not been negligent, which arguably is excluded under s 5D(3)(b). Before addressing the issue, it is convenient to consider the position in respect of claims for breach of contract and negligence. 2. A number of questions of causation arise in this case. The first is whether Alstom and UGL Rail relied on the advice given by Wilkinson Murray in 2003 and 2007. That question raises an issue of what s 5D describes as "factual causation". The second, which I have already mentioned, is whether it is appropriate for the scope of Wilkinson Murray's duty to extend to the losses said to arise from the advice Wilkinson Murray gave. The third is connected with the question whether UGL Rail acted reasonably in settling the claim made against it and, in effect, paying UGL Infrastructure a fixed sum to perform the rectification work. Those issues of causation are dealt with in that context. Did UGL Rail rely on Wilkinson Murray's advice? 1. Wilkinson Murray submits that UGL Rail has not established that it relied on Wilkinson Murray's advice. If it had relied on the 2003 advice, it would have installed two rows of panels on opposite sides of the tunnel instead of the single row of test panels it did. Similarly, Wilkinson Murray submits that UGL Rail did not rely on the 2007 advice. If it had done so, it would have installed three rows of panels, one of which would have been on the opposite side of the tunnel to the other two. 2. In considering the question of reliance, it is necessary to identify the loss in respect of which UGL Rail sues. That loss is the loss arising from the service of the defect notice in respect of the failure to meet the reverberation specification and THJV's claim that UGL Rail was responsible for that loss. The loss arose because the acoustic panelling installed by UGL Rail was inadequate. The question, then, is whether, in installing the panelling it did, UGL Rail relied on the advice given by Wilkinson Murray and whether, had it been given non-negligent advice, it would have acted on that advice so as to avoid the loss it says arises from service of the defect notice. If UGL Rail would have acted in the same way whether it received negligent or non-negligent advice, then it cannot be said that the negligent advice was a necessary condition of the loss in respect of which UGL Rail sues. The question of reliance in this case, then, depends not only on what UGL Rail did, but on what it would have done if it had been given non-negligent advice: see Tanna v Deutsche Bank (Asia) AG [1996] NSWSC 408 (upheld on appeal in Tanna v Deutsche Bank (Asia) AG (Court of Appeal (NSW), 15 September 1998, unrep)). That, in turn, raises the question of what form the non-negligent advice would have taken. It is convenient to start with that question. This is not a case where Wilkinson Murray, in order to avoid liability for negligence or for misleading or deceptive conduct, could simply have remained silent. 3. UGL Rail submits that, if Wilkinson Murray had exercised reasonable care and skill in 2003, it would have engaged in computer modelling using a software package such as Odeon and, if it had done so, it would have concluded that it was necessary to install 3000 mm of acoustic panelling per metre of tunnel (with the same absorption coefficients as the panelling actually installed), part of which would have to have been located on the opposite side of the tunnel. In making that submission, UGL Rail relies on extensive modelling conducted by Professor Kang using Odeon software. The results of Professor Kang's work are that 3000 mm of panelling was necessary and that it could be installed in a number of configurations. The configuration recommended by Professor Kang based on the work that he did that most closely resembles what was actually done was described by him as Simulation 1. In that simulation, 2000 mm of acoustic panelling was placed on the right-hand side of the tunnel roughly in the position of the panelling actually installed by United. A further 1000 mm was installed at a position approximately 180 degrees around the circumference of the tunnel from the 2000 mm of panelling. According to Professor Kang's simulation, if that amount of panelling had been installed in those positions, the reverberation specification would have been met at all frequencies when measured by reference to RT30 and at distances between sound source and receiver along the track ranging from 0 m to 20 m. 4. Wilkinson Murray did not serve any expert evidence taking issue with the computer simulations conducted by Professor Kang. It did seek shortly before the hearing to have a number of issues relevant to Professor Kang's simulations considered in an expert conclave between Professor Kang and Dr Burgemeister and to lead evidence on those matters at the hearing. However, I refused those applications on the basis that, if Wilkinson Murray wished to challenge the evidence given in Professor Kang's expert reports through evidence of its own, it should have filed that evidence in advance of the hearing. 5. Professor Kang was cross-examined on his simulations and, in its final submissions, Wilkinson Murray contended that Professor Kang's simulations were overly conservative for a number of reasons. 6. First, in conducting the simulations, Professor Kang took the absorption coefficients for the acoustic material at the test frequencies obtained by RMIT and discounted those by 20 per cent. He did so because those coefficients had been obtained in a test laboratory and his experience was that the absorption coefficients in practice were unlikely to match those obtained in laboratory conditions. That evidence strikes me as plausible. It is plain both from his reports and the evidence he gave that Professor Kang is an extremely knowledgeable and competent acoustic engineer who gave his honest and considered opinions on the technical issues in the case. I accept his evidence. 7. Second, Wilkinson Murray submitted that Professor Kang's report was overly conservative because Professor Kang assumed that the surface of the tunnels was smooth concrete, not rough concrete. It submitted that it was more appropriate to assume that the surface was rough concrete to allow for the effect of the various services that had been installed in the tunnel on the propagation of sound waves in the tunnel. In cross-examination, Professor Kang explained that he had chosen a scattering coefficient of 0.2 to take account of the services and that it was not appropriate to assume that the surface of the tunnel was rough concrete when, in fact, it was smooth concrete. He had conducted various sensitivity analyses and concluded that the choice of scattering coefficient had a limited effect on his results. Again, I accept Professor Kang's evidence. 8. Wilkinson Murray also submits that Professor Kang was overly conservative because he used RT30 rather than RT20 and determined the reverberation time for distances between sound source and receiver up to 20 m. I have already explained why in my opinion it was necessary to choose RT30 and to choose a variety of distances at which to take measurements. Professor Kang explained that he chose distances of up to 20 m because his modelling indicated that reverberation time increased for distances up to 20 m and beyond but that, by 20 m, the reverberant sound field had largely stabilised. No issue was taken with that evidence. Consequently, in my opinion, it was appropriate to use a distance of up to 20 m in order fairly to describe the reverberation in the tunnels. 9. The question remains whether, in order to have given non-negligent advice in 2003, Wilkinson Murray would have had to engage in computer modelling at that time. I do not think that there is a simple answer to that question. For the reasons I have given, I think that the preferable course would have been to engage in computer modelling. The tunnels were complex spaces from an acoustic point of view. The modelling undertaken by Professor Kang demonstrates that it would be necessary potentially to consider a large number of alternative solutions involving different quantities of acoustic panelling placed in various positions in the tunnel. It is likely to have been impractical to try all the possible combinations in the tunnel itself. On the other hand, computer modelling was not regularly undertaken in Australia at the time and Wilkinson Murray itself did not have the appropriate software and would either have had to acquire it or subcontract that part of the work. The fees charged by Professor Kang are not in evidence. However, having regard to the amount of work he did in preparing his own model, I think that it could be inferred that the costs of creating and testing a reliable computer model would be substantial. Having regard to those matters, it seems to me that it was open to Wilkinson Murray to advise Alstom/UGL Rail that there was no simple way of determining the amount of acoustic panelling that was required, that it was possible to conduct computer modelling but that Wilkinson Murray did not have the appropriate software, that computer modelling would be expensive and that an alternative approach may be to wait for the tunnel to be built and to rely on tests once that had happened. However, if advice had been given in those terms, Wilkinson Murray would also need to have said that, without computer modelling, it may be difficult to find the optimal solution. It could not have suggested that two panels would be sufficient. If it gave advice that it would be sensible to start with one panel, it would have to have given advice that a single panel may be wholly inadequate and that it may be necessary to install panels on both sides of the tunnel. 10. The starting point for considering what form non-negligent advice would have taken in 2007 is the advice actually given in 2003 and the fact that United had installed one row of test panels. In my opinion, the advice that an acoustic engineer would have given acting with reasonable care and skill in 2007 would have been similar to the advice that Wilkinson Murray ought to have given in 2003. In order not to have been negligent, Wilkinson Murray would have to have used the correct basis on which to report the reverberation times obtained with the one row of test panels. That would have required it to use RT30 and a range of distances between sound source and receiver up to 20 m. Reporting in that way, Wilkinson Murray would have concluded that the reverberation time far exceeded the specifications. It was open for it to say at that time that it would be necessary to carry out tests with additional panelling on both sides of the tunnel and to use a trial and error process. However, for the reasons I have given, if it adopted that approach, it would have been necessary to explain to UGL Rail that, because of the number of potential combinations, it may be difficult to find the optimal solution and that a more efficient way of doing that would be to conduct computer modelling. 11. Against that background, it is now possible to address the question of reliance. 12. The advice given in 2003 was that Alstom could comply with the reverberation specification by installing two rows of panels on the same side of the tunnel but that a reasonable course was to install one row of test panels to see whether that would be sufficient. The non-negligent advice would have been that it was not possible to determine the amount of panelling that should be installed without conducting computer simulations (or building a model) or testing on the actual tunnel; and that computer simulations would be preferable. The advice was given to Alstom. However, in circumstances where it agreed to an assignment of the Consultancy Agreement and that agreement was assigned to UGL Rail, Wilkinson Murray must be taken to have agreed that UGL Rail was entitled to rely on the advice that it had given. The chain of causation on which UGL Rail relies is somewhat complicated. Wilkinson Murray gave its initial advice. According to UGL Rail, Alstom planned to install one row of test panels on the basis of that advice. Subject to some changes in personnel, the same employees continued with those plans as employees of UGL Rail. The fact that they did not go back to read Wilkinson Murray's advice or find out what Mr Bridge had said to Mr Mortier does not alter the fact that what UGL Rail did in 2007 it did because of the advice given in 2003; and by agreeing to the novation, Wilkinson Murray agreed that it was entitled to do so. I accept this part of UGL Rail's analysis. It seems to me artificial to say that UGL Rail did not rely on the advice because it was given to Alstom when UGL's conduct was simply a continuation of Alstom's plans and Wilkinson Murray agreed to the assignment. 13. However, the question remains whether Alstom and UGL Rail would have acted differently in a way that would have avoided the loss if Alstom had been given the non-negligent advice. I am not satisfied that they would have. There is no evidence that Alstom/UGL Rail relied on the 2003 advice – and, in particular, the advice that two panels would be sufficient – in deciding how to configure the other services in the tunnel. It is possible that, if Alstom/UGL Rail had been given the advice that I have concluded they ought to have been given, they would have decided that it was desirable to engage in computer modelling. Mr Johnson gave the following evidence on that subject: If I had been told in May 2003 that in order to properly prepare a design to comply with the Reverberation Specification it was necessary for Wilkinson Murray to undertake additional computer modelling or further physical scale modelling (in addition to the fixed sum that Alstom were charged), then I would have agreed to additional computer modelling or physical scale modelling being undertaken. So long as I was satisfied that it was necessary to carry out computer modelling in order to prepare a design that met the Reverberation Specification, I would have agreed to Wilkinson Murray undertaking further modelling even though I considered modelling to be within the scope of the original consultancy allocation. 1. There is a question whether this evidence is admissible under s 5D(3)(a) of the CLA. In AI Mclean Pty Ltd v Hayson [2008] NSWSC 927 at [245], Bergin J (as her Honour then was) took the view that s 5D(3)(a) applied where the evidence was given by a director of a company about what the plaintiff company would have done. However, as Villa points out, it is difficult to reconcile that approach with the actual working of the section: see D Villa, Annotated Civil Liability Act 2002 (NSW) (2nd ed, 2013, Lawbook Co) at 152-3. The section appears to be restricted to cases where the person giving evidence is the person who suffered the harm. It does not obviously apply where the person suffering the harm is a corporation; and it is not clear that it can be used to exclude evidence given by any employee of what he or she might have done. 2. In my opinion, it is not necessary finally to resolve that issue. Mr Johnson's evidence is based on the proposition that the reverberation specification could only be complied with if Wilkinson Murray had undertaken additional computer modelling or physical scale modelling. Mr Johnson does not explain why testing on the tunnel itself would not have been acceptable. Consequently, little weight can be placed on his evidence. On the available evidence, it seems equally possible that, if Alstom had been given non-negligent advice, it would still have decided to install one row of tests panels and see what the results were before deciding what to do and UGL Rail would have continued to plan on that basis and would have installed a single row of test panels. That approach would have been consistent with Alstom/UGL Rail's obvious desire to save money. It was to be expected that they would have had some number of months from the time the first row of test panels was installed until the installation of the panelling had to be complete. There is no evidence from which it could be inferred that Alstom would have believed that it would have had insufficient time to conduct computer modelling or to install additional panelling if that is what it had to do. Consequently, it may well still have been prepared to see what the results were with a single test panel before deciding what to do and the likelihood is that UGL Rail would simply have proceeded to implement that decision. 3. The position is different in 2007. The final advice given in 2007 was to the effect that, although Wilkinson Murray could not guarantee that the installation of the two rows of panels as proposed would mean that the tunnel would satisfy the reverberation specification, based on the evidence and Wilkinson Murray's calculations, it would "just scrape in", and that, in order to guarantee the results, it was probably necessary to install a third row of panels on the other side of tunnel. The non-negligent advice would have been that additional panelling would need to be installed on both sides of the tunnel and that it was preferable to determine the optimal amount and location of panelling by computer modelling but it could be done by trial and error. 4. In my opinion, if UGL Rail had been told that two rows of panelling was inadequate, it would have agreed to further tests or computer modelling to determine what amount of panelling was required and it would have installed that panelling. It may, at that stage, have sought to renegotiate the specification that it had to comply with. Whatever the precise position, it would have taken steps to ensure that it complied with its contractual obligations while it had unrestricted access to the tunnel. It knew that it was bound by its contractual obligations. It knew that, failing everything else, THJV could have engaged someone else to perform the work and that it would be liable for those costs. It must have appreciated that the costs of the work would be greater once train services started. In those circumstances, it is inconceivable that it would not have taken steps to comply with its contractual obligations immediately. 5. Wilkinson Murray submits that UGL Rail did not rely on the advice it gave in 2007. It puts that submission in two main ways. First, it submits that UGL Rail did not follow the advice that Wilkinson Murray actually gave. It was told that Wilkinson Murray could not guarantee compliance if United simply proceeded with the two panels. Nonetheless, it decided to proceed with that configuration and, in doing so, it elected to take the risk of non-compliance itself. Second, it submits that it was not physically possible to install additional panels in the tunnel. 6. I do not accept either of those submissions. As to the first, it is true that UGL Rail did not do what Wilkinson Murray said was necessary to achieve compliance. Rather, it took the risk that the two panels would not be sufficient. However, it is one thing to take a risk believing that the risk is small. It is another thing not to take steps to achieve compliance knowing that compliance could not be achieved with two panels. As to the second, Wilkinson Murray relies on various statements made by employees of UGL Rail to the effect that space in the tunnel was limited and it was not possible to install additional panelling on the opposite side. However, those statements must be understood in context. They were made in a context where UGL Rail believed on the basis of the advice it had received from Wilkinson Murray that the tunnel was close to compliance with the two rows of panelling that had been tested. It was obvious that UGL Rail was putting pressure on Wilkinson Murray to come up with a solution that only involved two panels where it believed that it was close to compliance. The position would have been entirely different if UGL Rail had been told that the tunnels would fall well short of compliance with the two rows of panelling whatever their precise configuration, as subsequent events have demonstrated. The objective evidence demonstrates that there was room to install additional panelling on both sides of the tunnel. The scope of Wilkinson Murray's duty in 2007 1. Wilkinson Murray submits that the scope of its duty did not extend to the loss in respect of which UGL Rail sues – that is, the losses arising from non-compliance with the reverberation specification. In making that submission it points to a number of factors, including the following: * UGL Rail did not follow Wilkinson Murray's advice in 2003; * UGL Rail elected not to obtain advice from Wilkinson Murray on the design of the metal panels; * Wilkinson Murray's 2003 report was irrelevant to what UGL Rail decided to do in 2007; * UGL Rail did not accept Wilkinson Murray's advice in 2007 and decided to install panels throughout the tunnel before that precise configuration had been tested; * UGL Rail did nothing for eight months after it was told that the tunnels did not comply with the specifications; * the fees paid to Wilkinson Murray were relatively small; * the costs of appropriate software to do computer modelling would have been greater than the fees charged by Wilkinson Murray; * the Consultancy Agreement contemplated that Wilkinson Murray would only obtain professional indemnity insurance with a limit of $10 million; * Alstom allowed approximately $9 million to install acoustic panelling and the evidence suggests that it spent less than that doing so; * UGL Rail and UGL Infrastructure are related. The reality is that they made a sizeable profit out of what happened. 1. The last of these points raises questions concerning the reasonableness of the settlement and is dealt with later. A number of the other points are not correct. For example, on the findings I have made, it is not correct to say that UGL Rail did not rely on Wilkinson Murray's 2007 advice. But those points aside, it is difficult to see how any of the points made by Wilkinson Murray establish that it is inappropriate for the scope of its liability to extend to losses suffered by UGL Rail as a consequence of the failure of the tunnels to meet the reverberation specification. That was the very issue on which Wilkinson Murray was engaged to advise. It had expertise in that area which UGL Rail did not. It was paid a fee for the advice it gave. There is no merit, in those circumstances, in the suggestion that the scope of its duty did not extend to avoiding the very harm that its advice was intended to prevent. 2. In view of the conclusions that I have reached on the question of causation in relation to the contract and negligence claims, it is not necessary to consider what consequences follow if s 5D of the CLA does not apply to the claims based on the indemnities contained in the Consultancy Agreement or on breach of s 52 of the TPA. Even if s 5D applied to claims of those types, it would not prevent the Court from concluding that the loss claimed by UGL Rail was covered by the indemnities or occurred "by" the conduct of Wilkinson Murray. The estoppel defence and the implied term 1. Wilkinson Murray submits that, as a result of what happened, the parties adopted a common assumption in about 2003 that the design work undertaken by Wilkinson Murray was complete and that the parties would proceed by installing a quantity of acoustic material determined by them, which would then be tested. By not allowing Wilkinson Murray to complete its design work, it is alleged that UGL Rail is estopped from relying on the Consultancy Agreement. Put like this, Wilkinson Murray's estoppel defence echoes its claim that the services provided in 2007 were not provided in accordance with the Consultancy Agreement. 2. Alternatively, Wilkinson Murray submits that it was an implied term of the Consultancy Agreement that each party would do all things necessary to enable the other party to have the benefit of the contract and that UGL Rail breached that implied term in the following respects: * it contracted with Wilkinson Murray to prepare a Concept Design by June 2003, a Detailed Design by December 2003 and a Final Design by April 2004; * it installed half the quantity of acoustic material recommended by Wilkinson Murray and then adopted a "test and see" approach; * it treated the May 2003 report as a Final Design; * it installed other services in the tunnels before conducting any tests and did not plan for those tests when installing those services; * it told Wilkinson Murray that it was not possible to install a third row of panels on the other side of the tunnel. Put like this, the claim echoes Wilkinson Murray's complaint that UGL Rail did not follow its advice. 1. Neither of these defences has any merit. 2. As to the estoppel defence, Wilkinson Murray does not point to any evidence which establishes the existence of the common intention; and such a case was never put to UGL Rail witnesses. Contrary to Wilkinson Murray's contention, it did provide what was intended to be a final design in 2003. Alstom elected not to install panels in accordance with that design but, with Alstom's acquiescence, installed a test row of half the recommended panelling. For the reasons given earlier, the services provided by Wilkinson Murray in testing that test row were performed in accordance with the Consultancy Agreement. Wilkinson Murray breached its obligations under that agreement by testing and reporting on the test results in the way that it did. None of that could give rise to an estoppel or could relieve Wilkinson Murray of its obligation to exercise reasonable care and skill in the advice that it gave. 3. As to the breach of the implied term, it is not in dispute that the parties were under a duty to cooperate: see Secured Income Real Estate (Aust) Ltd v St Martins Investments Pty Ltd [1979] HCA 51; (1979) 144 CLR 596 at 607 per Mason J (with whom Barwick CJ, Gibbs, Stephen and Aickin JJ agreed); Concrete Pty Ltd v Parramatta Design and Developments Pty Ltd [2006] HCA 55; (2006) 229 CLR 577 at [14], [156]. However, it is very difficult to see how any of the matters referred to by Wilkinson Murray involve a breach of that duty. The first complaint seems to be that Alstom failed to cooperate because it did not follow the process contemplated by the Consultancy Agreement strictly. Neither party did. However, it is difficult to see how what happened amounted to a lack of cooperation which had the effect of depriving one party or the other of the benefits of the contract. The parties, in effect, agreed that it was appropriate to test half the recommended amount of panelling before proceeding with the installation of two rows of panels. In my opinion, UGL Rail was entitled to treat the May 2003 report as a final design. However, in fact, it did not. Instead, it agreed to further testing and modification of the panels until it was satisfied, on the basis of advice given by Wilkinson Murray, that the proposed design was likely to meet specification. How that involved a lack of cooperation is not clear. The services UGL Rail installed did not prevent the installation of additional panelling. UGL Rail may have told Wilkinson Murray that a third row of panelling could not be installed on the opposite side of the tunnel. Two points may be made about that. First, as I have said, the context in which that statement was made was important. It was made when, as a result of Wilkinson Murray's breaches, UGL Rail wrongly believed that it was likely the tunnels would meet specifications with the two rows of panels. Second, the installation of the other services and the statements made by UGL Rail did not interfere with Wilkinson Murray's ability to carry out tests and report accurately on the results, which Wilkinson Murray did not do. The reasonableness of the settlement 1. A person who settles a claim made by a third party which arises from a breach of a contractual duty owed to the person is generally entitled to recover from the contracting party in breach the amount of the settlement as damages for breach of contract provided the settlement is reasonable. This principle can be traced back at least to the 19th Century, but it has its modern genesis in the decision of the English Court of Appeal in Biggin & Co Ltd v Permanite Ltd [1951] 2 KB 314 ("Biggin"). In that case, the plaintiff bought a quantity of roofing adhesive called Permasec from the defendant and, to the knowledge of the seller, resold it to the Dutch Government to be used in reconstruction work following the end of the Second World War. The adhesive was defective. The Dutch Government commenced arbitration proceedings against the plaintiff, which settled the claim. The plaintiff then sought to recover the amount of the settlement from the defendant. At first instance, Devlin J held that it was not entitled to do so on the basis that the settlement was not a foreseeable consequence of the claim. Rather, it was a voluntary act of the plaintiff. He directed an enquiry into the actual damages caused by the defective adhesive. The Court of Appeal disagreed. It held that the settlement fixed the damages that the plaintiff was entitled to recover from the defendant for breach of contract and gave judgment for that amount in favour of the plaintiff. 2. The reasoning of the Court of Appeal has been criticised: for discussion, see W Courtney, "Settlements Following Breach of Contract" (2013) LMCLQ 157. However, the conclusion reached by the Court of Appeal is generally accepted as being correct; and it was followed, albeit in somewhat different circumstances, by a majority of the High Court in Unity Insurance Brokers Pty Ltd v Rocco Pezzano Pty Ltd [1998] HCA 38; (1998) 192 CLR 603 ("Unity Insurance"). In that case, the plaintiff arranged insurance through the defendant broker. The broker failed to disclose the plaintiff's claims history fully to the insurer. The insurer denied liability in respect of a claim on the basis of the non‑disclosure. The insured commenced proceedings against both the insurer and broker. It settled with the insurer and then claimed against the broker the difference between the amount of the settlement and the amount that it would have recovered if the broker had correctly disclosed its claims history. In Biggin, the question was whether the settlement determined the amount of the damages. In Unity Insurance, the question was whether the difference between the settlement and the amount that would have been recovered absent the breach determined the amount of the damages. The majority (Brennan CJ, McHugh and Hayne JJ; Gummow and Kirby JJ dissenting) held that it did. Just as it was not necessary for the plaintiff in Biggin to prove that the Dutch Government was entitled to succeed against it and the amount for which it was entitled to succeed, it was not necessary for the insured in Unity Insurance to prove that, if it had sued the insurer to judgment, it would have failed or recovered no more than the amount for which it settled. 3. Each of McHugh J and Hayne J referred to Biggin. McHugh J was critical of the reasoning of the Court of Appeal and the breadth with which the principle was stated. In his opinion, the case should be understood as an application of ordinary principles of causation and remoteness: [32] It is impossible, however, to think that their Lordships intended that money paid in reasonably settling a third party action following a breach of contract could be recovered from the defendant even though there was no causal connection between the settlement and the breach. But, even on this restricted reading of the judgments of the Court of Appeal, their Lordships' reasons are at least open to the interpretation that every reasonable settlement of litigation resulting from a breach by the defendant must be regarded as within the contemplation of the defendant. That proposition cannot be accepted. [33] Whether a settlement was within the contemplation of a defendant or a reasonable person in its position must depend upon the nature of the contract between the plaintiff and the defendant, their actual or imputed knowledge of the consequences of a breach, and the nature of the third party's claim against the plaintiff. As a general rule, a contract breaker must be taken to have reasonably contemplated that its breach may force the innocent party into litigation with third parties and that the innocent party may conclude that it is in its best interest to compromise the third party's claim. But it does not follow that the fact that it was reasonable for the plaintiff to compromise the claim against the third party necessarily means that the settlement was within the reasonable contemplation of the defendant. That is so even in those cases where the defendant's breach was proved to be causally connected with the settlement. Each case must depend upon its own facts. 1. Hayne J also doubted that the Court of Appeal should be understood as accepting a broad proposition that the amount paid under the settlement, if reasonable, should be taken as the measure of the plaintiff's damages. Like McHugh J, he thought the issue should be examined in terms of causation: [122] The fact that the dispute between insured and insurer may be resolved by agreement does not lead to any different result. The loss suffered by the insured, if the compromise is reasonable, is caused by the broker's breach of obligation. To the extent that policy is to be considered in answering the question whether the breach caused the loss, policy considerations reinforce the conclusion that the breach caused the loss. [123] No doubt the broker may be wholly excluded from any negotiations to resolve the differences between insured and insurer and it seems that that was the case here. But that does not mean that the broker is left to the mercies of the parties over which it has no control or that those parties may impose on the broker any liability that they choose. The settlement between insurer and insured must be reasonable. [Footnote omitted] However, it is apparent from what Hayne J says that he regarded the central question as one of reasonableness. If the settlement arose from a breach of duty and was reasonable, then, in the normal course of events, the settlement was caused by the breach of duty. 1. Brennan CJ did not discuss Biggin. However, he said that he was in "substantial agreement" with Hayne J and only pointed to some differences between them. In doing so, he still analysed the issue in terms of causation (at [3]): When a claim is met by an arguable defence, a compromise is a natural and foreseeable result. Therefore it is a natural and foreseeable result of whatever creates an arguable defence to a claim that the claim will be compromised. In the present case, whatever weakness there was in the insured's case against the insurer was attributable to the broker's negligent breach of its retainer or its negligence. The acceptance in settlement of the insured's claim against the insurer of a sum less than a full indemnity was something which occurred in the natural course of events or which was in contemplation of the parties at the time of the engagement of the broker or was reasonably foreseeable at the time of the broker's negligence. 1. Gummow and Kirby JJ dissented. Gummow J distinguished Biggin and similar cases. They were concerned with the question whether the plaintiff was entitled to recover an amount of a reasonable settlement following a claim against the plaintiff arising out of the defendant's breach of duty where the settlement could be regarded as something within the reasonable contemplation of the plaintiff and defendant as a serious possibility or not unlikely result of the defendant's breach of its obligations to the plaintiff. In contrast, in Unity Insurance, the question was whether it was within the reasonable contemplation of the insured and broker that the measure of the broker's liability would be determined by an agreement between the insured and the insurer: see id at [65]ff. Gummow J held that it was not. 2. Kirby J appears generally to have preferred the approach of Devlin J in Biggin, accepting only that a settlement would provide a cap on the damages that the plaintiff could recover: id at [89]. Kirby J accepted that cases where the claim is against a defendant who has given a contractual indemnity (as in the case where an insured under a liability policy settles a claim and then seeks to recover the amount of settlement from an insurer) may be an exception: id at [100]. 3. The principle that a reasonable settlement resulting from a breach of duty is recoverable as damages from the person in breach applies whether or not the plaintiff had any actual liability to the third party: General Feeds Inc Panama v Slobodna Plovidba Yugoslavia [1999] 1 Lloyd's Rep 688 ("General Feeds Inc"); Edwards v Insurance Office of Australia Ltd (1933) 34 SR (NSW) 88; Rail Corporation NSW v Fluor Australia Pty Ltd [2009] NSWCA 344 at [102] per Macfarlan JA (with whom Allsop P and Hodgson JA agreed). 4. There is no doubt that the principle applies equally where the contract includes an indemnity in respect of the relevant loss, although in that case the terms of the indemnity may affect the amount the contracting party is entitled to recover in respect of a settlement and the circumstances in which it is entitled to do so: BNP Paribas v Pacific Carriers Ltd [2005] NSWCA 72 at [13] per Handley JA. Similar principles also apply where the plaintiff's claim is in negligence: Unity Insurance; Rail Corporation NSW v Fluor Australia Pty Ltd at [102] per Macfarlan JA (with whom Allsop P and Hodgson JA agreed). In addition, they have been applied to claims for damages for contravention of s 52 of the TPA: see BHP Billiton (Olympic Dam) Corporation Pty Ltd v Steuler Industriewerke GmbH (No 2) [2011] VSC 659 at [216]-[234]. 5. Analysed in terms of causation and remoteness, the question whether a defendant should be held liable for the losses arising from a settlement raises slightly different questions depending on the cause of action on which the plaintiff relies. At common law, in the case of a claim for breach of contract, the question is normally stated in terms of whether the settlement was within the reasonable contemplation of the parties at the time the contract was entered into: see Hadley v Baxendale (1854) 9 Ex 341; (1854) 156 ER 145. That was the language used by the majority of the High Court in Unity Insurance. In the case of a claim of negligence (whether the claim is in contract or tort), the test is now to be found in s 5D of the CLA. In the case of a claim for damages under s 82 for contravention of s 52 of the TPA, the question is whether the loss was suffered "by" the conduct of the defendant. In the case of a contractual indemnity, the test of causation is determined by the indemnity; although, as I have explained, it is not entirely clear whether, in the case of a contractual indemnity, the principles are modified by s 5D. In any event, it was not suggested that these different formulations produced different results in this case. 6. The question of reasonableness is an objective one to be determined having regard to the material that was available to the parties at the time of settlement: Unity Insurance at [130] per Hayne J; General Feeds Inc at 692 per Colman J. The question is whether the settlement was reasonable, not whether the plaintiff acted reasonably in settling. Consequently, the fact that the plaintiff acted on the basis of legal advice is not determinative. As Brennan CJ explained in Unity Insurance (at [6]): The test of reasonableness is, as Hayne J says, an objective one. Evidence of the advice which the insured received to induce it to accept the settlement is not proof in itself of the reasonableness of the settlement advised. The factors which lead to the giving of the advice are factors relevant to the reasonableness of the settlement but the only relevance of advice given by the insured's legal advisers to settle is that it tends to negative the hypothesis that the insured acted unreasonably in accepting the settlement. See also at [35] per McHugh J, at [129], [135] per Hayne J; BHP Billiton (Olympic Dam) Corporation Pty Ltd v Steuler Industriewerke GmbH (No 2) at [229]ff per Habersberger J. For discussion, see W Courtney, "Settlements Following Breach of Contract" (2013) LMCLQ 157 at 166ff. The court will examine the negotiations that led to the settlement as well as the terms of the settlement as part of the process of determining whether the settlement was reasonable: see BNP Paribas v Pacific Carriers Ltd. 1. Put in this way, there are two distinct questions. One is whether the settlement was within the reasonable contemplation of the parties. The other is whether the settlement itself was reasonable. As Hayne J pointed out, the answer to the second question provides a mechanism by which the court balances two competing concerns. One is to encourage settlements and to make settlements easier to achieve by permitting two parties to a dispute to settle it without requiring them to obtain agreement from a third party who may be affected by the settlement. The other is not to put the third party (the defendant in the proceedings in question) in a position where its liability is determined by an agreement to which it is not a party. In terms of the test of causation stated in s 5D of the CLA, the first question involves a mixed question of factual causation and scope of liability causation. It is answered partly by asking whether the liability arising from the settlement would have occurred but for the defendant's negligence. However, it also raises the question whether there is a sufficient connection between the settlement and the negligence that it is appropriate to hold the defendant liable for the negligence. The second question raises the normative question whether it is appropriate for the negligent party to be held responsible for the whole of the settlement reached by the claimant. 2. In BNP Paribas v Pacific Carriers Ltd at [17], Handley JA stated that a party relying on a settlement as proof of loss must prove "not only that the result was reasonable but also that the negotiations were conducted with proper care and skill". Whether that is so, however, is unclear. If the settlement itself is reasonable, it is difficult to see why the negotiations must also be conducted reasonably. The fact that negotiations were not conducted reasonably may, on the other hand, be evidence that the settlement itself is not reasonable. 3. If the court concludes that the settlement was unreasonable, it may award as damages an amount that reflects a reasonable settlement: BNP Paribas v Pacific Carriers Ltd at [260]-[263] per Giles JA (with whom Sheller JA agreed on this point; Handley JA dissenting). 4. The principle has also been applied where the settlement covers a broader liability than the one for which the defendant is responsible. However, in that case, it may be necessary to apportion the settlement amount between the liability for which the defendant is responsible and other liabilities covered by the settlement: Bovis Lend Lease Ltd v RD Fire Protection Ltd [2003] EWHC 939 (TCC); (2003) 89 ConLR 169 at [81]; [95]‑[96] per HH Judge Thornton QC. 5. That approach is illustrated by the decision in General Feeds Inc. In that case, the owners of a vessel chartered the vessel to charterers for the carriage of a cargo of antioxidant treated fishmeal from Peru to China. The cargo arrived in a damaged state due to fire, heat and smoke. The insurers of the cargo owner made a claim for $2.4 million against the owners claiming that the cargo had been damaged by bad storage. The owners defended the claim on the basis that the damage was caused by the condition of the cargo at the time it was loaded and, in particular, the fact that the fishmeal had not been sufficiently antioxidant treated. Nonetheless, the owners settled the claim for $600,000. They then sought to recover that amount from the charterers on the basis that the cargo had been mis-described as antioxidant treated. That dispute was referred to arbitration. The arbitrators awarded the owners $400,000 on the ground that the damage to the cargo had been exacerbated by the owners' handling of it upon arrival. If the owners had handled it appropriately, the damage would have been limited to $1.6 million. On that basis, the arbitrators apportioned the settlement according to the contributions each of the charterers and owners made to the total claim of $2.4 million. On appeal, Colman J upheld the arbitrators' award on the basis that they had made no error of principle in the approach they adopted. 6. In the present case, in order to recover the damages it claims, UGL Rail must prove that: * it was reasonable to settle the claim that it had failed to comply with the reverberation specification set out in the Works Brief; * the terms on which it settled that claim were reasonable; * it was reasonable to pay UGL Infrastructure the lump sum it did in order to carry out the work required of it by the TIDC solution. 1. I am satisfied that it was within the reasonable contemplation of the parties (to state the test of causation in contractual terms) that UGL Rail would seek to settle a claim that it had failed to comply with the reverberation specification in the Works Brief. Wilkinson Murray knew that the reverberation specification came from the Works Brief. It knew that UGL Rail was required to comply with that specification in accordance with the Systems Subcontract between UGL Rail and THJV. UGL Rail and Wilkinson Murray must have expected that, if it was alleged that UGL Rail was responsible for the failure of the tunnel to meet the specification, UGL Rail would seek to settle that claim. There can be no doubt that UGL Rail settled the claim with THJV in response to an allegation that it was in breach of the Systems Subcontract. 2. The question whether the settlement between UGL Rail and THJV was reasonable raises two issues. One is whether it was reasonable to conclude that the costs of complying with the reverberation specification were greater than implementing the TIDC solution. The other is whether it was reasonable for UGL Rail to agree to undertake the work that it did in connection with the TIDC solution. 3. I am not satisfied that it was reasonable for UGL Rail to conclude that the costs of complying with the reverberation specification exceeded the costs of the TIDC solution. In submitting that it was, UGL Rail submitted – correctly – that the question was to be determined having regard to the information available at the time. However, there are two issues. One is what work was necessary to meet the reverberation specification. The other is the costs of that work compared to the costs of the TIDC solution. 4. The only detailed analysis that compared the costs of complying with the reverberation specification and the TIDC solution was undertaken by Mr Boyd. Mr Johnson did some rough calculations comparing the TIDC solution with various options that were put forward for complying with the reverberation specification; and he and Mr Campbell prepared a detailed costing of the TIDC solution. They did not do the same for any of the various proposals that were put forward for complying with the reverberation specification. Mr Johnson's principle calculation appears to have been done by reference to the solution proposed by Mr Tonin. However, Mr Tonin was retained by THJV; and it is unclear what instructions he was given concerning compliance with the reverberation specification. It may well have been in his client's interests to establish that compliance with the reverberation specification would be difficult. The same could be said of Heggies, who had been retained by TIDC. Mr Johnson described UGL Rail's own expert, Mr Challis, as "suitably vague" and his calculation of the costs of the solution proposed by Mr Challis seem little more than an educated guess. In my opinion, they did not form a reasonable basis on which to reach a conclusion about which solution was more expensive. 5. Mr Boyd compared the costs of what he called the Leo/Challis Option 2 solution with the TIDC solution. He undertook detailed calculations in relation to both solutions. It seems clear that he used the same basis for calculating the costs of both solutions so that the comparison of the costs appears to be reasonable. The difficulty, however, is that there is no evidence that the Leo/Challis Option 2 solution itself was a reasonable solution. Mr Challis was engaged by UGL Rail and Mr Leo was engaged by its insurers. However, it is not sufficient that UGL Rail and the insurers engaged experts and relied on what the experts said. UGL Rail must establish that the advice of the experts was itself reasonable. The position is no different than when considering the terms of settlement. It is not sufficient for the plaintiff to prove that it obtained legal advice in connection with the settlement and relied on that advice. It must prove that the advice itself was reasonable – that is, the court looks at the objective position, not at the reasonableness of the plaintiff's conduct. 6. In the present context, UGL Rail bears the onus of proving that its settlement was objectively reasonable. As part of doing that, it must prove objectively, on the basis of information available at the time, that the TIDC solution was cheaper than compliance with the reverberation specification. As part of doing that, it must prove the Leo/Challis Option 2 solution was a reasonable means of complying with the reverberation specification. It has not sought to do that. Rather, it has simply assumed that Mr Boyd was entitled to proceed on the basis that the Leo/Challis Option 2 solution set out what was reasonably necessary to comply with the specification. 7. In addition, there are reasons for believing that the Leo/Challis Option 2 solution was not a reasonable basis on which to comply with the reverberation specification. Although the Leo/Challis Option 2 solution is described in Mr Boyd's report, there is no report prepared by Mr Leo and Mr Challis in evidence setting out how they reached the conclusions they did. It is not clear whether they undertook any computer modelling or actual testing on the tunnel. Yet it is UGL Rail's case, which I accept, that that is what they would have to have done if their results were to have been regarded as reliable. Mr Challis did prepare an earlier report that explains how he arrived at the recommendations contained in that report. It is not clear to what extent the Leo/Challis Option 2 solution relied on that report. However, that report relied on extrapolating the average results obtained by Mr Heggie. Both experts in the present case accept that it was not appropriate to average results. It is unclear whether it was possible to extrapolate from the results obtained in the way that Mr Challis did; and Mr Challis seems to have had no regard to the placement of acoustic panelling in the tunnel in determining the quantity required even though it seems clear that the precise placement was important. If Mr Leo and Mr Challis had conducted computer modelling, then the likelihood is that their results would be similar to Professor Kang's. However, they concluded that considerably more acoustic panelling would need to be installed than that recommended by Professor Kang. UGL Rail has not sought to explain why their conclusion was reasonable having regard to the conclusions of Professor Kang, which I accept. 8. I am also not satisfied that the costs of the TIDC solution that were incurred by UGL Rail were greater than the costs that could properly be taken into account in complying with the reverberation specification. In its submissions, UGL Rail assumed that all it needed to do was prove that the TIDC solution was cheaper than the costs of complying with the reverberation specification and that the costs of implementing the TIDC solution were themselves reasonable. However, in my opinion, that is not correct. UGL Rail also needed to prove that the costs for which it claims were reasonably related to Wilkinson Murray's breach of duty. It has failed to do that. I say that for a number of reasons. 9. First and foremost, a straight comparison between the costs of complying with the reverberation specification and the costs of the TIDC solution is a false comparison from the point of view of Wilkinson Murray. Wilkinson Murray could only ever be liable for the costs incurred by UGL Rail as a consequence of Wilkinson Murray's negligent advice. If Wilkinson Murray had given the correct advice in the first place, then UGL Rail would have installed the correct amount of acoustic panelling in 2007. Assume, at least for present purposes, that that was 3000 mm per metre of tunnel. On that basis, UGL would always have been liable for the costs of 3000 mm of acoustic panelling per metre and the costs of installing that amount of panelling in 2007. Consequently, from Wilkinson Murray's point of view, the costs in 2008 or later of meeting the reverberation specification for which it would be liable were the additional costs of installing 3000 mm of panelling per metre less the panelling already installed. They were not the total costs of installing the additional panelling. UGL Rail would always have been liable for the costs of the panels and the costs of installing them at the same time as the original panelling was installed whether or not Wilkinson Murray had been negligent. From Wilkinson Murray's point of view, the true comparison was the costs of the TIDC solution compared to the additional costs of installing the required panelling late. The former would unquestionably be greater than that latter. The TIDC solution may have been more attractive to UGL Rail than the costs of meeting the reverberation specification because UGL Rail was always liable to meet those costs. On the other hand, Wilkinson Murray could never be liable for those costs. It was only ever liable for the additional costs of meeting the specification. By settling its obligation to meet the reverberation specification, UGL Rail was settling a liability that it had. But it was not settling a liability for which Wilkinson Murray was responsible. 10. Second, even if the first point is wrong, I do not accept that Wilkinson Murray should be held liable for all of the work done in implementing the TIDC solution when that solution also relieved UGL Rail and THJV of potential liabilities unconnected with Wilkinson Murray's negligence and satisfied TIDC's concerns with in-car noise (whether the fault of THJV or UGL Rail or not). Whatever the scope of UGL Rail's liability, why, it might be asked, should Wilkinson Murray be liable for UGL Rail's total costs of the TIDC solution when that solution provided benefits other than remedying the problem for which Wilkinson Murray was responsible? Looked at in this way, this is a case where the amount of the settlement should be apportioned because it covered matters for which Wilkinson Murray was not responsible. It should not matter that, absent a settlement, the costs for which Wilkinson Murray might ultimately be held responsible would be greater. The fact is that the parties did settle; and they settled more than the complaint that the reverberation specification had not been met. Consequently, even if the first point is wrong, it would have been necessary to apportion the costs of the TIDC solution between the problem for which Wilkinson Murray was responsible (reverberation control) and the other problems associated with in-car noise (concerned with the tracks and rails). 11. Third, I am not satisfied on the evidence that it was reasonable for UGL Rail to settle on the terms that it did. Mr Summers gave evidence that the settlement was far preferable to facing a claim for breach of the reverberation specification. If UGL Rail had not settled, it would have faced a claim that was likely to be greater than the settlement. There was no guarantee that the costed solution would work. UGL Rail faced the risk that its security bond would be called on and that THJV or TIDC would engage another contractor to complete the work, which would result in substantial damage to UGL Rail's reputation. All that may be accepted. The difficulty is that, if no settlement was reached, TIDC still faced a problem with in-car noise whether or not the reverberation specification was met; and THJV faced a possible liability in respect of that. Just as a failure to settle would have had serious consequences for UGL Rail, it would have had serious consequences for TIDC and THJV. It is to be expected, in those circumstances, that both TIDC and THJV would also have been under a lot of pressure to settle. Yet UGL Rail ended up bearing most of the costs of the settlement. No explanation is given in the evidence for how that came about. It may be that UGL Rail did all that it reasonably could to extract a more favourable settlement out of TIDC and THJV and, faced with the choice of accepting what was offered or losing the settlement altogether, it chose the former. However, in the absence of evidence of what UGL Rail did to extract a more favourable settlement, I am not satisfied that it did all that it reasonably could. 12. It might be said that it can be inferred that UGL Rail did the best it reasonably could. It was represented by competent legal advisers. It obtained expert advice. It faced the prospect of having to bear a significant proportion of the settlement itself and, consequently, it is to be expected that it would seek to maximise the contributions to the settlement that it could obtain from THJV and TIDC. However, I am not prepared to draw that inference in this case. UGL Rail was covered by insurance. Its interests were not necessarily in minimising the amount of its liability so much as minimising the gap between its liability and the amount of its insurance cover. TIDC was seeking to achieve a reduction of 15 dB in in-car noise. At most, compliance with the reverberation specification would have achieved half that. Yet it appears that UGL Rail ended up bearing the large majority of the costs of achieving TIDC's goal. In the absence of some explanation of the negotiations that led to that result, I am not prepared to conclude that the settlement was reasonable. 13. Lastly, I am not satisfied that the actual costs paid by UGL Rail in connection with the TIDC solution were reasonable. I say that for two reasons. 14. First, I am not satisfied that the amount of the costs was reasonable. UGL Rail led extensive evidence from Mr Boyd, Mr Johnson and Mr Campbell concerning the work that they did to estimate the costs of implementing the TIDC solution. The difficulty with their evidence, however, is that they give no explanation for how the actual costs of implementing the TIDC solution could be slightly more than half their estimated costs. I accept that the question whether the costs were reasonable is to be judged by reference to what was known at the time of settlement. However, at the time of the settlement, work had already begun on implementing the TIDC solution. The work was complete within approximately 20 weeks of the settlement. The large disparity between the actual costs and estimated costs over such a short period of time calls for a detailed explanation. However, none was given. Mr Johnson gave evidence that UGL Infrastructure improved its work practices once it started the work. However, if improved work practices explain the difference, then, having regard to the large disparity between the estimate and the costs, it seems to me that UGL Rail needed, in order to discharge its onus, to explain precisely what those work practices were, how they resulted in such large savings and why they could not have been anticipated at the time the costs estimates were prepared. 15. Second, I do not accept that it was within the reasonable contemplation of the parties that UGL Rail would reach a settlement with UGL Infrastructure by which UGL Infrastructure would agree to do all the work to comply with the TIDC solution for a fixed price calculated by reference to the amount that UGL Rail recovered from its insurer. The relationship between UGL Rail and UGL Infrastructure is unclear from the evidence. It is clear that UGL Infrastructure did the work that UGL Rail was contracted to perform under the original contract. However, that work was not done pursuant to any form of written contract; and no evidence was led concerning how the billing arrangements worked between UGL Rail and UGL Infrastructure. From Wilkinson Murray's point of view, it appears that no real distinction was drawn between the two companies. In my opinion, in those circumstances, it could not have been reasonably anticipated at the time that the Consultancy Agreement was entered into that, if any work needed to be done arising out of a breach of the reverberation specification, UGL Rail would enter into a fixed price contract with a related party to do that work rather than incur the actual costs of the rectification work. 16. Again, it was not suggested that a different conclusion should be reached insofar as UGL Rail's claim was based on the TPA. In some of its submissions, UGL Rail relied on the width of the indemnities given by Wilkinson Murray – and, in particular, the width of the connecting phrase "in connection with" used in both cls 6.1 and 18.1 – as supporting its claim. It pointed out that courts had frequently held that the expression "in connection with" is one of "broad meaning and wide connotation requiring merely a relationship between one thing and another": see QBE Insurance (Australia) Ltd v Wesfarmers General Insurance Ltd [2010] NSWSC 855 at [88] per Garling J; see also Drayton v Martin (1996) 67 FCR 1 at 32 per Sackville J. In my opinion, although the indemnities are undoubtedly wide, the words "in connection with" when used in the indemnities still require a causal connection between the loss in respect of which indemnity is sought and a breach by Wilkinson Murray of its obligations under the agreement. The parties could not have intended, for example, that Wilkinson Murray would be liable to indemnify UGL Rail for losses it suffered because the tunnels did not meet the reverberation specification even if that came about through no fault of Wilkinson Murray. Similarly, even assuming that the test of causation is not governed by s 5D of the CLA, it seems to me that it still requires that the settlement be reasonable before it can be said that the loss arising from the settlement occurred in connection with Wilkinson Murray's breach of the agreement. Clear words would be required to exclude that principle of causation. For those reasons, I do not think the result would be any different insofar as UGL Rail relies on the indemnities given by Wilkinson Murray. Damages 1. Wilkinson Murray submits that UGL Rail put its case on an all or nothing basis. Having failed to prove the damages it claims, it was entitled to nothing. 2. I do not accept that submission. It is clear from the decision in the Court of Appeal in BNP Paribas v Pacific Carriers Ltd that, if the court concludes that a settlement is not reasonable, it should nonetheless award an amount that reflects a reasonable settlement. For reasons I have explained, the position in the present case is more complicated because the amount that reflects a reasonable settlement as between UGL Rail and THJV does not provide a reasonable basis for determining Wilkinson Murray's liability. But that does not make the approach adopted by the Court of Appeal inapplicable. It is still possible to ask what amount reasonably reflects the loss that UGL Rail suffered as a consequence of Wilkinson Murray's negligence. That loss consists of two components. First, there are the additional costs that UGL Rail would have incurred if it had done the work necessary to comply with the reverberation specification at the time it did the work necessary to comply with the TIDC solution compared to the costs it would have incurred if it had done that work originally. Second, there are costs that it incurred in resolving the dispute concerning the failure to meet the reverberation specification. It is true that UGL Rail did not actually incur costs of the first type. However, it did incur costs of implementing the TIDC solution, which were far greater than those costs; and in implementing the TIDC solution it was relieved of any of the costs of complying with the reverberation specification. Consequently, I cannot see why UGL Rail should not be entitled in principle to recover costs of that type. 3. As to the first type of costs, neither party sought to lead evidence of what those costs might be. However, I am satisfied that UGL Rail would have incurred some costs of that type. It seems clear that the costs of installing acoustic panels would be greater in conditions where UGL Rail had limited access to the tunnels and was required to remove all its equipment each day compared to the position where it had unlimited access and could leave its equipment in place and would have been in a position where it could have installed all the required panelling at one time. In those circumstances, the Court must do the best it can to assess damages on the evidence it has, even if that involves a degree of guess-work: Commonwealth v Amann Aviation Pty Ltd [1991] HCA 54; (1991) 174 CLR 64 at 83 per Mason CJ and Dawson J, at 125 per Deane J; McRae v Commonwealth Disposals Commission [1951] HCA 79; (1951) 84 CLR 377 at 411-2 per Dixon and Fullagar JJ; Howe v Teefy (1927) 27 SR (NSW) 301 at 306 per Street CJ. 4. The task of the Court, then, is to estimate the additional costs of installing the additional panelling to meet the reverberation specification. The evidence from Professor Kang, which I accept, is that it would have been necessary to install 3000 mm of effective acoustic panelling per metre or tunnel length to meet the specification. The parties accept that approximately 1360 mm were installed. Consequently, the task is to estimate the additional costs of installing 1640 mm of acoustic panelling. There is evidence of the costs of installing the original two rows of panels. It is possible from that evidence to calculate the costs per metre of installing panelling at the time United had unrestricted access to the tunnels. There is also evidence of the costs of installing the panelling that formed part of the TIDC solution. Consequently, it is possible to calculate the costs per metre of installing panelling at the time United did not have unrestricted access to the tunnels. The difference between the two gives the additional costs per metre of installing panelling where access was restricted. On the findings I have made, it is those additional costs that could have been avoided if Wilkinson Murray had given non-negligent advice. 5. In a joint report, Mr Makin and Mr Young, the quantum experts retained by the parties, agreed that the labour costs of installing the first two rows of panels was $1,863,000. That agreement was reached on the basis of the actual labour costs of installation set out in an affidavit of Mr Kamil Erbel, who was, at the relevant time, the General Manager Finance of UGL Infrastructure. Mr Erbel was not cross-examined and I accept his evidence. Mr Young concluded that the plant costs were $125,000. Mr Makin concluded that they were $205,000. The difference of $80,000 reflected differences of opinion on the amount of rubbish that had to be removed and different assumptions made about the use of small tools, such as drills and drill bits. For the purpose of determining the installation costs of the two rows of panels, I think that it is reasonable to split the difference and assume that the plant costs were $165,000. Both experts agreed that the on-site overheads were $49,604 and that an allowance of 2.2 per cent should be made for off-site overheads. On that basis, the total costs of installing the first two rows of panels was $2,123,311.29, calculated as follows: Labour costs 1,863,000.00 Plant and equipment 165,000.00 On-site overheads 49,604.00 Total 2,077,604.00 Off-site overheads 45,707.29 (2.2% of above total) Total $2,123,311.29 1. The total length of the tunnels is 25,397 m. Excluding stations, the total length is 24,037 m. Assuming that panelling was installed throughout the tunnels other than in the stations, and assuming that 1.36 m2 of acoustic material was installed per metre of tunnel, the total amount of acoustic material that was installed as part of the installation of the first two rows was 24,037 x 1.36, making 32,690.32 m2. Consequently, the cost per square metre of installing the panelling was $2,123,311.29/32,690.32, making $64.95 per square metre. 2. Tendered during the course of the hearing was a spreadsheet setting out the actual costs incurred by UGL Infrastructure of installing the acoustic panelling that comprised the TIDC solution. The following items in that spreadsheet appear to be relevant to the costs of installing that panelling: Labour costs: Project management and supervision $882,708.81 Labour crew supervision 46,804.43 Engineering and drafting 47,996.19 Finance and administration 183,033.50 Open side wall panels 769,021.98 Behind tray wall panels 487,072.27 Below cable tray cladding 391,078.18 Above walkway panels 139,763.58 Consultants 2,790.00 $2,950,268.94 Plant costs: Small tools 170,924.32 PPE and medicals 39,037.18 Office consumables 18,091.79 Plant and equipment hire 78,134.94 Equipment enhancements for installation 68,632.29 Site vehicles and fuel 790,435.48 Telecommunications and IT 15,617.15 Site cleaning and security 46,356.42 $1,227,229.57 1. Allowing an amount of 2.2 per cent for off-site costs, the off-site costs were $91,904.97 (that is, ($2,950,268.94 + $1,227,229.57) X 0.022) and the total costs of installing the wall panels was $4,269,403.48 (that is, $2,950,268.94 + $1,227,229.57 + $91,904.97). 2. According to the Settlement Deed, UGL Rail was required to install the following amount of acoustic panelling: Zone Location Panel Height DN Tunnel Extent [m] UP Tunnel Extent [m] Panel Type A (Existing Panels) B Below Existing Panels 0.6m 7,318 4,505 Perforated metal-clad glasswool (2off) C Below Walkway 0.6m 7,276 7,989 Perforated metal-clad glasswool (1 off) 0.3m 9,701 10,652 Fibreglass fabric-clad glasswool (2 off) D Above Walkway 0.6m 2,205 117 Perforated metal-clad glasswool (2off) On that basis, UGL Rail was required to install the following amount of acoustic panelling in the tunnels: Zone B (7,318 + 4,505) x 0.6 x 2 = 14,187.6 Zone C(1) (7,276 + 7,989) x 0.6 = 9,159.0 Zone C(2) (9,701 + 10,652) x 0.3 x 2 = 12,211.8 Zone D (2,205 + 117) x 0.6 x 2 = 2,786.4 Total 38,344.8 1. Consequently, the costs of installing one square metre of effective panelling in 2009 as part of the TIDC solution was $4,269,403.48/38,344.8, which comes to $111.34 per square metre. On that basis, the additional cost of installing panelling in 2009 was $46.39 per square metre (that is, $111.34 - $64.95). 2. As I have said, according to Professor Kang, it was necessary to install 3000 mm per metre of acoustic material throughout the length of the two tunnels in order to meet the reverberation specification. Approximately 1360 mm had been installed. Consequently, it was necessary to install a further 1640 mm. Using the calculations set out above, the additional costs of installing that panelling would have been 1.64 x 24,037 x $46.39, making $1,828,725.35. 3. Although the figures I have used are very precise, the calculations necessarily involve an approximation. In particular, they make no allowance for the different costs of installing panelling in different parts of the tunnel. They also make no allowance for the efficiencies that might have been achieved if all the panelling had been installed at the same time. There is, however, no means of knowing what allowance, if any, should be made for considerations such as those. In those circumstances, in my opinion, the most reasonable approach is not to try to adjust the figure I have arrived at to take account of those factors. 4. The second type of costs are the costs that UGL Rail became liable to pay TIDC and THJV in connection with the dispute and its own legal costs of resolving the dispute. Those costs comprise the following: TIDC's costs of the dispute $1,000,000.00 THJV's costs of the dispute $867,390.84 THJV's supervision costs $1,251,048.33 UGL Rail's legal costs $358,717.85 1. It is not in dispute that UGL Rail incurred those costs. It became liable to pay the first three types of costs under the Settlement Deed. It is not disputed that its own legal costs in connection with the dispute were $358,717.85. The question is whether UGL Rail's liability for those costs can be said to be caused by Wilkinson Murray's breach of duty. 2. In my opinion, it is appropriate in principle to hold Wilkinson Murray liable for each category of these costs. Wilkinson Murray was responsible for giving UGL Rail advice to ensure that UGL Rail knew what amount of acoustic panelling it had to install in the tunnels to meet the reverberation specification. If it gave negligent advice with the consequence that UGL Rail installed the incorrect amount of panelling, it was to be expected that TIDC would serve a defect notice on THJV and that THJV would make a claim against UGL Rail, which is what happened. It is to be expected that each would incur costs in connection with that dispute and that TIDC would seek to recover its costs under the indemnity given to it by THJV and that THJV would seek to recover those costs and its own from UGL Rail. It is also to be expected that, if rectification work had to be done, that would require THJV to supervise that rectification work and it is to be expected in those circumstances that THJV would seek to recover the costs of supervision from UGL Rail. The relationship between TIDC, THJV and UGL Rail was known to Wilkinson Murray and it must have been within Wilkinson Murray's reasonable contemplation that claims of that type would be made if it gave negligent advice. 3. If the tunnels had complied with the reverberation specification, the likelihood is that no claim would have been brought against UGL Rail. I have already found that UGL Rail would have taken steps to comply with the reverberation specification if it had been given the correct advice by Wilkinson Murray. Consequently, UGL Rail would not have incurred the legal costs it did but for Wilkinson Murray's negligence. It is true that the resolution of the claim against UGL Rail became bound up with issues concerning in-car noise. However, from UGL Rail's point of view, the resolution of issues concerning in-car noise was the mechanism by which it resolved its liability for beach of the reverberation specification. There is no reason to think that the costs of resolving that dispute in some other way would have been less expensive. Consequently, in my opinion, it is appropriate to hold Wilkinson Murray responsible for all the legal costs incurred by UGL Rail in resolving the dispute. 4. The other three types of costs fall into a different category. In each case, it can be said that UGL Rail would not have agreed to bear those costs but for Wilkinson Murray's negligence. However, in each case, the costs also related to work that was unconnected to defects which arose from Wilkinson Murray's negligence. Even if Wilkinson Murray had not been negligent and the tunnels had complied with the reverberation specification, there would still have been a problem with in-car noise. TIDC would have investigated that problem. It would have made a claim against THJV and both TIDC and THJV would have incurred legal costs in connection with that dispute. 5. According to an email dated 17 March 2009 sent by Mr Burt to various employees of United, TIDC's costs included past and future management and supervision costs, together with external costs of lawyers and acoustic consultants. TIDC is said to have estimated those costs as being in excess of $3 million, which TIDC was seeking to recover from THJV. That estimate appears excessive. There is no evidence to suggest that TIDC was required to employ additional staff to deal with the dispute. Consequently, it is not at all clear whether it would have incurred additional management and supervision costs in connection with the dispute. There is no evidence of how the $3 million, or the $1 million for which TIDC settled, was made up. The likelihood is that part of it comprised a claim for past and future management and supervision costs. Without more information about the nature of those costs, I am not satisfied that they were properly recoverable. On the other hand, I accept that TIDC would have incurred substantial legal and consulting costs and it is likely that they would have been at least as much as the costs incurred by THJV. On that basis, I am prepared to accept that it was reasonable for THJV and UGL Rail to settle TIDC's claim on the basis that TIDC's costs were in the order of $800,000. However, I am not satisfied that it was reasonable to attribute all those costs to the failure to meet the reverberation specification. It is apparent that TIDC was principally concerned with in-car noise. The failure to meet the reverberation specification was an important contributing factor to in-car noise but it was not the only factor. Even if the tunnels had met the reverberation specification, there would have been a problem with in-car noise. The likelihood is that TIDC would still have served a defect notice on THJV and a settlement of that claim would have involved THJV agreeing to pay part of TIDC's costs. It is unclear whether THJV would have sought to recover those costs from UGL Rail. However, it is clear that UGL Rail would not, in those circumstances, have had a claim against Wilkinson Murray. 6. There is no obvious means of apportioning the amount payable to TIDC between the loss for which Wilkinson Murray was responsible and the loss for which it was not. The evidence is that TIDC was principally concerned with in-car noise. However, as I have said, a major contributing factor to in-car noise was the failure to meet the reverberation specification. There is no evidence to suggest that the work done by TIDC was concerned more with the investigation of the effect of rail dampers and grinding rather than the installation of additional acoustic panelling in the tunnel. In those circumstances, it seems reasonable to apportion TIDC's costs equally between the failure to meet the reverberation specification and other factors that contributed to in-car noise. On that basis, Wilkinson Murray is liable for half of $800,000 – that is, $400,000. 7. Applying similar principles, it seems to me reasonable that Wilkinson Murray should be held responsible for half of THJV's reasonable costs of the dispute. The total amount claimed by THJV is $867,390.84. That amount comprises $761,265.84 in legal and consultants' costs. There are invoices that support that amount; and I am satisfied that those amounts were reasonable and that, subject to what I say below, it was reasonable, as part of an overall compromise, for UGL Rail to agree to pay those costs. 8. The balance of the costs claimed by THJV consists of staff costs of $105,225 and miscellaneous costs of $900. In order for those costs to be recoverable, it would be necessary for UGL Rail to establish that those costs would not have been incurred but for Wilkinson Murray's breach of duty. It is not sufficient to establish that certain employees of THJV did work in connection with the resolution of the dispute. It would be necessary to establish that THJV incurred an additional liability to those employees because of that work or that those employees were prevented from engaging in other activities that would have earned THJV income. There is no evidence that that was the case. In the absence of that evidence, I am not satisfied that those costs were reasonable. 9. Again, it seems to me that THJV's costs related both to the failure to comply with the reverberation specification and the problems with in-car noise that were unconnected to that failure. It is not clear why, in those circumstances, UGL Rail agreed to bear all those costs. But whether it acted reasonably or not in doing so, not all those costs can be attributed to the breach for which Wilkinson Murray is responsible. Again, there is no basis for separating the costs between issues concerned with the failure to meet the reverberation specification and other issues. Accordingly, in my opinion, it is reasonable to hold Wilkinson Murray responsible for half those costs. On that basis, UGL Rail is entitled to recover $380,632.92 in respect of THJV's costs. 10. According to Mr Johnson, THJV originally claimed $2,159,637.47 in respect of its ongoing supervision of the TIDC solution and provided UGL Rail with a breakdown of how those costs were calculated. Mr Johnson reviewed those costs and concluded that they were excessive and, eventually, those costs were compromised at $1,251,048.33. Mr Johnson was in the best position to assess the reasonableness of those costs. It is apparent that he went through them with some care. In those circumstances, I am prepared to accept that the amount on which UGL Rail eventually agreed was reasonable. Again, however, those costs related to all the work of implementing the TIDC solution. For the reasons I have given, Wilkinson Murray should only be liable for half those costs. 11. On that basis, Wilkinson Murray's liability is calculated as follows: Costs of installing panels $1,828,725.35 TIDC's costs of the dispute (50%) $400,000.00 THJV's costs of the dispute (50%) $380,632.92 THJV's supervision costs (50%) $625,524.17 UGL Rail's legal costs $358,717.85 Total $3,593,600.29 Other defences 1. Wilkinson Murray raises three other defences to UGL Rail's claim. First, it relies on a limitation defence in respect of the 2003 advice. Second, it submits that UGL Rail's claim is an apportionable claim and should be reduced to take account of UGL Rail's responsibility for the loss in accordance with Pt 4 of the CLA or Pt VIA of the TPA. Third, it submits that UGL Rail was guilty of contributory negligence. Limitation defence 1. Having regard to the conclusions I have reached, Wilkinson Murray's limitation defence does not arise. However, in case those conclusions are wrong, I should say something about it. 2. The proceedings were commenced on 14 April 2011. It is not in dispute that the limitation period in respect of a cause of action for breach of contract and for negligence is six years from the time that the relevant cause of action first accrued: see Limitation Act 1969 (NSW), s 14. It was not alleged that a claim under the TPA was statute barred. A cause of action for breach of contract first accrues when the breach occurs: Gibbs v Guild (1881) 8 QBD 296 at 302 per Field J; Hawkins v Clayton [1988] HCA 15; (1988) 164 CLR 539 at 583 per Deane J; Nykredit Mortgage Bank Plc v Edward Erdman Group Ltd (No 2) [1998] 1 All ER 305 at 308 per Nicholls LJ. In the present case, the breach must have occurred no later than the end of May 2003 when Wilkinson Murray gave its advice. The proceedings were not commenced until 14 April 2011, well outside the six year limitation period. For that reason, the claimed based on breach of contract is statute barred. On the other hand, a cause of action based on a contractual indemnity first accrues on the date when the costs and expenses in respect of which the indemnity is given were first ascertainable: Wardley Australia Ltd v State of Western Australia (1992) 175 CLR 514 at 532-3. And a cause of action in negligence first accrues when the plaintiff suffers actionable damage: Hawkins v Clayton; Commonwealth v Cornwell [2007] HCA 16; (2007) 229 CLR 519 at [36]. The costs and expenses in respect of which the indemnity was given were not ascertainable before the Settlement Deed was entered into. UGL Rail did not suffer actionable damage at least until it relied on Wilkinson Murray's advice in 2007. In both cases, it is clear that the limitation period had not expired. Apportionment 1. Part 4 of the CLA relevantly provides: 34 Application of Part (1) This Part applies to the following claims (apportionable claims): (a) a claim for economic loss or damage to property in an action for damages (whether in contract, tort or otherwise) arising from a failure to take reasonable care, but not including any claim arising out of personal injury, (b) a claim for economic loss or damage to property in an action for damages under the Fair Trading Act 1987 for a contravention of section 42 of that Act (as in force before its repeal by the Fair Trading Amendment (Australian Consumer Law) Act 2010) or under the Australian Consumer Law (NSW) for a contravention of section 18 of that Law. … (2) In this Part, a concurrent wrongdoer, in relation to a claim, is a person who is one of two or more persons whose acts or omissions (or act or omission) caused, independently of each other or jointly, the damage or loss that is the subject of the claim. … 35 Proportionate liability for apportionable claims (1) In any proceedings involving an apportionable claim: (a) the liability of a defendant who is a concurrent wrongdoer in relation to that claim is limited to an amount reflecting that proportion of the damage or loss claimed that the court considers just having regard to the extent of the defendant's responsibility for the damage or loss, and (b) the court may give judgment against the defendant for not more than that amount. (2) If the proceedings involve both an apportionable claim and a claim that is not an apportionable claim: (a) liability for the apportionable claim is to be determined in accordance with the provisions of this Part, and (b) liability for the other claim is to be determined in accordance with the legal rules, if any, that (apart from this Part) are relevant. (3) In apportioning responsibility between defendants in the proceedings: (a) the court is to exclude that proportion of the damage or loss in relation to which the plaintiff is contributorily negligent under any relevant law, and (b) the court may have regard to the comparative responsibility of any concurrent wrongdoer who is not a party to the proceedings. (4) This section applies in proceedings involving an apportionable claim whether or not all concurrent wrongdoers are parties to the proceedings. 1. Similar provisions are contained in Pt VIA of the TPA. 2. In its List Response to Amended Technology and Construction List Statement, Wilkinson Murray pleads that UGL Rail's claim is an apportionable claim and that UGL Rail is a concurrent wrongdoer within the meaning of s 34(2) with the result that its claim should be reduced to an amount reflecting that proportion of the damage or loss claimed that the Court considers just having regard to the extent of Wilkinson Murray's responsibility for the damage or loss. 3. In Rennie Golledge Pty Ltd v Ballard [2012] NSWCA 376; (2012) 82 NSWLR 231, the Court of Appeal held that, on the proper construction of Pt 4 of the CLA as a whole, a plaintiff whose own acts or omissions have been a cause of his own loss or damage is not a "concurrent wrongdoer": at [15] per Basten JA, at [128]-[143] per Campbell JA (Barrett JA not deciding). That decision is binding on me. It is a complete answer to Wilkinson Murray's defence based on Pt 4 of the CLA and Pt VIA of the TPA. Contributory negligence 1. In the alternative, Wilkinson Murray contends that the damages claimed by UGL Rail should be reduced on the ground that UGL Rail was guilty of contributory negligence, relying on s 9 of the Law Reform (Miscellaneous Provisions) Act 1965 (NSW) or, in the case of the claim based on the TPA, s 82(1B) of the TPA. In its List Response to Amended Technology and Construction List Statement, Wilkinson Murray pleads that UGL Rail was guilty of contributory negligence for various reasons, including the following: 1) It failed to have regard to the advice and recommendations set out in the 2003 report; 2) It failed to provide TIDC or its acoustic advisors with Wilkinson Murray's reports; 3) It failed to verify the advice it was given; 4) It failed to ascertain whether TIDC and THJV would be satisfied if targets were met at most frequencies and were within 5 to 10 per cent of the specification at other frequencies; 5) It failed to arrange testing at a time when additional panelling could be installed in the tunnels and failed to configure the services in the tunnel so that additional panelling could be installed; 6) It failed to ensure that testing of the panelling was carried out in a way that was suitable for a tunnel rather than a room; 7) It installed panelling when it knew that that panelling may not be sufficient to meet specification. 1. In my opinion, none of these matters establish that UGL Rail was guilty of contributory negligence. As to the third and sixth points, UGL Rail did not have expertise in acoustic engineering. It is for that reason that it engaged Wilkinson Murray. It was entitled to rely on the advice it received from Wilkinson Murray concerning the amount of panelling that should be installed and how the testing should be carried out. Accordingly, it had no obligation to verify the advice that it received or to take steps to ensure that the testing was carried out properly. It relied, and was entitled to rely, on Wilkinson Murray to do those things. 2. As to the first and seventh points, UGL Rail's failure to rely on advice given by Wilkinson Murray, if there was such a failure, did not amount to contributory negligence. Either UGL Rail relied on the advice it received or it did not. If it did, then it has a claim. If it did not, then it does not; and the question of contributory negligence does not arise. 3. As to the second point, no reason was advanced by Wilkinson Murray for why UGL Rail should have passed its reports on to TIDC. UGL Rail's obligations were to THJV. Those obligations were to install sufficient panelling to meet the reverberation specification in the Works Brief. It was not part of that obligation to pass advice it received in order to comply with that obligation on to TIDC. 4. As to the fourth point, it appears to assume that the targets were met at most frequencies and were within 5 to 10 per cent of the specification at other frequencies. That was not the case. In addition, the complaint appears to be a complaint that UGL Rail failed to follow Wilkinson Murray's advice. For the reasons I have given, that does not amount to contributory negligence. 5. As to the fifth point, I have found that there was room to install additional panelling in the tunnel. 6. It follows that the defence of contributory negligence must fail. Orders 1. There should be judgment in favour of UGL Rail in the sum of $3,593,600.29. 2. It was not suggested that GST was payable on any amount that UGL Rail might be entitled to recover and no claim for GST was made on the amount claimed by UGL Rail. Consequently, I can see no reason to allow GST on the amount of the judgment. 3. UGL Rail claims interest from the date it entered into the Settlement Deed to the date of judgment. In my opinion, interest should run from the date it paid the amounts payable by it in respect of the settlement. That date or those dates do not appear to be clear from the evidence, although it seems that all payments had been made by the end of July 2009. Consequently, in my opinion, it is reasonable to allow interest from 31 July 2009 to the date of judgment. On that basis, UGL Rail is entitled to interest in the amount of $1,465,425.88. 4. I will hear the parties in relation to costs. 5. Consequently, the orders of the Court are: 1. Judgment for the plaintiff in the sum of $5,059,026.17; 2. Stand the matter over to a date to be fixed with my Associate for the hearing of any argument concerning costs. ********** DISCLAIMER - Every effort has been made to comply with suppression orders or statutory provisions prohibiting publication that may apply to this judgment or decision. The onus remains on any person using material in the judgment or decision to ensure that the intended use of that material does not breach any such order or provision. Further enquiries may be directed to the Registry of the Court or Tribunal in which it was generated. Decision last updated: 27 March 2015