Limit (No 3) Limited v ACE Insurance Limited

REIN J: The defendant, ACE Insurance Limited (“ ACE ”), issued an insurance policy to a joint venture consisting of Obayashi Corporation (“ Obayashi ”) and McConnell Dowell South East Asia Pte Limited (“ McConnell Dowell ”). “ The JV ”, as I shall refer to it, was engaged by PowerGrid Singapore Limited (“ PowerGrid ”) in August 2001 to design and construct two power transmission cable tunnels from the Senoko Power Station to Gambas Avenue in Singapore along Woodlands Avenue 8 (“ Woodlands Avenue ”).

The works involved, inter alia , the creation of shafts with diaphragm walls and extensive tunnelling. As a result of this work damage was caused to the property of third parties, including that of PowerGrid whose power cables already ran underneath the area next to Woodlands Avenue.

PowerGrid and the JV received claims from third parties and PowerGrid and the third parties sought indemnity from the JV.

The JV sought indemnity from ACE, and when ACE did not agree to meet the claim, from Limit (No 3) Limited (“ Limit ”), a Lloyds syndicate, and other Lloyds insurers (to whom I shall refer collectively as “ Lloyds ”) for which Limit acted as lead underwriter.

ACE did not indemnify the JV and Lloyds indemnified the JV. Lloyds now seeks “recoupment indemnity or contribution” from ACE on the basis that payments to the JV totalling S$13 million “operated to relieve ACE of its liability to the JV.”

ACE denies that it was liable to indemnify the JV and hence denies any liability to Lloyds. ACE also denies that Lloyds had any liability to indemnify the JV.

We are concerned here with three Lloyds policies. One for 2001 (“ the 2001 policy ”), one for 2002 (“ the 2002 policy ”) and one for 2003 (“ the 2003 policy ”). Nothing turns on any differences between the policies and I shall use the 2001 policy as the source of relevant clauses.

I received extensive and helpful submissions from counsel for both parties, the last of which submissions was received by email on 9 April 2009. The Issues

The issues which arise in relation to all these policies are: The ACE policy (1) What was the cause of the damage to the property of third party property owners (including PowerGrid)? (2) Did the cause (or causes) of damage identified in answer to [0] give rise to a liability in the JV to the third party property owners (including PowerGrid)? This question, it is agreed, is to be determined in accordance with the law of Singapore. (3) Have Lloyds established that the JV incurred the requisite “legal liability” within the meaning of the insuring clause, i.e. was a judgment, arbitral award or settlement required? (4) Was the damage to property giving rise to the liability arising from the cause referred to in [0] above, sustained within the ACE policy period? If not, does the ACE policy provide cover in respect of any legal liability of the JV as a result of this damage? This issue relates only to part of the damage to property the subject of the claim. (5) What is the proper construction of General Condition 14 (“ GC 14 ”) of the ACE policy and what is its effect on exclusion 5 (the professional services exclusion (“ the PSE ”))? (6) Did any liability of the JV arise out of loss of or damage to property caused by the removal or weakening of support so as to attract cover under sub-clause (iv)(b) of GC 14 of the ACE policy? (7) Did any liability of the JV arise out of a negligent act of the insured, its servants or agents so as to attract cover under sub-clause (iv)(c) of GC 14 of the ACE policy? (8) Did any liability of the JV in respect of any, and if so which, of the causes of damage referred to in the answer to [0] arise “out of the rendering of or failure to render professional services, during the policy period by the JV or by any other person whose acts the JV was legally liable including: (1) the preparation or approval of maps, plans, opinions, reports, survey, designs or specifications; and (2) supervisory inspection or engineering service.” within the meaning of the PSE (exclusion 5) of the ACE policy? (9) If so, and if any liability of the JV falls within the scope of GC 14, is ACE’s liability to indemnify the JV excluded by the PSE of the ACE policy? (10) Does the rule in Wayne Tank operate to exclude any liability of the JV where the cause was the result of the rendering of or failure to render professional services and some other cause? (11) As a matter of construction, is any liability of the JV in respect of the TBM/G4 problem covered by the ACE policy if this was an intentional act by the JV for which indemnity was claimed? (12) How many ‘occurrences’ were there and hence what limit and deductible, if any, apply if ACE is otherwise liable? The Lloyds policy (13) The same issue of causation arises as under the ACE policy set out in [0] above. (14) Have Lloyds established that the JV incurred the requisite “legal liability” within the meaning of the insuring clause, i.e. was a judgment, arbitral award or settlement required? (15) Was the JV unable to “obtain an admission of liability from ACE” within the meaning of clause 1.17 in Section 4 of the Lloyds policy? (16) If so, were Lloyds liable to indemnify the JV for any legal liability of the JV under Section 1 of the Lloyds policy? (17) If so, was any legal liability of the JV excluded from indemnity under Section 1 of the Lloyds policy by reason of: (a) clause 5.11 (the professional services exclusion)? (b) condition 7.3 (the reasonable precautions condition)? (18) Does the rule in Wayne Tank operate to exclude any liability of the JV where the cause arose out of the rendering of or failure to render professional services and some other cause? (19) Did sub-clause 1.17(b) of the Lloyds policy operate so that cover afforded under the Lloyds policy was the same as that provided under the ACE policy? If so, what was the extent of this cover having regard to proper construction of the PSE and GC 14 of the ACE policy? (20) Was Lloyds liable to indemnify the JV by reason of the ‘drop down cover’ provision in sub-clause 1.17(c) even if an exclusion in the Lloyds policy otherwise applied? (21) How many ‘ occurrences’ took place within the meaning of the Lloyds policy? (22) If not liable under Section 1, were Lloyds obliged to provide indemnity to the JV under Section 2 of the Lloyds policy? Further Issues (23) Assuming that both the ACE policy and Section 1 of the Lloyds policies require indemnity of the JV, does the payment by Lloyds of the amount of $13 million to the JV entitle it to recover from ACE and, if so, in what amount either by way of recoupment, indemnity or contribution? (24) The issue of quantum has been further complicated by several matters to which I shall later refer. The Properties and Infrastructure Damaged

Woodlands Avenue had underneath it (or almost so) electrical cables owned by PowerGrid, a sewage pipe owned by the Singapore Public Utilities Board (“ PUB ”) and a high-pressure gas pipeline owned by PowerGas Limited (“ PowerGas ”). Adjacent to the shaft and tunnels created by the JV are located a number of properties, the buildings and infrastructure (such as concrete aprons) of which the owners have claimed have suffered damage. Although originally the fact that the buildings and infrastructure had been damaged was not admitted, that is no longer an issue in the proceedings. It is clear that the damage arose as a result of the works and any faint argument to the contrary was not maintained. The JV Contract with PowerGrid

The JV entered into a contract with PowerGrid on 17 April 2001 (“ the PowerGrid contract ”). The PowerGrid contract can be described as a Design and Construct Contract. The purpose and scope of the project is described in the General Specification as: “(1) PowerSenoko Ltd is repowering the Senoko Power Station. The electrical power will be transmitted to substations at Woodlands Avenue 8, Tampines, Chua Chu Kang etc. To facilitate this transmission, a cable tunnel is required to be constructed beneath the existing cable reserve within/outside the Senoko Power Station, Admiralty Road West and Woodlands Avenue 8. This cable tunnel will, initially, accommodate a total of 8 Nos. 230kV and 2 Nos. 66kV power circuits which will be installed in phases. Eventually, these circuits will be upgraded to a total of 6 Nos. 400kV power circuits. (2) The scope of the Specification is to provide for the Works to be fully completed in every detail for the function designated. All items of construction, accessories, fittings, sundries, documentation, apparatus, or labour whether specified in detail or not but, which, in the opinion of the Engineer are usual or necessary for the satisfactory completion of the Works shall be deemed to be included in the Specification. (3) The Contractor shall so design and construct the Works that the overall performance requirements set out in Sections 4 to 11 of this General Specification are achieved.”

By clause 31 of the Special Conditions of Contract the JV was to be “responsible for the design, construction and maintenance of the Works, notwithstanding that the design submission may be accepted for construction”.

Clause 9.2.1 of the Material and Workmanship Specification – Section A imposed an obligation on the JV to “carry out the tunnelling operations with the utmost care and with the strictest precautions at all stages to prevent loss of ground, and to safeguard the adjacent properties against subsidence. Ground movements resulting from these works shall be restricted to a minimum.”

It was a term of the PowerGrid contract (General Condition 27) that, inter alia : “(1) Injury to Property – The Contractor shall be liable for and shall indemnify the Company and the Engineer against all liabilities, losses, claims, demands, proceedings, damages, costs, charges and expenses whatsoever in respect of any injury or damage to any property real or personal arising out of or in the course of or by reason of the execution and maintenance of the works unless it is shown to the satisfaction of [PowerGrid] or the Engineer that such injury, damage or loss was not due to any negligence, omission or default of the contractor, his servants or agents or of any subcontractor, his servants or agents. ….. (4) The Contractor shall forthwith and as a condition precedent to the commencement of any work under this Contract insure against the aforesaid risks or matters in the joint names of the Company, the Engineer and the Contractor with an insurance company to be approved by the Company in writing and for such amount as the Company may direct. (Please refer to the Appendix for the amount.) The Contractor shall make all payments necessary for the above purposes on the first day on which the same ought to be paid. The Contractor shall deposit the policy or policies for such insurance with the Company before the commencement of any work under this Contract and shall produce on demand to the Company the receipt of each of the said payments. Provided always that if the Contractor shall at any time fail to keep himself insured as aforesaid the Company may do all things necessary to effect or maintain such insurance and deduct such moneys expended by the Company for that purpose from any moneys due or to become due to the Contractor.”

General Condition 28 provided: “(1) The Contractor shall forthwith and as a condition precedent to the commencement of any work under this Contract take out at his own expense with an insurance company to be approved by the Company in writing a policy or policies of insurance indemnifying the Contractor, the Company and the Engineer from all liabilities arising out of claims by any and every workman employed in and for the performance of this Contract for payment of compensation under or by virtue of the Workmen’s Compensation Act or any other law amending or replacing such Act and from all costs and expenses incidental or consequential thereto. (2) The said policy or policies so taken out shall be deposited with the Company and the Contractor shall maintain it or them in full force and effect by payment of all premiums from time to time of the first day on which the same ought to be paid and until the completion of this Contract and upon demand the Contractor shall produce to the Company the last receipt for payment of such premiums.”

General Condition 29 provided: “(1) The Contractor shall forthwith and as a condition precedent to the commencement of any work under this Contract insure against loss or damage by fire or other causes all works and buildings constructed or in the course of construction in pursuance of or for the purposes of this Contract and all materials and other things delivered on to the site and approved by the Engineer and ready for incorporation in such works and buildings and shall keep the same insured until such works, buildings and things respectively are handed over to the Company. (2) The said insurance shall be effected with an insurance company to be approved in writing by the Company and in the joint names of the Company, the Engineer and the Contractor for the full value of the works executed together with all materials on site including any materials supplied by or the property of the Company; and the Contractor shall deposit with the Company the policies and the receipts for the premiums paid for such insurance. (3) In case of failure by the Contractor to effect or renew such insurance the Company may effect or renew such insurance and pay the premium in respect thereof and deduct the amount so expended from any moneys due or to become due to the Contractor. (4) In the event of loss or damage by fire or other causes to any works, buildings or things insured under this clause all moneys received by the Company under the insurance policies shall be paid by the Company to the Contractor by such instalments as the Company may deem reasonable and shall be applied by the Contractor in or towards the rebuilding, repair or replacement of the works, buildings, materials or things destroyed or damaged and the Contractor shall on receipt of such moneys proceed with all due diligence to effect such rebuilding, repair or replacement as aforesaid and shall have no claim to any payment in respect thereof beyond the payment to him of the said moneys; provided that in case of any loss or damage by fire or other causes as aforesaid the Company shall allow the Contractor such extension of time for the performance of this Contract as shall be just and reasonable.”

Special Condition 24 provided: “In addition to the provisions of Clauses 27, 28 and 29 of the General Conditions of Contract, the following provisions shall apply to the insurance to be effected by the Contractor: (1) The insurance shall be effected with an insurer approved by the Company and in terms complying with the specimen copies attached hereinafter. (2) In addition to Clause 27(4) of the General Conditions of Contract, the following minimum limits of indemnity shall apply i. For any one accident: S$50,000,000/- ii. For Period of Indemnity: Unlimited (3) The specimen insurance is the minimum requirement stipulated. The Contractor shall satisfy himself of the adequacy of the insurance therein and shall increase the coverage if he deems necessary.”

Annexed to the PowerGrid Contact as Appendices A and B were the specimen wordings contemplated by Special Condition 24. Appendix B was forwarded to ACE by the JV’s broker AON Risk Services Singapore (Insurance Brokers) Pte Ltd (“ AON ”) and it became incorporated in the ACE policy as GC 14. Some wording found in Appendix A was also forwarded by AON to ACE but these wordings were not incorporated in the ACE policy. One of the wordings sent but not incorporated was a wording described as “RIBA CLAUSES 19(2)(a) COVER”: see Exhibit L at p 1-355. The exclusions found within the wording forming part of Appendix A with the heading “EXCEPTIONS” were not forwarded by AON to ACE. The content on the Appendices is relied on by ACE in a manner I shall describe below. The ACE policy

The following are relevant aspects of the ACE policy (with emphasis added). (1) The ACE policy describes itself as “Liability Insurance”: see Exhibit L at p 1-220. It names PowerGrid, the JV, “Sub-contractors, Co-contractors, Suppliers in any tier”, Development Resources Pte Ltd/Mott-MacDonald Singapore Pte Ltd (whom I shall refer to as “ the Engineer ”) and “[a]ny other interested party connected with the contract and for whom [the JV] is required to provide insurance for their respective rights and interests.” (2) It describes the project as: “ Design and Construction of Transmission Cable Tunnel from Senoko Power Station to Gambas Avenue including Equipment, Building, Verification/Access Shafts and Associated Mechanical & Electrical Equipment.” (3) It defined the Period of Insurance as: “ Whole period of project from 17 April 2001 to 27 May 2003 (Both dates inclusive) followed by a 24 month Maintenance Period. ” (4) There was a limit of liability of $5 million for “any one Occurrence or series of Occurrences arising out of one event”. (5) The coverage was: “The Insured will pay to or on behalf of the Insured all sums which the Insured shall become legally liable to pay for any compensation in respect of: 1. Personal Injury 2. Property Damage Occurring within the Territorial Limits during the Period of Insurance as a result of an Occurrence happening in connection with the Project.” (6) ‘Property Damage’ was defined as: “(a) physical injury to or loss or destruction of property including loss of use therefrom; (b) loss of use of property which has not been physically injured or destroyed provided that such loss of use of caused by an Occurrence.” (7) ’Occurrence’ is defined as “ Any one loss, disaster or casualty or series of losses, disasters or casualties arising out of the one event .” (8) There were a number of exclusions for example, a “Vehicles” exclusion for any liability for claims caused by any vehicle covered by a statutory insurance, a Libel and Slander exclusion and a Employer’s Liability exclusion. The PSE was in the following terms: “ Professional Indemnity This policy does not apply to personal injury, property damage or advertising injury arising out of the rendering of or failure to render professional services, during the policy period by any Named Insured or by any other person for whose acts to Named Insured is legally liable including: - (a) the preparation or approval of maps, plans, opinions, reports, surveys, designs or specifications and (b) supervisory inspection or engineering services .” (9) GC 14 was in the following terms: “The clauses contained in this special endorsement shall not in any way be deleted, altered or amended by other provisions in the policies and any other subsequent endorsements and where contradiction arises, this special endorsement shall precede. (i) PowerGrid Ltd and the Consultants M/s Development Resources Pte Ltd shall be insured in joint names with the Contractor or any sub-contractor as a party to the policy. Each of the parties comprising the Insured shall be considered as a separate and distinct unit and words "the Insured" shall be considered as applying to each party in the same manner as if a separate policy had been issued to each of the said parties and the Insurers hereby agree to waive all rights of subrogation or action which they may have or acquire against any of the aforesaid parties arising out of any accident in respect of which any claim is made hereunder provided nevertheless that nothing in this clause shall be deemed to increase the Limit of Indemnity in respect of any one occurrence as stated in the Schedule. (ii) Any employee of PowerGrid Ltd and the Consultants M/s Development Resources Pte Ltd shall not be deemed to be a person engaged in the service of or acting on behalf of the Insured notwithstanding that PowerGrid Ltd and the Consultants M/s Development Resources Pte Ltd are named as Insured under the policy. (iii) Property belonging to PowerGrid Ltd shall not be deemed to be property belonging to or in the charge or under the control of the Insured or any servant or agent of the Insured except for that part of the property temporarily occupied by the Insured for work therein. (iv) The Insurers shall indemnify the Insured in respect of injury, illness, loss or damage covered by or in connection with or arising from: a. any lift elevator hoist crane or other lifting machinery owned or used by the Insured or for the maintenance of which the Insured is responsible; b. loss of or damage to any property or land or building caused by vibration or by the removal or weakening of support (including damage to property occasioned by or resulting from any such loss); c. any negligent act of the Insured, servants or agents of the Insured; d. bursting of any pressure parts of any steam boilers or any economizer, any vessels or apparatus (including steam turbine or engine or other steam driven machinery) intended to operate under steam pressure. subject to the limits of indemnity for any one occurrence as stated in the Schedule.” The Lloyds policy

The Lloyds policy (which I should note it was agreed is governed by Victorian law) had the following relevant provisions (with emphasis added). (1) It named as the insured McConnell Dowell but this was apparently extended to the JV by endorsement. (2) It had a US$12 million cover for any one occurrence for Section 1 and a US$18 million in the aggregate. (3) It had a US$12 million liability for any one claim for Section 2 limited to US$18 million in the aggregate. (4) Section 1 was a General and Product Liability cover requiring Lloyds to indemnify the JV against liability of the JV to pay damages for compensation (including claimant’s costs, fees and expenses) in respect of: (i) injury to any person; (ii) damage to property; and (iii) advertising injury. (5) Clause 5 had a number of exclusions such as Vehicles, Aircraft, Employer’s Liability, Asbestos and: “ 5.11 Professional Indemnity liability arising out of the rendering of or failure to render professional advice or service by the Insured or error or omission connected therewith, but this Exception 5.11 does not apply to the rendering of or failure to render professional medical advice by medical persons employed by the Insured to provide first aid and other medical services on the Insured’s premises or work sites.” (6) Section 2 was a Professional Indemnity section by which Lloyds agreed to indemnify the JV “against any claim made against the Insured and notified to the Insurer during the Period of Insurance in respect of any civil liability incurred by or on behalf of the Insured in the conduct of the Profession.” (7) ‘Profession’ for Section 2 was defined as: “2.4 Profession means: the business of the Insured arising out of the activities of professionals, not being limited to all project managers, builders, site superintendents, quantity surveyors, estimators, managers, engineers, foremen, architects, designers, technicians and technical service providers of whatever description and any other person who exercises any independent judgement of a professional nature in the course of their work.” (8) Clause 3.8 of Section 2 provides: “3.8 Loss Mitigation Expenses The Insurer will subject to the terms, Conditions, endorsements and Exceptions of this Policy, indemnify the Insured against costs and expenses necessarily incurred in respect of any action taken to mitigate loss or potential loss that otherwise would be the subject of an indemnifiable Claim under this Section 2. The onus of proving a claim under this Extension shall be upon the Insured, who will be obliged to obtain prior approval from the Insurer during the Period of insurance of their assumption of liability and probability of eventual Claim, in the event of mitigation action not being timorously undertaken.” (9) Clause 4.8 of Section 2 provides: “This Section does not provide any indemnity against any Claim: ….. 4.8 Bodily Injury/Property Damage arising from or attributable to any Damage or injury defined in Section 1 Definitions 4.5 and 4.7 and more specifically paid under Section 1 of this Policy, unless arising out of advice, design, specification or the performance of or omission to perform a professional duty when this Section 2 will apply.” (10) There was a ‘Difference in Conditions Cover’ clause in the following terms: “1.17 Difference In Conditions Cover It is acknowledged that it is customary for the Insured, or for other parties on behalf of the Insured (including joint venture partners principals and contractors) to effect, on behalf of the Insured, insurance coverage specific to a particular risk, project or agreement. Such insurance is described as Underlying Insurance. Furthermore, the Insurer, or others on behalf of the Insurer, may issue policies of Insurance in overseas jurisdictions, for the purpose of local legal requirements, servicing, premium payment and claims handling, referred to as Local Policies. In circumstances where Underlying Insurance or a Local Policy is applicable, this Policy shall be deemed to be the Master Policy. It being understood and agreed that: (a) In the event of the Insured being indemnified by a Local Policy or Underlying Insurance in respect of a claim for which indemnity is available under this Policy, the insurance afforded by this Policy shall be excess insurance over the applicable limit of indemnity of the Local Policy or Underlying Insurance. Coverage under this Clause shall not apply unless and until the Local Policy or Underlying Insurance shall be obligated to pay the amount of the Underlying Limit on account of a Claim or Claims, which would, but for the limit of indemnity of the Local Policy or Underlying Insurance, be covered hereunder; (b) if such Local Policy or Underlying Insurance covering the Insured against their legal liability in respect of damage to property or personal injury to third parties or employees, provides indemnity to the Insured by virtue of its scope of cover, definitions or conditions in respect of legal liability, costs and expenses which are not provided under the terms, Conditions and Exceptions of this Master Policy, then the cover afforded under Section 1 only of this Master Policy shall provide such Indemnity to the same extent as provided by the Local Policy or Underlying Insurance. Notwithstanding the provisions of this clause (b), Exceptions under Section 5 – Exceptions Applicable to All Sections, shall always apply to any Indemnity provided hereunder. (c) should any such Local Policy or Underlying Insurance, by virtue of its scope of cover, definitions, conditions or limits of liability, not indemnify the Insured in whole or in part in respect of such legal liability, costs and expenses as herein provided, this Master Policy, subject to its terms, Conditions and Exceptions, shall provide indemnity to the extent that such indemnity is not provided by the terms and conditions of such Local Policy or Underlying Insurance. In the event that the Insured cannot obtain an admission of liability from the insurer of an Underlying Insurance and/or Local Policy and/or the Underlying Insurance and/or Local Policy fails or is reasonably likely not to indemnify the Insured, then the Insurer of this Master Policy shall be obligated to indemnify the Insured and defend any actions. The Provisions of this clause are subject always to the terms, Conditions and Exceptions of this Policy. PROVIDED THAT: (i) the Limit of Liability under this Master Policy shall be reduced by an amount equal to the indemnity recoverable under any Local Policy. Notwithstanding the number of separate Insureds under all Local Policies, the aggregate liability of the Insured under this Master Policy to the Insured’s jointly and severally shall not exceed the Limits of Liability specified herein. (ii) in the event of cancellation of an Underlying Insurance or Local Policy or reduction or exhausting of the limits of indemnity thereunder, this Policy shall: (a) in the event of reduction, pay in excess if the reduced Underlying Limit; (b) in the event of cancellation or exhaustion, continue in force as Underlying Insurance. (iii) in respect to indemnity provided under clauses 1.17(b) and 1.17(c) only, the Insured shall be obligated to pay the Deductible for each claim.” (11) General Condition 7.3 of the Lloyds policy provided that: “The Insured shall (a) take reasonable precautions to prevent or minimise liability and maintain efficient ways, works, machinery, fencing and plant and shall make reasonable endeavours to comply with all statutory obligations and regulations imposed by any authority.” The Project Background

The area in which the excavation and tunnelling occurred consists of Bukit Timah Granite with, in much of the relevant area, an alluvium of “Kallang Formation” made up of “E” type clays and “F” type sands. Unit E clays occur as “peaty clay with a soft consistency” and are “relatively compressible”. The F1 sand unit “represents a layer of relatively high permeability or a layer with good drainage characteristics”: see Dr Redman’s report of 14 December 2007 (“ Exhibit E ”) at pp 8 – 9. The bedrock is either “G1” (fresh to slightly weathered rock) or “G2” (moderate to highly weathered rock). Between the E/F layer and bedrock there is much residual “G4” material which is highly weathered rock that is closer to soil than rock: see Dr Pells’ report of 10 September 2008 (“ Exhibit 4 ”) at p 15, and it is described by Mr Embery (the JV’s Construction Manager) as “decomposed rock”: see T733.30. A depiction of the general area of the Kallang Formation in two zones is found in Exhibit 1 in yellow: see also Figures 1 and 6.2 of Exhibit 4 and Exhibit 10. The precise boundaries and content of the Kallang Formation appeared to be quite difficult for the experts to determine: see T774. A number of the buildings which suffered damage are located above the Kallang Formation.

Twin service tunnels (an east and a west tunnel) were to be constructed with a length of approximately 1.7 kilometres and a width each of about 3.7 metres. At Woodlands Avenue there was to be constructed a substation and related facilities. The tunnels were to terminate near the junction of Woodlands and Gambas Avenues and a shaft was constructed there known as the Gambas shaft. Another shaft, at a point further north along Woodlands Avenue, which shaft was known as the Woodlands shaft, was also constructed. Exhibit 1 contains a useful depiction of the project. An aerial photograph also gives some indication of the proximity of buildings to the Gambas and Woodlands shafts: see Figure 2.2 of Exhibit 4. To excavate the main tunnel a refurbished Tunnel Boring Machine manufactured by The Robbins Company (“ Robbins ”) and known as a “double shield TBM” was obtained by the JV (“ the TBM ”): see Figures 5.8 and 5.9 of Exhibit 4 and photo 2 annexed to Exhibit B.

The project commenced in 2001 with the creation of the Gambas shaft followed by the Woodlands shaft. Two lower adits were created between the base of the Woodlands shaft and what was to be the east tunnel (“ EA1 ” and “ EA2 ”), and two upper adits were created between the Woodlands shaft and the west tunnel (“ WA1 ” and “ WA2 ”). The only upper adit in respect of which water flows are relevant is WA2 and reference to an “ upper adit ” in the balance of these reasons will be a reference to WA2. I have understood references in expert reports to the “upper adit” to mean WA2 unless otherwise specified.

The TBM commenced the main tunnelling work on 10 March 2002 and it became entombed on 7 June 2002 at chainage 1141 following significant over-excavation in G4 material (this over-excavation is known as the “ TBM/G4 problem ”).

Prior to excavation of the shafts the JV had available to it a soil investigation report from Kiso-Jiban Consultants Co Ltd (“ Kiso-Jiban ”). A firm of geotechnical engineers retained by the JV PB Merz McClellan Pte Ltd (“ PBMM ”) proposed some additional field and laboratory tests: see Exhibit R: Court Book (“ CB ”) vol 3.1 at pp 3-103, 3-113 and 3-159 – 3-191, which the JV engaged Kiso-Jiban to carry out. PBMM were retained “to provide professional design and support services” in relation to the project: see para 13 of Mr Embery’s affidavit of 6 March 2008 (“ Mr Embery’s first affidavit ”). The report of Kiso-Jiban is found at CB vol 3.1 at pp 3-45 – 3-78. In October 2001 further bore hole investigations were conducted by Kiso-Jiban the results of which were provided on 1 November 2001 and appear at CB vol 3.1 at pp 3-3550 – 3-3562. A depiction of the ground in elevation as revealed by the Kiso-Jiban November results is found at Figure 5.10 of Exhibit 4 at p 39. A cross-section is found of that at Figure 7.5 of Exhibit 4. That diagram shows that the information available by November 2001 indicated that the east tunnel would be bored through a mixture of G4 and G2, hence the phrase “mixed face” material, which appears in the evidence.

The diaphragm walls in the Woodlands and Gambas shafts were constructed by Sambo Tosfoc Pte Ltd (“ Sambo ”) pursuant to a subcontract with the JV: see CB vol 2.1 at pp 2.1.1 – 2.1.85. The walls consist of a number of concrete panels inserted into narrowly excavated slits before removal of the soil within the shafts: see Exhibit 3 and Figure E of Dr Pells’ report of 19 February 2009 (“ Exhibit 6 ”).

The Gambas shaft had an overall depth of approximately 50.7 metres and an external diameter of about 12.8 metres. Pneumatic and standpipe piezometers (P1R, P2, S1R and S2) were installed in the vicinity of the Gambas shaft. The purpose of the piezometers was to monitor changes in groundwater conditions. Their locations are depicted in Figure 4 of Exhibit E and at CB vol 3.26 at 3-19423A.

Similar arrangements were made for the Woodlands shaft (P3, P4, S3 and S4). Their locations are depicted in Figure 4 of Exhibit E and at CB vol 3.26 at 3-19423A. The Woodlands shaft had an overall depth of 56.8 metres and an external diameter of about 15 metres.

The JV also arranged ground settlement monitoring – initially above the PowerGas pipeline which ran underneath Woodlands Ave: see the GS markings, in particular GS84 – GS90, which are broadly opposite the Woodlands shaft and adits: see Exhibit 1 and CB vol 3.11 at p 3-7053 (which details settlement in that region). All settlement monitoring was conducted by registered surveyors, Wisecan Engineering Services Pte Ltd (“ Wisecan ”).

In relation to the Gambas shaft the installation of the diaphragm wall commenced on 13 July 2001, the wall was completed on 23-25 August and toe grouting commenced on 27 August. Solid rock was reached on 27-31 October 2001: see ACE’s chronology and Table 2 of Dr Redman’s report of 10 November 2008 (“ Exhibit F ”).

In relation to the Woodlands shaft, the installation of the diaphragm wall commenced on 1 August 2001 and was completed on 3 September 2001. “Rock material” was reached on 28 November 2001: see ACE’s chronology at p 17 and Table 1 of Exhibit F.

The adits from the Woodlands shaft to the east and west tunnels underneath Woodlands Avenue are depicted in Exhibit 1 and more detail is shown on Exhibit 2. The work on the adits either commenced in early December 2001: see Mr Embery’s first affidavit at para 26, the affidavit of Mr Embery of 5 November 2008 (“ Mr Embery’s second affidavit ”) at para 30 and see CB vol 3.7 at 3-4571, or on 19 January 2002: see Figure 6 to Exhibit F. The document at CB vol 3.9 at p 3-5626 is evidence that work related to the construction of the adits commenced on 10 January 2002 and the central pillar wall between the adits was cast on 19 January 2002. That document, and also a document at CB vol 3.7 at p 3-4697, support the conclusions that work on the adits proper commenced on 19 January 2002: see also ACE’s chronology at pp 25 and 28.

I have referred to the existence of the Kallang Formation, and to the presence of G4 material, both of which are relevant to what occurred. Another geological feature of significance is the existence of “dykes”. A “dyke” is a near vertical protrusion of volcanic material into fractures in the parent rock – here granite. Dykes are commonly found in this part of Singapore and an example of one can be seen in Exhibit 13, and see also Exhibit 4 at para 13. There was evidence that the intrusion of volcanic material often has significant impact on the integrity of the granite nearby due to the intense heat of the intruded material when it enters the existing formation. A dyke was encountered at chainage 1456 (“ the TBM/dyke problem ”). Dr Pells, in Figure Z of his report of 16 December 2008 (“ Exhibit 5 ”) shows how water ingress to the east tunnel increased dramatically when the dyke was encountered at chainage 1456 in June 2003; see also Exhibit E at paras 228 – 230.

The presence of another dyke on one side of the proposed site for Woodlands shaft was established by the results from bore hole NB8: see Exhibit 1, designated by the symbol within the ring of the shaft diaphragm wall. Exhibit 13 shows a dyke in adit WA1 above the gas pipeline. The precise extent of the dyke (or dykes) in or near the Woodlands shaft is not clear. Mr Embery identified, on Exhibit 1, the two locations at which the dyke was observed – within Woodlands shaft and above the gas pipeline in adit WA1 – and he drew on Exhibit 1 an extension of the dyke. He showed more detail of the observed dyke above the gas pipeline on Exhibit 2. Dr Redman (one of Lloyds’ experts) and Dr Pells (one of ACE’s experts) agreed that the dyke ran through the Woodlands shaft: see Exhibit 2. On Exhibit 2, which put the Woodlands shaft and adits in focus, it will be seen that Dr Redman thought originally that a dyke went through the shaft and along WA1 (as depicted in light blue), and Dr Pells thought a dyke might have gone through both WA1 and WA2 (depicted in red). Dr Pells noted in Section 3.2 of Exhibit 6 that there was an absence of logging records in relation to adit WA2 so there is no certainty about whether the dyke which Dr Pells believes went through the WA1 and WA2 adits in fact did so, but there is agreement that a dyke (or dykes) did intersect with the Woodlands shaft and with WA1.

Another aspect of the geotechnical conditions, which is connected to what I have already described, is the permeability of the G4 material. Permeability ratings were given by PBMM with G4 rated as having good drainage similar to clean sands and sand gravel mixtures: see Exhibit E at p 7.

Both parties have retained experts who have given evidence in the proceedings both in the form of reports and orally. Lloyds rely on the evidence of Dr Redman, a geotechnical engineer, and Mr Cooper, a tunnelling expert, and on Mr Bruce Embery. Mr Embery was the Construction Manager engaged by the JV. He is a civil engineer by training and has had extensive experience in excavation work. He was both a lay and expert witness and in relation to his evidence as an expert he acknowledged and agreed to be bound by the Expert Code of Conduct. Lloyds also relies on the evidence of Mr Jewell a hydrogeologist who carried out hydrological modelling with a view to demonstrating the likely effect of lowered groundwater. ACE relies on the evidence of Dr Pells, a geotechnical engineer, and Mr Sternath, a tunnelling expert.

There were some aspects of the matter on which Dr Redman did not feel able to comment due to a lack of expertise so that the contest is not always between Dr Redman and Dr Pells but sometimes between Mr Cooper on the one hand, and Dr Pells and Mr Sternath on the other. ACE submitted that on some topics covered by Dr Pells there was no contrary expert opinion advanced by Lloyds. Dr Redman’s Approach

Dr Redman divides up the damages and repair work into four separate incidents although he accepts that they are not all wholly discrete. I summarise his explanation for the incidents as follows. Woodlands shaft (1) The JV excavated the Woodlands shaft to enable access to the areas beneath Woodlands Avenue to be tunnelled. (2) The design required the excavation or cutting of narrow channels followed by the installation of a diaphragm panel wall into those channels before removing the bulk of the soil within the shaft wall. (3) The design required the sub-contractor to anchor the base of the diaphragm wall into G1/G2 rock beneath the wall and to carry out grouting work beneath the base of the diaphragm wall to a level of 11 metres. (4) After the diaphragm wall was installed and the soil removed so as to create a shaft, groundwater levels in the adjacent area dropped. (5) The main reason that the groundwater levels in the adjacent area dropped was because: (a) water leaked through the diaphragm wall; and (b) water flowed into excavation below the diaphragm wall at and within the toe grouted zone. (6) The reason groundwater leaked through the diaphragm wall was because of defects in the construction of the diaphragm wall namely gaps between panel joints and breaks in the panels. (7) Dr Redman was of the view that the diaphragm wall did not extend to the design depth: see Exhibit F at paras 36 – 40, in respect of Woodlands shaft. This is based on the assumption that bore hole NB8 reflects the ground conditions in other parts of the shaft. (8) The loss of groundwater at the top levels of the soil lead to depressurization of clays (E type clays) which abound in the Kallang Formation and the draining of F type material which is fluvial sand: see Figure 3 of Exhibit E which depicts the layers of soil and their relative levels. (9) The depressurization of the clays and the flow of sand caused consolidation of the soils in the vicinity of the shaft. (10) Buildings in the vicinity of the shaft cracked and infrastructure sagged or dropped as a consequence of the consolidation of soils beneath them. The Gambas shaft (11) The same problems at work in respect of the Woodlands shaft occurred in respect of the Gambas shaft save that much less damage occurred as a result. The Byora Precision building was damaged partly as a consequence of this settlement. Dr Redman, in his first report (Exhibit E), attributes the cause of the fall in groundwater pressure principally to the construction defects and to the exposure of the ground at the toe: see para 159 of Exhibit E. He also regards as probably a secondary reason the “continuing excavation works for the TBM launching tunnels”. He attributes damage in or after March 2002 to the TBM advance: see Exhibit E at paras 168 and 176, and says in Exhibit F at para 122 that “the main settlement in the vicinity of Gambas shaft is that which takes place once the TBM is launched”. The TBM/G4 problem (12) There was encountered in the tunnelling between Gambas shaft and Woodlands shaft a mixed geological profile, that is, there was G4 and G1/G2. The TBM was particularly designed for dealing with hard rock and when the JV encountered G4 material this lead to over-excavation with loss of more soil material than was desired, the ingress of mud and water into the main tunnel and collapse of the overburden. (13) The loss of soil and water in the tunnel lead to consolidation of the material above the tunnel and consequential problems to the neighbouring buildings on the surface and to infrastructure, including a high-pressure gas pipeline, laid in the ground close to Woodlands Avenue, and in the road itself. (14) Dr Redman is of the view that the JV recognised the problem of mixed material. He was of the view that there was poor control on the ground during tunnelling. The TBM/dyke problem (15) At chainage 1456 a dyke of volcanic rock was found to be present. The excavation of the dyke and material adjacent to it lead to groundwater being lowered with consequent depressurization due to water inflows into the tunnel in the period April to July 2003. Depressurization of the clay residual soils and the Kallang Formation lead to settlement of soil in the vicinity of the dyke which lead to cracking and damage (principally) to the following buildings and infrastructure: (i) Jin Fu Gong Temple (“ Jin Temple ”); (ii) Wu Lan Lian Sheng Temple (“ Sheng Temple ”); and (iii) the PUB sewer which experienced further settlement. (16) Dr Redman and Dr Pells were in agreement that “ground settlement occurred in conjunction with the advance of the TBM from April 2003 up until the dyke feature was encountered on or about 19 June 2003. The settlement occurred as a consequence of groundwater lowering/depressurization due to water inflows into the tunnel. Depressurization occurred in both clay residual soils and in the Kallang Formation. The relative proportions are not clear.” See Exhibit B at pp 3 – 4. Dr Pells’ Approach

Dr Pells’ approach is that there were not four incidents but really one, namely, inadequate design because the JV’s design did not provide adequately for groundwater and soil conditions that in his view should have been anticipated before excavation commenced.

He also sees the cause of settlement in each of the buildings or infrastructure affected as being linked to the excavation of adits and the use of an inappropriate TBM with an inadequate soil grouting program, not as a consequence of the excavation of the Woodlands shaft and the Gambas shaft.

In dealing with the four items described by Dr Redman it must be borne in mind, therefore, that Dr Pells does not accept the characterisation of what occurred as four incidents.

Dr Pells agrees with [0], [0], [0] and [0]. So far as [0] is concerned he disagrees that water leaked through the diaphragm wall to any significant degree, although he accepts that there were construction defects in the diaphragm wall.

Dr Pells disagrees with [0] and [0]. He says there is no evidence of depressurization of the E clays. He says that there was a close temporal relationship between the TBM/G4 problem and the damage to buildings in the vicinity. Dr Redman explains the temporal gap between the drop in groundwater and depressurization as being what one would expect.

Dr Pells agrees with [0].

Dr Pells disagrees with [0] as he says that even if the diaphragm walls did not extend as deeply as specified they did extend to rock. He says the grouting work specified by the JV was insufficient as it consisted only of one ring of grouting whereas, in his view, two rings, at least, were required.

Dr Pells’ position in relation to the Gambas shaft is the same as his position in respect of the Woodlands shaft.

Part of Dr Pells’ criticism of the JV is that it did not design the adits and tunnels to be excavated at a greater depth. Had the JV done so, he says, it would have required tunnelling through hard rock – G1 or G2 – for which the TBM was suitable and not G4 or mixed face material for which it was not suitable. He points to the Kiso-Jiban material as depicted in the diagrams as showing that by November 2001 the JV was informed of the problem which it faced but that it did nothing to avoid that problem. Mr Sternath’s evidence was to the same effect: see T1129 – 1136.

Dr Pells criticises the absence of an adequate modelling of the conditions and the failure of the JV to design a program of probing ahead of the excavation work and to grout ahead with testing for permeability before and after grouting. He views the JV as having embarked upon excavation without any adequate criteria for soil stabilisation. He is critical of the JV for leaving probing to the discretion of the Construction Manager. The Construction Manager (Mr Embery), Dr Pells says, decided not to probe ahead of excavation by the TBM and permitting Mr Embery to do so was also a significant design flaw.

Dr Pells also draws attention to the nature of the dyke material in the context of the Woodlands shaft excavation, noting that the fractured granite for five metres below the dyke intersection was of high permeability according to a Kiso-Jiban report: see Exhibit 5 at p 8. Dr Pells prepared a model of the Woodlands shaft (“ Exhibit 14 ”) which, on one side depicts his reading of the geological conditions, and on the other, depicts Dr Redman’s reading. As I have noted, Dr Redman did accept that there was a dyke through the Woodlands shaft. Evidence about the proximity of the roof of the upper adit (WA2) to G4 material is consistent with Dr Pell’s reading. No evidence was called in reply to refute Dr Pell’s conclusion that the geological conditions were more complex than those assumed by Dr Redman, as shown on the side of the model marked ‘Pells model’. Credibility of Witnesses Mr Embery

ACE submitted that the opinions expressed by Mr Embery “should be given little, if any, weight, and his evidence on controversial issues of fact should be viewed with great caution and should only be accepted where corroborated by contemporaneous documentation.” See ACE’s submissions of 10 March 2009 at para 245.

This submission was supported by a number of detailed examples in addition to the submission that Mr Embery was often non-responsive in his answers and that, at times, he “was unable to articulate any sensible basis for some of the assertions and opinions contained in his affidavits”: see ACE’s submissions of 10 March 2009 at para 247. One of the examples of this on which ACE relied was Mr Embery’s assertion in his affidavit that significant ground settlements had taken place around Woodlands Avenue by the end of 2001 when the settlement monitoring data did not support that.

In part, as ACE submits, Mr Embery’s problems stemmed from the fact that he had not undertaken any detailed review of the documentation and was largely relying on his recollection from seven years earlier. As the focus of what occurred became narrower – e.g. whether or not the settlement commenced before 30 December 2001 – his recollection was of limited assistance. Very few photographs were available to him and those that were produced at the hearing were of a poor quality. I accept the criticisms of Mr Embery’s reliability and largely for the reasons advanced by ACE, which I set out below, in addition to the difficulty I have just mentioned: (1) He sought to create explanations for his assertions that on further scrutiny did not have any support: his assertion that minutes might not be an accurate record of what was said: see T697 (and this in a context where, in his affidavit, he had said nothing about the inaccuracy or unreliability of records: see Mr Embery’s first affidavit at para 12) and his putting forward that as an explanation when at T698.15 – 17 he subsequently appeared to accept that the particular document was not inaccurate; his assertion that the Engineer had agreed to proceed with excavation without investigation of G4 material: see T685.7, which he said was contained in a document that he had seen recently: see T686.43. The document that he was talking about was produced (a letter of 4 June 2002 from the JV to the Engineer: see CB vol 3.13 at p 3-8050), and not only did it not support his claims, but it was to the contrary of his contentions because in the letter the Engineer instructed the JV to “immediately stop further excavation to east bore tunnel until such time” as the JV has submitted various details of proposed work. He would not agree that the letter did not support his contention. (2) He asserted that he had obtained confirmation of the suitability of the TBM when no such document was produced. The JV’s letter to the Engineer referred to at [231] below, was not consistent with that contention. (3) He said, initially, that there had been no significant water seepage into the adits but it became apparent that his view was that the JV had managed it properly – his initial response at T600.4 – 10, T601.44 – 47 and T603.3 – 606.44, should be contrasted with his later evidence at T616.22 – 34. (4) His affidavit was silent on the warnings given to him by the Engineer and as to the basis for his allowing the excavation to continue notwithstanding risks which it appears must have been apparent to him. (5) He said that he had only expected 10 metres of G4 material: see T652.30, or 20 metres: see T658.40. The document at CB vol 3.10 at p 3-6266 refers to anticipated G4 material between chainage 1100 and chainage 1200. He said there were other documents relied on: see T655.4, but the only document identified in re-examination was CB vol 3.5 at p 3-3732 which points only to an absence of a G4 problem at chainage 1180, but there had already been G4 problems at chainage 1110: see Exhibit 1, so there had been 30 metres of G4 problems by the time of the entombment of the TBM. (6) On several occasions, which I refer to below, Mr Embery put forward explanations for the JV’s conduct or as to its understanding that he soon resiled from: see for example, [240(6)] and [240(7)]. Other Witnesses

So far as the other witnesses were concerned, there was no attack on Mr Sternath, Mr Jewell or Mr Funge, whose evidence was directed to establishing that there had been justified complaints of damage by third parties and the costs to the JV of the repairs. There was some limited criticism of Dr Redman by ACE but I do not think that Dr Redman’s impartiality as an expert witness was successfully impugned. There was, in fact, agreement that Mr Sternath was an impressive and reliable witness, although ACE pointed out that he did not have the expertise on geotechnical matters that Dr Pells had. Dr Pells

Lloyds launched a significant attack on the credit of Dr Pells based on: (1) His refusal to concede that in his first report (Exhibit 4) he was addressing his comments to the causes of ground settlement resulting in damage to property: see T970 – 974. (2) That in his second report (Exhibit 5) Dr Pells asserted that he had received further materials which he relied on in producing his second report when the documents referred to were received by him prior to his first report: see T969.41 – 47, T985.19 – 26. (3) That he changed his view on the mechanism of settlement when Dr Redman indicated agreement with it: see T1462. (4) His assertion that his comments that there had been no systematic grouting included in the design for the toe area were not based on a failure to appreciate that there had been grouting specified for the toe area but rather was a description of inadequate grouting: see T990 – 991. Dr Pells had agreed in a conclave of experts held on 18 February 2009 that: “There is no standard book or industry standard that gives a definition of the term systematic grouting.” See Exhibit B at p 25. In cross-examination Dr Pells said that the standard text in undergraduate engineering would demonstrate the meaning of the term. He was asked to produce the text the following day and he said that his briefcase containing the texts had been stolen: see T1083.8 – 11. (5) Dr Pells had given, without an adequate basis, a figure of 100mm settlement in the G4 material out of a total settlement of 300mm, but the following day at T1031 – 1032 he explained that he had arrived at the 100mm figure in error.

ACE in para 14 of its submissions of 19 March 2009, responded to these criticisms saying that Dr Pells was not refusing to concede the content of his first report (Exhibit 4), that there was no inconsistency in Dr Pells saying that additional factual information has come to the writer’s attention, and that revisiting material previously provided was not inconsistent so far as Dr Pells’ use of the 100mm figure is concerned. ACE described the error corrected the following day as trivial, with no bearing on the technical issues in the case.

Lloyds asserts that Dr Pells was an advocate for ACE and had reached conclusions on the basis of inaccurate or questionable information. ACE responded to this by asserting that the basis for revisions of Dr Pells’ views was “transparent” from his report and that the inaccurate or questionable information was never identified. I accept that experts may change their minds and it is wholly appropriate that they do so when new information is brought to their attention or they realise that they have fallen into error. Having said this, I did find Dr Pells’ change of opinion rather dramatic and somewhat surprising given that not only had he accepted Dr Redman’s view on the subject but provided his own graphic explanation saying: “inflows from this permeable zone, which was not designed to be cut off by the diaphragm walls, was akin to pulling the plug out of a bathtub. Such a process effectively would cause depressurization of the compressible soil horizons even if the sand layers were cut off by the diaphragm walls.” See Exhibit 4 at p 63. It became clear that his sole reason for abandoning the theory that he had embraced in his first report (Exhibit 4) was the result of piezometers in the vicinity of the Woodlands shaft: see Exhibit 5 at pp 54 – 56, which showed no depressurization, and he maintained his explanation for the ground problems in connection with the TBM/dyke problem as “a consequence of groundwater lowering/depressurization due to water inflows into the tunnel”: see Exhibit B at p 3. Mr Sternath, called by ACE, supported Dr Redman’s theory about depressurization.

The fact that Dr Pells had agreed that there was no common meaning of ‘systematic grouting’: see Exhibit B at p 25, but yet asserted that such a meaning could be found in any standard textbook without such a book ever being produced (his explanation that his briefcase had been stolen does not explain why such a text was never subsequently produced in the remaining seven days of the hearing) and his position on depressurization, left me with a degree of concern about his impartiality but he clearly has extensive knowledge of his field and I did not feel that overall he was an unreliable witness as a result of these matters and the other matters in respect of which criticism has been levelled by Lloyds. I think the evidence of Dr Pells and Dr Redman has been hampered by a number of problems: (1) the passage of time; (2) the absence of any detailed photographic record relevant to the matters in issue; (3) shortcomings in the site documentation; (4) the difficulty of attempting to reconstruct what occurred from documents and only one witness whose recollection was not assisted by detailed re-familiarization with all the records and whose reliability as a witness is in doubt; (5) the potentially complex structure of the geological conditions themselves; and (6) the delay between consolidation within the ground, impact on structures and observance of the impact on structures, as well as the possible confluence of causes. Mr Cooper

In relation to Mr Cooper, ACE submitted that apart from his evidence on general standards to be applied in relation to tunnelling, Mr Cooper’s admitted lack of expertise with TBMs and the location of the tunnels and his assumptions concerning the conditions rendered his evidence of little assistance: see ACE’s submissions of 10 March 2009 at paras 264 – 278.

I think that there was a variation between the assumptions made by Mr Cooper and the facts as established. Mr Cooper’s view that the TBM was suitable for G4 conditions without substantial surface pre-grouting was at odds with the views of Mr Sternath and Dr Pells and is contradicted by the JV’s own letter to the Engineer in May 2002 and the JV’s subsequent approach: see Construction Execution Procedures (“ CEP ”) dated 21 April and 23 April 2003 at CB vol 3.21 at pp 3-14647 et seq and 3-14786 et seq, which included modifications being made to the TBM, and see Mr Embery’s acceptance that it was not suitable at T657.29. I accept ACE’s submissions in this regard and prefer the evidence of Mr Sternath and Dr Pells on matters relating to the TBM/G4 problem where it is in conflict with Mr Cooper. ISSUE 1: What was the cause of the damage to the property of third party property owners (including PowerGrid)?

Lloyds provided a document entitled ‘Plaintiff’s Summary of Rectification Costs”. It is helpful to use that document to set out the claims for each of the four incidents. (1) Woodlands shaft Claim Description Amount 1;1.1 Excessive Settlement around high pressure gas pipe line (100%) $3,109,806.91 1;1.2 Damage to PV 1900mm diameter water main atpower gas off take station (100%) $167,060.10 1;1.3 AEM Tech Building (50%) $37,518.20 1;1.4 Repair of 300mm medium pressure gas pipeline(100%) $75,158.94 1;1.5 Pacific Vinitex Building (100%) $126,231.18 1;1.6 Hup Fatt Brothers Building (100%) $951,573.68 1;1.7 PowerGrid 66 kv joint bays 4 and 5 (80%) $623,732.02 1;1.11 Permanent lifting of T2 and T4 high pressure gaspipeline at junction of Woodlands 8 and 9 (100%) $337,548.05 1;1.13 Road repairs in Woodlands Avenue 8 (90%) $102,951.11 1;1.14 Recharge wells etc (80%) $659,977.50 1;1.15 Repair to monsoon drains and canal (100%) $6,890.68 1;1.16 Repair to road formation on Woodlands IndustrialPark Avenue 8 (100%) $22,057.00 6; Part 1 Deep Sewer Repair MH2 – MH 4 (50%) $759,027.40 TOTAL $6,979,532.77 (2) Gambas shaft Claim Description Amount 1;1.10 Byora Precision (10%) $2,476.33 1;1.14 Recharge Wells etc (10%) $83,997.13 TOTAL $86,473.46 (3) TBM/G4 problem Claim Description Amount 2; Part 1 Temporary Repair to PUB deep sewer line (100%) $1,771,958.13 2; Part 2 Compensation grouting works to PUB deep sewer line(100%) $344,017.95 2; Part 3 Permanent Repair to PUB Deep Sewer Line (100%) $1,538,372.35 1;1.13 AEM Tech Building (50%) $37,518.20 1;1.17 PowerGrid 66 kv joint bays 4 and 5 (20%) $155,933.00 1;1.18 and 1.8.1 Excessive settlement around 66 KV joints 4,5 and 6(100%) $149,822.76 1;1.19 Expose high pressure gas pipeline (100%) $1,170.24 1;1.10 Byora Precision (90%) $22,286.93 1;1.13 Road repairs in Woodlands Avenue 8 (10%) $11,439.01 1;1.14 Recharge wells etc (10%) $82,797.18 TOTAL $4,116,315.75 (4) TBM/dyke problem Claim Description Amount 6; Part 1 Deep Sewer Repair MH2 – MH 4 (50%) $759,027.40 6; Part 2 Intergraphica (100%) $232,383.66 6; Part 3 Jin Fu Gong (100%) not yet repaired 6; Part 4 Wu Lan Lian Sheng Temple (100%) $61,020.00 6; Part 6 Settlement of 66 kv joint neal lamp pole 16 alongWoodlands Avenue 8 (100%) $151,373.40 6; Part 7; 1.6 Rectification at Yee Lee Oil (100%) $8,900.00 6; Part 8; 1.7 Dismantle fencing, gate and hack concrete kerb(100%) $3,108.80 TOTAL $12,397,935.26

There is no disagreement that buildings and structures owned by others and the PowerGas pipe, PowerGrid cable joints and Woodlands Avenue roadway were all damaged by virtue of groundwater loss and/or soil collapse associated with the construction work for the project.

In relation to the TBM/G4 problem there is no dispute that it was caused by the use of the TBM to excavate G4 material with resultant over-excavation, groundwater loss and collapse of soil above and around the excavation.

The disputes in relation to causation are as follows: (1) What aspect of the works caused the structures listed in [60(1)] and [60(2)] to be damaged? (2) As to the cause of the damage described under TBM/dyke problem in [60(4)], there is no dispute that this occurred because the excavation using the TBM caused groundwater loss and consequent ground consolidation, but there is a dispute as to whether that arose by reason of a failure of the JV to do something that ought to have been done by it (including following its own procedures). (3) As to whether all of the problems, including the TBM/G4 problem, are really all part of the same failure of the JV – i.e. as Dr Pells maintains, a failure to revise its design for the works by alteration of the route or depth of the main tunnel and its adits and/or use a different TBM and/or include an adequate pre-excavation modelling and testing program. (4) Whether in relation to the TBM/G4 problem, Lloyds has failed to establish that the JV took reasonable precautions. (5) Whether there was one occurrence or more than one occurrence, and if so, how many. The Woodlands and Gambas shafts

The first dispute centres largely on whether the significant groundwater loss (which it is agreed occurred) arose because of construction defects in or at the base of the diaphragm walls of the Woodlands and Gambas shafts. Lloyds claims that there were construction defects, namely defects in the walls themselves and a failure of the contractor to take the walls down to a sufficient depth.

Dr Redman accepted that the ingress of water through the upper adit and by virtue of the construction of it and was a contributory cause of groundwater loss: see Exhibit E at paras 39, 76 and 77. Since Dr Pells attributes the cause of settlement of adjoining property (prior to commencement of the main tunnel excavation in March 2002) to water loss in the excavation of upper adit (rather than the Woodlands shaft) there is, in fact, a measure of agreement.

There is no dispute that there were defects in the diaphragm walls through which water leaked and that they are to be characterised as construction defects: see Exhibit D at para 43, CB vol 3.3 at 3-1987 and T550 – 553. There was, however, dispute as to whether the water flow through the walls was significant and as to whether the problems at the toes were construction defects and whether, if they were, the loss of groundwater at the toe and through the diaphragm walls had caused subsidence in adjoining property.

I am satisfied on the balance of probabilities, for reasons I will explain, that the loss of groundwater at the toes was not a result of construction defects and it was not suggested that water loss through the upper adit was a consequence of construction defects.

There was extensive debate between the experts about the precise mechanism by which soil beneath the damaged structures came to settle, which included minute examination of the data revealed by piezometers, their reliability and how that data should be interpreted. Dr Redman’s theory was that the clay layer in the Kallang Formation around the shafts had become depressurized as a result of the significant degree of groundwater loss, with the water table dropping below the clay layer. Mr Jewell, by his hydrological modelling, supported Dr Redman’s theory as to how loss of pressure in the clay over time affected surrounding conditions as far afield as the affected structures. Dr Pells rejected the modelling as having any relevance: see Exhibit 5 at para 23, at p 29 and at section 3.3 – 3.4. There were assumptions made by Mr Jewell in his modelling that were themselves contentious – such as the degree of permeability of the G4 and ignoring the presence of the dyke at the Woodlands shaft (see Dr Redman’s evidence at T840.8), but what the modelling supports, in a broad sense, is the conclusion that loss of groundwater at the shafts can affect soil some distance away, which Mr Sternath accepted and which Dr Pells must have accepted as feasible in concurring originally with Dr Redman’s theory.

Initially Dr Pells agreed with Dr Redman’s theory but he changed his mind when he examined more closely the readings from piezometers in close proximity to the Woodlands shaft. There was, he found, no evidence of depressurization from piezometer P4 in the clay layer prior to commencement of the work on the adits and there was other evidence from piezometers P8, FBH-12A and FBH-14A also in soft clay that showed little depressurization: see Exhibit 5 at pp 47 – 49.

The contest was intense and included a full-scale attack on the credibility of Dr Pells and a muted attack on that of Dr Redman. The significance of the argument on the question of whether construction defects in the diaphragm walls and the claimed construction defect in the toe area had lead to the damage to third party property, rather than work on the adits, was related to whether an exclusion clause in the ACE policy (and a similar clause in the Lloyds policy) applied. I do not think it is necessary to determine whether the groundwater loss impact was on the clay layer or the other layers in order to resolve this issue. In his second report (Exhibit 5 at p 59), Dr Pells, in answer to the question: “What was the cause or causes of the ground settlement that occurred during the project (1) during the excavation of the Gambas and Woodlands shafts?”, wrote: “Due to the lowering of groundwater pressures (depressurization) leading to compression of soils but not including consolidation of the soft clay horizon within the Kallang Formation.”

Although Dr Pells made clear that he did not accept depressurization of the clay layer as an explanation, he did appear to accept that poor design of the shaft toe with seepage into the shaft (as well as in the adits) also produced compression of soils and damage: see Exhibit B at p 4.

Mr Sternath (ACE’s tunnelling expert) was of the view that there had been groundwater loss from the shafts which was likely to have caused damage to the structures. He accepted Dr Redman’s view that there had been a failure to take the Woodlands diaphragm wall to the correct depth (and seemed unable to accept, even as a hypothesis, any contrary conclusion: see T1127 – 1128). Dr Pells’ contention that the grouting at the toe was inadequate provided an explanation for groundwater loss that, if correct, strengthened the view that excavation work at the base of the Woodlands shaft was responsible for groundwater loss.

Mr Sternath was an impressive and impartial expert witness. He is not a geotechnical engineer but he is someone with extensive practical tunnelling experience and he expressed a view as to the interrelationship between the excavation of the shaft and the damage to property that accorded with Dr Redman’s views. He provided, at p 5 of Exhibit 11, an explanation of the process that is instructive: “1. Soft Soil is subject to settlements in case of changes in the groundwater (GW) table as the weight of soil with GW and without GW is different. 2. The pores in soft soil are initialled filled with GW and are able to bear the ground load with less deformation. 3. When the pores fall dry they will be compressed and the soil will deform (consolidate) such leading to surface settlements. 4. Different kinds of soils react to dewatering with more or less settlement. Particularly vulnerable for consolidation are soft clays and peat as they exist in the “Kallang Formation” surrounding the Woodlands Shaft area. 5. A further serious source of ground settlements is ground loss during excavation operations. Ground loss means the volume of soil which is extracted from the ground in an uncontrolled manner. 6. This can happen when the soil is left unsupported or due to GW inflow into the construction site when file particles are carried with the GW flow. 7. The volume deficit will lead to uncontrolled ground settlement. Ground loss during excavation can be avoided by choosing adequate support and sealing methods.” See Exhibit 11 at p 5. At T1149.10 – 1150.6 Mr Sternath indicated the distance at which an effect could be expected horizontally, particularly in permeable soils. The properties affected were within that zone.

ACE’s attempt to resist the connection between the excavation of the Woodlands shaft and the damage claimed by Lloyds to have resulted, included drawing attention to the lack of photographs (depicting ponding or pumping machinery being used) and records relating to the alleged problem in the shaft. It also drew attention to Mr Embery’s evidence that leaks in the diaphragm wall and in the toe were attended to promptly and effectively. I think there is some force in this, but nor are there records establishing the extent of water removed from the shafts prior to February 2002 and Mr Embery explained that there was no point in undertaking measurement of water in the shafts prior to completion of excavation: see T545.6 – 22. There is clear evidence of groundwater loss before the commencement of excavation of the adits (i.e. 19 January 2002): see Figures 6 and 10 of Exhibit F and PSM1245.04 to Exhibit 5.

Mr Embery saw water flows at the toe and attempts were made to deal with those flows which took some time to implement. Once groundwater levels had dropped they would not be reinstated by grouting or plates, but rather, the flow would be stemmed. Recharge wells, used to reinstate or attempt to reinstate prior groundwater levels, were not installed until February 2002. There is a further reference to the importance of water loss at the toe: see T586.14 – 41, T589.50 – T590 and T594, where Mr Embery made clear that he regarded the water loss through the toe as more significant than through the walls: see also T530.49 – 531.2. The JV was on notice of the prospect of water ingress, there being a statement that “water inflows may occur at unknown locations of varying severity” in the tender documents provided to the JV: see Exhibit 5 at p 36. The excavation was at least 40 metres below the water table: see Exhibit 5 at p 36. As Dr Redman noted, everyone would have known that “there was a lot of water in the region”: see T804.13. I think that the evidence of loss of groundwater from monitoring results referred to in Dr Redman’s second report (Exhibit F), Mr Embery’s evidence as to his observations and the work done to deal with the problems, the note of the agreement of the experts that as at 18 January 2001 there was a water flow into the shaft of 0.04m³/minute (or 0.7 litres/second) (see Exhibit D at para 4.3) and Dr Redman’s evidence (leaving aside his theory of a failure to take the diaphragm wall of Woodlands shaft down far enough) support the finding that I make that there was significant loss of water through the Woodlands shaft and particularly at the toe of the shaft because it was at a deeper level than the defects in the wall itself and in an area of rock or material affected by the presence of the dyke, and that the groundwater loss, by reason of its impact on surrounding soils, lead to consolidation beneath the properties damaged. The likelihood is that the same process occurred at the toe of the Gambas shaft but at a reduced flow because the conditions were less problematic and there was no dyke. Whilst there was no adit construction from the shaft there were ‘commencement’ tunnels there.

I turn to explain why I do not regard the problem of water flow at the base of the shafts to be a construction defect.

There was no contemporary evidence put forward to establish, other than in respect of defects in the diaphragm wall itself, that Sambo did not do what it was required to do under its contract – either in the form of an observation or complaint. Dr Redman refers to an assertion in a letter from the JV to ACE’s agents that there was “improper interface sealing between the toe of the diaphragm wall and the rock foundation”: see Exhibit E at para 73. I have not been able to identify the letter from which that quote was taken (and it is not found in the letter of the stated date at CB vol 3.16 at p 3-10589) but I shall assume it is contained in one of the 53 volumes of the court book. That statement is quite consistent with a failure in design as opposed to construction.

Dr Redman explains his theory that the anchorage requirements of a connection between the base of the wall into G2 and the grout of 11 metres into G1 were not achieved in paras 36 – 40 of Exhibit F and see Diagram 2, 4 and 5 annexed to that report. The theory is based on extrapolation from the results of bore hole NB8 in the case of Woodlands shaft. In respect of Gambas shaft, Dr Redman at paras 155 – 160 of Exhibit E, concluded that diaphragm wall leaks and leakage through the ground at the exposed toe were both part of the cause of the 10 metre loss of groundwater, with more emphasis given to the diaphragm wall, but see also Exhibit F at paras 88 – 95 and his conclusion at paras 121 – 123 that the main settlement in the vicinity of the Gambas shaft occurred once the tunnelling commenced, which explains his approach to apportionment in [60(2)].

The explanation preferred by Dr Redman for what occurred is plausible, but the considerable difficulty I have is that Mr Embery, who was called in Lloyds’ case, not only gave no support to the contention that Sambo was in breach of its contractual obligations, but actually gave evidence to the contrary: see T542.5 – 50, where he said that the diaphragm wall was properly constructed to finish in rock. If ‘rock material’ in the case of the Woodlands shaft site diary was intended to denote some difference between rock and what was encountered, the JV was fully aware of the fact. Very little attention was given in the case to the same issue in respect of Gambas shaft, but again, Mr Embery’s evidence did not support a failure by Sambo to reach the design depth.

There is apparently no documentary material which supports the contention that Sambo had not constructed the Woodlands shaft wall to the correct depth or carried out the work at the toe other than in accordance with its contractual requirements. Dr Redman accepted that there was no documentary support for his conclusion and that the results of NB8 may not have been representative of the whole shaft: T798 – 799.

It should be noted that by agreement made between the JV and Lloyds and between the JV and ACE, both parties had access to all documents held by the JV. Although much material has been obtained and tendered, these records make scarce reference to the problems which the JV experienced.

Dr Pells, at p 14 of Exhibit 5, explained why Dr Redman’s theory was flawed, as it was based on the NB8 results without any recognition of the impact of the dyke and the construction documentation.

There is an important piece of evidence at T556.49 – 557.31, where Mr Embery explains that the grouting at the toe of the Woodlands shaft diaphragm wall was not as effective as that at the Gambas shaft. Woodlands was a deeper shaft than Gambas shaft but the Gambas shaft had no dyke in or near it. There is no suggestion by Mr Embery that he observed inadequacies in the depth to which Sambo took the diaphragm wall for either shaft or failure to complete the works in accordance with the contractual requirements. The JV carried out the grouting work and so had ample opportunity to observe the work done by Sambo. There is no suggestion that the grouting was ineffective because the JV employees did not attend to the grouting at Woodlands shaft with the same level of care as at Gambas shaft. The existence of a dyke at the base of the Woodlands shaft and the obvious fluctuations in soil materials are more likely to have been the cause. Dr Pells, Mr Sternath and Mr Cooper all agreed that the presence of a dyke would (or could, in the case of Mr Cooper) be a significant matter. That much less damage resulted from Gambas shaft, which also had, on Lloyds’ case, leaks through the diaphragm wall points to factors other than defects in the wall being more significant. I have noted that Dr Redman accepted in his second report (Exhibit F) that most of the settlement near the Gambas shaft was a result of the TBM tunnelling, thus reducing the significance of water loss in the Gambas shaft.

There is thus identified a cause of groundwater loss – i.e. ineffective grouting for the conditions encountered – which is more likely to have been a result of a design defect rather than a construction defect. If there was a failure to take the wall to a specified depth – as Dr Redman and Mr Sternath thought – the JV, I infer, was aware of it and implicitly approved a change to its original design. Either the diaphragm wall was not designed to a sufficient depth or the grouting regime imposed did not deal sufficiently with the nature of the geological conditions that were encountered. In a letter of 4 January 2002 the Engineer referred to the poor quality of exposed wet rock at the base of the shaft, pointing out that the quality was not “Moderately Strong” and expressing criticism of the work of Kiso-Jiban: see CB vol 3.7 at p 3-4538. Whether the geological conditions that were encountered ought to have been appreciated is relevant to the question of whether the JV was negligent, but at present, the question is what caused the damage to adjacent properties.

Mr Cooper: see T938 – 942, accepted that a proper design would take into account the presence or likely presence of a dyke whether in the vicinity of the shaft or the adits; and would establish a level of permeability by test packers and involve pre-grouting. He thought that the design was adequate but it does not appear that the presence of the dyke was specifically catered for in the Woodlands shaft and there was no pre-grouting for the adits near the shaft or further away, nor grouting to stabilize the surrounding soil at the shaft prior to excavation.

The evidence of Mr Embery referred to in [75] above and further at T586 – 587 is also relevant to another matter which is whether there was significant water loss through the upper adit. Mr Embery sought to minimize the extent of water loss through the upper adit by initially seeking to resist the accuracy of what had been described by PowerGrid’s representative Engineer in a letter of February 2002 (see CB vol 3.10 at p 3-6154) as a significant quantity of water still flowing through the rock face of the upper adit and its having caused “continuous loss of groundwater and severe ground settlement”: see also T599 – T600, but see T603.30 – 39, T604.32 – 34, T605, T606.20 – 24 and T616.22 – 34 and Exhibit 1 where Mr Embery has marked the area of water flow in the upper adit. The roof of the upper adit was very close to the known G4 material: see Exhibit E at paras 74 and 75, Exhibit F at para 64, Exhibit 4, q 1, p 75 and section 3.5 of Exhibit 5, and see also the dewatering results from Woodlands shaft from February 2002 at PSM 1245TR2 Figure P of Exhibit 5. I recognize that Mr Sternath, although he thought the upper adit was not located at a sufficient depth, saw this defect as unconnected with water loss, but I am satisfied on the balance of probabilities that there were also water flows in the upper adit, and that these were not the result of construction defects, but rather, as Dr Pells contends (see Exihibt 5 at p 63), were due to the failure to design a method of preparation of the fractured permeable rock around the adits to reduce or eliminate water flows. Documents 90-145 and 90-146, which formed part of MFI 14, suggest that the water loss through the upper adit was much less than from the Woodlands shaft, but both the shaft and the adit were located in or close to the dyke.

I find that water losses in the upper adit contributed to the general loss of groundwater with consequential impact on adjoining property, which I find was otherwise substantially caused by the flow of water at the toe of the Woodlands shaft and, far less significantly, by leaks through the diaphragm wall.

The presence of the dyke, at least in the shaft area, the fractured quality of the rock at the base of the shaft, the high water table, the high level of permeability of ground conditions were all matters which pointed to the need for extensive precautionary steps to be taken if groundwater loss was not to be experienced. If Dr Pells is correct and a second dyke also went through the upper adit, this would only increase the need for caution, but it is not necessary to determine if it did.

I should note that in submissions Lloyds suggested that even if no failure by Sambo to take the diaphragm walls to rock was established, other construction defects could not be ruled out – the onus being on the insurer to establish that the cause of the damage was one excluded by a relevant exclusion clause. The onus does lie on the insurer to establish that the exclusion clause applies but the insured (and here Lloyds) has to establish that the damage was caused by its work. ACE pointed out that Lloyds is not the insured, but a party who seeks contribution or recoupment. ACE submits that Lloyds has the onus of establishing that it was obliged to pay out the JV claim. I accept that submission. Lloyds sought to establish the cause by calling Mr Embery to establish water loss through the walls and at the toe, but, as I have noted, Mr Embery does not support the existence of any construction defect in the toe area. The only possible construction defects which were identified by Dr Redman were the failure to take the toe to the required depth and defects within the walls.

Dr Redman agreed in cross-examination that the damage reported by Pacific Vinitex as at 31 December 2001 was not attributable to the works: see T765 – 760, but I do not accept ACE’s contention that he concluded that none of the damage to that building was referable to the works and there is evidence from Mr Funge (see Mr Funge’s affidavit dated 6 March 2008 at para 89) about further damage in May 2002 and later, which is clearly referable to the works: see Exhibit 1 together with CB vol 3.11 at p 3-7053, and which shows the settlement at the PUB gas pipeline as at 14 January, 23 March and 2 August 2002 in the proximity of the Woodlands shaft and close to the group of buildings of which Pacific Vinitex is one.

I am satisfied on the balance of probabilities that groundwater loss causing damage to the items listed in [60(1)] and [60(2)] was caused, at least in part, by: (1) excavation of the Woodlands shaft, Gambas shaft and the adits; (2) diaphragm wall leaks in the shafts due to construction defects; (3) leakage of water at the toe of the diaphragm walls arising because of the poor geological conditions at the base, including a dyke, in the case of Woodlands shaft, and the insufficiency of the grouting program called for by the design in the known or likely geological conditions which did not involve grouting before excavation and testing for permeability; and (4) in the case of Woodlands shaft, also leakage of water through the upper adit in conjunction with the presence of the dyke due to the positioning of the upper adit through or close to a dyke and in a position without adequate hard rock cover.

I am satisfied on the balance of probabilities that the groundwater losses arising out of [0] – [0] impacted upon neighbouring ground with consequential damage to property. The TBM/G4 problem

I deal with the categorisation of the TBM/G4 problem at [189] below. The TBM/dyke problem

The dyke at chainage 1456 (“a major weathered dyke with weak rock and significant water inflows”: Exhibit E at para 230) was encountered on 19 June 2003. This is described by the parties as the “ the TBM/dyke problem ”. There was no dispute between the experts that groundwater loss caused by the tunnelling led to third party damage. There was, by the time of the conclave at least, agreement by the experts that the groundwater loss caused by the TBM tunnelling in the vicinity of the dyke did not commence when the TBM hit the dyke, but rather that it intensified at that point: see Exhibit E at paras 227 – 261, and see Exhibit B at p 3. On or about 11 June 2003 the water flow into the tunnel was 250 litres per minute: see T714, and see CB vol 3.23 at p 3-15689 which shows more than 150 litres per minute around 15 May 2003 and provides details for April to June 2003.

The argument in relation to the TBM/dyke problem had two aspects. There was a dispute as to the reason for the failure of the JV to prevent the groundwater loss, and secondly what damage had occurred before 27 May 2003. If ACE’s contention that the policy period ended on 27 May 2003 is correct (I deal with this contention below) then any damage caused by the incident after 27 May 2003 would not be the subject of losses under the ACE policy.

In para 227 of Exhibit E, Dr Redman said: “The TBM was re-launched from the Northern Adit at Woodlands shaft on 19 March 2003. Prior to the re-launching of the TBM an assessment was made of the hazards associated with groundwater inflows and procedures established to adopt probing and grouting from the TBM and/or face to identify and control inflows. A contingency plan existed for the implementation of grouting from the surface if required. Ultimately the surface grouting option was utilised when the TBM was stopped on or about 19 June 2003.”

The contingency plan only mandated surface grouting ‘if required’ and it appears that it was not regarded by the JV as necessary until 19 June 2003.

There were three strands to ACE’s case in relation to the TBM/dyke problem and these are outlined at T1232 – 1233: (1) that the water flows into the tunnel even before the dyke was struck were inconsistent with proper grouting being carried out; (2) that the JV had in place a regime which required it to follow a grouting procedure when flows created by test holes exceeded a particular level and that regime was not followed even though levels were exceeded; (3) the project ought to have utilised a different alignment for the tunnel which would have avoided the problem.

So far as the third aspect is concerned there is no clear evidence of how the TBM/dyke problem itself could have been avoided. There is no direct evidence of how deep the dyke went and whether a tunnel at a lower level would have avoided the problem. There is, on the other hand, no evidence that the JV experienced problems with the west tunnel, which was lowered and this lends some support to the proposition that it would have been an effective solution. Mr Cooper’s evidence: see Exhibit K at para 39, was that lowering the tunnel may not have eliminated the potential to intersect with the dyke and he thought that a failure to effect a redesign of the vertical alignment was not inconsistent with the obligations of a reasonably competent contractor. I have noted, however, that the JV did redesign the depth of the west tunnel. So far as the second point is concerned, there was evidence that the program for dealing with groundwater loss was extensive and Mr Sternath seemed to agree that it was a comprehensive scheme. He did however, say that: “A competent tunnelling contract would have taken action to encircle the location and shape of the dyke in advance and would have implemented appropriate measures to mitigate its effects on the mining process. Relying on probing and grouting from inside of the TBM was far too little to tackle this problem properly.” See p 24 of Exhibit 11. Mr Embery did say that Kiso-Jiban had recommended installing a recharge well, which was done: see his first affidavit at para 56, but it is clear that this did not prevent the significant flow of water into the main tunnel.

The JV prepared two CEPs, dated 21 April and 23 April 2003: see: CB vol 3.21 at pp 3-14647 et seq and 3-14786 et seq, which ACE claims were not followed. Lloyds disputes that the JV failed to follow its own program, which required test holes to be drilled and water flow out of them measured. If that is correct then the program, however comprehensive it may appear to have been, has been shown to be flawed, and that leads to the conclusion that the JV’s design for the works was defective. Since there was significant water inflow into the tunnels (even if that inflow did not arise through the test holes) then the design did not sufficiently anticipate and guard against the prospect of water loss: see CB vol 3.23 at p 3-1689, the documents at pp 90-72 – 90-79 which were part of MFI 14 and T714 for details of the water flow. Dr Pells gave evidence as to where the dyke could be anticipated: see Figures 4.2 and 6.2 of Exhibit 4, and Mr Sternath said the JV “had to have known of the risks of hitting a dyke or similar formation”: see Exhibit 11 at p 12. That was not challenged by Dr Redman. Mr Embery admitted foreknowledge of the presence of a dyke at around chainage 1450: see Mr Embery’s first affidavit at para 56.

Dr Pells’ criticism of the JV, supported by Mr Sternath, in failing to adequately prepare the ground to prevent water loss before excavation in the region of this dyke is, I think, made out. Mr Cooper did not deal with the extent of water loss actually experienced and I do not read his report at paras [36] – [39] as expressing any view as to why the JV’s program failed to prevent the significant groundwater loss that occurred.

In short then, the JV’s proposed means of dealing with the higher permeability zone in the region of chainage 1456 was ineffective to prevent significant groundwater loss and consequent damage. Given the history of the problems at the Gambas and Woodlands shafts and the entombment of the TBM, the JV was clearly aware of the risks of groundwater loss and in my view, the failure to introduce a regime of grouting from the surface in advance, if relocation of the tunnel was not viable or deemed unnecessary, points to a failure in the design of the project. ISSUE 2: Did the cause (or causes) of damage give rise to a liability in the JV to the third party property owners?

The parties each retained an expert in respect of Singapore law. Lloyds retained Mr Quentin Loh SC and ACE retained Mr Alvin Yeo SC. There was little or no disagreement between them and they helpfully prepared a joint report in respect of the law of nuisance in Singapore.

Their agreement is contained in their joint report of 23 February 2009 (“ Exhibit O ”), which makes reference to Mr Loh’s report of 14 February 2008 (“ Exhibit M ”) and Mr Yeo’s report of 16 September 2008 (“ Exhibit 15 ”), and most of what follows is in their own words and taken from that material, but using new consecutive numbering, omitting duplication and removing personal pronouns: (1) Singapore law generally adopts the English common law on nuisance. Hence there are cases where judges refer to English cases on the English common law relating to nuisance as if it was part of Singapore law. (2) There has been one exception, the Singapore Court of Appeal has refused to follow the English decision of Dalton v Angus (1881) 6 App Cas 740 (a case on the right of support of a building on a piece of land from a neighbouring plot of land): see Xpress Print Pte Ltd v Monocrafts Pte Ltd [2000] 3 SLR 545 (“ Xpress Print ”). (3) In this case, the liability of the contractor to the owners of properties affected by their dewatering of the ground raises the issue of whether the English common law as laid down in Acton v Blundell (1843) 12 M. & W. 324; 152 ER 1223, Langbrook Properties Ltd v Surrey County Council [1969] 3 All ER 1424 (“ Langbrook ”) and Stephens v Anglian Water Authority [1987] 1 WLR 1381 applies in Singapore. (4) In refusing to follow Dalton v Angus , the Singapore Court of Appeal said at [37] of Xpress Print (omitting citations): “Further we are of the view that the proposition that a landowner may excavate his land with impunity, sending his neighbour's building and everything in it crashing to the ground is a proposition inimical to a society which respects each citizen’s property rights and we cannot assent to it. No doubt the trial judge felt constrained by the authority of Dalton v Angus and Lee Quee Siew’s case , but this court is entitled to depart from those cases, and therefore does not suffer from such impediment. In the event, we are of the opinion that the current state of affairs cannot be allowed to persist. The question is therefore not whether the principle applied by the trial judge in the court below should be rejected, for it clearly must, but rather how far the duty of the landowner should extend. ... We believe that the true legal justification for the right of support is the legal principle encapsulated in the Latin maxim sic utere tuo ut alienurn non laedas , which translates in English to: use your own property in such a manner as not to injure that of another. The importance of that principle is compounded in Singapore in view of our land use pattern, whereby all land available for commercial, industrial or residential purposes is used to a high intensity. The damage that might be caused if landowners were lackadaisical in their excavation works could be astronomical, not to mention the cost in human lives or injury to property. ... Once the primacy of the principle forbidding landowners to use their property o ( sic ) the injury of others is accepted, we think that there is scant justification for the 20-year gestation period for a right of support in respect of a building .” (5) The tort of nuisance is a separate cause of action from negligence in Singapore and negligence is not a necessary element in an action based on nuisance. (6) The person liable for a nuisance is the actual wrongdoer, whether or not he is in occupation of land from which it emanates. (7) On the assumption that the third party property damage occurred as a result of interference stemming from the construction site, the parties potentially liable in nuisance are the occupiers of the construction site. (8) The Singapore Courts have recognized "occupiers" as "he who has the immediate supervision and control and the power of permitting or prohibiting the entry of other persons": see the Singapore High Court decision of Wong Jin Fah v L&M Prestressing Pte Ltd & Ors [2001] 4 SLR 529. (9) The occupiers of the construction site are the employers and the contractors, namely PowerGrid and the JV. (10) Whether a person can successfully rely on the defence of independent contractor to avoid liability for nuisance is dependent "on whether he could reasonably have foreseen that the work he had instructed the independent contractor to do was likely to result in a nuisance": see Clerk & Lindsell, para 20-53 at p 1191. (11) It is not an answer to a person accused of nuisance in relation to the right of support for buildings to plead that it was caused by his independent contractor: see Clerk & Lindsell, paras 20-110 at p 219, see also Afro-Asia Shipping Co. Pte Ltd v Da Zhong Investment Pte Ltd [2004] 2 SLR 117. (12) There is no legislation equivalent or similar to the amendment in s 177 of the Conveyancing Act 1919 (NSW). (13) Singapore has generally adopted the English common law position with respect to the tort of nuisance. Nuisance has been defined as "a condition or activity which unduly interferes with the use or enjoyment of land": see the Singapore High Court decision of Tesa Tape Asia Pacific Pte Ltd v Wing Seng Logistics Pte Ltd [2006] SGHC 73 at [5]. (14) An actionable nuisance arises where there is undue interference with the neighbour's right to support their land resulting in subsidence in the surface of the neighbour's land and material injury to property results. This position is endorsed in the Singapore Court of Appeal landmark case of [ Xpress Print ] at [51], where it was held that there is: "a right to support from the adjoining soil, which in practical reality translates into a correlating duty of the adjoining landowner not to cause damage to his neighbour’s land by excavating or otherwise removing his land without first securing alternative means of support. Further, the right is infringed as soon as, but not until, damage is sustained in consequence of the withdrawal of that support. We would observe that what is prohibited is an active interference with the support which causes damage." (15) Although liability for nuisance has generally been regarded as strict, the principle of reasonable use is available as a defence: see OTF Aquarium v Lian Shing Construction Co Pte Ltd (Liberty Insurance Pte Ltd, Third Party) [2007] SGHC 122 at [23] – [26]. This means that if it can be proved that the use that caused the damage/disturbance was reasonable, the principle operates to preclude liability for any consequential damage. (16) On the assumption that the third party property damage was caused by the JV’s construction works, PowerGrid will be liable to such third party property owners for compensation for damage caused to their property under s 31 of the Electricity Act: para 7 of Exhibit O. (17) On the assumption that damage to and/or prevention of use of any property of the PUB was caused by the JV’s construction works, either PowerGrid or the JV will be liable to the PUB under the strict liability provision of s 48 of the Public Utilities Act: see para 8 of Exhibit O.

Section 31(3) of the Electricity Act provides: “The licensee referred to in subsection (1) shall pay compensation to any person who suffers any damage or loss of value to his property as a result of any work performed by the licensee under subsection (1) of such amount as may be agreed between the licensee and the person, and such compensation may include an annual payment for the use of any land or immovable property for the licensee’s purpose.”

Section 48 of the Public Utilities Act provides: “Compensation for damaging property of Board 48. (1) Any person who removes, destroys or damages, whether wilfully or otherwise, any property belonging to the Board or hinders or prevents the property from being used or operated in the manner in which it is intended to be used or operated shall, in addition to any penalty for which he is liable for an offence under this Act, be liable to pay compensation for the damage he has done and the compensation shall be recoverable by civil action before any court of competent jurisdiction. (a) Any court before which a person is charged with an offence under this Act may assess the compensation payable under this section and may make an order for payment of the same. (b) Any order made under subsection (2) may be enforced as if it were a judgment in a civil action.”

ACE objected to the receipt of [0] and [0] notwithstanding that both Mr Loh and Mr Yeo agreed with these conclusions.

In my view the joint views of the experts on both matters are admissible based upon the expressed assumptions and in my view the assumption in each case has been made out. Section 80 of the Evidence Act 1995 permits evidence about the ultimate issue but in any event the experts are addressing the question of whether Singaporean legislation would apply or not on the assumed facts.

The JV was not an adjoining landowner but rather a contractor working on the land owned by another person. The Singapore Court of Appeal’s imposition of a duty on a landowner not to cause damage to his neighbour’s land by excavating or otherwise removing his land “without first securing alternative means of support” (see [0] above) would apply a fortiori to a contractor who excavates the land of the owner in such a fashion as to remove support.

That English authority to the contrary would not be followed in Singapore can be seen in the Court’s statement in Xpress Print that the principle sic utere tuo ut alienum non laedas , i.e. “use your own property in such a manner as not to injure that of another” was the true legal justification for the right of support. In Langbrook , which involved withdrawal of support consequent upon pumping of water from adjacent land, Plowman J held that the Latin maxim set out above was not applicable. Langbrook was followed by Giles J (as he then was) in Xuereb v Viola (1989) 18 NSWLR 453.

I accept the joint expert opinion as reflecting the law of Singapore, not only because it is relevant evidence on which I am entitled to rely but also because it appears logically consistent with Xpress Print : see also Loh Siew Keng v Seng Huat Construction Pte Ltd [1998] SCHC 197 per Chan Seng Onn JC at [225] – [230].

No case was advanced by ACE that the use by the JV was a reasonable use: see T405 and T1165, and given the conclusions I have reached as to the failure to take steps available to the JV to deal with the risk of groundwater loss and that the use would not be one which has to be tolerated under a principle of “give and take or live and let live”, per Lawton LJ in Kennaway v Thompson [1981] QB 88 at 94 as cited Clerk & Lindsell on Torts , 19th ed (2006) Sweet & Maxwell at para 20-35 at p 1182, I do not think that a reasonable use defence was available to the JV or Lloyds: see Clerk & Lindsell on Torts and J G Fleming, The Law of Torts , 9th ed (1998) LBC Information Services at p 464. This is also consistent with the approach that the JV took in dealing with the third party claims.

It follows that in my view the JV was liable to the claimants in nuisance. I also accept the experts’ views on s 31(3) of the Electricity Act and s 48 of the Public Utilities Act, but it is not necessary to consider these legislative provisions further. ISSUE 3: Has the JV incurred a ‘legal liability’ within the meaning of the insuring clause?

The ACE policy requires that ACE pay to or on behalf of the JV all sums which the Insured “shall become legally liable to pay for any compensation”. ACE argues that it incurs no liability to pay any monies to the JV until the JV has had imposed on it a judgment, arbitral award or has entered into a formal settlement. It is agreed that there has been no judgment or arbitral award.

Lloyds argues that there was liability in the JV to the claimants, that work was carried out by the JV at its cost to repair the damaged buildings and infrastructure, and that the amounts so incurred constitute the compensation which the JV has had to give to the claimants, so no judgment, arbitral award or formal settlement is required. It was agreed that should the Court hold that a formal settlement is required by the wording, Lloyds’ argument that ACE’s conduct led the JV to believe that ACE was content for the JV to undertake the repair work and created an estoppel, would need to be considered in a separate hearing.

ACE relies on Penrith City Council v Government Insurance Office (1991) 24 NSWLR 564 (“ Penrith ”) to support its contention that it had no liability to indemnify the JV until judgment, arbitral award or formal settlement with the third party claimants is entered into.

In Penrith the GIO had issued a policy indemnifying the Council for breach of professional duty by reason of any negligent act, error or omission, whenever or wherever committed. Prompt notification of a claim or notice of a claim was required as the insurer was entitled to take over conduct of defence of the claim. A third party (“ Mitora ”) made a claim by way of cross-claim against the Council and on 19 October 1983 the Council sought indemnity under the policy. The GIO by its letter of 29 December 1983 refused to indemnify the Council. In 1984 after further correspondence, the GIO indicated that it maintained its position. The Council settled the claim by the third party on 2 April 1990 for $65,000 and it then sought to recover that amount and its legal costs in defending the third party action. The settlement was made without the consent of the GIO and it was not asked to consent. On 21 December 1990 the Council commenced proceedings against the GIO and the GIO asserted that the Council’s claim was statute barred as it had been brought six years after the cause of action accrued. This argument required determination of which of a number of possible dates was the date of accrual of the cause of action.

It was accepted that the policy did call for payment of legal costs by the Council but that the Council had never made a demand for such costs to be paid.

Giles J (as he then was) at 568E held that although the Council was entitled to indemnity “when the Mitora claim was made against it, or when it gave notice of the claim to the defendant, in the sense that it was entitled to the benefit of the defendant’s promise to indemnify it against the claim.” The Council’s cause of action against the GIO was for unliquidated damages for breach of contract. His Honour then considered when the breach occurred. His Honour held that it did not occur when the insurer failed to take over conduct of the defence because the GIO was not obliged to take over the defence; it was not in breach when the claim was made because receipt of a claim by an insurer is not a breach by an insurer. Further since the amount of liability of the Council could not be determined until there was either “agreement, arbitration or judgment” the GIO was not in breach of its obligation to meet that liability. Giles J said at 570 – 571: “The making of a claim or the deemed making of a claim, rather than the establishment of liability, determined when the defendant was exposed to a risk, but it was still necessary that a liability be established in order that performance of its promise be required of the defendant.”

There were three further dates which were identified as possible alternative dates: (1) the date of settlement by the Council; (2) the date on which the GIO again informed the Council it would not indemnify; or (3) the date the Council has paid the amount of $65,000.

All three dates were after 21 December 1984, and hence within the limitation period. His Honour said at 571B-C: “It does not matter whether [the Council’s] cause of action accrued on [date (1), date (2) or date (3)]. All were after 21 December 1984. Liability may be established by reasonable settlement as well as by arbitration or judgment (see Edwards v Insurance Office of Australia Ltd (1933) 34 SR (NSW) 88 at 94 and 98, 51 WN (NSW) 13) and on one view there was a continuing refusal to provide indemnity so that the defendant was in breach when liability was established. On another view, it was necessary that the defendant be informed of the settlement in order that it could properly be said that it was required to pay the $65,000 to Mitora or the plaintiff in performance of its promise to indemnify the plaintiff against the claim, so that its breach did not occur until it confirmed its refusal of indemnity on 12 October 1990. Or it may have been necessary that the plaintiff pay Mitora and seek reimburse­ment — as is shown by Wren v Mahony (1972) 126 CLR 212 at 229, the fact that the indemnity is against claims does not mean that the defendant was required to pay Mitora or to pay the plaintiff before it had paid Mitora. I express no preference for one or another of these views. It is enough that there was no breach completing the plaintiff's cause of action until after 21 December 1984.”

Giles J referred to a number of cases supporting his approach that the date of breach of contract was the relevant date. Cases establishing that the insured needs to suffer a loss before the insurer became liable were, he thought, not inconsistent with his approach. I proceed on the basis that Penrith was correctly decided.

The policy in Penrith was a claims made policy and not, as the ACE policy is, an occurrence policy. I think the passage cited in [119] above makes clear the context in which his Honour was considering what was required was the ‘breach’ by the insurer. In my view ACE became liable to indemnify the JV when the JV became liable to the third parties. The quantum of liability could not, however, be determined at that stage. ACE had full knowledge of the claims made by the third parties, the work done by the JV and the JV’s assertions of the costs of their work by mid-2005: see Exhibit A and in particular, pp 6-1590 – 6-1601. Whilst there is now an argument about what the amount should be, that will be determined in these proceedings.

ACE says however, that without a judgment, arbitral award or formal settlement it has no liability to indemnify. The third part claimants sent letters to the JV complaining that the JV’s works caused damage to their property. The JV’s works had, in fact, caused damage to their property. The JV took steps to repair that damage at its cost. No proceedings were brought by the third party claimants. Edwards v Insurance Office of Australia Ltd (1933) 34 SR (NSW) 88, to which Giles J referred, is a decision of the Full Court of the Supreme Court of New South Wales. The insured had indemnity against legal liability in respect of a motor vehicle which he owned and which his brother was driving at the time of an accident in which two persons were injured. The injured persons sued the insured and his brother, and the insured was left to defend himself because the insurer repudiated liability (wrongfully as it turned out). The question was whether the insured needed to prove his negligence or only rather that the settlement he had reached with the injured persons was reasonable. Davidson and Halse Rogers JJ, with whom Stephens J concurred, held that the insured did not have to establish that the driver had been negligent but only that the settlement was reasonable: see also Drayton v Martin (1996) 67 FCR 1 at 36 per Sackville J and Distillers Co Bio-Chemicals (Aust) Pty Ltd v Ajax Insurance Co Ltd (1970) 130 CLR 1 at 9 – 10 per Menzies J.

In the present case, unlike Edwards , Lloyds, rather than simply asserting that it was reasonable for the JV to settle the third party claims, has sought to establish successfully, that the JV was liable to the third parties. On the other hand it has not established that it entered into a formal settlement with the third parties.

The ACE policy does not itself provide that ACE is only required to pay the JV when a judgment, arbitral award or settlement has been entered into and Edwards makes it clear that judgment or arbitral award are not essential to the establishment of an insurer’s liability to indemnify. I have noted that in Penrith Giles J used the words “reasonable settlement”, but earlier he used the words “agreement, arbitration or judgment”.

The ACE policy required ACE to pay “all sums which the Insured shall become legally liable to pay for any compensation in respect of … property damage…”. I have found that the JV was legally liable to compensate the third parties and in my view the repair work which the JV undertook was a form of compensation, which the JV provided to those parties and then has endeavoured to quantify. I infer that it was agreed that the work would be accepted by the third party claimants in lieu of any monetary payment.

It follows that, in my view, the JV did incur a legal liability within the meaning of the insuring clauses. ISSUE 4: Was the damage sustained within the ACE policy period?

For ACE to have been required to indemnify the JV, the 'property damage' had to have occurred within the "period of insurance". The "period of insurance" is defined as the period between 17 April 2001 to 27 May 2003, followed by a 24 month maintenance period. Lloyds argues that the policy period is a period of four years. I do not accept that submission. The damage must occur within the 24 month period of works unless the damage has been caused by something done within the maintenance period. It was agreed that the maintenance period did not commence before 27 May 2003. It follows, in my view, that damage which occurred to buildings and structures after 27 May 2003 is not covered by the policy. There is evidence that the JV effected a new policy effective from 28 May 2003 with a different insurer. I do not think this is relevant, but it does at least demonstrate that there is nothing extraordinary about the conclusion I have reached.

It is then necessary to consider when it was that each of the buildings and structures (using Dr Redman’s list) suffered damage as a result of the tunnelling works that resumed in April 2003: (1) PUB deep sewer (worsening); (2) Intergraphica; (3) Jin Temple; (4) Sheng Temple; (5) 66 kv joint neal lamp pole 16; (6) Yee Lee Oil; and (7) Fence, gate, kerb.

There is some evidence which supports the conclusion that the additional damage to the PUB sewer occurred before 27 May 2003. First there is the fact that two blockages were suspected by 24 April 2003: CB vol 3.24 at pp 3-17211 – 3-17213, 3-17209 – 3-17210. Mr Funge prepared a report: CB vol 3.24 at pp 3-17195 – 3-17196, referring to damage as at 24 April 2003. Dr Redman assessed the cause as 50% due to the Woodlands shaft and 50% to the TBM/dyke problem.

The dyke at chainage 1456 was not struck until June 2003 after the expiry of the policy period, but there was significant ground surface settlement along the gas pipeline between 9 April and 4 June 2004: see Exhibit J at p 46. The sewer runs in the same direction as the gas pipeline. I find on the balance of probabilities that the worsening of the sewer drop had occurred by 27 May 2003 even though the PUB did not complain until 8 August 2003.

ACE accepts that damage to the Sheng Temple and Intergraphica occurred before 27 May 2003: see paras 35, 44 – 45 of ACE’s submissions in reply on third party damage.

So far as the Jin Temple is concerned, Lloyds noted that the owners of the Jin Temple had informed the JV that they had noticed cracks about three months before 13 June 2003. Given the significant settlement around the gas pipeline and significant flows of water, it is quite consistent with cracking having occurred (whether noticed or not) before 27 May 2003. I find on the balance of probabilities that this damage occurred within the policy period.

Items (5), (6) and (7) are very small and no submissions were advanced in respect of them. Dr Redman was unable to say when the damage to these items occurred.

Accordingly I am persuaded that damage to items (1) to (4) in [130] occurred within the policy period, but not items (5), (6) and (7) in [130]. Item (1), however, was partly a result of the TBM/G4 problem and would be excluded for the reasons explained earlier. ISSUE 5: What is the effect of GC 14 in the ACE policy?

There is not only dispute as to what is the proper construction of GC 14 and particularly how it fits with the PSE (exclusion 5) but there is a dispute as to how the Court is to go about the process of construction of the ACE policy. Lloyds contends that GC 14 swept away the exclusions including the PSE and ACE contends that such an interpretation, even if available on a strict construction of GC 14 (which ACE disputes) ought not be accepted having regard to the background to the formation of the insurance contract. ACE has sought to tender material constituting, principally, correspondence between the JV’s broker, AON and the JV, and between AON and itself to which it claims the Court ought have regard. Lloyds objected to the tender of this material but agreed that its relevance could be determined by this judgment. ACE also wishes to rely on the contents of the Appendices to the PowerGrid contract.

It is agreed that the relevant law as to construction of the ACE policy is that of Singapore, and that the recent case of Zurich Insurance (Singapore) Pte Ltd v B-Gold Interior Design & Construction Pte Ltd [2008] SGCA 27 (“ Zurich ”), a judgment of the Court of Appeal of Singapore, is the relevant binding authority.

The judgment in Zurich comprehensively reviewed the law relating to construction of contracts. It is a judgment of considerable depth and learning but helpfully, the judgment sets out at [131] – [133] the approach to be taken to the admissibility of extrinsic evidence in Singapore and the canons of construction which remain relevant.

The Court approved, as non-exhaustive, the summary of principles of construction (listed by Mr Gerard McMeel in his article ‘Prior Negotiations and Subsequent Conduct – the Next Step Forward for Contractual Interpretation?’ (2003) 119 LQR 272), i.e. (omitting internal paragraph references): “ The aim of construction First, the aim of the exercise of construction of a contract or other document is to ascertain the meaning which it would convey to a reasonable business person. The objective principle Secondly, the objective principle is therefore critical in defining the approach the courts will take. They are concerned usually with the expressed intentions of a person, not his or her actual intentions. The standpoint adopted is that of a reasonable reader. The holistic or 'whole contract' approach Thirdly, the exercise is one based on the whole contract or an holistic approach . Courts are not excessively focused upon a particular word, phrase, sentence, or clause. Rather the emphasis is on the document or utterance as a whole. The contextual dimension Fourthly, the exercise in construction is informed by the surrounding circumstances or external context. Modern judges are prepared to look beyond the four corners of a document, or the bare words of an utterance. It is permissible to have regard to the legal, regulatory, and factual matrix which constitutes the background in which the document was drafted or the utterance was made. Business purpose Fifthly, within this framework due consideration is given to the commercial purpose of the transaction or provision. The courts have regard to the overall purpose of the parties with respect to a particular transaction, or more narrowly the reason why a particular obligation was undertaken. Lawful effect Sixthly, a construction which entails that the contract and its performance are lawful and effective is to be preferred. Contra proferentem Seventhly, where a particular species of transaction, contract, or provision is one-sided or onerous it will be construed strictly against the party seeking to rely on it. Avoiding unreasonable results Eighthly, a construction which leads to very unreasonable results is to be avoided unless it is required by clear words and there is no other tenable construction. Specially negotiated terms Ninthly, a specially agreed provision should override an inconsistent standard provision which has not been individually negotiated. Generally provisions versus precise provisions Tenthly, a more precise or detailed provision should override an inconsistent general or widely expressed provision.” (emphasis added in original)

The Court, it said: (1) is required to bear in mind the limits found in ss 94 – 96 of the Singapore Evidence Act, discussed at [69] – [77] of the Zurich judgment; (2) may be aided by ss 97 – 99 of the Singapore Evidence Act, discussed at [78] – [79] of the Zurich judgment; and (3) “must remain mindful of the fact that in the end, commercial parties should have as much control over interpretative method as they do over other terms of the contract” – words found in Mr McMeel’s article, but again, endorsed by the Court.

The Court of Appeal summarised the approach to be taken (omitting internal paragraph references): “[132] Synopsis To summarise, the approach adopted in Singapore to the admissibility of extrinsic evidence to affect written contracts is a pragmatic and principled one. The main features of this approach are as follows: (a) A court should take into account the essence and attributes of the document being examined. The court's treatment of extrinsic evidence at various stages of the analytical process may differ depending on the nature of the document. In general, the court ought to be more reluctant to allow extrinsic evidence to affect standard form contracts and commercial documents. (b) If the court is satisfied that the parties intended to embody their entire agreement in a written contract, no extrinsic evidence is admissible to contradict, vary, add to, or subtract from its terms (see ss 93-94 of the Evidence Act). In determining whether the parties so intended, our courts may look at extrinsic evidence and apply the normal objective test, subject to a rebuttable presumption that a contract which is complete on its face was intended to contain all the terms of the parties' agreement. In other words, where a contract is complete on its face, the language of the contract constitutes prima facie proof of the parties' intentions. (c) Extrinsic evidence is admissible under proviso ( f ) to s 94 to aid in the interpretation of the written words. Our courts now adopt, via this proviso, the modern contextual approach to interpretation, in line with the developments in England in this area of the law to date. Crucially, ambiguity is not a prerequisite for the admissibility of extrinsic evidence under proviso ( f ) to s 94. (d) The extrinsic evidence in question is admissible so long as it is relevant, reasonably available to all the contracting parties and relates to a clear or obvious context. However, the principle of objectively ascertaining contractual intention(s) remains paramount. Thus, the extrinsic evidence must always go towards proof of what the parties, from an objective viewpoint, ultimately agreed upon. Further, where extrinsic evidence in the form of prior negotiations and subsequent conduct is concerned, we find the views expressed in McMeel's article and Nicholls' article persuasive. For this reason, there should be no absolute or rigid prohibition against evidence of previous negotiations or subsequent conduct, although, in the normal case, such evidence is likely to be inadmissible for non-compliance with the requirements set out at [125] and [128]-[129] above. (We should add that the relevance of subsequent conduct remains a controversial and evolving topic that will require more extensive scrutiny by this court at a more appropriate juncture.) Declarations of subjective intent remain inadmissible except for the purpose of giving meaning to terms which have been determined to be latently ambiguous. (e) In some cases, the extrinsic evidence in question leads to possible alternative interpretations of the written words (i.e., the court determines that latent ambiguity exists). A court may give effect to these alternative interpretations, always bearing in mind s 94 of the Evidence Act. In arriving at the ultimate interpretation of the words to be construed, the court may take into account subjective declarations of intent. Furthermore, the normal canons of interpretation apply in conjunction with the relevant provisions of the Evidence Act, i.e., ss 95-100. (f) A court should always be careful to ensure that extrinsic evidence is used to explain and illuminate the written words, and not to contradict or vary them. Where the court concludes that the parties have used the wrong words, rectification may be a more appropriate remedy. [133] It remains to be noted that, while we are aware that the impetus towards the contextual approach is in part due to the Europeanisation of the English common law, this shift also accords with common sense and logic. We also note that in several other common law jurisdictions such as Australia, New Zealand and Hong Kong, the contextual approach now prevails over the traditional approach (for an admirably lucid and comprehensive overview of the current Australian position, see J J Spigelman, "From text to context: Contemporary contractual interpretation" (2007) 81 ALJ 322). The traditional approach does not accord with ordinary commerce. Under that approach, the courts, faced with a wealth of text but a dearth of context, have often attributed to contracting parties artificial objective intentions that are divorced from reality. For the reasons given above, the contextual approach is here to stay in Singapore. It is also intriguing to note that, while meretricious symmetry or convergence is not the holy grail of legal jurisprudence, the adoption of the contextual approach to contractual interpretation is conceptually broadly similar to the purposive approach which our courts now adopt vis-a-vis statutory interpretation.”

The facts of Zurich were these. A builder entered into a contract with a property owner to carry out maintenance repairs and some additions and alterations to a business centre on the property. A clause of that building contract required the builder to obtain a policy of insurance indemnifying the owner, the builder and all subcontractors against damage to persons and property, for workers compensation and fire. A director of the builder asked a broker to arrange for the necessary cover under the building contract. Zurich issued a Contractors All Risks Insurance policy which covered physical loss or damage to the items entered into Section I and third party liability under Section II. One of the exclusions excluded liability consequent upon loss of or damage to property belonging to the building owner. A fire caused by the negligence of one of the builder’s subcontractors damaged the owner’s premises and it obtained a judgment against the builder. The builder claimed on its policy against Zurich and Zurich sought to rely on the exclusion clause.

The trial judge held that the exclusion clause operated and hence that Zurich was not liable to indemnify the builder. On appeal to the High Court of Singapore that decision was overturned, it being held that the exclusion clause was inoperable having regard to the genesis of the policy. On a further appeal, the Court of Appeal held that since the policy represented the complete agreement between the parties the parole evidence rule applied and no extrinsic evidence was admissible to construct, vary, add or subtract from the terms of the policy. For extrinsic evidence to be admissible it had to be “relevant reasonably available to all the contracting parties and relate to a clear or obvious context”. In a normal case such evidence, it was said, was likely to be inadmissible for non-compliance with the requirement sent out at [125] and [128] – [129] of the Zurich judgment.

The Court at [129] endorsed the views of Lord Hoffman in Investors Compensation Scheme Limited v West Bromwich Society [1998] 1 WLR 896 (“ Investor Compensation Scheme ”) as qualified by Lord Hoffman in Bank of Credit and Commerce International SA v Ali [2002] 1 AC 251: see [55] – [59] of the Zurich judgment.

It will be noted that Lord Hoffman, in the third paragraph of his elucidation in Investor Compensation Scheme at 912 said: “The law excludes from the admissible background the previous negotiations of the parties and their declarations of subjective intent. They are admissible only in an action for rectification. The law makes this distinction for reasons of practical policy and, in this respect only, legal interpretation differs from the way we would interpret utterances in ordinary life. The boundaries of this exception are in some respects unclear. But this is not the occasion on which to explore them.”

The Singapore Court of Appeal said at [62] (citations and internal paragraph references omitted): “First, the scope and quantity of extrinsic evidence that can be used in the exercise of interpretation has now increased. It should be noted, however, that the prohibition on evidence of the prior negotiations of the contracting parties and their declarations of subjective intent apparently remains applicable (see Lord Hoffmann's third proposition in Investors Compensation Scheme ; such evidence may, however, be admitted and used to interpret a contract after the court has determined that it contains a latent ambiguity. The restriction on evidence of prior negotiations has been said to arise out of the principle that the parties' intentions have to be objectively ascertained [see Prenn v Simmonds [1971] 1 WLR 1381]. It has come under fire in England (see Gerard McMeel, "Prior Negotiations and Subsequent Conduct – The Next Step Forward for Contractual Interpretation?" (2003) 119 LQR 272 ("McMeel's article") and Donald Nicholls, "My Kingdom for a Horse: The Meaning of Words" (2005) 121 LQR 577 ("Nicholls' article"); note also in our local context Art 8(3) of the United Nations Convention on Contracts for the International Sale of Goods (1980), which is applicable in Singapore by virtue of s 3 of the Sale of Goods (United Nations Convention) Act (Cap 283A, 1996 Rev Ed)). (We will deal with this point in greater detail at [132] below.) The second, and by far more important, impact of the Investors Compensation Scheme restatement is the change in the way in which contractual interpretation is to be carried out. From Lord Hoffmann's fourth proposition in this restatement, it appears that ambiguity is no longer a prerequisite for the admissibility of extrinsic evidence. Furthermore, the court may conclude that the "wrong words or syntax" (see Investors Compensation Scheme at 913) have been used, i.e. the court may substitute what it deems to be the correct language or syntax for that used in the contract. This creates an immediately apparent tension with the thin definition of the parol evidence rule as articulated at [33] above.”

As I have noted, the Court at [132] of Zurich, set out above, indicated agreement with the views expressed in Mr McMeel’s article ([140] supra ) and D Nicholls, ‘My Kingdom for a Horse: The Meaning of Words’ (2005) 121 LQR 577. The Court rejected the proposition that there should be an absolute prohibition against evidence of previous conduct (or subsequent conduct) but noted that it will normally be inadmissible for non-compliance with the requirement set out at [125] and [128] – [129] of Zurich . It is thus necessary to see what these requirements are. “[125] Turning to the first issue, we endorse Lord Hoffmann's view that extrinsic material is admissible if: (a) it is relevant – i.e., it would affect the way in which the language of the document would have been understood by a reasonable man [see BCCI v Ali [2002] 1 AC 251 at [39]; and (b) it was reasonably available to all the contracting parties (see Investors Compensation Scheme at 912. Such material is not confined to conventional empirical facts (such as the existence of an object or the occurrence of an event), but can include the state of the law, as in cases in which the court takes into account that parties are unlikely to have intended to agree to something which is unlawful or legally ineffective [see BCCI v Ali [2002] 1 AC 251 at [39]]. However, it must always be borne in mind that the purpose of interpretation is to give effect to the intention of the parties as objectively ascertained . Objective ascertainment of the parties' intentions (known as "the objective principle") is the cornerstone of the theory of contract and permeates our entire approach to contractual interpretation. …… [128] Crucially, the context of the contract must be clear or obvious. This was emphasised in [ Sandar Aung v Parkway Hospitals Singapore Pte Ltd [2007] 2 SLR 891], where the court stated (at [29]) that it would: “... go so far as to state that even if the plain language of the contract appears otherwise clear, the construction consequently placed on such language should not be inconsistent with the context in which the contract was entered into if this context is clear or even obvious ...” [emphasis added] The court added that (ibid): “It might well be the case that if a particular construction placed on the language in a given contract is inconsistent with what is the obvious context in which the contract was made , then that construction might not be as clear as was initially thought and might, on the contrary, be evidence of an ambiguity.” [emphasis added in bold italics] [129] We have already emphasised the importance of contractual certainty. In our view, the benefits of adopting, via proviso (f) to s 94, the contextual approach to contractual interpretation (viz, flexibility and accord with commercial common sense) will be maximised and its costs (viz, increased uncertainty and added litigation costs) minimised if, as a threshold requirement for the court's adoption of a different interpretation from that suggested by the plain language of the contract, the context of the contract should be clear and obvious. This is not a call for a retreat to literalism. In our view, this threshold requirement to some extent addresses the central weakness of contextualism - uncertainty. It is necessary and desirable to lay down this threshold requirement in order to achieve the right balance between commercial certainty and the imperative of giving effect to the objective intentions of the contracting parties.” (emphasis added in original)

I would note that the importance of context in construing contracts is well recognised in Australian law: see Pacific Carriers Ltd v BNP Paribas (2004) 218 CLR 451 at 462, and see Wilkie v Gordian Runoff Ltd (2005) 221 CLR 522 at 528 – 529 and Allstate Exploration NL v QBE Insurance (Australia) Ltd [2008] VSCA 148, in the insurance context.

I have had regard to the material upon which ACE seeks to rely. It demonstrates that ACE originally intended the professional exclusion to be included and that ACE charged no additional premium for the addition of GC 14. It contains no detail of conversations between ACE’s employees and AON’s employees (or any other agent acting on behalf of the JV) and I have a very real doubt as to whether the material advanced meets the criteria for admissibility laid down in Zurich (and which lead to rejection of the material sought to be relied on in that case) because it does not affect the way the language used would have been understood by a reasonable man, but assuming it is admissible it does not make clear or shed any light on how the parties were intending to treat GC 14 and the PSE.

Lloyds contends that GC 14 overrides the PSE because a 'contradiction' arises between GC 14 and the PSE. The PSE excludes liability where it arises out of the rendering of professional services and GC 14 says the policy will indemnify the JV, inter alia , for liability for damage to property caused by the weakening of support and any negligent acts of the JV.

Read literally, GC 14(iv) provides cover for "injury, illness, loss or damage covered by or in connection with or arising from" the matters enumerated and does not use the words 'liability for'. Lloyds submits that the opening words of (iv) should be read as: "The Insurers shall indemnify the insured in respect of liability for injury, illness, loss or damage covered by or in connection with or arising from..."

ACE disputes this interpretation and submits that the clause should be read without the addition of any words and hence read as a property damage extension (which ACE accepts would permit PowerGrid to recover under the policy). ACE submits that if that is not accepted, the words should not be read as restricting the ambit of the PSE.

If GC 14 is read as Lloyds submits it should, it would produce the result that none of the exclusions would apply where there had been negligence but do apply where there has been no negligence unless it involved a loss of support. ACE contends that it would be a remarkable thing if a policy were to include a whole raft of exclusions and then by inclusion of another clause effectively remove the need for any such exclusions, which is the consequence of accepting Lloyds’ contentions.

Lloyds call in aid a principle of construction that it should be accepted that a clause which, on its face, provided cover, should be seen as having work to do rather than as mere surplusage: see Bowling v Wienert [1978] 2 NSWLR 287 at 291, where Lee J said: “The very purpose of the extension is to give the protection of indemnity to those engaged in water skiing, aquaplaning or the like, and to hold otherwise would be tantamount to saying that the defendant had obtained an extension of the indemnity which left him in no better position than he would have been if the extension was not there.” ACE submits that if the words ‘liability for’ are not read into the clause then that principle is satisfied since GC 14 can be viewed as a property damage cover extension. Lloyds did not contend that if these words were excluded, any claim made by the JV would fall within the ACE policy.

Lloyds submit that the words “illness” and “injury” do not sit comfortably with the notion that GC 14 was really a property damage cover extension and they draw attention to the fact that the ACE policy is otherwise a public liability policy and so describes itself. The PSE itself does not actually refer to ‘liability for’ (or ‘claims for’), but Lloyds accepted that the opening words underneath “Exclusions”: “The Insurer shall not be liable for claims in respect of:” did apply to the PSE (exclusion 5).

The opening words of GC 14: “The clause contained in this Special Endorsement shall not in any way be deleted, altered or amended by other provisions in the policy and any other subsequent endorsements, and when contradiction arises, this Special Endorsement shall precede.” were said by Lloyds to be quite clearly a clause providing for GC 14 to be paramount. In my view the words “deleted, altered or amended” do not apply to exclusions and general conditions, because exclusions and general conditions do not delete, alter or amend GC 14. It is said, however, “where contradiction arises” does bring in the exclusion clause since, for example, the PSE excludes cover for liability arising out of the provision of professional services but GC 14 does not. On one view “where contradiction arises” means where contradiction arises because a clause expressly seeks to delete, amend or alter what is contained in GC 14, so that there is no contradiction merely because an exclusion clause already in place excludes liability in certain cases.

ACE also sought to draw attention to the RIBA clause in Appendix A. The argument was that GC 14, if ambiguous, ought to be read in a narrow way because it, as a Special Endorsement in Appendix B to the PowerGrid contract, was designed to deal with the exclusion from cover in respect of damage caused by the negligence of the contractor when subsidence had occurred, i.e. when RIBA cover and RIBA exclusions were utilised.

Thus paraphrasing the argument it is as follows: (1) Appendix B to the PowerGrid contract was a clause commonly included in construction contracts to offset the exclusion of cover contained in the RIBA clauses; (2) regard should be had to the fact that RIBA policy and Appendix B were both included in the PowerGrid contract as specimen policies; and (3) the Special Endorsement in Appendix B should be viewed as qualified because RIBA policies were common and both parties should be treated as having knowledge that Appendix B was only intended as a qualification to RIBA and not to introduce more widespread coverage.

This leads on to an attempt by ACE to explain why the Special Endorsement (or GC 14, as it became) contained the express provisions about negligence and why it was important to emphasise that it took precedence over another clause which excluded negligence: see Gold v Patman [1958] 2 All ER 497 in which it was held that a builder was not in breach of an obligation to obtain insurance when he obtained insurance to cover his own liability for his negligence. The difficulty with this explanation is that what was sent to ACE did not include any negligence exclusion, the RIBA exclusions or “exceptions” (see CB vol 2.1 at p 2.150) having not being sent to ACE. I do not accept that one could construe GC 14 by reference to clauses that were not communicated by the JV to ACE. I do not understand the Court of Appeal in Zurich to permit examination of material that was known to one party and not forwarded to the other in the process of negotiation of the policy wording.

I am of the view that the words ‘liability for’ should be read into GC 14(iv) because: (1) the policy is a public liability policy; (2) the introductory part of GC 14 speaks of other provisions of the policy and none of those provisions relate to property insurance; (3) the words “injury and illness” do not sit comfortably with a policy insuring property; (4) the word “damage” is not qualified by the word “property”; and (5) Appendix B to the PowerGrid contract from which GC 14 was taken, is titled “Special Endorsement attaching to and forming part of the public liability policy no. ___” (emphasis added).

I think there is force in ACE’s submission that it is an odd thing to include general provisions and exceptions only to then reinstate cover not normally provided by a public liability policy and I have not found construction of GC 14 an easy matter.

I think it is significant, however, that the form of GC 14 is identical to Appendix B of the PowerGrid contract and that its source was known to both parties, with the inference that it was something the JV was seeking to have incorporated because PowerGrid wanted it incorporated. That is the commercial context in which the need for this clause arose and the importance is that PowerGrid had an interest in obtaining protection from claims against itself as well as the ability to claim against a co-insured. Against this background there is more scope for acceptance that the policy is not fully coherent internally: see for example Netherlands Insurance Co Est 1845 Ltd v Karl Ljungberg & Co AB, The Mammoth Pine [1986] 3 All ER 767 at 771 per Lord Goff. GC 14 has work to do in that: (1) it contains a ‘waiver of subrogation’ sub-clause; (2) it contains a sub-clause which deals with how property of a co-insured is to be viewed when it is in control of the co-insurer; and (3) it contains a sub-clause which deals with how employees of PowerGrid and the supervisor under the PowerGrid contract are to be regarded. So far as (1) is concerned, there already is a waiver of subrogation clause within the policy: see General Conditions 5 and 6, Exhibit L at p 1.228.

Sub-clause (iv), however, expands the scope of GC 14 beyond these matters and it is difficult to see why these matters were expressly included if GC 14 was not intended to grant to the insureds a more expansive cover than existed without the sub-clause. I agree that it is unlikely that the sub-clause was intending to turn the policy into a workers compensation or motor accident policy but it is not difficult to accept that ACE understood that the JV wanted cover in respect of liability as a result of loss of or weakening of support in any event. The express inclusion of liability for these items (irrespective of negligence) and of liability for negligence in respect of other matters has a purpose if it is accepted that liability arising from the rendering of, failure to render or negligent provision of professional services would not otherwise be covered if it lead to loss of support or subsidence. I regard the words “shall precede” as overcoming any narrowing of the policy by exclusions relevant to public liability. On the approach I take, the liability to PowerGrid of the JV, otherwise excluded by the PSE, is protected by GC 14, a clause which PowerGrid expressly required the JV to have included, presumably in its own interest. ISSUE 6: Was there removal or weakening of support?

ACE also argued that GC 14 (iv)(b) did not apply because there was no 'removal or weakening of support'. Mr Simpkins argued that there was no evidence that the buildings, aprons, pipes etc could not be held up by the soil that was underneath these structures. They may have dropped down to a new level, he submitted, but then could still be 'supported' by the soils. In my view there is no substance to this argument. The PUB sewer pipe, to take one example, was resting on soil. When that soil consolidated and dropped, forcing the pipe to sag to accommodate the changed conditions, it lost support, albeit temporarily. The loss of support is what caused it to fracture or at least be endangered and hence it required remedial work. Similarly, when a building's footings dropped down they dropped down as a result of a loss of support: see Dr Redman’s evidence at T813.45 – 816. Mr Sternath’s analysis at p 5 of Exhibit 11, which I have set out at [73] together with his comment “The property damages were caused by ground settlements resulting from uncontrolled outflow of groundwater and soil material into the Gambas and Woodlands shafts as well as the Woodlands adits and the eastern TBM tunnel during different stages of excavation” confirms the correctness of this view. In my view, these events are properly characterised as a removal or weakening of support. ISSUE 7: Was there any negligent act by the JV?

As I am of the view that all of the JV’s liability arose out of a weakening of support which rendered the JV liable to third parties in nuisance it is not necessary to determine whether or not the JV was negligent. I have no doubt that it was negligent at the least, in relation to the TBM/G4 damage but I do not think it is necessary to go beyond that. In my view the PSE in the ACE policy and its equivalent in the Lloyds policy do not require negligence to be established to be operative. ISSUE 8: What is the effect of the Professional Services Exclusion in the ACE policy?

ACE bears the onus of establishing that this exclusion applies. ACE submits the PSE operates as follows: (1) There must be a liability in the JV to a third party for the policy to provide cover. (2) The liability must, relevantly, arise out of damage to property. (3) The damage in question must be linked to “a particular activity of the Insured or of another person for whom the Insured was responsible or there was some particular activity” See para 110F of ACE’s submissions of 10 March 2009. which the Insurer could have carried out or could have had carried out which would have prevented the damage. (4) The liability of the JV to the third parties arose because the JV’s design of the tunnelling works and its execution of the tunnelling works were both defective. (5) Design is a ‘professional service’. (6) Execution of tunnelling work involved “formulation and implementation of complex construction procedures and decisions to be taken in response to the continuation of construction in conditions actually encountered.” See para 129 of ACE submissions of 10 March 2009. (7) It is not necessary for ACE to establish negligence on the part of the JV. If liability is imposed on the JV and the cause is the rendering of professional services or failure to render a professional service, that is sufficient. If negligence is required, says ACE, then negligence is established. (8) ACE accepts that if Lloyds were able to establish that the cause of the damage claimed to arise in respect of the Woodlands shaft works were a result of construction defects in the diaphragm wall or construction defects in the toe area of the wall then the PSE would not apply but it asserts that the damage which arose has not been established to have arisen as a result of construction defects or at least not solely as a result of construction defects.

ACE points out that “arising out of” are words of wide import: see Government Insurance Office (NSW) v RJ Green & Lloyd Pty Ltd (1966) 114 CLR 437, although it should be noted that interpretation was influenced by the statutory context. I accept ACE’s contention that the PSE does not require ACE to establish negligence. I think this only serves to demonstrate the potentially far-reaching effect of the exclusion.

Lloyds asserts that the PSE has no operation because: (1) The cause of damage, and hence liability, were construction defects, in relation to the Woodlands and Gambas shafts, that there was no negligence in relation to the TBM/dyke problem and that the TBM/G4 problem was not due to the rendering or failure to render professional services. (2) The PSE ought to be read as limiting itself to liability to clients with whom the JV had a professional relationship – not third parties such as those who occupy premises adjoining Woodlands Avenue or who owned sewer lines or power cables under Woodlands Avenue.

It will be seen that there is a significant factual dispute relevant to the operation of the PSE that links closely to the question of why the difficulties which have transpired occurred.

Lloyds submits that the PSE ought to be read narrowly and it calls in aid Vero Insurance Ltd v Power Technologies Pty Ltd [2007] NSWCA 226 (“ Vero ”) and Fitzpatrick v Job t/as Jobs Engineering [2007] WASCA 63 (“ Fitzpatrick ”), which Vero applied . If the interpretation Lloyds contends for is upheld it would not be necessary to determine the answer to the question of whether the liability of the JV arose out of the rendering of or failure to render a professional service, except in the case of PowerGrid’s claim.

In Vero , the NSW Court of Appeal Beazley JA (with whom Campbell JA and Harrison J concurred) held that the insured’s liability did not arise out of a professional duty owed to a worker who was injured at a power station but rather arose out of breach of a duty of care owed to him in negligence. The injured plaintiff has been engaged to work at a power station removing fly ash fibres released where the insured carried out maintenance and repair work on boiler tubes that were insulated with asbestos and maintained by the insured in conjunction with others.

The exclusion clause in Vero was in the following terms: “The Company shall not be liable for: (1) Claims arising out of a breach of the duty owed in a professional capacity by the Insured”.

Beazley JA referred to Chemetics International Ltd v Commercial Union Assurance Company of Canada (1984) 11 DLR (4th) 754 and Fitzpatrick . Beazley JA accepted the approach of Buss JA in Fitzpatrick (which approach had the concurrence of Steytler P) which was to hold that the exclusion of professional indemnity cover in the context of a products liability policy was limited to claims made against the manufacturer/designer “arising out of breaches of duty owed to persons who had retained it to perform work or services in the course of its business”: see [147] of Vero . Buss JA had said in Fitzpatrick at [270]: “ The exclusion does not extend to breaches of duty owed to third parties who may suffer foreseeable loss or damage as a result of negligent acts or omissions by Jobs Engineering in designing, manufacturing or supplying machinery and equipment, including the negligent failure to give advice of the kind which it should have given to V & D Ridolfo. My construction of exclusion 10(a) is consistent with the evident object of the products liability cover, namely, to provide indemnity, of real and not negligible value, in respect of claims for personal injury and property damage caused by defective goods and property designed and manufactured by Jobs Engineering, and put into circulation within Australia. ” (emphasis added) And Beazley JA said at [148] – [149]: “In my opinion, the approach taken by Buss JA in Fitzpatrick should be applied here. The exclusion clause considered here is sufficiently identical to that in Fitzpatrick and the type of policy involved was relevantly similar, in the sense that it was to provide specific cover to the respondent for claims for sums which it became legally liable to pay for bodily injury. The construction given to the exclusion clause by the Western Australian Court of Appeal enables the policy to provide the indemnity which the parties undoubtedly intended the policy to have. To adopt a different construction would have significantly undermined the commercial purpose of the policy. Insurance policies are rarely confined in their terms by state boundaries and it is important that the clauses which are relevantly the same should be given the same construction. Finally, this Court should apply the decisions of a court of equal status unless the Court considers that the decision is wrong and should not be followed: see Australian Securities Commission v Marlborough Gold Mines Limited (1993) 177 CLR 485 at 492; [1993] HCA 15; Farah Constructions Pty Ltd v Say-Dee Pty Ltd (2007) 81 ALJR 1107; [2007] HCA 22 at [135]. The policy was a public liability policy. It excluded claims arising out of a breach of duty owed in a professional capacity by the respondent. The respondent did not owe a professional duty to Mr Barlow. Rather, it owed him a duty of care as a third party in respect of whom it was reasonably foreseeable might suffer damage as a result of its negligence in designing and supplying plant and equipment to Delta and its predecessors and in the manner in which it carried out maintenance on that plant and equipment.”

ACE points out that Vero is not binding on me first because I am required to apply Singaporean law and secondly because the wording of the exclusion clause in Vero was “claims arising out of a breach of the duty owed in a professional capacity by the Insured”, whereas the ACE policy speaks in terms of claims “arising out of the rendering or failure to render professional services”. In my view, although the ACE clause is wider because it does not require a breach of duty, this, in my view, is not a material difference.

ACE and Lloyds each asserted that its contentions gained support from the Court of Appeal’s decision in Transfield Services (Australia) v Hall [2008] NSWCA 294 (“ Transfield ”). Adventure Training Systems Pty Ltd (“ ATS ”) had had the responsibility of maintaining a rope course that was utilised in a naval reserve training program and during the use of which a naval reservist was injured as a consequence of rusted wire forming part of a safety strop.

The insurer relied on an exclusion clause excluding liability for: “10.4.1 the rendering of or failure to render professional advice or services by the Insured or any error or omission connected therewith. 10.4.2 advice, design formula or specification given for a fee.”

The trial judge had held that ‘professional advice or service’ meant advice or service that could normally be expected to be provided by a professional engineer and that work in the nature of a routine maintenance did not amount to providing professional advice or service. At [191] – [192] Campbell JA, indicating his agreement with the trial judge’s approach, said: “No features of the particular insurance policy in question are pointed to as aids to construction of this phrase. In construing any exclusion, the language of which admits of more than one interpretation, the court takes into account both the contra proferentem principle (which is well established in insurance law, whatever its status might be in other areas of contractual interpretation: North v Marina [2003] NSWSC 64; (2003) 11 BPR 21,359), and the principle that it would not give effective business operation of contract if an exclusion clause inappropriately circumscribed the cover provided by the insuring clauses: Fitzpatrick v Job [2007] WASCA 63; (2007) 14 ANZ Ins Cas 61-731 at [270]. This court has recently applied Chemetics and Fitzpatrick v Job, in Vero Insurance Ltd v Power Technologies Pty Ltd [2007] NSWCA 226; (2007) 5 DDCR 206 at [147]-[149]. The High Court refused special leave to appeal from the decision in Vero on 7 March 2008. Upholding the trial judge's decision concerning construction of this exclusion is comfortably within the language of the clause, and would be in accordance with the contra proferentem principle. There is no relevant point of distinction between the present case and Vero , so far as the construction of the exclusion is concerned. In these circumstances there is no need to rely on the principle whereby exclusion clauses should not be construed so as to circumscribe inappropriately the insuring clauses. In my view the trial judge’s decision on this question of construction was right.”

Campbell JA agreed with the trial judge’s conclusion that ATS had given advice for a fee and hence that the insurer’s liability was excluded. McClellan CJ at CL (with whom Beazley JA concurred) disagreed and held that the second part of the exclusion clause did not operate. In the course of dealing with the second part of the exclusion clause Campbell JA said at [210]: “In Vero , the engineers had not breached any duty owed in a professional capacity at all – there was nothing wrong with the boilers that they designed and installed. Rather, the claim arose out of a duty of a different character, namely an ordinary common law duty of care. The exclusion that is in question in the present case depends upon the manner in which the liability against which indemnity is sought arose ( viz , from the giving of advice for a fee). It would involve narrowing the exclusion, in a way for which there is no textual support, if it were restricted to liability owed only to the narrow class for which Mr Campbell contends. ” (emphasis added)

McClellan CJ at CL agreed that the first part of the exclusion had no application but said at [239]: “The word “advice” in the exclusion clause should be interpreted bearing in mind what was said by the High Court in Darlington Futures Ltd v Delco Australia Pty Ltd (1986) 161 CLR 500 at 510: “The interpretation of an exclusion clause is to be determined by construing the clause according to its natural and ordinary meaning, read in light of the contract as a whole, thereby giving due weight to the context in which the clause appears including the nature and object of the contract, and, where appropriate, construing the clause contra proferentem in case of ambiguity.” The policy was effected for the purpose of indemnifying ATS with respect to risks in carrying out its business. That business included the maintenance of facilities which it had supplied. Excluded from the indemnity was any liability for professional advice or advice given for a fee. Also excluded are “design, formula, or specification given for a fee”. Providing a “design, formula or specification” is to provide a specific form of advice. Common to each of the matters excluded is the provision of information in the nature of intellectual property. This is quite unlike the task involved in carrying out a maintenance inspection (which included various “repairs”) of existing equipment and providing an assurance that it is safe for it to be used. The exclusion clause would only operate to exclude liability for advice given for a fee where that advice was given pursuant to a separate professional engagement of ATS in the nature of a consultancy distinct from its maintenance function. ” (emphasis added)

It will be observed that ATS had a professional liability to Transfield, not to the injured plaintiff; the Court regarded the liability to the plaintiff as within cover because it did not arise out of a separate professional engagement of ATS in the nature of a consultancy. This aspect focused upon what was meant by ‘advice for a fee’. The Court was unanimous that the reasoning in Vero applied to the first part of the clause. Campbell JA, in dissent, certainly did not accept the argument that liability to Mr Hall was not caught by the exclusion because he was not the client: see the passage from [210] of Transfield highlighted at [179] above. The last passage of McClellan CJ at CL’s judgment at [239] of Transfield set out at [180] above did not, in coming to a contrary view, expressly embrace the argument put on behalf of Mr Hall, and probably does not do so implicitly. I accept therefore that Transfield is not authority in support of Lloyds’ argument but given the entire Court’s acceptance of Vero and Vero ’s acceptance of what had been said by Buss JA in Fitzpatrick at [147] (highlighted at [174] above), I do not regard Transfield as authority against the proposition advanced by Lloyds.

I proceed on the basis that Singapore law would recognise and adopt the approach of avoiding an inappropriate circumscription of cover.

Reference was made to GIO General Ltd v Newcastle City Council (1996) 38 NSWLR 558. The GIO policy provided cover for the Council in respect of (a) public liability for personal injury and damage to property, (b) product liability, and (c) professional liability. There was an endorsement which excluded cover for claims in respect of “the rendering of or failure to render professional advice or services” except as provided in (c). The first issue on appeal was whether the claim fell within (a). Kirby P (with whose judgment Powell JA concurred) held that it did not and hence that it fell within (c). The majority were of the view that the endorsement, in any event, required the claim to be brought under section (c). Sheller JA disagreed that absent the endorsement the claim did not fall within (a), but agreed that the endorsement placed the claim within (c). It was held that in the context of a professional indemnity insurance written for a local council, the term ‘professional’ involves no more than advice and services of a skilled character according to an established discipline: per Kirby P at 568F – 569A, with whom Powell and Sheller JJA agreed. Kirby P made it clear that the context was important to how widely the term ‘professional’ was defined, and see Suncorp Metway Insurance Ltd v Landridge Pty Ltd [2005] VSCA 223 per Nettle JA at [27], cited with approval in Vero per Beazley JA at [145]. All members of the Court held that s 40 of the Insurance Contracts Act 1984 (Cth) did not apply to excuse the insured’s failure to give notice of the claim within the requisite period, and that latter point was the subject of the successful appeal to the High Court.

In the Court of Appeal it was not contended by the GIO that the claims against the Council did not fall within the policy but rather only that they did not fall within (a) – thus it was the insurer who wanted the claim to fall within the description ‘professional’ so that it would be a claim in respect of which notice was required to be given and the insured, whilst arguing that the term ‘professional’ did not cover the claim, had no interest in contending that only claims by its ‘clients’ were covered; as Kirby P pointed out at 567G, given his conclusion that section (a) was not available, the Council would favour the conclusion that it fell within (c). I do not think this case offers any guidance on whether an exclusion of liability for design should be construed as restricted as Lloyds here contend.

In Fitzpatrick Buss JA was pointed out that the term ‘professional’ in a professional indemnity policy did not necessarily bear an identical meaning in an exclusion clause of a public or products liability policy” see Vero at [145]. I think this observation is apt here too; part of the reason for that it that is that exclusion clauses are to be read contra proferentum in the case of ambiguity: Provincial Insurance Aust Pty Ltd v Consolidated Wood Products (1991) 25 NSWLR 541 and Transfield per Campbell JA at [192] and McClellan CJ at CL at [239].

In forming a view as to the ambit of the exclusion I think it is significant that the cover provided was public liability cover, and that the underlying activities were to be provided pursuant to a design and construct contract. The liability and also everything that is specifically mentioned is required to arise out of the rendering (or failure to render) a professional service, which entails the provision of a service to a client. The JV did not render a service to anyone other than PowerGrid. It is true that the JV had professional indemnity cover from another source, but leaving aside PowerGrid the property owners here were not clients of the JV and the existence of other cover cannot control the terms of the cover provided by the ACE policy.

In my view an interpretation of ‘professional services’ that is not restricted to professional clients of the JV would cut back the cover dramatically and pays insufficient regard to the obvious commercial purpose of the policy and I adopt the approach taken in Fitzpatrick , Vero and Transfield , which is to read down the clause as dealing with liability to a client.

It follows that the only liability which is potentially excluded by the PSE is the liability of the JV to PowerGrid.

In my view, the JV is liable to PowerGrid in contract (see [13] above) and in nuisance. Part of the reason for that damage, even on Lloyds’ case, was the TBM/G4 problem. Mr Donaldson conceded at T1229 that there ought to have been more probing and grouting, and see Dr Redman at T816.46 – 49 and T817. Dr Pells noted that there had been 629 m³ of over-excavation in this region: see Exhibit 4 at p 41. I deal with the TBM in more detail in relation to Issue 17 below, but, in short, the JV proceeded with excavation using an unsuitable TBM in circumstances where it was on notice of the risk involved and took inadequate steps to deal with the problem by either lowering the depth of the tunnel or ensuring that the ground around and in front of the TBM was adequately grouted before the TBM was set to further excavation. Mr Sternath’s evidence at T1133, which I accept, explained the JV’s design criteria were flawed because it was on notice that there was a high probability that there was not five metres of rock cover above the tunnel, did not alter the depth of the east tunnel (as they did in relation to the west tunnel), did not change the TBM and did not check and treat the ground in advance of excavation: see T1138 – 1139 and see Exhibit 11 at pp 17-20. Dr Pells’ criticisms encompass these points as well. Dr Redman was also critical of the JV in relation to its failure to adequately investigate and assess the ground conditions and inadequate development of contingency plans to deal with the potential hazard that the mixed-face condition presented: see T845.43 – 847.15 and CB vol 8.2 at p 8-520. I find, therefore, in relation to the TBM/G4 problem (as with the Woodlands shaft) that the liability to PowerGrid arose out of the rendering of or failure to render a professional service to PowerGrid. ISSUE 9: If the PSE and GC 14 applies, is ACE’s liability to indemnify excluded by the PSE in ACE policy?

I have dealt with the effect of GC 14 under Issue 5 at [137] to [164] above and I have concluded in relation to GC 14 that its effect is to override the PSE. It therefore follows that the liability to PowerGrid is not excluded by operation of the PSE. ISSUE 10: Does the rule in Wayne Tank operate to exclude any liability of the JV where the cause was the result of both professional and non-professional services?

The “ Wayne Tank ” principle (which derives from Wayne Tank & Pump Co Ltd v Employers Liability Assurance Corp Ltd [1974] 1 QB 57) is that where loss or damage (or liability) arises by reason of two or more causes, one of which is excluded under the policy, then the policy does not respond: see Peterson v Union de Assurance Ltd Paris IARD [1985] ANZ Ins Cas 61-24 (“ Peterson ”), Baulderstone Hornibrook Engineering Pty Ltd v Gordian Runoff Limited [2008] NSWCA 243.

ACE’s submissions pointed out that there was no evidence of what Singaporean law was on this topic and that accordingly, I should apply the law of New South Wales, a proposition which I accept. The submissions helpfully summarised the principles from Peterson (and no contrary submissions were advanced by Lloyds) as: “Assuming that there are at least 2 causes of the loss, and one of them is not excluded by the policy then the rules of construction set out in [ Peterson ] summarised below may need to be applied: (a) if one of the causes is excluded, the loss is also excluded irrespective of whether or not the other cause or causes are insured; and (b) if one of the causes is insured and the other cause or causes are neither Insured nor excluded, the loss is covered.” See ACE’s submissions dated 11 March 2009 at para 81. And summarised the relevant principles relating to proximate causes as: “(a) Proximate cause does not mean the latest, but the direct, real or common sense, dominant, operative or efficient one ( National & General Insurance Co Limited v Chick [1984] 2 NSWLR 860); (b) In any event, the term “arising from” requires something less than a relationship of proximate cause and imports a more general notion of causation irrespective of whether the term is used in an insuring clause or an exclusion clause. (c) It is not always possible to single out one effective or dominant cause of a loss and the Court should avoid straining “to find a dominant cause if… there are two causes both of which can properly be described as effective causes of the loss…unless one cause is clearly more decisive than the other, it should be accepted that there are two causes of the loss and no attempt should be made to give one of them the quality of dominance” ( Wayne Tank ). (d) It is proper for a Court to conclude that on a common sense view of the facts there can be two causes which are equal or at least nearly equal in their efficiency in bringing about the damage ( JJ Lloyd Instruments Limited v Northern Star Insurance Co Limited (“Miss Jay Jay”) [1987] 1 Lloyd’s Rep 32); (e) There may be two or more causes in the sense that where there are two or more causes equal, or about equal, in effect the insured is entitled to recover if one of the causes is an insured peril. There may be more than one cause of sufficient significance. Causes can be regarded as having approximately equal efficiency ( Waterwell Shipping Inc v HIH Casualty & General Insurance Limited [1987] Aust Torts Reports 81-444). (f) “…where each of two or more causes may realistically be described as an efficient and, in that sense, proximate cause, prefers common sense to the artificial exercise of attempting to decide which of two or several is the winner of what in fact may be a dead heat.” ( HIH Casualty & General Insurance Limited v Waterwell Shipping Inc [1998] 43 NSWLR 601).” See ACE’s submissions dated 11 March 2009 at para 82.

I accept that the Wayne Tank principle leads to the exclusion of liability for any damage arising in part from an excluded clause. I note that Dr Redman attempted an apportionment of responsibility between incidents. I thought he was quite candid as to the basis of his apportionment and his approach seemed to me to be quite reasonable, but in any event, as long as an excluded event had some part to play in the liability imposed, then that is sufficient to exclude that item of claim. I shall deal with the consequence of this conclusion later in the judgment. ISSUE 11: Was the TBM/G4 problem an intentional act?

There is a principle of insurance law that an insured cannot recover from the insurer in respect of damage or liability that arises out of a deliberate or intentional acts: see Beresford v Royal Insurance Co Ltd [1938] AC 586, Fire & All Risks Insurance Co Ltd v Powell [1966] VR 513, Hanover Insurance Group v Cameron 298 A 2d 715 (1973), Aetna Casualty and Surety Co v Freyer 411 NE 2d 1157 (1980) and Clayton v Mutual Community General Insurance Pty Ltd (1995) 64 SASR 353. In my view the principle is not enlivened in relation to the TBM/G4 problem (which was the only incident in respect of which this assertion was made) as there is no basis upon which it could be concluded that Mr Embery or the JV intended to bring about the collapse of the overburden, and/or damage to property. ISSUE 12: How many ‘occurrences’ were there?

I have explained the difference in approach of Dr Redman and Dr Pells to the number of incidents.

The question of what is an ‘occurrence’ and what is ‘one event’ is one which has arisen in a number of cases. ACE referred to the Dawson’s Field Arbitration (March 1972), a decision of Mr Michael Kerr QC (as he then was), Kuwait Airways Corporation v Kuwait Insurance Co SAK [1996] 1 Lloyd’s Rep 664 and Scott v Copenhagen Reinsurance Co (UK) Ltd [2003] 2 All ER (Comm) 190, to which I would add SR International Business Insurance Co Ltd v World Trade Centre Properties LLC F Supp 2d 2002, a case concerning the World Trade Centre attack of 11 September 2001.

ACE submits that for a series of facts to be described as an event there must be unity of time, unity of location and unity of cause, but this involves “the question of degree of unity in relation to cause, locality, time and if initiated by human action, the circumstance and purpose of the persons responsible” per Mr Kerr QC at (as he then was) in the Dawson’s Field Arbitration . In Kuwait Airways Corporation v Kuwait Insurance Co SAK , Rix LJ (as he then was) said: “An occurrence is not the same as a loss, for one occurrence may embrace a plurality of losses. Nevertheless, the losses’ circumstances must be scrutinised to see whether they involve such a degree of unity as to justify their being described as, or as arising out of, one occurrence. The matter must be scrutinised from the point of view of an informed observer placed in the position of the Insured. In assessing the degree of unity regard may be had to such factors as to cause, locality and time, and the intentions of human agents.” and: “A plurality of losses is to be regarded as a single aggregated loss if they can be sufficiently linked to a single unifying event by being causally connected with it. The aggregating function of such a clause is antagonistic to a weak or loose causal relationship between losses and the required unifying single event. This is the more easily seen by acknowledging that, once a merely weak causal connection is required, there is in principle no limit to the theoretical possibility of tracing back to the causes of causes.”

ACE submitted that the causal relationship between the losses did not have to be one of proximate cause.

Although I accept ACE’s contention that the JV’s approach to the work can be described as involving a failure to adequately deal with prospectively poor ground conditions with likely abundant water flows and consequential groundwater loss and loss of soil and overburden, I do not accept that this can properly be described as ‘the event’. In my view, the event is an event in either tunnelling or excavation – that is what led to the loss of groundwater and/or overburden. Further, in my view, the TBM/G4 problem arose out of quite a different cause (i.e. tunnelling through G4 material in the main tunnel to chainage 1141) at a different time and location to each of the Woodlands shaft, Gambas shaft and TBM/dyke problems. There are only two items claimed for Gambas shaft and Dr Redman saw the TBM/G4 problem as contributing to both, so there is no damage ascribed to both the Gambas and Woodlands shaft excavations. The Woodlands shaft and Gambas shaft excavations took place at different times and, although close, sufficiently different locations. The Woodlands shaft involved a dyke and construction of adits close by, neither of which was a factor at the Gambas shaft. I do not think losses to the two items claimed for Gambas shaft arose out of the same event as Woodlands shaft. ISSUE 13: Cause of damage under Lloyds policy.

I have dealt with this issue in relation to the ACE policy: see above at [60]. ISSUE 14: Has the JV incurred a ‘legal liability’ within the meaning of the insuring clause of Lloyds policy?

I have dealt with this issue in relation to the ACE policy: see above at [103]. ISSUE 15: Was the JV ‘unable to obtain an admission of liability’ from ACE?

Clause 1.17 of the Lloyds policy has given rise to a number of issues. The first is that Lloyds claims that the JV was unable to obtain an admission of liability. ACE, of course, does not even now admit that it was liable to indemnify the JV and indeed has denied that it was: see para 15(c) of the Amended Commercial List Response to Further Amended Summons.

Lloyds points to correspondence in June and July 2005 passing from the JV to ACE which elicited no admission of liability.

The relevant chronology of events, relied on by Lloyds in its submissions dated 9 March 2009 at para 116, is as follows: (1) On or about 12 January 2002, the JV notified AON that a third party had advised the JV that it had suffered property damage. A copy of the letter from the JV to AON dated 12 January 2002 is located at [CB vol 6.1 at pp 6-1 – 6-7]. This letter was forwarded by AON to ACE by facsimile on 14 January 2002, a copy of which is located at [CB vol 6.1 at pp 6-8 – 6-10]. (2) On or about 15 January 2002, ACE acknowledged by email to AON receipt of the claim notification forwarded to it by AON on behalf of the JV on 14 January 2002. A copy of this email is located at [CB vol 6.1 at pp 6-11 – 6-12]. (3) On or about 7 February 2002, ACE notified AON by email that it had appointed Crawfords to investigate the damage to the property of third parties notified by the JV. A copy of this email is located at [CB vol 6.1 at p 6-18]. (4) Thereafter, the JV advised ACE of any further concerns regarding ground settlement and damage to third party property and sought indemnity for the costs it had and was incurring to rectify such damage. (5) In or about September or October 2003, the JV claimed indemnity under the plaintiffs’ insurance policies for the costs incurred in relation to the rectification of third party property damage. (6) On or about 13 November 2003, AON received from ACE a letter asserting that the JV’s claims were not covered by the ACE policy, on the grounds of an exclusion excluding property damage arising out of the rendering of or failure to render professional services. A copy of this letter is located at [CB vol 6.2 at p 6-799]. (7) On or about 18 November 2003 AON informed ACE that the JV did not agree with the decision regarding indemnity. A copy of the letter from AON to ACE dated 18 November 2003 is located at [CB vol 6.2 at p 6-803]. (8) Thereafter followed correspondence between the JV, AON and ACE in relation to the ACE policy. (9) On or about 18 November 2003, AON sent a letter to [Limit] enclosing the letter from ACE to the JV dated 13 November 2003 (at CB 6.812). The letter from AON relevantly states: “…the ACE stance clearly alters the position for DA Constable; from one of ‘watching brief’ over a primary Public Liability Insurer, to the potential insurer of resort…” (10) On or about 12 April 2005 the JV sent to AON a detailed submission in support of the proposition that the JV was entitled to indemnity from ACE. AON forwarded that submission to ACE. (11) On 24 June 2005 ACE sent AON a letter repeating its position that, by reason of the professional services exclusion and the alleged failure by the JV to take reasonable precautions to prevent damage, it was not liable to indemnify the JV. (12) By letter of 8 July 2005, [Limit] wrote to AON confirming indemnity, and in about July 2005, [Lloyds] agreed to make an interim payment to the JV of SG$10 million, minus the deductibles on the policies. (13) In about October 2006 [Lloyds] entered into a written agreement with the JV, finally resolving the JV’s claims on the Lloyds policy. The settlement required [Lloyds] to pay to the JV a further SG$3 million, which amount was paid to AON on behalf of the JV in August 2006.

ACE did not dispute this chronology. ACE’s argument is that since it had no liability to indemnify in the absence of a judgment, arbitral award or settlement it had no obligation to indemnify and the JV could not seek an admission to indemnify. In my view, there is no separate point here – the JV has established that it was unable to obtain an admission from ACE and clause 1.17 was enlivened. ISSUE 16: If so, was Lloyds liable to indemnify the JV under Section 1?

Whether Lloyds was liable to indemnify the JV depends on the answer to some issues common to both policies. I have dealt with those common issues in the context of the ACE policy. Whether an exclusion contained only in the Lloyds policy was available to Lloyds depends on the approach taken to clause 1.17. Lloyds claims that clause 1.17 (the Difference in Conditions clause) is to be construed as follows. If the insured cannot obtain an admission of liability from an underlying insurer then Lloyds will indemnify the insured on the same terms and conditions as the underlying policy save for the exceptions under Section 5 noted in sub-clause 1.17(b), with the consequence that Lloyds “follows the form” of the ACE policy and exceptions of the Lloyds policy, other than those in Section 5, do not apply. Of particular significance is the “reasonable precautions” exception contained in the Lloyds policy but not in Section 5, because there is no “reasonable precautions” requirement in the ACE policy and because ACE asserts that Lloyds could have relied on that provision if the claim was a Section I claim and Lloyds was not therefore liable under its policy in respect of the TBM/G4 damage and therefore cannot claim against ACE in respect of it. ACE submits that the Lloyds policy does not “follow the form” of the ACE policy in circumstances where the underlying insurer fails to meet a claim.

The structure of clause 1.17 (leaving aside the material in Exhibit 1 at p 1-41) is as follows: (1) Where the cover provided by Lloyds and the underlying policy is identical but for the amount – the Lloyds policy is an excess policy and will only respond if the amount of indemnity required exceeds the limit of the underlying policy: sub-clause 1.17(a). (2) Where the cover for legal liability provided by the underlying policy is more extensive than the Lloyds policy the Lloyds policy will, under Section 1 of the Lloyds policy, provide cover to the same extent as the underlying policy – but subject to the exceptions under Section 5 of the Lloyds policy: sub-clause 1.17(b). (3) Where the cover under the underlying policy does not indemnify the insured (in whole or in part) in respect of legal liability, costs and expenses as covered under the Lloyds policy, the Lloyds policy “shall provide indemnity to the extent that such indemnity is not provided by the terms and conditions” of the underlying policy: sub-clause 1.17(c).

The question is whether the words which follow sub-clause 1.17(c), i.e.: “In the event that the Insured cannot obtain an admission of liability from the insurer of an Underlying Insurance and/or Local Policy and/or the Underlying insurance and/or Local Policy falls or is reasonably likely not to indemnify the Insured, then the Insurer of this Master Policy shall be obligated to indemnify the Insured and defend any actions.” which I shall refer to as “ the First Addendum ”, and followed by “The Provisions of this clause are subject always to the terms, Conditions and Exceptions of this Policy.” which I shall refer to as “ the Second Addendum ”, are to be read as part of sub-clause 1.17(c) or as applicable more generally.

Lloyds argues that the First Addendum has to be read with sub-clause 1.17(b). It argues that if the reasonable precautions clause applies (contrary to its contention) then the coverage under the Lloyds policy is not as wide as the ACE policy so that coverage under ACE policy terms and conditions prevail. It follows on this argument, that if the Lloyds policy excludes cover for a failure to take reasonable precautions, but the ACE policy does not, then the ACE policy is wider than the Lloyds policy and hence cover is available on the ACE terms pursuant to sub-clause 1.17(b).

There are some problems with the construction advanced by Lloyds. (1) As a matter of layout the First Addendum and the Second Addendum appear to be a continuation of sub-clause 1.17(c). The proviso appearing beneath the Second Addendum is clearly intended to be a proviso to the whole clause. (2) The First Addendum seems conceptually close to what is dealt with in sub-clause 1.17(c) since sub-clause 1.17(c) deals with situations were there is no cover or less cover than the Lloyds policy due to the terms of the underlying policy and the First Addendum deals with a situation where there is no, or no effective, cover because the underlying insurer has refused or is unable to pay. Sub-clause 1.17(b) deals with a situation where there is more cover in the underlying policy than the Lloyds policy. (3) It would be surprising that an insurer would refuse cover for a failure to take reasonable precautions but waive such a restriction merely because the underlying insurer did not have such a term. (4) It is not Lloyd’s case that the reasonable precautions clause or clause 5.11 of the Lloyds policy apply.

ACE submits that sub-clause 1.17(c) is the only relevant sub-clause. One problem with the competing construction advanced by ACE is that the Second Addendum uses the phrase “this clause” when it should, if ACE is correct, say “this sub-clause”. This is answered I think, by the use in the proviso of the word ‘clause’ in “clauses 1.17(b) and 1.17(c)”. The second point is that sub-clause 1.17(c) uses “subject to its terms, Conditions and Exceptions” and the Second Addendum says “subject always to the terms Conditions and Exceptions of this Policy”, which is repetitive. I think the answer is that the Second Addendum was making it clear that the First Addendum was also subject to the terms, Conditions and Exceptions of the policy. I agree with ACE’s submission that the effect of sub-clause 1.17(b) is to expand excess cover where the underlying policy is more generous than Lloyds’, but that otherwise the Lloyds policy provides indemnity in its own terms when dropping down to replace missing cover or operating as an excess cover.

To interpret the First and Second Addendum as part of sub-clause 1.17(c) avoids the conflict between the Second Addendum and the last sentence of sub-clause 1.17(b) – since it provides that only the exceptions under Section 5 are to apply to that clause.

Sub-clause 1.17(b) applies in its terms when the local policy is more expansive that the Lloyds policy, and without any issue of insolvency or refusal of the local insurer to indemnify. Although I doubt it was intended, it is possible that both (b) and (c) apply in a particular situation – i.e. the local policy is wider in its cover but there is a refusal to indemnify. That, on Lloyds’ case, is the position here. Accepting that that is so, if both (b) and (c) are called into play, then the exceptions relevant to (c) must also be relevant since (c) is at least relied on in part. I do not think that the limitations in (c) can be ignored because (b) is relied on as well.

I should add that [0] raised a further problem. Lloyds asserts that the clause 1.17 applies based on sub-clause 1.17(b). To make good that proposition it needs to be asserted that the ACE policy provided more extensive cover than the Lloyds policy. The only basis on which it can be said that the ACE policy provides more extensive cover than the Lloyds policy is if the ACE policy has no reasonable precautions clause and the Lloyds policy does and it applies. It is not part of the Lloyds’ case that the reasonable precautions clause operates here (and it never has asserted that it did). Given that, I find it difficult to see how Lloyds could have been entitled to rely on sub-clause 1.17(b) to make the payment if the only basis that sub-clause 1.17(b) is enlivened is a basis that would preclude any payment being required.

In my view, the First and Second Addendums are to be read as part of sub-clause 1.17(c) and hence the exclusions in Section 1 of the Lloyds policy are applicable. ISSUE 17: If so, was that liability excluded under Section 1 by the professional indemnity exception in clause 5.11 or the reasonable precautions condition in condition 7.3? Clause 5.11

All that I have said concerning the PSE is applicable to the Lloyds clause 5.11. There was no submission that the Court’s determination as to the width of the PSE should be different in respect of the Lloyds clause 5.11. The PowerGrid claim is excluded by clause 5.11 and there is no equivalent to GC 14 in the Lloyds policy. Thus the PowerGrid claim is excluded. Reasonable Precautions

ACE accepts that “reasonable precautions” clauses, such as condition 7.3 of the Lloyds policy, are to be read in a restrictive way. A clause of that kind was the subject of consideration by the Victorian Court of Appeal in CGU Insurance Ltd v Lawless [2008] VSCA 38 (“ CGU v Lawless ”). Redlich JA at (with whom Maxwell P and Neave JA concurred) referred with approval to the test articulated by Diplock LJ in Fraser v B N Furman (Production) Ltd (1967) 3 All ER 57 at 61 (‘ Fraser ”) which was approved in Albion Insurance Company Ltd v Body Corporate Strata Plan No 4303 [1983] 2 VR 339 (“ Albion ”) at 342 – 343: “What, in my judgment, is reasonable as between the insured and the insurer, without being repugnant to the commercial purpose of the contract, is that the insured, where he does recognise a danger should not deliberately court it by taking measures which he himself knows are inadequate to avert it. In other words, it is not enough that the employer’s omission to take any particular precautions to avoid accidents should be negligent; it must be at least reckless, that is to say, made with actual recognition by the insured himself that a danger exists, and not caring whether or not it is averted. The purpose of the condition is to ensure that the insured will not, because he is covered against loss by the policy, refrain from taking precautions which he knows ought to be taken.” And to the test articulated by McGarvie J, with whom Young CJ and Anderson J agreed, in Albion at 344 – 345: “the condition will protect the insurer where there is a failure to take reasonable precautions which is due to ‘a deliberate decision to expose himself to the risk of bodily injury occurring, because he did not care whether it occurred or not.’… The insured is in breach of the condition if, through a lack of concern and desire to prevent bodily injury, he deliberately adopts a course of action or inaction which he realises exposes him to the risk of someone being injured by the recognised danger.”

Redlich JA also drew attention to the fact that McGarvie J had held that the insured may rely on evidence of the circumstances to show that whatever errors of judgment may have led to particular reasonable precautions not being taken, it was not due to a lack of desire and concern to prevent bodily injury, and to the fact that similar views are to be found in Legal and General Insurance Australia Ltd v Eather (1986) 6 NSWLR 390 (“ Eather ”), adopting that: “nothing was said in Eather which in my opinion requires any qualification of the test as stated in Fraser and adopted in Albion .”

Redlich JA then said (omitting citations): “[17] The test is wholly subjective. A failure to take reasonable precautions will occur only where there is a deliberate course of action or inaction which the insured realises exposes him to the risk of someone being injured by the danger which has been recognised. Hence the phrase ‘a deliberate decision to court the danger’. The insured might establish compliance with the condition by showing one or more of the following things: (1) There was no recognition of the danger or the extent of the danger of bodily injury; (2) Particular precautions would not have been reasonable in the circumstances; (3) No particular precaution was considered or it was not regarded as reasonable or practicable in the circumstances; (4) The failure to take the precautions was not due to a lack of desire and concern to prevent bodily injury.”

I should note that Fraser was accepted by both parties in CGU v Lawless as providing appropriate guidance. In this case both Lloyds and ACE accepted that CGU v Lawless authoritatively laid down the approach that should be taken, save that Mr Simpkins pointed out that [17] should be used as an example and not as a necessary checklist and Mr Donaldson also relied on the words of Brownie AJA in VACC Insurance Ltd v BP Australia Ltd [1999] NSWCA 427, where his Honour said at [58] that a reasonable precautions clause “must be construed having regard to the commercial purpose of the insurance contract: an insured may be held to be negligent, in not foreseeing some risk, or in not taking appropriate steps to avert that risk, but he insures in respect of his liability arising from that finding, and what is "reasonable" as between himself and his insurer is different to what is "reasonable" as between himself and someone injured in consequence of his negligence. To constitute a breach of the condition, he must have recognised the danger, and deliberately courted it, by taking measures he knew to be inadequate to avert it; or his conduct must have been "at least reckless, that is to say, made with actual recognition by [him] that a danger exists, and not caring whether or not it is averted" : [ Fraser v Furman at 61]; Albion Insurance Co Ltd v Body Corporate Strata Plan Number 4303 [1983] 2 VR 339 at 344-345; and [Eather] . The test is a subjective one, personal to the insured .” (emphasis added)

CGU v Lawless also makes reference to Verna Trading Pty Limited v New India Assurance Limited [1991] 1 VR 129, a decision of the Full Court of Victoria which held that the insured bears the onus of establishing that it did use reasonable precautions, a position which has been accepted by the NSW Court of Appeal in Eather , applying Kodak (Australasia) Pty Ltd v Retail Traders Mutual Indemnity Insurance Association (1942) 42 SR (NSW) 231; 59 WN 197.

I proceed upon the basis that if an insured establishes that it did not know of the relevant risk it cannot be said that it courted the risk. I accept too that if an insured, aware of the risk, subjectively believed that steps to be taken by it would be reasonable to exclude or minimise the risk that would satisfy the condition. In this sense then, the test is subjective, although there are limits to acceptance of evidence that the insured believed a particular fact or matter, which limits may equate to a requirement for the belief to be reasonable. Where no action is taken by an insured to deal with a known risk I do not think that the condition will be readily satisfied by an assertion by the insured that it thought there was little likelihood of the risk materialising: see for an example of this, a finding of failure to establish compliance with the clause in the area of motor vehicle theft: Devco Holder Ltd and Burrows & Paine Ltd v Legal and General Assurance Society Ltd [1993] 2 Lloyd’s Rep 567 at 570 – 572. There is an acceptance in the authorities that the precautions which are said to not be reasonable must be viewed in the context of the particular policy: see Eather at p 396B per Kirby P.

There was a further point raised by ACE which pointed to Mr Embery’s evidence that it was not he alone, on behalf of the JV, who decided to press on with the excavation of G4 and no other evidence was called on this. ACE submitted that Lloyds has therefore not established what the JV thought. I am inclined to the view that Mr Embery’s involvement with the project was so significant that evidence from him, alone, is capable of establishing that reasonable precautions were taken, but for reasons which will be become apparent, I do not think it is necessary to resolve that question and I proceed on the basis that it is not a defect in Lloyds’ proof that only Mr Embery was called.

The question then is whether Lloyds has established that the JV (through Mr Embery) was not reckless or, more pertinently in this context and based on Brownie AJA’s formulation set out above at [220], that the JV did not deliberately court the risk because (1) the JV was not aware that excavation through G4 entailed a risk of loss of groundwater and overburden with consequential risk to infrastructure in the vicinity of the works; or (2) the JV being aware of the risk, did not fail to take measures it knew were required to avert the risk; or (3) that such measures as were available and not taken were not appropriate or necessary to avert the risk. (See Mr Donaldson’s formulation at T1494.26 and Mr Simpkins’ formulation at T 1530.39 – 40 which are similar, but neither of which is couched in a form that places the onus on the JV.)

In relation to the TBM/G4 problem (and treating it as a discrete matter from the other three incidents) there is no disagreement that this arose because of the excavation of the tunnel by the TBM beneath Woodlands Avenue. The TBM was entombed on 2 June 2002 at chainage 1114M, and was recovered in January 2003. It did not recommence work until March 2003.

In dealing with the TBM/G4 problem, in his first affidavit, Mr Embery said that: (1) In order to minimise groundwater flows into the tunnel excavation, grouting was carried out to a point approximately 50 metres in front of the TBM launch position (at para 51) but that once the TBM was launched it was not possible to probe drill directly in front of the TBM as it could only be done at an angle (at para 52). (2) The TBM experienced difficulty when excavating though G4 which was encountered from about 15 April (at para 53). (3) Cutter motor clutches burnt out due to the attempt to bore though soft material (at para 55). (4) “Problems were also encountered from time to time with water and mud flowing from the face of the excavation through the TBM” (at para 56). (5) “The [JV] attempted to address the problem of the soft material through grouting but problems continued and progress was slow up until about 17 May 2002 when a substantial quantity of water and mud flowed through the TBM. Various measures were taken to seek to address the problems, including chemical grouting, microfine cement and the use of hydrophylic material but water, mud and soft material continued to hamper progress. Excavation continued on 24 May 2002 but was finally halted on 7 June 2002 after another substantial flow of mud and water through the TBM” (at para 57).

In his second affidavit Mr Embery explained why, in his opinion, continued excavation with the TBM in G4 was reasonable. These factors were: (1) The TBM was designed to facilitate probing and drilling ahead of the excavation face (at para 40) and could successfully negotiate such material (at para 41). (2) The available information indicated that the distance through which G4 was likely to be encountered was to be limited (at para 42). (3) Prior to the launch of the TBM from the Gambas shaft the JV prepared a CEP which was submitted to the Engineer for approval and Mr Embery said: “I am certain that the TBM would not have been launched without the approval of DPRL” (at para 43) and then said: “I recall the CEP provided for probe drilling in advance of the excavation and if probe drilling revealed any water ingress, then pre-grouting was to be carried out. In my opinion, that was a reasonable approach” (at para 44). (4) He reiterates that prior to the launch of the TBM probing and grout holes were completed up to 50 metres in front of the TBM launch position and grouting was carried out after to reduce possible water flow (at para 45). (5) Until the TBM encountered problems with probe drilling it was not necessary to probe ahead. Probe drilling was abandoned shortly after the launch of the TBM (at para 46). (6) The striking of G4 at chainage 1135 on 15 April 2002 was anticipated and “once we hit the G4, probe drilling commenced again” (at para 47). (7) Further bore holes were drilled from the surface and from the face of the TBM in an attempt to calculate the estimated length of the G4 material: “boreholes were drilled, and recharge wells were installed in the boreholes after they were drilled in accordance with Kiso-Jiban’s proposal.” (8) Mr Embery refers to what he has said in para 57 of his first affidavit (see above) and says that Mr Farkhard of the Engineer was on site every day and could see what the JV was doing. (9) He then said: “The [JV] did not ignore the risk of uncontrolled ground loss and settlement, nor did it proceed with the excavation through the G4 material knowing that ground settlement and associated damage was likely to occur. It believed, in my opinion, reasonably, that with the grouting measures being undertaken, the G4 could be traversed without serious incident. Unfortunately, and for reasons which remain unclear to me, it did not successfully negotiate the G4 material.” (at para 51)

So far as the suitability of the TBM is concerned, I accept Mr Sternath’s view that it was not suitable. It was designed for excavation in hard rock and not soft rock. It had no pressure bearing plate and could not deal with mud and water flows: see Exhibit 11 at paras 5 – 10. The TBM had detachable probes mounted from the rear which had a restricted range and which were in fact detached early on in the progress of the TBM. The TBM was, after the entombment and recovery, modified for its recommencement of tunnelling: see T959.46 – 50. Mr Cooper expressed a different view in his report, but his view was based on assumptions about the nature of the ground and location of the G4 material (i.e. that it was made up of relatively impermeable clays at that level, that there were only a few metres of G4 and that the TBM could support soft ground and propel forward in soft ground, which the JV had already found not to be the case) and he had not been involved in the selection of a TBM where it would encounter G4 material: see T935.40. Dr Pells agreed with Mr Sternath, and Mr Embery accepted, now at least, that the TBM employed was not suitable. Mr Cooper’s evidence at T959 – 960 is actually consistent with the conclusion that the TBM in use as at 2002 was not suitable.

In relation to the anticipated distance of G4 material, Mr Embery initially said he thought there were 10 metres of G4 material and later 20 metres: see T658.40. The only report identified indicated that G4 could be expected in a 100 metre section, it did not support the conclusion that only 10 or 20 metres of G4 material would be anticipated.

In relation to the CEP, Mr Embery’s affidavit referred to the CEP, but he admitted that it was not followed: see T649.25 – 48.

By reference to the involvement of the Engineer, Mr Embery sought to show that the Engineer had approved of what the JV was doing. There are two significant letters sent by the Engineer to the JV, the first on 7 May 2002 and the second on 2 June 2002. In the first letter DPRL expressed the view that the TBM was designed for excavation in rock and “[i]f the nature of the weathered granite in the face changed to something which was more sandy, that the joint venture might be unable to prevent wash-in of the material through and around the TBM and that that is something which could result in unacceptable ground loss or possible collapse”: see CB vol 3.12 at p 3-7608.

That letter was prescient as it was received shortly before extensive G4 was encountered in the tunnel: see T660.42. Mr Embery said he had himself made enquires and caused other enquires to be made about the TBM but there was no corroboration of that fact. In responding to the letter of 7 May 2002, the JV explained how it had come to select the TBM as the JV’s interpretation of available information at the time of tender was “that the TBM was capable of working in conditions where the G4 material was above the tunnel horizon and not through the G4 material which is the current situation.” See CB vol 3.12 at p 3-7655.

I think that this letter indicates a knowledge of the unsuitability of the TBM at the time of excavation and Mr Embery’s assertion that he had obtained information that indicated its suitability for G4 excavation was a failed attempt to overcome that inference.

Mr Embery asserted at T655.25 – 39 that the enhanced risk of groundwater loss was not well understood at the time: see also T657.20 – 35, but see T660.38 – 661.9 and T673 – 674, and the 7 May 2002 letter from the Engineer, itself, which made clear what the consequences of excavation in G4 were.

The second letter, dated 2 June 2002, referred to “our numerous correspondence raising our concern over the prevailing situation” and the JV’s “inability to prevent unacceptable ground settlements”: see CB vol 3.13 at p 3-8050, and required immediate cessation of all work on the east tunnel until submissions for future work had been provided to the Engineer. This was relied on by Mr Embery to support his contention (see T606.43 and T703.28) that the Engineer permitted excavation of the G4 material without investigation. The Engineer’s letter of 2 June 2002 does not support Mr Embery’s contention although he refused to admit that it did not: T703.30 – 34.

I should note that the minutes of a meeting between the JV, PowerGrid and the Engineer’s representative held on 10 May 2002 contained the following item: “10. In response to [the Engineer’s] query as to whether the TBM operation and/or blasting should be stopped while the steps to mitigate the settlement problem are being implemented, [the JV] stated that TBM operations should continue so that it could quickly get through the potentially dangerous G4 material. Blasting work should also continue so that the waterproofing and permanent lining could be quickly installed to prevent further water drawdown. 11. Kiso-Jiban added that treatment of the peaty clay layer by grouting was not recommended due to the unpredictable effects of such a measure. 12. [the JV]/Kiso-Jiban stated that it could not stop the ongoing settlements and that the most effective action to prevent further damage to the utilities was to expose and support them at the critical locations.” See CB vol 3.12 at p 3-7649.

It is clear that, by receipt of the Engineer’s letter of 7 May 2002, Mr Embery was alive to the unsuitability of the TBM and the significant risk of wash-in of material through and around the TBM with the prospect of “unacceptable ground loss or possible collapse”, and that the G4 material was potentially dangerous. Mr Embery agreed that the Engineer was concerned about G4 before G4 was encountered in the tunnel: see T600 – 601.

I set out an important segment of cross-examination of Mr Embery at T655.16 – 658.9: “Q. Mr Embery, you accept, don't you, that tunnelling through G4 material can create difficulties that need to be attended to which are different from those encountered when one is boring through hard rock? A. Yes. Q. There are enhanced risks of groundwater loss, aren't there? A. That was not our assumption at the time. Q. You accept, don't you, there are enhanced risk of groundwater loss, don't you? A. Based on historical information, yes. HIS HONOUR Q. Did you mean by that that you now know that or you that you knew that at the time? A. I now know that from both our project and others in the area. Q. What about at the time, didn't you know that then? A. No, it was not well understood at the time. Q. By you? A. By us by the project, if you want. SIMPKINS Q. And there were enhanced risks of subsidence also, aren't there? A. Through excavating G4, no. Q. Well, can I put this to you: if you attempt to excavate G4 using a hard rock TBM, there are enhanced risks of subsidence, aren't there? A. Based on hindsight, yes. Q. That's also something, is it, that you did not appreciate at the time? A. It's not something that we believed was an issue at the time. Q. I want to focus on your state of mind, not something that you appreciated at the time. A. No, no. It was a collective decision on what the team understood at the time. Q. Well, is it correct to understand, therefore, that before the entombment of the TBM, not appreciating those matters that I just asked you about, you were largely unconcerned about the prospect that G4 material would be encountered by this hard rock TBM? A. No, we were not concerned. Q. And is it correct to understand that you did not have any concern at any time prior to the entombment of the TBM? A. In the period immediately prior to it, yes, we had some very serious concerns. Q. Now, the TBM was entombed, is this right, on about 7 June 2002? A. Yes, yes. From memory, yes. Q. When you refer to the period immediately prior to that, what period are you identifying? How long before that date? A. It was approximately six to eight weeks prior when we encountered G4 until we got to the point of overexcavation. Q. So, correct me if I may be wrong, but the concern that you developed arose when the TBM had actually encountered the G4 in the tunnel, not before? A. No, not before. Q. And your concern arose out of what you could observe about the difficulty that was being encountered in using this hard rock TBM to tunnel through G4 material; is that right? A. Yes. Q. Now, is it right that one of the matters that you considered in 2001 was the suitability of the TBM for excavating the G4 material? A. Yes. Q. Can you explain, please, why it was that that was a matter that occupied your consideration in 2001? A. As the additional geotechnical data started to arrive, we then started to analyse what extent of G4 we might have and what that material would behave like. Q. That was something that you were a participant in, but I gather from your answer not the sole participant. So there were others, were there, engaged in considering this matter? A. The project team was considering that as an issue, and generally, yes. Q. The view that you had, as I understand it, in 2001 was you expected about 10 metres of G4 material? A. Yes. Q. Is that a view that you adhered to throughout 2001? A. Not a significant departure from that, no. Q. Do you mean by that that there was some departure from the assessment of 10 metres? A. As we gained more data, the extent of the G4 became a little more obvious. But the information was still very unreliable and so it was very much a guesstimate based on the data. Q. Was the information unreliable because you didn't have enough borehole results? A. Yeah. Well, yes. In hindsight, we needed boreholes at 2 metre centres, which was quite clearly not the normal way of obtaining information. Q. If there had been boreholes at 2 metre centres, it's your opinion, is it, that the true extent of the G4 material and its location would have been properly understood before the TBM encountered it? A. Yes. Q. And that if the risks of groundwater loss or subsidence had been appreciated, the tips [steps semble ] could have been taken before hitting the G4 to progress those risks; is that right? A. If the extent had been understood at that time, yes. Q. And may I take it that if you had had knowledge of the true extent of the G4 material and properly understood the risks, there were things that you would have caused to be done that in fact weren't done before the entombment of the TBM; is that right? A. On hindsight, yes. Q. What are the things you would have done? A. In hindsight, probably would have not advanced the machine into that zone, or and/or required a substantial amount of grouting, pregrouting. Q. When you refer as an alternative to the prospect of the machine not being advanced, are you contemplating the possibility that there might have been a redesign which could have been undertaken and avoided, or minimising the length of the G4 material which was to be dealt with? A. At the time, a redesign was not an option available to us. It became an option afterwards. Q. So that I understand your answer A. A redesign at the time was not an option. Q. I understand that answer, but I want to go back to the one before. You said at least one possibility was that the TBM wouldn't be advanced. Now, I'm trying to follow up from that answer to understand what you contemplated would happen. I'm assuming you didn't think the project would stop and never continue under any form. What do you think the consequence of not advancing would be? A. The TBM was a critical part of the project in terms of programme on the project. The options would have included to tunnel through the zone and divert the machine through at a later date, which is effectively what was done to rescue the machine.”

Mr Embery said that had he known how much of the tunnel had G4 there he would not have continued to excavate G4 material with the TBM without some other action being taken: see T678.45 – 50. Realignment of the tunnel was an option adopted in relation to the other tunnel (the west tunnel): see T679.30, and it had even earlier been considered in respect of the east tunnel (in 2001) because of the knowledge that G4 material was going to be encountered: see T679.30. Mr Embery also said grouting from the surface was an option to make the G4 material less permeable and more stable: see T680, and could have included a grout and test to determine permeability: see T680.

Mr Embery accepted that: (1) he was aware that G4, if infiltrated with water, would be unstable and potentially dangerous with a risk of ground loss: see T673; (2) no investigations were carried out to determine what water or water flow existed in the G4 material: see T674.18; (3) the JV decided to excavate the G4: see T674; (4) water in the G4 had become evident by 10 May 2002: see T675.19 – 25; (5) the JV had the option of adding grout or otherwise attending to the G4 material at the tunnel face in order to ensure there was no ground loss: see T675.40 (although he says he only appreciated this with the benefit of hindsight); (6) he rejected the idea that further investigations could be conducted to establish the presence and quantity of water: T675.50 but then at T676.5 agreed it certainly could have been done; and (7) the Engineer had in 2001 and 2002, regularly criticised the JV for failing to carry out systematic grouting: see T698.21. Mr Embery initially asserted that there was systematic grouting but then it was put to him that it had not been carried out because his view was that it would take months and the program was too tight, and he then agreed with that: see T698.21.

At T675.5 – 35 there was the following series of questions and answers: “Q. Let me get this right. Your view about potential danger was that, in certain circumstances, excavation of the G4 material could cause ground loss? Have I got it right so far? A. In certain circumstances, yes. Q. The certain circumstances you envisaged were those in which there was a sufficient quantity of water or water flow in the potentially dangerous G4 material; correct? A. Yes. Q. You proceeded, after 10 May 2002, to excavate the potentially dangerous G4 material without any specific investigation as to whether the water or water flows within that created a danger? A. Yes. But that's based on the fact that there was no water flow when the G4 was first struck. Q. The G4 was first struck some while before 10 May 2002, wasn't it? A. That's right. Q. And, well prior to 10 May 2002, water had been encountered in the G4 material, hadn't it? A. In the period from when it was first struck, 10 May, yes, there was water becoming evident. Q. Is this right, that although water was becoming evident, no specific investigation was carried out at any time after 10 May 2002 and before the entombment of the TBM to determine whether that water was present in the G4 material in sufficient quantities to make it potentially dangerous to excavate? A. No, there wasn't. And the key reason for that being that the longer you delayed, the more the risk increased.”

There are several points to be made about this explanation: (1) Implicit in it is a recognition of risk – that is, the situation was so precarious that the JV just had to keep going. That is not consistent with anything said by Mr Embery in his affidavits. (2) Mr Sternath, whilst accepting that in some circumstances such an approach might be justified, expressed the view that what Mr Embery had said was “totally wrong”: see T1120.40. He explained why that was so: “A. What he said is totally wrong. Q. Why? A. Because this G4 this is special term for this to describe this type of soil is unstable, particularly under some pressure of up to 50 metre of water table. This groundwater would make the soil liquid and unstable. And as the joint venture used a poor hard rock TBM which could not support the face and the circumstances, this assertion is wrong because this TBM entered deliberately into stable [unstable semble ] conditions where ground loss was necessary consequence.” See T1120.40 – 1120.49. (3) Mr Sternath was cross-examined at T1146.28: “Q. But where someone carrying out tunnelling works finds themselves in a position where they are in unstable ground A. Yes. Q. They frequently have to make a judgment A. Yes. Q. About whether it's best to stop or keep moving; correct? A. Yes, correct. Q. And that there are times when it's right to try and get the tunnel advanced before the poor material deteriorates further? A. Correct, yes. Q. It's a matter of judgment? A. Yes, but under certain limits. HIS HONOUR Q. What are those limits? A. The limits are that that you are mining in a reasonable stable ground. What you could expect here is loose soil with 4 to 5 metre water pressure, except you must the not enter at all. You have to perform probe drills in advance. Here, it was considered having 40 metre long probe drills, you set probe drill, you would have detected this and you could have stopped earlier in safe conditions. If you are in this unsafe conditions, then you have to observe the quantity of material which you are excavating from the tunnel. And when this is exceeding theoretical volume, then you have to stop this as this limit is exceeded. And as far as it happened here, on a very short distance, they have over excavated 627 cubic metre. In such case, you cannot continue.” See T1146.40 – T1147.8. (4) If the position was so precarious at 1 June 2002, the letter of the Engineer of 7 May 2002, which warned of the danger four weeks before, had been ignored. The response of the JV was noted in minutes of 10 May 2002 between the JV and the Engineer: see above at [236]. (5) The minutes referred to at [236] do not support the contention that the JV believed that the tunnelling work had to continue because there was a risk of collapse if it did not proceed as at that date. (6) In any event the presence of G4 material was identified well before 10 May and certainly by April.

I have referred to the contents of para 57 of Mr Embery’s first affidavit. Mr Embery did not refer in chief to CB vol 3.22 at 3-15022 – 3-15027 but in re-examination Mr Donaldson sought to ask him detailed questions about those pages and objection was taken by Mr Simpkins to that course: see T739 – 740. Those pages are described as “Detailed Report Outlining Sequence of Events between 15 April 2002 [the date G4 was discovered] and 7 June 2002 [the date the TBM stopped]”. The document records for 4 May the Engineer’s request for probing. There are references to probing through the manhole on 4 and 9 May, and on 17 May (without mention of the manhole). The manhole was closed on 29 May, it would seem, as a result of the flow of ‘muck’ on the previous day. On 13 May there was water leaking at a rate of 50 litres per minute. The records show that there was grouting close to the face on a number of days, but the point is that there was no proper preparation of the ground in advance of the TBM. No reference was made to any documents containing details of what the probes revealed.

Mr Embery referred in para 57 of his first affidavit to a number of steps which the JV took to deal with the problem of the G4 and I have set out that paragraph above at [0]. To the extent that chemical grouting and hydrophilic material was used to attempt to stabilise the flows coming through or close to the face of the TBM, they appear to be dealing with only the immediate and pressing problem, rather than being a systematic grouting program, as Mr Embery’s answer at T675.40 confirms. The point is that there was no proper preparation of the ground in advance of the TBM. Mr Sternath explained that: “all this grouting facilities you have from within the TBM are very limited. On the surface you can - you can use state of the art equipment to do it properly, you can check your grouted body, whether it is really fulfilling the requirements. Also it is very difficult to do it from a TBM.” See T1141.20 – 24.

Mr Cooper’s evidence at T960.14 – 40 was: “Q. If contrary to that assumption what was visible was a very significant portion of the face of the G4, would that send alarm bells? A. Yes, because going back to the geotechnical interpretations, the view that was expressed to the joint venture was the two metres of G4 material immediately above the interface was of a high clay content and less permeable. If the bulk at the face was G4 material, then they would be into the highly permeable material and then you have got an issue of ground flowing which is, in a tunnelling environment, is something you avoid. Q. And would that link to, if there was an oozing of soil or mud, would that be indicative of a higher water content? A. Yes, but oozing in itself is not the problem. It is when it starts to run. You can control oozing but once the ground is running and bringing a lot of material with it, it is very hard to control, particularly if you don't have doors and you have got to be very careful about grouting in that environment as well because you finish up grouting your whole tunnel boring machine in so Q. It has got no room to move? A. No, no. So grouting once the ground starts to move, it is very difficult to grout. Q. If the ground starts to move, is that a time when prudence would dictate that you stop? A. Yes. Q. And work out A. What you are doing [going semble ] to do. Q. a more extensive programme? A. Yes, yes.”

Mr Embery’s evidence overall threw up three explanations for why the JV continued with the G4 excavation: (1) There was no cause for concern because the TBM was appropriate; (2) There was no cause for concern because adequate steps were being taken to consolidate the soil; and (3) There was cause for concern, but the only means of dealing with it was to press on.

It was submitted by Mr Donaldson (see T1494.5 – 1496.4) that the cross-examination of Mr Embery was insufficient to found a submission that he or the JV had not taken reasonable precautions – that is, that he knew that the steps he was taking were inadequate. Mr Donaldson pointed to paragraphs including para 57 of Mr Embery’s first affidavit and to Mr Embery’s evidence in re-examination: see T537.1 – 29, T738. The onus rests on the JV to establish that it did not proceed with the excavation without taking adequate steps to avert the risk, and the cross-examination was directed to establishing awareness of the risks and a failure to take steps to avert the risks. It was put to Mr Embery that he was reckless in proceeding with the TBM excavation through G4, which he denied. I accept that that deals with only one part of the Fraser formulation but given the detailed explanations given by Mr Embery I do not think that it can be said that he was not given a fair opportunity to defend his actions.

I have set out at [225] – [246] above, the factual matters relating to Mr Embery’s decision to proceed with the excavation of G4. In my view he was on notice of the risk of over-excavation and aware that there were steps that could be taken to deal with the problem, i.e.: (1) redirecting the route of the east tunnel (just as was done with the west tunnel); (2) delaying excavation until a more suitable TBM was purchased or hired; (3) ascertaining the extent of the G4 material, the level of water and the degree of permeability before continuing and determining how ground loss could be avoided or minimised; and (4) halting the excavations to permit grouting work to stabilise the G4 material.

Returning to the matters set out in [227] above as the factors which Mr Embery asserted made the excavation reasonable: (1) The TBM was not designed to work in the G4 material (as the JV itself recognised prior to the collapse). (2) The available information did not indicate that the distance of G4 was likely to be as limited as Mr Embery asserted. (3) The Engineer had warned of its concerns and the CEP which provided for probe drilling in advance of excavation was not followed: see Mr Embery’s evidence at T649.25 – 48. (4) The bore holes that were dug did not show a limited distance of G4 – what they did indicate was the absence of solid rock: see T670.32 – 35. (5) Mr Embery’s explanation as set out at [227(6)] is quite inconsistent with his assertion that he was relying on the need to get through dangerous material quickly (and see the minutes referred to at [236]). His assertion that even as at the date of his second affidavit he did not know why the over-excavation had occurred is difficult to accept.

This leaves Mr Embery’s assertion that he had to press on to avoid collapse. I have already dealt with that earlier. The fact that Mr Sternath, Dr Pells and Mr Cooper (to a degree) agreed that it was not appropriate to press on does not establish that Mr Embery did not genuinely believe that that was the best course although that evidence is relevant to the question of his belief. I have referred to the difficulties with Mr Embery’s credibility and this, coupled with his failure to assert in either of his two affidavits that that was the reason for pressing on and his propounding of reasons that were not consistent with it, quite apart from the fact that he had already been warned of the danger before any crisis of that kind arose, and the evidence of the experts to which I have just referred – particularly that of Mr Sternath – leave me unpersuaded that he believed that there would be a collapse if he did not press on with the tunnelling.

I find that Mr Embery was aware of the risk of collapse with consequent settlement and was aware of the steps available to remove or reduce the risk, which he failed to take, and I am not persuaded that his failure to take appropriate steps was reasonable or arose out of a genuine belief that they were not needed or could not be instituted. Hence my conclusion is that the JV breached the requirements of condition 7.3 of the Lloyds policy in relation to the TBM/G4 problem. Lloyds did not contend that it could rely on having acted reasonably in paying the claim of the JV, even if the reasonable precautions exclusion was available or potentially available to them, it being accepted that if Lloyds were not liable under Section 1 of their policy they would have been liable under Section 2: see T1212 – 1213. ISSUE 18: Does the rule in Wayne Tank operate to exclude any liability of the JV where the cause was the result of both professional and non-professional services?

I have already dealt with this in the context of the ACE policy in Issue 10. ISSUE 19: Did sub-clause 1.17(b) of the Lloyds policy operate so that cover afforded under the Lloyds policy was the same as that provided under the ACE policy? If so, what was the extent of this cover having regard to proper construction of the PSE and GC 14 of the ACE policy?

I have dealt sub-clause 1.17(b) of the Lloyds policy in Issue 16 and I have dealt with the construction of the PSE and GC 14 of the ACE policy in Issues 5, 8 and 9. It is not apparent that there is any additional element in this issue that had not already been addressed above. ISSUE 20: Was Lloyds liable to indemnify the JV by reason of the ‘drop down cover’ provision in sub-clause 1.17(c) even if an exclusion in the Lloyds policy otherwise applied?

I have dealt with this in Issue 16. ISSUE 21: How many ‘occurrences’ were there for purposes of Lloyds policy?

I think it was accepted by the parties that the same approach to ‘occurrences’ in the ACE policy would apply to the Lloyds policy. ISSUE 22: If not liable under Section 1, were Lloyds liable under Section 2?

It will be recalled that in relation to the ACE policy Lloyds argued that the ACE PSE was overcome by GC 14. In relation to clause 5.11 of the Lloyds policy (“ the Lloyds PSE ”) it was contended that if ACE was liable then Lloyds was liable by virtue of sub-clause 1.17(b). Lloyds also argued that if the Lloyds PSE applied then the claim fell within Section 2 of the policy: see T1212.39 – T1214.19. The Section 2 claim was clearly, in this context at least, an alternative claim relied on should the Lloyds PSE apply and the ACE PSE, although applicable in its terms, was overcome by GC 14.

The Plaintiff’s Reply which relied on Section 2 was not, however, limited to the Lloyds PSE, but included a reference to clause 7.3 of the Lloyds policy (i.e. the reasonable precautions condition). It pleaded that if clause 7.3 was held to operate then Section 2 cover was available. Clause 7.3 is expressed to be applicable only to Section 1.

Little was said about the Section 2 claim and what was said seemed to minimise its significance: see for example T158, T1212 - T1213.

ACE mounted a number of arguments concerning Section 2. First there are those set out at T1380 which I shall summarise as: (1) There was no explanation by Lloyds as to why there was Section 2 cover. (2) Was there a demand for compensation made against the insured? (3) Was the claim notified to Lloyds? (4) Was the payment made by Lloyds made for a claimed loss under clause 1.1 of Section 2 or for the insured’s costs for mitigation loss not yet the subject of a claim – clause 3.8 of Section 2? (5) If payment was made under clause 3.8 of Section 2: (a) would the loss otherwise have been the subject of an indemnifiable claim for compensation; and (b) did the insured obtain the prior approval of Lloyds for the mitigation work? (6) If part of what was paid was payment under clause 3.8, that was not cover which ACE agreed to give under its policy so no contribution is available. Some further arguments were advanced subsequently (see reasons for judgment of 30 September 2009) and they are: (1) Sub-clauses 1.17(b) and (c) do not operate in relation to Section 2 of the policy: (a) As a matter of construction clause 1.17 (the Difference in Conditions clause) was not intended to give drop down cover for Section 2. It is asserted that the second paragraph of sub-clause 1.17(a) is clearly intended to relate to Section 2 cover because of the use of the words “Claim or Claims” with a capital “C” and that the paragraph is otiose if it did not have this purpose and that conclusion in relation to sub-clause 1.17(a) means that only sub-clauses 1.17(b) and 1.17(c) apply to Section 1 of the Lloyds policy. ACE asserts that sub-clauses 1.17(b) and (c) which use the phrase “legal liability costs and expenses” are directed to Section 1 cover and not Section 2 cover. (b) That the correspondence passing between Lloyds’ agent and the JV’s agent establishes that it was not intended that Section 2 cover be the subject of the drop down provisions. The correspondence relied on is identified and I have had it marked ‘Exhibit S’. Exhibit S was by agreement admitted subject to relevance. (c) That the statement of deductibles in the Lloyds policy (Exhibit S vol 2 at p 1.6) points to Section 1 as the only section having deductibles for clause 1.17 because the only reference to clause 1.17 of the policy is under the heading ‘Section 1’. (2) Lloyds has not proven that Section 2 responds (this is really the same as [259(1)]).

Section 2 provides cover for any claim made against the JV during the period of insurance “in respect of any civil liability incurred by or on behalf of” the JV in the conduct of its profession. Profession is defined widely and “Claim” is defined as: “2.1 Claim means any of the following: (a) a writ, statement of claim, summons, application or other originating legal or arbitral process, cross-claim, counter claim or third or similar party notice served on the Insured; or (b) an oral or written demand for compensation made by a third party against the Insured .” (emphasis added)

In dealing with [259](1), (4), (5) and (6) I observe that the First and Second Addendum of sub-clause 1.17(c) imposed a liability on Lloyds to indemnify the JV if ACE failed to admit liability (or could not and would not pay) but “subject always to the terms, Conditions and Exceptions of this Policy”. The proviso thus imposed two additional requirements: (1) the liability of the JV had to fall within the terms of the Lloyds policy; (2) the liability of the JV had to not be excluded by virtue of a condition or exception to the Lloyds policy. There was no disagreement that “Exceptions” include exclusions.

These requirements, I think, carry with them a need to determine whether the claim falls within Section 1 or Section 2 of the policy so that it can be determined whether the liability in question falls within one or other section and so that any relevant conditions and exceptions can be identified. I accept Lloyds’ submission that the exceptions and conditions which apply to claims that would fall under Section 1 (if they were over $15 million) are the exceptions and conditions expressed to apply to all claims under the policy and those specific to claims under Section 1, and that the conditions and exceptions which apply to Section 2 are those expressed to apply to all claims under the policy and those specific to Section 2 claims. It follows that if a claim would fall within Section 2 (were it for an amount in excess of $15 million) the reasonable precautions condition 7.3 is irrelevant to that claim and is also irrelevant to a claim paid under sub-clause 1.17(c) when the claim is for an amount equal to or less than $15 million.

If a claim falls within coverage under Section 2 of the Lloyds policy and is not excluded by any condition or exception I do not think it is necessary to determine whether it would be payable under one provision of Section 2 or another. In any event given the fact that claims were made on the JV by third parties and given that I have found that there was a liability in the JV, I am of the view that the JV had no need to resort to clause 3.8 to obtain payment.

In relation to [259(7)(b)], the discussion of principle in relation to pre-contractual correspondence: see [137] – [150] above, applies equally here. Even if the material were admissible it sheds no light on the meaning of clause 1.17 and could not do so because the correspondence is not directed to whether there should be drop down cover in relation to the ACE policy but rather to another quite different policy.

In relation to [259(7)(a)] I do not think that it is at all clear that sub-clause 1.17(a) is intended to deal only with Section 2, but in any event although I accept that the use of the phrase “Claim or Claims” in sub-clause 1.17(a) stands in contrast to “legal liability in respect of damage to property or personal injury to third party or employees” in sub-clause 1.17(b) and “such legal liability, costs and expenses” in sub-clause 1.17(c): (1) sub-clause 1.17(a) uses the word “claim” without a capital ‘C’ in the first paragraph; (2) when it refers to “Claim or Claims” it does so in relation to “Claim or Claims” which the underlying insurer is required to pay; (3) sub-clause 1.17(b) specifically limits itself to Section 1 which neither (a) or (c) do; (4) clause 1.17 is found in a section headed “Section 4 – General Conditions Applicable to All Sections” and there is no obvious reason why the drop down cover should only be available for Section 1 claims and not Section 2 claims – the express exclusion in sub-clause 1.17(b) reinforces the view that sub-clauses 1.17(a) and (c) were not so limited; (5) even if sub-clause 1.17(a) was read as excluding drop down claims under Section 2, it does not follow that sub-clause 1.17(c) should be read in that way unless the use of the phrase “legal liability, costs and expenses” were the key – if that were so there would have been no need to expressly refer to Section 1 in sub-clause 1.17(b).

In relation to [259(7)(c)] the ‘Drop-Down Deductibles’ section (clause B.2) refers expressly to both Section 1 and 2 and refers to a deductible of USD30,000 for “each Claim or Occurrence”. The word “Occurrence” is applicable to a Section 1 claim just as “Claim” is applicable to a Section 2 claim. Clause B.2 dealing with “Each and Every Deductible” applies generally and Clause B.3 “Drop-Down Deductibles” applies when the Aggregate Deductible is exhausted. Why clause B.2 should refer to Section 1 and in that context refer to Difference in Conditions and use the word “Claim” and clause B.3 not refer to Difference in Conditions and use “Claim or Occurrence” is not clear but it is at most ambiguous and I do not think it supports ACE’s contention that Section 2 cover was not available for clause 1.17, particularly having regard to the express limitation of sub-clause 1.17(b) not repeated in sub-clauses 1.17(a) and (c).

In relation to [259(2)] it was agreed that there was no legal process issued by the third parties whose property was damaged but there was no dispute that PowerGrid and third parties had made demands on the JV, and these demands were in my view, by their nature, claims for compensation and I have held that there was legal liability to these third parties (including PowerGrid): see [3], [113] and [125] above. Details of the correspondence relevant to each of the third party claims are found in Appendix 2 to the plaintiff’s written submissions of 9 March 2009.

In relation to [259(3)] there is no doubt that the claims of the third parties were notified to Lloyds. The claims were notified at different times and it does not matter in which year they were notified because the Lloyds policies were in place for each of the three years.

It follows that for claims in excess of $15 million arising out of liability incurred by the JV as a result of the excavation works by reason of acts or omissions of the JV in the conduct of its business as a professional as defined, Lloyds was obligated to indemnify the JV under Section 2, and that for claims of less than $15 million arising out of liability incurred by the JV as a result of the excavation works by reason of acts or omissions of the JV in the conduct of its business as a professional as defined, and for which ACE was liable but would not admit liability, Lloyds was obligated to indemnify the JV under the clause 1.17.

It is not uncommon for insurance policies with more than one section to address the question of whether a claim that falls under one section can be maintained under another section, particularly if an exclusion in one section is operative but there is no such clause in this policy. The PSE in Section 1 is, in effect, dealing with overlap because it is excluding from Section 1 cover for professional services claims, and clause 4.8 in Section 2 of the Lloyds policy, which is the only other clause which deals with overlap, and which is somewhat oddly worded, was not relied on by ACE. It is clear, and Lloyds accepts, that for Lloyds to be able to recover under clause 1.17 the liability had to be such as would fall within the Lloyds policy as well as the ACE policy. Clause 1.17, however, does not in terms require that the liability under the underlying policy be of the same kind as the liability under the Lloyds policy for clause 1.17 to operate, and I do not think it can be implied that the basis liability of Lloyds under its policy had to be identical to that of ACE under its policy for clause 1.17 the operate.

It follows that Lloyds have established that Section 2 cover is available for the claims made by the JV. ISSUE 23: Recoupment, indemnity or contribution?

I have concluded that ACE was required to indemnify the JV under the ACE policy and that Lloyds was required to indemnify the JV under the Lloyds policy for at least some portion of the liability incurred by the JV.

It is a principle of insurance known as contribution that when two insurers cover the same risk and the first insurer indemnifies the insured pursuant to the terms of the first insurance policy, the law recognises a right in the first insurer to recover a proportionate share of the amount paid out from the second insurer who has by reason of the payment made by the first insurer been relieved of the burden that would otherwise have been imposed on it to pay the insured’s claim.

The payment by the first insurer “is made for the benefit of both, and, contribution is equity”: Albion Insurance Co Ltd v Government Insurance Office (NSW) (1969) 121 CLR 342 at 346 (“ Albion v Government Insurance Office ”) per Barwick CJ (with whom McTiernan and Menzies JJ agreed). In Albion v Government Insurance Office Kitto J at 349 – 350 (with whom Windeyer J agreed) said that contribution had at is base two principles: (1) that persons who are under coordinate liabilities to make good the one loss must share the burden pro rata; (2) an insured although insured under more than one policy and able to seek indemnity from any one of the insurers may not recover more, in total, than a single reparation for the loss so that the position between insurers is analogous to that between sureties.

Priestley JA in Australian Eagle Insurance Co Ltd v Mutual Acceptance (Insurance) Pty Ltd [1983] 3 NSWLR 59 at 64, on a right of contribution on the ground of double insurance, said that he thought it useful to ask: “(a) did the two insurers have a common burden; and (b) if the insured were to be paid under both policies would he be paid twice is respect of the same damage.”

Contribution between insurers is thus an example of the operation of wider principles: see Albion v Government Insurance Office , and is an important aspect of the law of sureties: see Meagher, Gummow and Lehane, Equity: Doctrines and Remedies, 4th ed (1992) Butterworths at 10-005 – 10-015 and O’Donovan and Phillips, The Modern Contract of Guarantee , 3rd ed (1996) LBC Information Services at 12.1100 – 12.1200.

When a surety pays out a creditor pursuant to a guarantee the surety is entitled to pursue other guarantors who have not paid their proportionate share: see Dering v Earl of Winchelsea (1787) 1 Cox 318; 29 ER 1184. The surety is also entitled to pursue the debtor whose debt he guaranteed: see Israel v Foreshore Properties Ltd (in liq) (1980) 54 ALJR 421 at 424. There is a significant difference in the two recoveries. When a surety claims from co-surety on the same guarantee the surety cannot recover all of what he has paid but must retain a proportionate share. If the debtor does have the ability to pay, as against him the surety is entitled to a complete indemnity: see McColl’s Wholesale Pty Ltd v State Bank of NSW [1984] 3 NSWLR 365 at p 376 per Powell J (as he then was) for a convenient summary of the law.

There are other situations where the party who has paid a debt to a third party is entitled to recover all of what he has paid from the party primarily liable. For example, in Brook’s Wharf and Bull Wharf Ltd v Goodman Bros [1937] 1 KB 534 at 545 (“ Brook’s Wharf ”) a warehouseman who had paid customs duty on goods imported because it was compelled by law so to do, was held entitled to recover the money paid from the importer. Another example is where a tenant (“A”) who had assigned a lease to ‘X’ who had then assigned the lease to ‘Y’ was held entitled to recover from ‘Y’ the money that ‘A’ was compelled to pay the landlord: see Exall v Partridge (1799) 101 ER 1405 and England v Marsden (1866) LR 1 CP 529.

Lloyds claims that it is entitled to recover from ACE all of the money it has paid to the JV because ACE was liable to indemnify the JV and did not do so.

Lloyds, although an insurer, put its case primarily as one of recoupment because it paid the JV when ACE was required to pay and had not done so.

The position here is complicated by the following factors: (1) Lloyds were excess insurers to the extent that ACE indemnified the JV. (2) Lloyds also agreed to cover the JV if the JV could not obtain an admission of liability (and on my finding it could not obtain such an admission). (3) When the liability to a third party arose Lloyds had no liability to indemnify the JV save, contingently, to the extent that ACE was not liable to indemnify the JV. (4) When ACE refused to admit liability to indemnify the JV Lloyds became liable to indemnify the JV for the same risk. (5) When Lloyds indemnified the JV it removed a burden from ACE to indemnify the JV, pursuant to an obligation that it had only because ACE would not meet its obligation to the JV.

In essence Lloyds’ argument is: (1) ACE has been relieved of a burden to indemnify the JV by the payment of Lloyds. (2) It would be just and equitable to require ACE to reimburse Lloyds all of the amount paid by Lloyds because Lloyds’ liability to pay the JV only arose because ACE did not meet its obligations under its policy. (3) This is similar to the approach taken in contribution cases but with the added aspect that Lloyds was supposed to be an excess insurer and not a primary insurer unless ACE was not liable (not relevant to this aspect) or if ACE refused to admit liability and its failure to do so was a breach of ACE’s obligations under the ACE policy. (4) If it is just and reasonable to require ACE to pay to Lloyds the money that Lloyds paid to the JV it does not matter if that is expressed as a claim for contribution with a level of contribution equal to 100% or recoupment, the entitlement to such a payment by ACE is the same. Whilst contribution often involves equal contribution, it does not always require equal contribution eg: Government Insurance Office (NSW) v Crowley [1975] 2 NSWLR 78 at 82 – 83, GRE Insurance Ltd v QBE Insurance Ltd [1985] VR 83 at 103 – 104 and Drayton v Martin (1996) 67 FCR 1 at 38.

ACE resists Lloyds’ claim. It claims that Lloyds’ case does not meet the requirements for contribution nor the requirements for recoupment for reasons I shall outline.

ACE submits that if it became liable to indemnify the JV when third parties suffered damage to their property (the casualty) then at that point in time ACE and Lloyds were not co-insurers of the JV because Lloyds only became a co-insurer when the JV could not obtain an admission of liability, at the latest in 2005. ACE, as I have outlined above, does not accept that it has ever become liable to indemnify the JV, but this present argument proceeds on the basis that it is held that ACE did become liable to indemnify the JV.

ACE pointed to the decision in AMP Workers Compensation Services (NSW) Ltd v QBE Insurance Ltd [2001] NSWCA 267; (2001) 53 NSWLR 35 (“ AMP v QBE ”), which is authority for the proposition that the relevant time to assess whether there is joint insurance is at the time of the casualty: see [17], and has been followed in Workcover Queensland v Suncorp Metway Insurance Ltd [2005] QCA 155 at [52] per Jerrard JA (with whom McMurdo P and Douglas JA agreed) and Lumley General Insurance Ltd v QBE Insurance (Australia) Ltd [2008] VSC 216. At the time of the damage to third party property ACE became liable to indemnify the JV. Lloyds did not, at that point, become liable for any amount below the threshold indemnity because Lloyds’ obligation to indemnify in respect of the cover under the underlying policy only arose when the JV failed to obtain an admission that ACE would indemnify. ACE submits that Lloyds cannot assert that it became liable at an earlier time because that would only be on the basis that its cover was less extensive than that provided under the ACE policy.

Given that ACE was on risk from the moment of casualty and Lloyds was only contingently liable on the failure of ACE to indemnify it, I accept that the two insurers were not jointly liable as at the time of casualty. Where one insurer is liable to indemnify and remains liable to indemnify and subsequently another insurer becomes liable there may be some room for argument as to the applicability of AMP v QBE , but I proceed on the basis that contribution as a remedy in the insurance context is not available as a remedy for this reason. A second issue is whether contribution is available where the party seeking contribution claims entitlement to 100% of the amount. Commercial and General Insurance Co Ltd v Government Insurance Office (NSW) (1973) 129 CLR 374 at 380 – 381 is a significant obstacle to such an argument, but even if theoretically possible, such a claim takes the matter out of the realm of a ‘rateable’ apportionment as applied in the case of double insurance and hence would also propel Lloyds’ claim into one more akin to recoupment than contribution. This last point is the mirror image of the fact that if ACE had met its liability to the JV it would have had no basis for a claim on Lloyds for contribution – pointing to a lack of reciprocity of obligation.

The case of Ruabon Steamship Co Ltd v London Assurance [1900] AC 6 is authority for the proposition that the fact that ‘A’ obtains a benefit from the action of ‘B’ for which ‘B’ has incurred expense does not, without more, give ‘B’ a right to recoupment or contribution from ‘A’. That can be accepted but that principle does not address the present problem.

There are a number of cases which make it clear that the fact that both ‘B’ and ‘C’ may be liable to ‘A’ does not necessarily lead to the conclusions that if ‘B’ is required to pay ‘A’, ‘B’ can then pursue ‘C’ for contribution: see Scholefield Goodman & Sons Ltd v Zyngier [1986] AC 562 and Craythorne v Swinburne (1807) 14 Ves Jun 160 (cases involving a surety for a surety) and see Re Gasbourne Pty Ltd [1984] VR 801. In Burke v LFOT Pty Ltd [2002] HCA 17 a vendor and its solicitor, held liable for misrepresentations inducing purchase, were held not entitled to contribution from the negligent solicitor for the purchaser, since equitable contribution is founded on concepts of fairness and justice and any payment to the party making the misrepresentation by the purchaser’s solicitor would amount to unjust enrichment: see [22] per Gaudron ACJ and Hayne J and see McHugh J at [25], [38] and [44] – [49]; see also Street & Halls v Retravision (NSW) Pty Ltd (1995) 56 FCR 588 and Cockburn v GIO Finance Ltd (No 2) [2001] NSWCA 177 at [27], [36], [78] – [80] and cases cited there.

ACE also contends that if Lloyds’ claim is viewed as a claim for recoupment it fails because it does not meet the requirement laid down for such a claim and calls in aid a statement of principle found in Goff and Jones, The Law of Restitution, 6th ed (2002) Sweet & Maxwell at para 15.001 in which it is stated: “To succeed in his claim for recoupment, the plaintiff must satisfy certain conditions. He must show: (a) that he was compelled, or was compellable, by law to make the payment; (b) that he did not officiously expose himself to the liability to make the payment; and (c) that his payment discharged a liability of the defendant.”

Mason and Carter in their text Restitution Law in Australia (1995) Butterworths at [631] express the requirement of a claim in recoupment to be: “(a) the need for the plaintiff’s payment to relate to a liability falling upon the plaintiff and the defendant; (b) the plaintiff’s payment benefited the defendant; and (c) absence of the plaintiff’s officiousness.” The learned authors reject a fourth requirement that the plaintiff’s payment must have discharged the defendant’s liability: at [631].

ACE submits that: (1) The only reason Lloyds was required to indemnify the JV was because it had contractually promised to do so – this is not, says ACE, a payment under compulsion. (2) Lloyds, by offering to pay the JV if ACE did not, was acting officiously within the meaning of the authorities: see Owen v Tate [1975] 2 All ER 129. (3) Lloyds, by paying the JV, did not ‘discharge’ ACE’s liability but rather, simply precluded the JV bringing a claim against ACE because it had already been indemnified by Lloyds.

In Owen v Tate , the defendants obtained a loan from a bank, which was secured by a charge by way of legal mortgage on the property of a Ms Lightfoot. Ms Lightfoot became concerned that the deeds to her house were being held to secure the defendants’ loan and she approached Mr Owen, who agreed to assist. Owen deposited £350 with the bank and signed a form of guarantee by which he guaranteed payment of up to £350 due by the defendants. He did not consult the defendants before doing so. The bank applied the £350 in repayment of the defendants’ debt. Owen then demanded payment of the £350 from the defendants.

Scarman LJ noted that the defendants had protested to the bank when the bank indicated that it would release Ms Lightfoot’s security and replace it with Owen’s security and Owen had provided the money to assist Ms Lightfoot and without notification to the defendants. Scarman LJ was not satisfied that it was just and reasonable in all the circumstances to grant Owen a right to reimbursement. He stated at 135 what he saw as the true principle of the matter thus: “If without an antecedent request a person assumes an obligation or makes a payment for the benefit of another, the law will, as a general rule, refuse him a right of indemnity. But if he can show that in the particular circumstances of the case there was some necessity for the obligation to be assumed, then the law will grant him a right of reimbursement if in all the circumstances it is just and reasonable to do so. In the present case the evidence is that the plaintiff acted not only behind the backs of the defendants initially, but in the interests of another, and despite their protest. When the moment came for him to honour the obligation thus assumed the defendants are not to be criticised, in my judgment, for having accepted the benefit of a transaction which they neither wanted nor sought.”

In the course of his judgment (in which Stephenson LJ and Ormrod LJ agreed, adding observations of their own), Scarman LJ noted that there are exceptions to the rule that a volunteer who makes a payment on behalf of another cannot obtain repayment. He also referred at 131 to the words of Lord Wright in Brook’s Wharf : “The obligation imposed by the Court simply under the circumstances of the case, and on what the Court decides is just and reasonable having regard to the relationship of the parties. It is a debt or obligation constituted by the act of the law, apart from any consent or intention of the parties or any priority of contract.”

In Owen v Tate , Scarman LJ cited a passage from the first edition of Goff & Jones, The Law of Restitution, (1966) Sweet & Maxwell, at p 207 in which reference was made to the need for both parties to have been subject to a common demand for money for which, as between the parties, the defendant is primarily liable. At least by the fourth edition (1998), the learned authors had placed the requirement for the common demand in the category of contribution and not recoupment, noting at p 390 that it was enough that the party who paid was ‘compelled’ to pay a debt to the third party – “it does not matter that there is no common liability to be sued”: at p 390. Mason and Carter challenge the view of common liability even for contribution: see [619] – [620], noting that in Australia, commonality of burden or loss is what is required.

In Brook’s Wharf a warehouseman which had been required to pay import duty on goods that had been stolen was held entitled to recover the impost from the importer. In Glebe Island Terminal Pty Ltd v Continental Seagram Pty Ltd (The Antwerpen) (1993) 40 NSWLR 206 Sheller JA at 224 with whom Cripps JA agreed said of the case: “The proposition was that the importer was and remained liable for duty as at the date of importation. The obligation imposed by the Acts upon the custodian was ancillary and did not supersede the importer's liability. The claim was one for money paid by the custodian under compulsion to the use of the importer which remained ultimately responsible for the payment: Moule v Garrett (1872) LR 7 Ex 101 at 104. In modern terms the claim was for restitution or fair and just compensation for a benefit accepted or retained unjustly: Pavey & Matthews Pty Ltd v Paul (1987) 162 CLR 221 at 255-256.”

In Moule v Garrett (1872) LR 7 Ex 101 it was held that an assignee of a lease was obliged to indemnify the original assignee of the lease who had been required to pay money to the landlord as a result of breaches by the incumbent lessee; and even where the assignee had not taken the assignment from the original lessee. In the course of submissions Willes J at 103 made reference to the principle that: “where two persons are under an obligation to the same performance, though be different instruments, if both share the benefit which forms the consideration they must divide the burden; if one only gets the benefit he must bear the whole: Dering v Earl of Winchelsea (1787) 1 Cox 318, 29 ER 1184.”

There are several other cases which follow the approach in Moule v Garrett : Duncan Fox & Co v North and South Wales Bank (1880-81) 6 App. Cas. 1 at 11 – 12 (“ Duncan Fox ”), Becton Dickinson UK Ltd v Zwebner [1989] QB 208, Selous Street Properties Ltd v Oronel Fabrics Pty Ltd [1984] 1 EGLR 50; (1984) 270 E.G. 643.

Duncan Fox , was concerned with a bill of exchange accepted by the defendant and endorsed by the plaintiff and the entitlement of the acceptor to securities held by the acceptor of the bill the debtor’s bank. In the course of his speech, the Lord Chancellor Lord Selborne said at p 13: “It appears to me that these principles of Equity are not less applicable to cases of the third class – cases in which there is, strictly speaking, no contract of suretyship, but in which there is a primary and secondary liability to two persons for one and the same debt, by virtue of which, if it is paid by the person who is not primarily liable, he had a right to re-imbursement or indemnity from the other – than to those of the second class, in which there is a contract for suretyship to which the contractor is not a party. To this third class of cases, the right of an indorser against an acceptor of a bill of exchange may most properly be referred. The liability of the indorser to the holder is, by the law merchant, conditional, and (as was said by Mr Justice Buller in Tindal v Brown [(1786) 1 Term Rep 167]) “only secondary”; but when the conditions required by the law are fulfilled, it becomes absolute, and is that of a principal; and the indorser’s right, if he pays the holder, to recover over against the acceptor is not founded on any agreement between him and the acceptor (who is as likely as not to be a stranger without any communication with him before the indorsement), but is established by the same law.”

Here Lloyds and ACE both became liable to indemnify the JV but ACE primarily so and Lloyds only because ACE would not meet its liabilities. Lord Blackburn at 19 said: “I think that though the indorser of a bill is not exactly a surety for the acceptor, or a co-surety with those who are sureties for the acceptor, yet he stands in a position sufficiently analogous to that of a surety to bring him within the principle of Dering v Earl of Winchelsea [ (1787) 1 Cox 318; 29 ER 1184].” and these words and those of Lord Selborne, although of course dealing with a different subject matter, point to an approach that has resonance here.

In Lumley General Insurance Ltd v Oceanfast Marine Pty Ltd [2001] NSWCA 479 (“ Lumley ”), the NSW Court of Appeal had to consider whether Lumleys were entitled to prove in the winding up of Oceanfast in respect of a commercial bond for $5 million, which had been provided by Lumleys to Adsteam in support of Oceanfast’s obligations under a shipping contract, and met when Oceanfast defaulted. Adsteam had lodged a proof of debt for its losses under the contract without any deduction for the $5 million received from Lumleys pursuant to the bond. The trial judge (Austin J) had held that the principle in Moule v Garrett applied notwithstanding that the agreement between Lumleys and Oceanfast did not expressly provide for repayment by Oceanfast of any monies paid by Lumleys. He also held, however, that the absence of any agreement by Oceanfast that payments made by Lumleys would reduce the debt to Adsteam meant that Adsteam could prove in the winding up for the full $15 million claimed, and that Lumleys could not.

The Court of Appeal, by a majority (Priestley and Giles JJA) upheld the appeal. Priestley JA noted that if the principle of Moule v Garrett applied (and his Honour thought it did: see [6]) then Lumleys’ payment did discharge the debt of Adsteam by that amount and there was no impediment to Lumleys’ claiming the $5 million. Giles JA noted that it had been accepted on the appeal that Moule v Garrett applied and he held that by the payment of the $5 million by Lumleys thus “Oceanfast obtained the benefit of the payment by discharge of its liability” which “was the occasion for equity to impose on it the burden of recoupment”: see [163].

Beazley JA (in dissent) regarded the absence of any contractual right in Lumleys to recover from Oceanfast to be of critical significance and her Honour held that Lumleys was not entitled to prove in the winding up – in effect, Lumleys took the risk that Oceanfast would not be able to meet its obligations to Adsteam and that Adsteam’s right to claim the balance from Oceanfast should not be reduced by a competing claim by Lumleys (see [89] for a concise exposition of the competing claims).

It will be observed that Priestley JA was of the view that Moule v Garrett applied and that Priestley and Giles JJA were both of the view that the payment by Lumleys discharged the debt to Adsteam by that amount (as, it was pointed out by Priestley JA, was the trial judge). There was no suggestion that the fact that Lumleys’ payment was consequent upon a commercial agreement and in satisfaction of an obligation thereunder, meant that there was no right of recoupment.

A useful discussion of the question of benefit and advantage is found in O’Donovan and Phillips, The Modern Contract of Guarantee ([276] supra ) at 12.110: “the mere fact that the guarantor’s payment confers a benefit upon the debtor does not always ensure the success of the guarantor’s restitutionary claim for a reimbursement where the benefit was unwanted or unnecessary. Certainly, where the debtor derives no advantage from the payment, the guarantor will not be entitled to a reimbursement. The position is different if the payment made also relieves the payer from her or his own liability. For example, where two persons are liable for the same debt, and, as between them, one is primarily liable and the other is secondarily liable, the former is obliged to indemnify the latter if the latter discharges the liability. This restitutionary right to an indemnity is not founded on a request for payment and there is no question of the secondary obligor officiously exposing herself or himself to the liability to pay the principal debt: the secondary obligor is, after all, liable in her or his own right. The right to an indemnity arises simply because the secondary obligor discharges the liability of the primary obligor. This principle allows a lessee, who had assigned the lease, to obtain an indemnity after payment to the lessor, not only against the assignee, but also against a surety for the assignee. The indemnity arises despite the absence of privity of contract between the lessee and the surety for the assignee.”

As was pointed out in Cockburn v GIO Finance Ltd (No 2) [2001] NSWCA 177 contribution and recoupment are siblings. In Karacominakis v Big Country Developments Pty Ltd [2000] NSWCA 313 at [239] Giles JA said: “The basis is described in Mason and Carter, Restitution Law In Australia , 1995 at 208 as recoupment, derived together with contribution from considerations of natural justice (as to contribution see Albion Insurance Co Ltd v Government Insurance Office (NSW) (1969) 121 CLR 342 at 350-2 per Kitto J). The authors say - “Rights of contribution and recoupment derive from a single source, namely the injustice of the defendant having had its burden relieved by the plaintiff. But a right of recoupment differs in its application from contribution because there is no `equality': rather the respective positions of P and D are such that it is just that P should throw the whole burden of P's liability to X upon D's shoulders. If it were otherwise, D would be seen to have received an unjust benefit (that is, the effective release of the burden to X) at the expense of P who bore it. [Cf Pavey v Matthews Pty Ltd v Paul (1987) 162 CLR 221 at 256-7 per Deane J.] Thus, in the standard guarantee situation a guarantor (P), who is forced to meet the creditor's (X's) claim, is entitled as against the principal debtor (D) to recoupment in full for the outlay. This is one example of a broader principle, as Lord Wright MR demonstrated in Brook's Wharf and Bull Wharf Ltd v Goodman Bros: [[1937] 1 KB 534 at 544, cited with approval by Walsh JA in Armstrong v Commissioner of Stamp Duties (1967) 69 SR (NSW) 38 at 47. See also The Pindaros [1983] 2 Lloyd's Rep 635. The essence of the rule is that there is a liability for the same debt resting on the plaintiff and the defendant, and the plaintiff has been legally compelled to pay, but the defendant gets the benefit of the payment, because his debt is discharged either entirely or pro tanto, whereas the defendant is primarily liable to pay as between himself and the plaintiff.””

This passage was approved by Giles JA, with whom Handley and Stein JJA agreed.

I think the authorities to which I have referred amply support the view that recoupment is available where it is just and equitable to permit it.

In Esso Petroleum Co Ltd v Hall Russell & Co Ltd (The Esso Bernicia) [1989] AC 643, the Esso Bernica, whilst under the control of tugs collided with mooring dolphins at a jetty and the ship and the jetty suffered damage. Bunker oil escaped in large quantities. The owners of the vessel sued the shipbuilders and the Council (who were responsible for pilotage in the harbour). Amongst the damage were payments that Esso had been required to pay to crofters, whose livestock had been affected, and operators of the terminal. Those sums were paid under a voluntary agreement known as Tanker Owners Voluntary Agreement Concerning Liability for Oil Pollution. The shipowners sought to recover the payment made to the crofters.

The House of Lords, per Lord Goff with whom Lord Keith, Lord Brandon and Lord Templeman concurred, held, relevantly, that Esso could not proceed in its own name to pursue the shipbuilder because there was no contract of indemnity between Esso and the crofters and the payment to the crofters did not discharge the shipbuilder’s liability to the crofters: see 663B. Lord Jauncy, who reached the same conclusion, pointed out that Esso was under no general duty in law to the crofters and the payments made were entirely gratuitous. The crofters could have sued the shipbuilder notwithstanding the payments: see 678A-B.

Lord Jauncy said at 678D: “What has caused these payments to be made? In my view they were made because Esso has chosen, by entering into and remaining a party to TOVALOP, to assume a voluntary obligation to the crofters and not because of any alleged negligence on the part of Hall Russell [the shipbuilders].”

The insurance contract which Lloyds entered into with the JV was not the voluntary assumption of an obligation but a contract made for consideration, namely the premium.

Reference was made to Lucas v CMT Construction & Metropolitan Tunnels (1986) 4 ANZ Ins Cas 60-752 (“ Lucas ”). In that case, a workers compensation insurer (“ NEM ”), which had paid $32,000 to an injured worker, sought to recover that amount from the worker’s employer (“ CMT ”). The worker had obtained a judgment against CMT and CMT had sought and obtained indemnity from the motor vehicle insurer. In fixing the amount payable by the employer to the worker, the Court had taken into account the workers compensation payments made on behalf of the employer by NEM. The claim for subrogation was held to be misconceived and the decision may be seen as an application of the principle in Sydney Turf Club v Crowley (1972) 126 CLR 420 at 421 approving Sydney Turf Club v Crowley [1971] 1 NSWLR 724 since the employer, it was held, had no remaining rights against the motor vehicle insurer or the employee. NEM argued that CMT’s liability for workers compensation had been discharged but its claim for recoupment was also rejected. ACE relies on his Honour’s comment at 75,536 that “[P]ayments of compensation by NEM were never made under compulsion of law but were made under a contract of insurance. I have considered a number of the authorities cited by Goff and Jones in Chapter 12 but I have been unable to find a case near to the case now before the Court” which appears to offer some support for ACE’s position because of the characterisation of obligations to pay pursuant to an insurance contract as voluntary obligations, but this comment must be seen in the context of a claim by an insurer to recover from an insured employer where the employer had not received any amounts back from the worker for those payments. CMT had its liability to its employee discharged by virtue of NEM’s payment on its behalf but if it was required to repay NEM it would have had no benefit from the contract of insurance. I do not think that Lucas assists ACE. Of course, entry into a contract is voluntary, but I do not think that makes payments made by an insurer to an insured pursuant to a contract voluntary in the relevant sense when an insurer, who has met its liability to an insured, seeks to recover from another insurer. If it did then a co-surety (or co-insurer) would not be able to pursue another co-surety (or co-insurer). I have referred to Lumley which is another example of payments made pursuant to a contract, and see also GRE Insurance Ltd v QBE Insurance Ltd [1985] VR 83 at 103 – 104, referred to below.

Another case to which reference was made was State Bank of Victoria v Parry (1990) 2 ACSR 15. Parry was the Chairman and Chief Executive of PCL and a director of Kynes Pty Ltd (“ Kynes ”) and other companies in the Parry Group. Parry gave a guarantee to the State Bank of Victoria in respect of a facility provided by the bank to Kynes. The purpose of the facility was to refinance a loan used by Kynes to purchase shares and options from PCL. Parry was held liable to the bank for in excess of $8 million. He claimed indemnity or contribution from PCL asserting that he had paid the monies in discharge of his duties as a director of PCL. The substantial benefit of the guarantee had been provided to Kynes, not PCL, and his claim for recoupment was rejected. I do not think this case assists ACE.

In my view this present case, even if not falling within the description of contribution, is sufficiently close as to enable some aspects of the approach taken in relation to contribution in the insurance context to be relevant when looking at the case as one for recoupment, i.e.: (1) Just as there is no question that a second insurer who pays a claim can recover from another insurer even where the first insurer has no knowledge of the existence of the other policy there is no requirement for Lloyds to prove that ACE requested that the JV enter into the Lloyds policy or knew that the JV proposed to do so (there is, as it happens, evidence that ACE did know of the Lloyds policy). (2) In contribution, it is accepted that the payment by one insurer discharges the burden of the second insurer and it should be accepted here. I reject ACE’s assertion that it was not discharged by Lloyds’ payment. I note that in para 170 of its submissions of 10 March 2009, ACE agreed that it “obviously obtains the benefit of having its liability discharged” but as this discharged Lloyds’ own liability that made a difference. I do not accept that contention – if Lloyds was not meeting a liability of its own then it might have been said that it was an officious intermeddler. (3) The essence of contribution in insurance law is that it is just and reasonable for the party relieved of liability to the insured to pay its share to the insurer which has paid. That same underlying principle here requires consideration of whether the liability for ACE was a primary obligation or not. (4) In GRE Insurance Ltd v QBE Insurance Ltd ([313] supra ), it was held that an insurer who paid more than the rateable proportion it was required to pay pursuant to principles of construction was entitled to recover from the second insurer the excess paid to the insured: per McGarvie J at 102, per Murray J at 95 – 96: “But the appellant did not succeed in its denial of liability and if it now can succeed upon the basis that the respondent is not entitled to contribution simply because it paid out the whole loss it will indeed be profiting by its own wrong, as it turned out, in wrongly denying primary liability. In the absence of clear and compelling authority I do not think this Court should be ready to countenance such a result.” (5) In my view the law of contribution recognises that there is nothing ‘officious’ about an insurer covering a liability that is covered under another policy by another insurer. An agreement by an insurer to meet a liability imposed on another insurer when that second insurer refuses to meet its obligations seems to me to be a commercial arrangement with positive ramifications for commerce and the construction industry, rather like credit or mortgage insurance, and I do not accept ACE’s argument that Lloyds only has itself to blame for entering into a policy of that kind and meeting its obligations under it, or that Lloyds, by providing conditional cover, was acting “behind [ACE’s] backs”: see T1394.5. ACE’s position has a distinct lack of merit to it and none of the cases relied on by ACE support, in my view, such an outcome. (6) At the time that the JV sought payment of its claim from both ACE and Lloyds and ACE refused to pay, the answer to both of the questions posed by Priestley JA in Australian Eagle Insurance Co Ltd v Mutual Acceptance (Insurance) Pty Ltd ([275] supra ) would have been yes, even if until the refusal the answer to the first question would have been no.

In my view, it is just and equitable to order recoupment of any liability of which ACE was relieved by Lloyds’ payment. ISSUE 24: Quantum

ACE, by its Scott Schedule response, admitted that a total of $13,341,252.32 had been incurred by the JV. That amount included an allowance for the Jin Temple of $800,000, which amount, it was indicated at the commencement of the hearing, has not been paid. It was accepted that the amount which ACE should be taken to have admitted was therefore $12,541, 252.32.

I held in an interlocutory judgment that ACE was permitted to withdraw the admission, but in limited respects, namely as to: (1) the cost of labour or other services provided by persons who were otherwise employed by the JV; (2) the cost of acquisition, hire and/or use of plant and equipment owned or leased by the JV (other than costs limited to the difference between the acquisition and sale costs of plant and equipment acquired for the purpose of performing the work the subject claims of the JV and later sold to third parties); (3) the cost of investigation, monitoring and preservation of property (other than costs in the nature of payments to third parties); and (4) Whether ACE could limit the amount for which it was liable to the amounts assessed by Mr Davies, Lloyds’ assessor. Lloyds disputes that Mr Davies’ figures could be held to be final because after he assessed the amounts payable, which came to a total significantly less that the amount claimed by the JV, the figure of $13 million was agreed upon. In any event that argument remains to be determined.

Leaving aside the Jin Temple, and the arguments advanced by ACE on liability, the maximum amount for which ACE may be liable is $12.6 million. This amount could be reduced to $9 million and possibly further reduced by amounts identified in (1), (2) and (3), even assuming all of the elements of the Lloyds claim were made out. I say ‘possibly’ because it may be that the amount allowed by Mr Davies or some part of it did not include the items which ACE will be permitted to contest.

It was agreed that the matters (1), (2), (3) and (4) of [318] will need to be referred to a referee and that attention will need to be given by me to how that aspect of the case should be handled after I have handed down this judgment. Conclusion

It follows that in my view: (1) Lloyds was not liable to indemnify the JV under Section 1 of the Lloyds policy for: (a) any of the items listed under [60(2)] Gambas shaft; (b) any of the items listed under [60(3)] TBM/G4 problem; or (c) AEM Tech Building, PowerGrid 66 kv joint bays 4 and 5, road repairs in Woodlands Avenue 8 and recharge wells etc listed under [60(1)] Woodlands shaft, because liability for these items was due wholly or in part to causes excluded by condition 7.3 of the Lloyds policy, and also, in the case of the PowerGrid damage to 66 kv joint bays 4 and 5 at [60(1)] , by clause 5.11 of the Lloyds policy. (2) Lloyds was liable to indemnify the JV in respect of the items not covered under Section 1 of the Lloyds policy, and referred to in (1) above, by virtue of Section 2 of the Lloyds policy. (3) ACE is not liable for settlement of 66 kv joint neal lamp pole 16 along Woodlands Avenue 8, rectification at Yee Lee Oil and dismantling fence, gate and hack concrete kerb listed under [60(4)] TBM/dyke problem, because the damage, and hence liability of the JV, has not been established to have occurred on or before 27 May 2002. (4) ACE is not liable to indemnify the JV for the items listed in (3) above and Lloyds therefore cannot recover for payments made in respect of those items, but ACE is liable for the balance of payments, set out in [60] above, made by Lloyds subject to determination of the remaining issues of quantum which I have identified in [318], in respect of those items which have not been excluded.

I will stand the matter over to a suitable date for directions as to the balance of these proceedings. I will also give the parties an opportunity to bring to my attention any matter requiring determination which I have not dealt with other than the issue of quantum.