BAILII [Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]

England and Wales High Court (Commercial Court) Decisions


You are here: BAILII >> Databases >> England and Wales High Court (Commercial Court) Decisions >> Aegis Electrical and Gas v International Services Co Ltd & Anor [2007] EWHC 1762 (Comm) (25 July 2007)
URL: http://www.bailii.org/ew/cases/EWHC/Comm/2007/1762.html
Cite as: [2007] EWHC 1762 (Comm), [2008] Lloyd's Rep IR 17

[New search] [Context] [Printable version] [Help]


Neutral Citation Number: [2007] EWHC 1762 (Comm)
Case No: 2005 368

IN THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
COMMERCIAL COURT

Royal Courts of Justice
Strand, London, WC2A 2LL
25/07/2007

B e f o r e :

MR. JUSTICE ANDREW SMITH
____________________

Between:
AEGIS ELECTRICAL AND GAS
Claimant
INTERNATIONAL SERVICES COMPANY LIMITED
- and -
CONTINENTAL CASUALTY
COMPANY


Defendants

____________________

Andrew Hunter (instructed by Clyde & Co.) for the Claimants
Colin Wynter QC (instructed by Davies Lavery) for the Defendants

Hearing dates: 2, 3 and 4 My 2007,

____________________

HTML VERSION OF JUDGMENT
____________________

Crown Copyright ©

    MR JUSTICE ANDREW SMITH :

    Introduction

  1. This is a dispute about whether the claimant reassured, to whom I shall refer as "the Syndicate", can recover from the defendant reinsurer, to whom I shall refer as "CCC", in respect of two settled claims arising from incidents at an oil refinery owned by Coastal Aruba Refinery Company ("CARC"), a subsidiary of El Paso Energy Corporation ("El Paso"). One claim was for damage that the Visbreaker Unit at the refinery sustained on 9 April 2001, and is referred to as the "Visbreaker loss". The other related to damage to the D1AReactor that occurred on 6 November 2001, and I shall refer to that as the "D1AR loss". The claimant is a member of Lloyd's Syndicate 1225 for the years 2001 and 2002 and claims as a representative member of the syndicate. The claims are made under a facultative reinsurance policy underwritten by CCC, who are American insurers in the CAN group. CCC deny that either loss falls within the scope of the reinsurance.
  2. The Refinery

  3. The refinery is on the island of Aruba in the South Caribbean. It was founded by Exxon in the 1920's and used by them to process crude oil, mainly imported from Venezuela and Mexico, until the 1980's when they closed it down. The facility was bought by CARC in 1989, who reopened it in November 1990.
  4. The refinery comprised a "Visbreaker", a vessel into which crude oil was pumped, and in which it was heated and so caused to "crack" into different hydrocarbon products.
  5. The refinery also had four hydrotreaters, including the D1A reactor unit. Its purpose was to remove sulphur from the oil feedstock (kerosene, diesel and gas oil streams) by introducing hydrogen. As a result of a catalytic reaction, the reactor produced low-sulphur hydrocarbons and also hydrogen sulphide as a by-product. The unit's central component, reactor R-2501, was a steel vertical pressure vessel, 12'6" in diameter and 82'9" in height. The shell itself was 1.8" thick, and it had stainless steel internal cladding. There were circumferential bands or welds running around the outside of the vessel, which were presumably designed to reinforce the structure. There were platforms at different levels of the outside wall. Inside the reactor were a number of metal beds which held the catalyst: CCC's expert witness, Mr. Peter Moore, understood that there were three such beds and the Syndicate did not suggest otherwise. It was designed to operate at a pressure of 670 psig and a temperature of 790 degrees Fahrenheit, and it had temperature alarms that were triggered if the temperature reached 825 degrees.
  6. The Visbreaker loss

  7. The details of the Visbeaker loss are not important to what I have to decide. It suffices to say that during routine maintenance carried out by plant personnel, oil leaked while a strainer on a pump was being cleaned and hot equipment in the vicinity ignited it.
  8. The D1AR loss

  9. I must examine the damage to the D1A reactor, and in particular the R-2501 vessel, in more detail because if the interpretation of the reinsurance cover advanced by CCC is correct, I must decide whether its proximate cause was "explosion" (within the meaning of an attachment to the reinsurance slip).
  10. A number of reports were prepared after the incident, and these provide the main evidence about what happened.
  11. i) First, there is an undated report presented by the insured, which has been referred to as the "CARC" report. Some photographs of damage to the reactor unit were presented with it, and the photographs have captions, some of which are difficult to understand. It is impossible to say whether the photographs illustrate all the damage or only examples of it. (The report is headed "Draft Incident Report Form", but no later version of it is in evidence and I was told that it is the final version.)
    ii) Secondly, there is a report of Engineering Design & Testing Corp. of Houston ("EDT") dated 9 November 2003, which was prepared for the Syndicate and El Paso's other underwriters. The Syndicate also put in evidence a graph which had been provided to them by EDT in the course of the trial. It is not entirely easy to interpret, but apparently it traced measurements, taken at hourly intervals during the incident, of the temperature measured at various points inside the vessel, of the pressure inside the vessel and of the flow of hydrogen into the vessel.
    iii) Thirdly, there is a report dated 9 January 2004 that was written by Bateman Chapman Limited ("BCL"), who were engaged as loss adjusters by the Syndicate and other underwriters of the direct cover.

  12. However, the picture that can be derived from these reports and the photographs is pretty sketchy. I was assisted in interpreting them by Mr. Peter Moore of Minton, Treharne & Davies Limited, a metallurgist who was called by CCC to give expert evidence. I am grateful for his evidence, but he was clearly hampered by the limited information available to him. I should add that there are also in evidence a draft report dated 6 December 2001 written by the loss adjusters McLarens Toplis N A Inc. ("McLarens Toplis"), but Mr. Moore was not asked to consider that report and neither party relied upon it.
  13. On 3 November 2001 the whole Aruba refinery suffered a power cut because of a short circuit in a turbo-generator. The DlA reactor unit closed down, and because no hydrogen was being pumped to sweep the unit, a quantity of some hydrocarbons remained in the reactor but it is impossible to say how much. At 8.30 hours on 5 November 2001 the operators attempted a "hot start up": that is to say, they went about reheating the reactor before resuming operations. At the start of this operation the pressure in the unit was 380 psig and the temperature of the catalyst beds was between 575 and 625 degrees Fahrenheit. The start up involved re-pressurising the reactor with hydrogen and increasing it to operating temperature. In the course of this, as a result of a combination of circumstances, excessive heat was generated from exothermic hydrogenation reactions. (Apparently hydrogen was contaminated with carbon monoxide and carbon dioxide, the catalyst was particularly active and there was a particularly potent mix of hydrocarbons. It was suggested that the hydrogen might have reacted with carbon monoxide and carbon dioxide in the presence of a nickel based catalyst.) A temperature alarm by way of the first catalyst bed sounded at 23.44 on 5 November 2001 and other alarms sounded on occasions thereafter, but were apparently ignored by the operators.
  14. The pressure built up during the start-up operation to about 460 or 480 psig just before the first alarm sounded. At 24.00 on 5 November 2001 it increased to 506 psig and, according to EDT, thereafter it was controlled at that level throughout the incident.
  15. EDT's graph shows that at about that time the temperature by way of the first catalyst bed went up to some 1,400 degrees Fahrenheit. The temperature by way of the top of the second bed increased to more than 800 degrees. Different temperatures were recorded at different levels in the vessel, the hottest areas presumably being where the exothermic reaction was at its most intense. At about 4.00am on 6 November 2001 the flow of hydrogen into the vessel was temporarily stopped (or very much reduced), and apparently there was a corresponding drop in the high temperatures by way of the first catalyst bed and the top of the second catalyst bed. However, when the flow of hydrogen was resumed an hour or two later, these temperatures increased again.
  16. EDT reported that shortly before 20.26 hours on 6 November 2001 the wall of the unit bulged about three feet below the top of the third catalyst bed, and longitudinal cracks and cracked welds appeared on and near the bulge. At 20.26 a fire was reported, and at about this time the temperature by way of the top upper catalyst beds dropped rapidly. The fire in the reactor was put out by depressurising the reactor and injecting nitrogen into the unit.
  17. EDT's graph appears to show a dramatic increase in the hydrogen flow at around this time, but it cannot be seen exactly when this occurred.
  18. The evidence about the damage to the reactor is also imprecise. What is clear from a photograph of the vessel, or at least of the lower part of it, is that its basic structure remained intact. I should expand upon some aspects of the damage.
  19. EDT's report referred to "a split in the vessel shell as well as a bulge above the second platform". The photographs show what appears to be a split in the shell running straight and vertically. The captions refer to it as a "leak" and describe the crack as being 4 inches long and at the level of the third catalyst bed. It was, as Mr. Moore pointed out, above the third circumferential weld. There is no evidence of the width of the split, and although it appears to gape somewhat more than other cracks, its general appearance is similar. The caption states that the shell bulged around the crack, but there is no photographic or other evidence of further collateral damage by way of the split.
  20. I infer that this was the only place where the shell of the vessel was punctured. There is no specific evidence of this, but no other split or rupture was mentioned in the reports or observable in the photographs. I think it likely that if there had been any other such damage, it would have been mentioned and photographed. I conclude that the shell was split right through only at the point of the 4 inch rupture.
  21. EDT described further cracking and bulging of the vessel wall, and said that the damage was "caused by cracks on the vessel wall and bulging of the vessel. The cracks started because the temperatures exceeded the design temperatures of the vessel. The cracking was caused by overheating and associated build-up of pressure." There were also cracked welds. The location of some of this damage was described as being two feet above the first platform which was 17 feet above the bottom head (which was near the bottom of the vessel) or three feet below the third catalyst bed. This was probably near where the wall split.
  22. As I interpret the photographs and the captions to them, a number of them show vertical cracks to and near the third circumferential weld, and they show a cracked ring support near the third circumferential weld. At one point Mr. Andrew Hunter, who represented the Syndicate, submitted that there was cracking all over the vessel, but there is no evidence that the damage was so widespread. However, as I interpret the evidence including the photographs, the cracking was not confined the area where the vessel split. It seems to me that Mr. Moore fairly summarised the evidence thus: "...it is apparent that multiple linear longitudinal cracks occurred in the vessel close to the location of the leak, although some damage is evident at other locations". He described the cracking as widespread, but it does not seem to me that there is any firm basis for this conclusion: all I need conclude, and all that I can conclude on the evidence, is that it was not confined to one area.
  23. The cracking was predominantly (although not exclusively) vertical (or "longitudinal", as it was described). It was Mr. Moore's evidence that, "This longitudinal cracking is probably due to the material of the vessel bulging outwards due to internal pressures exceeding the strength of the vessel walls. The linear longitudinal nature of the cracking of this vessel, as opposed to either circumferential or meandering cracking, is typical of pressure overload." Notwithstanding the evidence from EDT that the pressure was controlled, I accept this evidence. However, I have not found evidence of bulging in places other than where the wall ruptured, and although the position is not at all clear I conclude on the balance of probability that the wall noticeably bulged only in that area, and elsewhere the bulging that Mr. Moore describes was minor.
  24. As I have said, a fire was reported at 20.26 on 6 November 2001. Mr. Moore's evidence was that there could not have been a fire in the reactor without an oxidising agent, and, while the vessel contained flammable gases and vapour, there was no oxidising agent in the vessel until the wall was punctured, when air would have been introduced. Again, I must draw conclusions from what scant evidence there is, and although the evidence about what was in the vessel is exiguous, I accept Mr. Moore's reasoning and conclude that the fire probably occurred after the rupture and other cracking of the wall.
  25. There was reference in the Syndicate's evidence to internal damage to the vessel. Mr. Richard Foulger, their claims manager, said that the increase of temperature inside the vessel "caused the various internal structures to melt and buckle". Mr. Joe McMahon, the author of the BCL report, also referred to internal damage. However, there is no evidence in the reports that are in evidence of damage of the kind that Mr. Foulger described, and it is obscure where he obtained this information. Inevitably an incident of this kind would cause some internal damage to the vessel, but there is no evidence upon which I can rely about its nature or extent.
  26. Apart from EDT's account of a report of bulging and cracking to which I have referred, there was no evidence about what was seen or heard by contemporaneous observers. There is no account of a bang or loud explosive noise, nor of flames or a flare visible from where the vessel was ruptured. Admittedly there is no direct evidence to the contrary, and Mr. Colin Wynter QC, who represented CCC, argued that it is not known whether anyone was in the vicinity at the time. However, apparently someone did report the bulging and cracking, and so, it seems to me, someone was in the vicinity, and I infer that if anything dramatic had been observed, it would have found its way into the reports.
  27. The vessel was overheated, at least in some areas, for some 20 hours. As Mr. Moore explained, the heat weakened the wall of the unit. As the heat and therefore the chemical activity increased, EDT's graph indicates that the pressure in the vessel was controlled. It seems likely that the actual damage took place over a short period before the fire. This was Mr. Moore's inference from the number and nature of the cracks: "with the large size of some of these cracks and the sheer number of the cracks, for all these cracks to merely have travelled up and down without travelling through the vessel wall, would imply to me that there is a relatively short period of time in which the cracking had been taking place". I accept that evidence and, while it is impossible to be at all precise the period, I conclude that it probably all took place within no more than a minute or two before 20.26 on 6 November 2001. It might have been associated with the sudden increase in the flow of hydrogen shown on the EDT graph at around this time, but this part of the evidence is unexplained and it seems to me too speculative, and unnecessary, to draw this conclusion on the balance of probabilities.
  28. Once the wall was ruptured, hydrogen was expelled from the vessel. This involved, as Mr. Moore opined and as I accept, a sudden release or outpouring of gas, and was accompanied by what Mr. Moore described as an "abrupt noise", and there was a shock wave from the point of the rupture. It is impossible for me to make any specific finding about the volume of the noise or about the power or intensity of the shock wave, beyond saying that on the one hand I accept Mr. Moore's evidence that, bearing in mind the high temperature in the vessel and recognising that the pressure was about equivalent to 35 atmospheres, they were "quite marked", but, as I have said, the noise was not so dramatic as to cause anyone in the vicinity to remark upon it.
  29. CCC also argued that I should conclude that there was an expulsion of boiling oil or vapourised oil forced from the vessel when it ruptured. I am unable to accept that argument. There is no evidence about how much oil there was in the vessel during the start-up operation, and without evidence about this there is, it seems to me, no proper evidential basis for this conclusion.
  30. The direct cover

  31. The insured under the cover written by the Syndicate was El Paso and associated entities, but I shall refer to them simply as "El Paso". It was by way of a "Package Policy", and it was designed to "wrap around" El Paso's cover with OIL, the energy mutual insurer. The Syndicate was the second of three leading underwriters, taking a line of 7.5% that was written down to 5.5125%. The risk was underwritten by the Syndicate on 1 i January 2001. As far as is material, the cover was for a period of 12 months to be agreed by the leading underwriter, and the agreed starting date was 29 January 2001.
  32. The slip stated, "all Wording/Clauses as per the expiring El Paso placement as applicable, or as agreed by the three Leading Underwriters hereon". In the event apparently no wording was agreed until 6 November 2003, long after the reinsurance cover was written and the losses had occurred.
  33. The insured "Interest" comprised seven sections of which that directly relevant in this case is section VI which was for "onshore property", including business interruption, and more specifically section V1(A), which concerned onshore property other than property in the course of construction, installation, testing, erection and assembly. The conditions relating to section VI(A) were stated in the slip to be "as per the expiring El Paso wording". The wording for the cover was not agreed by 26 January 2001, when CCC wrote the re-insurance cover. Eventually, long after the losses that give rise to this litigation, it was agreed in terms that were materially identical to the expiring cover.
  34. The insuring clause in section VI(A) for property stated, "This policy insures ... Real and Personal Property" and the wording went on to identify specific items that were included or excluded. Under the heading "Perils Insured" it stated "This policy insures all risks of physical loss or damage to property insured", and had special provisions about earthquake and flood. Specific perils were excluded. The terms "accident" and "object" were not used either in the insuring clause or in defining the perils insured against, but they were used in two other clauses of the wording for section VI(A):
  35. i) In clause 10 of the wording there is a provision headed "Hazardous Substances Coverage", which is in these terms:

    "If, as a result of an accident, property is damaged, contaminated, or polluted by a substance declared by a government agency to be hazardous to health, insurers shall be liable under this endorsement for the additional expenses incurred for cleanup, repair or replacement, or disposal of that damaged, contaminated or polluted property, including any resultant damage for business interruption. Asbestos material removal resulting from an "accident" as defined in this policy is limited to the coverage provided in this clause."
    ii) Clause 21 was concerned with suspension, and read as follows:
    "Upon the discovery of a dangerous condition with respect to any object, any representative of the insurers may immediately suspend the insurance with respect to an "accident" to said "object" by written notice ... Insurance so suspended may be reinstated by insurers..."

  36. The definitions in clause 30 of the wording included this definition of "accident":
  37. "Accident shall mean a sudden and accidental breakdown of an object or a part thereof and resulting in physical damage that necessitates repair or replacement of the object or part thereof."

    The term "object" was defined as

    "any boiler, fired or unfired pressure vessel, refrigerating or air conditioning system, piping and its accessory equipment, and any mechanical or electrical machine or apparatus for the generation, transmission or utilization of mechanical or electrical power including any electronic computer or electronic data processing equipment and its associated media".

  38. The wording for on-shore business interruption provided insurance for loss resulting from necessary interruption of business caused by insured loss or damage to property, and the words "accident" and "object" were not used in that wording.
  39. The reinsurance

  40. The reinsurance was underwritten by CCC on 26 January 2001. The slip described the "Type" of cover as "Machine Breakdown, Boiler Explosion and Business Interruption following for Onshore Assets only". The assured was identified as El Paso and other associated persons. The period was "12 months at 1st February 2001 or date to be agreed Leading Reinsurer plus 30 days extension if required". The interest was, "In respect of the Machinery and Equipment of the Original Assured plant(s) Onshore and/or property as original". The limit was "USD 15,000,000 each and every loss, combined single limit for Material Damage and Business Interruption (Material Damage Limited to USD 5,000,000 each and every loss). In excess of Deductibles".
  41. The conditions included: "To follow the terms, clauses, conditions, exceptions and settlements of the original policy wording as far as applicable hereto".
  42. CCC's underwriter, Ms. Christine Dennis, when initialling the slip, wrote on it the words, "Subject to B & M terms & conditions attached pages 1-3". A document of three (unnumbered) pages was attached to the slip. It was headed, "Boiler & Machinery Coverage Defined", and it was signed by Ms. Dennis. I shall refer to it (simply as a convenient label and not to describe its legal effect) as the "Additional Conditions". It contained two clauses by way of definitions of "Accident" and "Object". However, those terms were not used in the reinsurance slip.
  43. "Accident" was defined in the Additional Conditions as meaning "a sudden and accidental breakdown of an Object or a part thereof, which manifests itself at the time of the occurrence by physical damage that necessitates repair or replacement of the Object or part thereof. It then provided that "accident shall not mean or include loss or damage" resulting from specified events or of specified kind, including "(f) from explosion other than: (1) explosion of the parts of steam boilers, steam turbines, steam engines, steam pipes interconnecting any of the foregoing or gas turbines; or (2) moving or rotating machinery or parts of same when such loss is caused by centrifugal force or mechanical breakdown ... (p) caused by or resulting from a peril insured elsewhere in the reinsured policy".
  44. Object was defined as "any boiler, fired or unfired pressure vessel, refrigerating or air conditioning system, piping and its accessory equipment, and any mechanical or electrical machine or apparatus used for the generation, transmission or utilization of mechanical or electrical power", and there were then identified various things that the terms "Object" did not mean or include.
  45. The dispute about the meaning and effect of the Additional Conditions to the slip is essentially this: CCC say that they defined and limited the extent of the reinsurance cover to claims arising from "Accidents" as defined. The Syndicate say that the impact of the definitions is upon the meaning of the words "Accident" and "Object" when they appear in the underlying cover provided by the Syndicate to El Paso.
  46. There was some evidence about how CCC came to write the reinsurance to which I should refer. As I have said, it was underwritten by Ms. Dennis, who at the relevant time worked in New York as Vice President and Marketing Manager for Boiler and Machinery Underwriting with CAN Commercial Lines. It was broked to her by Glenrand Risk Solutions, a division of Glenrand Marsh Limited, which now trades as Glencairn Limited and to whom I shall refer as "Glencairn".
  47. Mr. Nick Cook of Glencairn wrote to Ms. Dennis about this risk on 9 January 2001, and explained that the Syndicate was looking for reinsurance only in respect of onshore property. He sent her a draft slip for the reinsurance cover and a cover note for the direct cover, which referred to the conditions for the insurance of the on-shore property being "As per the expiring El Paso wording". Ms. Dennis did not ask for the wording of the policy that the Syndicate was writing or for the expiring wording, and she was not provided with it. On 25 January 2001 Mr. Cook told Ms. Dermis that they had a firm order for the reinsurance cover and invited her to underwrite, which she did by scratching the slip.
  48. Thus, Ms. Dennis wrote the reinsurance without seeing the wording of the direct cover and without seeing the wording of the expiring El Paso wording referred to in the cover note and the slip for the direct cover. I infer that Glencairn were aware of this: it would have been unrealistic for them to suppose that she would have obtained those wordings otherwise than from them.
  49. The claims under the direct cover

  50. The Syndicate learned of the Visbreaker loss on about 10 April 2001, and CGU, the first leading insurer, appointed McLarens Toplis as loss adjusters.
  51. The Syndicate learned of the D1AR loss on about 16 November 2001, when it was reported in these terms: "During startup following the power outage there was an internal temperature excursion in the RX-2501 reactor. The RX-2501 vessel ruptured as a result. Following the rupture, the reactor caught fire between the 2nd and 3rd catalyst beds...".
  52. In September 2003 El Paso presented claims under the direct cover in respect of both the Visbreaker loss and the D1AR loss, and also in respect of a smaller loss relating to the failure of a steam turbine ("the turbine loss") on 26 September 2001. I have not seen that claim document, but I understand that the claim in respect of the D1AR loss was supported by the CARC report to which I have referred (or possibly an earlier version of it). It stated that,
  53. "The damage was caused by cracks on the vessel wall and bulging of the vessel. The cracks started because the temperatures exceeded the design temperatures of the vessel. The design temperatures were exceeded because of four causes:
    • There were very high temperatures in the catalyst beds,
    • No action was taken on the alarms,
    • The temperatures were not monitored,
    • There might have been some faulty transmitters."
  54. Although the underwriters had appointed McLaren Toplis as loss adjuster in respect of the Visbreaker claim, they also appointed BCL to advise about the D1AR loss, and by December 2001 it seems that BCL had become the leading adjuster, and they prepared reports on the three losses, including the report about the DlAR claim.
  55. El Paso sought to aggregate the losses from the Visbreaker loss, the turbine failure and the D1AR loss on the basis that none of the three incidents would have occurred but for the Visbreaker loss, and so they argued that they should bear only one deductible. By an agreement dated 30 December 2003 the Syndicate and other underwriters agreed to compromise the three claims for a total of US$103 million. The agreement was governed by the law of Texas. Its recitals stated that El Paso submitted "a detailed claim to Underwriters dated September, 2003...", and that the underwriters "contest certain aspects of coverage under the facts presented and the terms and conditions of the Policies". It contained a provision that the agreement "is a compromise of disputed claims and shall not be construed as an admission or concession by any Party that any coverage under the Policy or obligation to pay exists or does not exist". It said nothing about the basis upon which the Syndicate (or any underwriter) agreed to compromise the claims or to pay the settlement sums.
  56. The evidence of Mr. McMahon was that he did not recall any issue about whether the D1AR loss fell within the direct cover, stating that the loss was clearly covered as the direct policy provided all-risks cover. His concerns were directed to how the loss related to earlier incidents at the refinery and whether underwriters might have a subrogated claim against a third party, a hope that was disappointed when it was learned that the operators whose errors led to the loss were employed by the insured.
  57. The BCL report referred to and rejected the insured's contention that the D1AR loss had a direct causal link with the turbine failure and should not attract its own deductible. It went on to state that "...operator error led to serious damage to D1A reactor when exceedingly high temperatures were generated during a process runaway", a runaway taking place, as Mr. McMahon described it in his evidence, when reactions between hydrogen used to pressurise the unit and the material in it continued uncontrolled and generated large amounts of heat. His evidence was that he regarded the extremely high temperatures as the proximate cause of the loss, expressing the opinion that the exothermic reaction between oil and hydrogen led to "the overheating of the vessel to the extent that the internal organisation of the structure collapsed and the exterior of the vessel weakened to the point that it could no longer withstand the pressures it was designed to contain." Mr. McMahon's evidence was that he did not give any consideration to whether the explosion exclusion in the reinsurance cover applied.
  58. It was Mr. Foulger's evidence that in settling the claim he "had regard to the various reports provided by BCL [and others], as well as participating in various market meetings", and as a result he was "satisfied that the losses claimed by the assured were valid claims under the Direct Policy, and that the settlement negotiated was a proper amount". He explained that his understanding from the reports, and in particular the BCL report, was that the loss "arose due to the heating of the D1AReactor unit, which followed a chemical reaction between the oil left in the reactor during a shut-down and the hydrogen used to re-pressurise the unit when the reactor was started-up again. The increase in temperature caused the various internal structures to melt and buckle, and the over heated exterior of the vessel weakened to the point that it could no longer contain the pressures that it was designed to withstand, resulting in it cracking and bulging".
  59. Mr. Foulger said that when settling the claim with El Paso, he did not consider whether the D1AR loss was from explosion. He did not consider the terms of the Syndicate's reinsurance cover, although he also gave evidence that, if when settling the loss he had considered whether the loss was the result of explosion, he would have concluded that it was not. There was no need for him to do so when considering the position under El Paso's insurance cover, and he rightly did not have regard to the Syndicate's reinsurance protections.
  60. I accept Mr. Foulger's evidence about why he entered into the settlement agreement and his thinking when he did so.
  61. The claims under the reinsurance

  62. CCC were first advised of the Visbreaker loss, the turbine loss and the D1AR loss by an email from Glencairn dated 5 March 2003, but they were then given little detail of the losses. On 9 April 2003, the Syndicate told CCC that they would not be pursuing the Visbreaker loss "as it is not boiler and machinery". On 12 January 2004 the Syndicate confirmed that the Visbreaker loss was "not a boiler and machinery matter" but stated that the D1AReactor loss (described as "reactor cracked and split") and the generator loss were.
  63. CCC refused to settle the D1AR loss, initially contending that the cause of the loss was "operator error" and so excluded from the definition of "Accident" in the Additional Conditions because it was a loss "caused by or resulting from a peril insured elsewhere in the reinsured policy". They no longer pursue this argument. As for the Visbreaker loss, CCC contended in their pleaded case that the Syndicate are estopped by the correspondence from asserting that the claim was covered by the reinsurance, but they abandoned that contention during the trial.
  64. The "Additional Conditions"

  65. I first consider the issue about how the Additional Conditions affect the scope of the reinsurance cover.
  66. The Additional Conditions were referred to by Ms. Dennis on the slip as ""B[oiler] & Machinery] terms & conditions". They are clearly not a complete set of standard terms and conditions, but on their face they are designed to relate to boiler and machinery insurance.
  67. CCC submit that the Additional Conditions are effective to restrict the cover provided by the reinsurance to losses from "Accident" (as defined) in respect of such property as falls within the definition of "Object"
  68. The Syndicate respond to this by pointing out that the statements in the reinsurance slip about "Type" and "Interest" do not use the term "Accident" or the term "Object". Therefore, they argue, the definitions of those terms in the Additional Conditions cannot define or impinge upon those provisions of the reinsurance contract. The only statement about what part of the direct all-risks cover of onshore property was included in the reinsurance is found in the provision about the Type of reinsurance cover, "Machinery breakdown, boiler explosion and business interruption following for onshore assets only". Accordingly the reinsurance applies to all losses, provided the loss was the result of an event that was either the breakdown of machinery or the explosion of a boiler. However, there is the provision that the reinsurance will "follow terms, clauses, conditions, exceptions ...of the original policy wording as far as applicable" to the reinsurance cover, and the Syndicate say that the Additional Clauses are to be understood to relate to terms so introduced into the reinsurance cover. The direct cover too does not use the term "accident" or "object" in the definition of cover, nor, I add, in defining the perils insured. On the Syndicate's argument, the words can relate only to the relatively incidental provision of clauses 10 and 21, and they are so to be construed.
  69. In support of this argument the Syndicate say that this is the only analysis of the relationship between the reinsurance slip and the Additional Conditions that makes sense of all the contractual language and that, since this analysis does not lead to a result that is commercially absurd or so unreasonable that the court is driven to search for some other meaning, it is to be adopted.
  70. As a matter of impression, it would be surprising first that the parties should decide to tinker with the effect of the follow the terms provision in the reinsurance cover in these two minor respects, the more so since there is no indication that Glencaird and Ms. Dennis otherwise showed such attention to detail, and secondly that they introduced such detailed changes so obliquely.
  71. However that may be, the basis of the Syndicate's argument is that their interpretation precisely respects the parties' language. In my judgment, it does not do so. The Additional Conditions are headed "Boiler & Machinery Coverage Defined". If the Syndicate's argument were accepted, they would not, to my mind, "define" the extent of the cover relating to boilers and machinery in any natural sense of the word. They would merely modify two relatively minor terms of the cover. I consider that CCC's interpretation of the reinsurance contract gives a role to the Additional Conditions that can naturally be regarded as a definition of coverage, and I prefer it. I add as a minor footnote that the definitions in the Additional Conditions do not precisely fit with the usage in clauses 10 and 21 of the wording of the direct cover in that in the Additional Conditions the terms "Accident" and "Object" consistently have capital initial letters and this is not so when the terms are used in the wording of the direct cover.
  72. I do not overlook the Syndicate's further arguments that any uncertainty should be resolved in their favour because:
  73. i) The "follow the terms" provision in the reinsurance slip militates in favour of a match between the terms of the insurance slip and the reinsurance slip, and therefore any mismatch between them should be minimised.
    ii) Any ambiguity is to be resolved contra proferentem and therefore against the reinsurers, both because they proposed the Additional Conditions for inclusion in the contract (and were proferens in contrahendo) and because they seek to rely upon the exception in the Additional Conditions (and are proferens coram judice): see Youell v Bland Welch. [1992] 2 Lloyd's Rep 127 at p.134.

  74. I am not persuaded by these arguments. As for the former, on either view the effect of the Additional Clauses is to modify the extent of the reinsurance cover, and while I would acknowledge the force of an argument that a mismatch should be avoided if it can be, I see no good reason, given that the Additional Conditions do introduce a mismatch, to construe them so as to minimise this. As for the invocation of the contra proferentem principles, I do not see any ambiguity in the effect of the Additional Clauses that justifies its application in either or both of its guises.
  75. I have thus far considered the interpretation and effect of the Additional Conditions without regard to the relevant background (or factual matrix) of the contract, and have preferred CCC's arguments about the meaning and effect of the Additional Conditions. However, this conclusion, it seems to me, is confirmed by a further consideration.
  76. As I have concluded, the parties were aware when they made the contract that Ms. Dennis had not seen the wording of the relevant section of the direct cover, or the expiring wording on which the new wording was to be based. Mr. Hunter argued that this is not properly to be considered when construing the contract because the parties' knowledge that Ms. Dennis had not seen the wording of the direct cover (either by way of wording actually agreed for the direct cover or by way of the wording for the expiring cover that it was to reflect) derived from the negotiations for the reinsurance, and it is well established that what the parties said or did in the course of negotiations is not a legitimate aid to construction. I am unable to accept this submission. There is properly a distinction between negotiating stances adopted by the parties and what they said and did during negotiations, and the knowledge to be attributed to the parties when they made the contract, whether or not the knowledge (or ignorance) resulted from the course of the negotiations. I consider that the limits of Ms. Dennis' knowledge about the direct cover, of which Glencairn were aware, are part of the background that can properly be considered when interpreting the slip including the Additional Conditions attached to it.
  77. The fact that Ms. Dennis did not know the terms of the underlying cover, and therefore did not know whether and if so in what context the terms "Accident" and "Object" were there used supports the conclusion that the purpose of the Additional Conditions was not to define those terms wherever they happened to be used in the wording that was adopted for the relevant section of the direct cover. It is unrealistic to suppose that Ms. Dennis would have introduced these definitions into the reinsurance contract without having any idea about where they would apply, and unrealistic to suppose that Glencairn understood her to be doing so.
  78. I add that Mr. Wynter urged me to find comfort, if not support, for CCC's arguments in the negotiations and subsequent conduct of the parties. Although such comfort might there be found were it permissible, these are, of course, illegitimate aids to construction and I ignore them.
  79. Conclusion about the Visbreaker loss

  80. The Syndicate accept that if CCC's case about the impact of the Additional Conditions upon the reinsurance contract is correct, CCC are not liable to them in respect of the Visbreaker loss.
  81. The dispute about the D1AR loss

  82. CCC say that the D1AR loss falls within the exclusion of loss or damage "from explosion", and therefore does not arise from "Accident" as defined in the Additional Conditions and is not within the reinsurance cover. They accept that but for this exclusion they would be liable for the DlAR loss.
  83. There is no dispute that the words "loss or damage ...from explosion" apply when, and only when, explosion is the proximate cause of the loss, and CCC's contention is that explosion was the proximate cause of the relevant loss.
  84. Here the burden of proof is upon CCC. This was questioned by Mr. Wynter and he drew my attention to the discussion in MacGillivray on Insurance Law (2003) 10th Ed. at para 19-7. However, the burden of proof, in my judgment, is dictated by a simple application of the rule that the burden lies upon the party who asserts a proposition and not the party who denies it: see Joseph Constantine SS Ltd v Imperial Smelting Corp Ltd.. [1942] AC 154 at p. 174 per Visc. Maugham. That said, I do not doubt that, as is stated in McGillivray (loc cit), "the prime duty of a court is to make up its mind how the loss occurred; the rules about burden of proof only need be resorted to when the evidence is insufficient for the court to make a decision".
  85. The Syndicate meet CCC's argument that the D1AR loss is the result of explosion with two submissions. First, they say that by agreeing to the "follow the settlements" clause CCC have precluded themselves from advancing this argument. Secondly, they say that in any event the proximate cause of the loss was not explosion, or at least that CCC have not shown that it was. Although I appreciate that the latter argument might be said to be logically posterior, it is convenient to consider it first.
  86. Was the D1AR loss "from explosion"?

  87. CCC say that damage to the D1AR unit was the result of explosion, in the sense of the term defined in the Shorter Oxford Dictionary as follows: "...the action of going off with a loud bang, or of bursting under the influence of suddenly developed internal energy".
  88. There is no dispute the question turns upon the meaning of the term "explosion" as a matter of ordinary usage. Neither party contends that I should seek some scientific, technical or specialist meaning of the word. Moreover, although the word "explosion" was used elsewhere in the Additional Conditions and indeed in the wording of the direct cover, no argument was advanced that this informs the meaning of the word in the exception on which CCC rely.
  89. CCC rely upon the evidence of Mr. Moore to support their submission that the D1AR loss was caused by explosion. His opinion is that there was a rapid and violent release of pressure from the vessel when it was ruptured. This was in part a release of hydrogen, and to that extent the explosion was a straightforward pressure explosion. He also identified another aspect of the explosion as a "boiling liquid expanding vapour explosion" or "BLEVE", involving a rupture or loss of containment resulting in a reduction in pressure and hence rapid boiling and vaporisation of oil, followed by an abrupt release of liquid at a temperature above its boiling point. The rupture is an intrinsic part of an explosion, whether or not it be a BLEVE: without it there can be no sudden loss of containment and no release either hydrogen or of boiling liquid and escape of vaporised oil or of anything else.
  90. Mr. Moore's opinion that there was a BLEVE was challenged in the course of his cross-examination on two fronts.
  91. First, it was suggested that he used the term too broadly, and BLEVE is not an appropriate term to describe any rupture of a vessel where there is a release of boiling liquid or vapour. The Syndicate put in evidence articles written by a Professor A M Birk of Queen's University, Kingston, Ontario, who distinguished a BLEVE from "ruptures with transient jet releases". According to Professor Birk, "For a BLEVE to happen there must be a combination of a weakened tank state and enough liquid and vapour energy in the tank to drive the failure the full length of the tank".
  92. I accept, as did Mr Moore, that there are two respectable views about what amounts to a BLEVE, and it is of interest that the term is not universally adopted as a term appropriate to describe all ruptures caused by internal pressure accompanied by an expulsion of liquid or vaporised contents of the ruptured vessel. However, in the end, the importance of this evidence is limited. As I have said, I am not concerned the appropriate usage of the term BLEVE as a matter of scientific or technical usage, but with what amounts to "explosion" as a matter of ordinary usage.
  93. Secondly, the Syndicate say that there is no proper basis for Mr. Moore's argument that this event had the characteristics of what he would call a BLEVE. I have concluded that it has not been shown that boiling or vaporised oil was expelled from the reactor, and therefore I accept that submission. Moreover, if there had been a BELVE, there would, as Mr. Moore explained, have been a plume of fire or flare, and, as I have concluded on the balance of probabilities, there was none.
  94. In my judgment, therefore; CCC have not established that there was a BLEVE in any technical sense of the term, but it is neither necessary nor sufficient for them to do so. What matters is whether they have shown that there was explosion in the natural and ordinary sense of the word, and if so whether it was the proximate cause of the loss. The Syndicate dispute both these points.
  95. I am not persuaded that there was any explosion at the D1AR unit on 6 November 2001. In the end, the question is perhaps much one of impression. To my mind the word connotes violence and manifest violence of a kind which has not been shown to have been present in this case, and it connotes a shattering destruction which too is absent. There is no evidence of any blast damage or other significant damage outside the reactor. I have concluded that, although the noise associated with the rupture would have been marked, it was not dramatic. There is no evidence of any puncture to the wall of the reactor other than a clean 4 inch tear at one point on the wall of the reactor, the area of which, I calculate on the imprecise assumption that the diameter of the vessel was 12'6" for its whole height, was some 3,250 square feet.
  96. This conclusion seems to me consistent with the decision and judgment of Mr. Justice Staughton in Commonwealth Smelting Ltd. v Guardian Royal Exchange Insurance Ltd., [1984] 2 Lloyd's Reports 608 and the judgments of the Court of Appeal, [1986] 1 Lloyd's Reports 121, upholding his decision. In that case, a piece of machinery called as blower was destroyed when its steel impeller and casing shattered into bits, making large holes in the walls of the building that housed it. Mr. Justice Staughton identified (at p.612) as characteristics of an explosion that it is (i) a violent event, (ii) a noisy event, and (iii) an event "caused by a very rapid chemical or nuclear reaction, or the bursting out of gas or vapour under pressure". I am not persuaded that the event in this case had the first two qualities to the extent characteristic of an explosion. I do not understand that Lord Justice Parker (with whose judgment the other members of the Court of Appeal agreed) adopted a significantly different approach from Mr. Justice Staughton to interpreting the term "explosion". Lord Justice Parker (at p. 126) said that, "It is, I think, conceivable that one might say that some of the damage ... was at any rate contributed to by an explosion, namely the explosion of the air from its confinement within the casing, the casing having been shattered by the entirely independent operation of the failure of the impellor". That observation was made in the context of a case where the violence of the event was obvious and of an appeal where the principal criticism of Mr. Justice Staughton was (see loc cit p.125) "that he adopted a dictionary meaning of the word which was more a scientific meaning than a popular meaning, and that he went behind what was the obvious manifestation, which was one of an explosion, to look at the cause which lay behind the explosion".
  97. However, even if I am wrong in this conclusion, the Syndicate have another argument: they rightly point out that CCC have to show that the explosion was the proximate cause of the loss, that is to say the physical damage to the unit. (El Paso were, of course, insured against business interruption as well as physical damage, but that cover was for business interruption following physical loss caused by an insured risk.).
  98. This leads to the question what precisely CCC seek to characterise as "explosion". If they isolate the rupture of the wall from the other bulging and cracking, they cannot show that the rupture caused the physical damage generally. Mr. Moore agreed that the BLEVE (as he characterised the rupture) occurred after the cracking and bulging. He must be right in as much as cracking and bulging occurred before the wall was split and pressure released.
  99. The only way that CCC can present their case, therefore, is to argue that all the cracking and bulging, including the split in the wall, are to be taken as a single event and that event as a whole is to be characterised as "explosion". However, this does not seem to me a natural use of language. Taken as a whole, the physical damage was to be characterised as bulging and cracking. Indeed, this is the predominant description of it in the reports. Certainly one crack did go right through the wall and as a result hydrogen was released, but, to my mind, it entirely artificial to elevate this incidental aspect of the damage to become its defining feature.
  100. I therefore reject CCC's argument that the D1AR loss was loss from explosion within the meaning of the Additional Conditions, and it follows that the Syndicate's claim in respect of the DlAR loss succeeds.
  101. The "follow the settlements" provision

  102. In view of this conclusion, the Syndicate do not need to rely upon the follow the settlements provision. I would have rejected the Syndicate's argument about this, and, having heard full argument, I should give my reasons.
  103. The Syndicate say that the factual basis upon which they settled the claim was that the proximate cause of the loss was that "... operator error led to serious damage to D1A reactor when exceedingly high temperatures were generated during a process runaway" and therefore, the argument goes, on the basis that the proximate cause was not explosion. They then argue that the "follow the settlements" provision prevents CCC from disputing the factual basis upon which they settled the claim made by El Paso in respect of the DlAR loss, and that CCC are therefore confined to arguing that upon that factual basis the loss, as a matter of law, did not fall within the reinsurance cover.
  104. The factual premise of the argument is therefore that the Syndicate compromised the claim on the basis that the proximate cause of the D1AR loss was not explosion. The Syndicate do not rely upon the Settlement Agreement to establish this, and indeed there is nothing in the Settlement Agreement that they could rely upon. Nor do they rely upon the specific nature or basis of the claim made by El Paso that was settled. The argument is that what matters is the Syndicate's basic reason for compromising El Paso's claim as they did. (At one point, Mr. Hunter went so far as to submit that CCC could not dispute anything in the loss adjusters' reports but that argument was not developed. It would not have assisted the Syndicate to establish a factual basis for their argument.)
  105. 8S. Mr. Foulger's evidence was that when settling the El Paso claims he relied upon the BCL report. I have set out the most relevant parts of it. It recited the argument of El Paso that the proximate cause of the loss was the turbine failure on 26 September 2001, and rejected that contention. It said that there was serious damage to the reactor "when exceedingly high temperatures were generated", but it said nothing of the manner in which the high temperatures caused the damage: that is to say, it did not consider whether they brought about an event that was properly to be described as "explosion".

  106. I therefore conclude that the Syndicate have not shown that the basis upon which they settled the claim was that, as a matter of fact, the loss was not from explosion. All that they have shown is that they compromised the claim without specifically recognising that the loss was the result of explosion, but there was no reason for them to consider whether the chemical reactions and associated heat and pressure manifested themselves in, and amounted to, explosion. What they recognised did not exclude the possibility that the loss resulted from explosion.
  107. In any case, even if the Syndicate had established the factual premise for their argument, I would have rejected their submission that the follow the settlements provision precludes CCC from contending that the proximate cause of the loss was explosion.
  108. A reinsurer is prima facie obliged to indemnify the reassured for losses that fall within the scope of both the insurance cover and the reinsurance cover, and not otherwise. CCC's contention is that the D1AR loss falls outside the reinsurance cover, and the question is whether the follow the settlements clause provides the Syndicate with an answer to that contention.
  109. The Syndicate's proposition that CCC are bound by the basis of which they (the Syndicate) "recognised" the settlement with El Paso reflects, and I would suppose derives directly or indirectly from, a much-cited passage from the judgment of Robert Goff LJ in The Insurance Co of Africa v Scor (UK) Reinsurance Co Ltd. [1985] 1 Lloyd's Rep 312 at p.330, but there the notion of what the reassured "recognised" referred to the claim that he decided to pay (in whole or in part) and not to his thinking when he did so. The relevant passage of the judgment is this:
  110. "In my judgment, the effect of a clause binding reinsurers to follow settlements of the insurers, is that the reinsurers agree to indemnify insurers in the event that they settle a claim by their assured, i.e. when they dispose, or bind themselves to dispose, of a claim, whether by reason of admission or compromise, provided that the claim so recognized by them falls within the risks covered by the policy of reinsurance as a matter of law, and provided also that in settling the claim the insurers have acted honestly and have taken all proper and businesslike steps in making the settlement. ... In particular, I do not read the clause as inhibiting reinsurers from contesting that the claims settled by the insurers does not, as a matter of law, fall within the risks covered by the reinsurance policy; but ... I do consider that the clause presupposes that the reinsurers are entitled to rely not merely on the honesty, but also on the professionalism of insurers, and so is susceptible of an implication that the insurers must have acted both honestly and in a proper and businesslike manner."

  111. In the Scor case, the reinsurers did not reassure only some of the risks covered by the direct insurance but all of them: in that sense at least, the insurance and the reinsurance were fully back to back. By agreeing to the follow the settlements provision, the reinsurers agreed that if the reassured settled a claim by their insured, they (the reinsurers) would not dispute the settlement provided it was made honestly and in a proper and businesslike manner. Subject to the proviso, the reinsurers agreed not to challenge such a settlement directly and therefore implicitly agreed not to challenge it indirectly either: that is to say, not to challenge the factual basis for it under the guise that, while not questioning whether the claim fell within the direct cover, they did not accept that it fell within the reinsurance cover and required the reassured to prove that it did. Since the covers were back to back, it followed that the reinsurers could question whether the loss was within the reinsurance cover consistently with their obligation to follow settlements was only by raising legal arguments.
  112. The Syndicate seek to apply this reasoning here although the reinsurers reinsured only some of the risks that fell within the original cover and in that sense the insurance cover and the reinsurance cover are not back to back. Consequently it does not follow from the fact that a loss falls within the original cover that it falls within the reinsurance cover or would do so subject only to legal questions about what the reinsurance covers, and for the reinsurers to question whether a loss settled by the reassured under the direct cover falls as a matter of fact within the reinsurance cover does not necessarily challenge the insurers' decision to pay or to compromise the claim, and so to recognise it to that extent as one that falls within the direct insurance cover.
  113. The starting point in considering the Syndicate's argument is the wording and meaning of follow the settlement provision itself because, as I need hardly say, every case depends upon the proper construction of the particular reinsurance contract. As Lord Mustill pointed out in Hill v Mercantile & General Reinsurance plc, [1996] 1 WLR 1239 at pp.l251F-1252B, a follow the settlements provision is to be interpreted against the background of two rules,
  114. "First, that the reinsurer cannot be held liable unless the loss falls within the cover of the policy reinsured and within the cover created by the reinsurance. Second, that the parties are free to agree on ways of proving whether these requirements are satisfied. Beyond this, all problems come from the efforts of those in the market to strike a workable balance between conflicting practical demands and then to express the balance in words.
    These practical demands can be seen most easily in the context of traditional reinsurance, where the party reinsured is the insurer under a contract made directly with the person whose property or other interest is at risk. Two impulses act in opposite directions. The first is to avoid the investigation of the same issues twice; and, moreover, an investigation on the second occasion by a reinsurer whose knowledge of what happened when the risk was written, and whose facility for investigating the claim, are inferior to those of the direct insurer. The second impulse, acting in the other direction, is to ensure that the integrity of the reinsurer's bargain is not eroded by an agreement over which he has had no control.
    This conflict is quite easily managed when the insurance and the reinsurance are on the same terms and where the parties are essentially co-adventurers; for example, in participatory reinsurance, or facultative reinsurance with a large retention. Here, the interests of the direct insurers and the reinsurers are broadly the same, and it is not imprudent for the reinsurers to put themselves unconditionally in the hands of the reinsured for the settlement of claims which will be passed on to them.
    The problems are more acute when ... the terms of the successive policies are not the same ... For example, in the former case it may well happen that a claim under the direct policy does not require the determination of issues which are crucial to liability under the reinsurance: as happened in the "constructive total loss" cases like Chippendale v Holt, (1895) 1 Com. Cas. 197; and indeed in the present case where there can be no doubt that the loss, whatever exactly it was, fell within the direct contracts, whereas this was not necessarily the case under the reinsurances."

  115. In this case, the follow the settlements obligation is "to follow ... settlements of original policy wording as far as applicable". The syntax is governed by the follow the terms limb of the provision, but I interpret the term "settlements of original policy wording" to refer to a settlement under the original policy wording, and there is no dispute that the settlement of the claim for the D1AR loss was such a settlement. It is not clear whether the adjectival phrase "as far as applicable hereto", that is to say applicable to the reinsurance cover, qualify as a matter of grammar the word "settlements" (and "terms, clauses, conditions, exceptions") or qualify "original policy wording", but on either view it reflects that the reinsurance cover is not fully back to back with the direct cover, and its effect, as I interpret it, is to restrict the effect of the follow the settlements obligation to settlements to which the reinsurance cover is applicable.
  116. This leads to the question whether the phrase "as far as applicable" restricts the obligation to settlements of claims in respect of losses of El Paso's property resulting from risks covered by the reinsurance, or whether it suffices that the settlement made by the reassured would be covered by the reinsurance on the basis of the facts recognised by the reassured. This question must be answered, and effect must be given to the phrase, bearing in mind that the subject of the reinsurance cover was El Paso's property, and not claims made by the Syndicate: see Toomey v Eagle Star. [1994] 1 Lloyd's Rep 516 at p.523 and Wasa International Ins. Co. Ltd. v Lexington Ins Co.. [2007] EWHC 896 (Com) at para. 33.
  117. The Syndicate, as I understand it, argue for the latter interpretation. Building on the observations of Lord Mustill that a follow the settlement provision can be seen as an agreement by the parties as to how it can be proved that the loss falls within both the insurance and the reinsurance covers, they contend that on the true interpretation of the follow the settlement provision the reassured can rely upon the basis of the settlement to order to show that that method of proof is "applicable". This seems to me to be a circular argument: effectively it prays in aid the follow the settlement provision in order to establish that the follow the settlement provision is applicable. It seems to me that the natural meaning of the qualification "as far as is applicable" is that the follow the settlement provision applies when the risk is one covered by the reinsurance, and that to interpret it more narrowly is to elevate the first of Lord Mustill's "impulses" at the expense of the second in a way that does not properly respect the wording of the parties' agreement.
  118. However, even if the Syndicate are not prevented from relying upon the follow the settlements provision because it is qualified by the phrase "as far as applicable", I do not consider that the Syndicate can rely upon it to preclude CCC from disputing the basis on which they settled the claim, or the facts that they recognised as the basis for settling it. It is perhaps worth again going back to the wording of the follow the settlements provision itself and observing that CCC agreed to follow settlements, not the basis of settlements or the Syndicate's reasons for settlements.
  119. It is also worth recalling Lord Mustill's observation in Hill v Mercantile & General Reinsurance Co plc. [1996] 1 WLR 1251H-1252A that where the interests of direct insurers are broadly the same, it can be seen that it is "not imprudent for the reinsurers to put themselves unconditionally in the hands of their reinsured for the settlement of claims which will be passed on to them". The implication of the Syndicate's argument is that CCC "put themselves unconditionally in the hands of their reinsured" not only as to whether the claim fell within the direct cover about which they had similar interests and as to deciding whether to recognise facts that bear upon that question, but as to matters relevant only to whether the claim fell within the reinsurance cover, about which the Syndicate and CCC had directly opposite conflicting interests.
  120. For these reasons I would not accept the Syndicate's argument on this point unless it is supported by compelling authority. I must therefore consider whether there is such authority. The distinction drawn by the Syndicate between factual matters and legal questions reflects a distinction found in Scor, in Hill v Mercantile and General Reinsurance Company and in Assicurazioni Generali SpA v CGU International Insurance plc. [2003] EWHC 1073 (Comm), [2003] 1 Lloyd's Rep. 725 per Mr. Gavin Kealey QC sitting as a Deputy High Court Judge, and [2004] EWCA Civ 429, [2004] 2 All ER (Comm) 114 per the Court of Appeal. However Scor was a case in which, as I have pointed out, the insurance cover and the reinsurance cover were fully back to back, and the reinsurers were confined to challenging whether the claim fell within the reinsurance cover on a legal basis because a factual challenge would necessarily involve challenging whether the settlement was within the insurance cover. As far as Hill is concerned, I understand the distinction when made by Lord Mustill to be drawn from the particular wording of the follow the settlements provision in that case, which was subject to a proviso that the settlements binding upon the reinsurers were to be "within the terms and conditions of the original policies and/or contracts and within the terms and conditions of this reinsurance". Lord Mustill said at p.l252H, "the crucial words are "within the terms and conditions" of the original policies and of the reinsurance". To my mind these draw a distinction between the facts which generate claims under the two contracts, and the legal extent of the respective covers: the purpose of the distinction being to ensure that the reinsurers original assessment and rating of the risks assumed are not falsified by a settlement which, even if soundly based on the facts, transfers into the inward or outward policies, or both, risks which properly lie outside them." I therefore do not consider that the Syndicate's argument is supported by the distinction between factual matters and legal questions identified in Scor or in Hill v Merchant and General.
  121. In Assicurazioni Generali SpA v CGU International Insurance plc, the insurance cover and the reinsurance cover were back to back: thus Tuckey LJ (loc cit at para. 1) as able to formulate the first question that arose on the appeal was "To what extent does the first proviso [sc the first proviso identified by Robert Goff LJ in the passage in his judgment in Scor that I have cited, loc cit at p.330] preclude reinsurers from raising coverage issues relating to terms which are the same on both the insurance and the reinsurance contracts".
  122. As I understand it, it was in this context that Mr. Gavin Kealey QC in his first instance judgment engaged upon consideration of what the reinsurer was bound to accept because of the reassured's settlement and what points remained available to him to argue as falling with Robert Goff LJ's first proviso. To this end, he considered in what circumstances a settlement made under the direct cover should be taken to involve the reassured accepting (by settling the claim) a matter relevant to the settlement under the direct cover so that the reinsurer would be challenging the basis of the settlement under the direct cover if he were permitted to dispute it in the guise of questioning whether the claim fell within the reinsurance cover. Mr. Kealey pointed out that for this purpose it is not enough to look only to the settlement itself but it is necessary to look behind it in order to discern, when a payment was made despite an argument against the claim being available to the reassured, the recognition of the claim reflected a recognition of facts which the reinsurer too should be bound to acknowledge or whether the reassured had settled because he failed to appreciate that there was a legal defence. Of course, the reassured could hardly claim that the settlement reflected that he recognised the factual basis of the claim rather than overlooked a legal question if the claim was not arguably one that could overcome a legal defence. Mr. Kealey developed his examination of this distinction in his judgment at paragraph 51 where he explored the distinction drawn by Mr. Justice Evans in Hiscox v Outhwaite (No 3). [1991] 2 Lloyd's Rep 524 between settlements of claims about there was no (or no realistic) dispute and settlement where there was a genuine dispute about whether the losses fell within the insurance cover as a matter of law. This, I think is why in the Court of Appeal in the Assicurazioni case Lord Justice Tuckey (loc cit at para 18) stated the position as follows:
  123. "... none of the earlier cases have considered how the first proviso [in Scor] works in practice in a case such as this. Here the judge has done so. By reference to the words "the claim so recognised" he has concluded that the insurers do not have to show that the claim they have settled in fact fell within the risks covered by the reinsurance, but that the claim which they recognised did or arguably did. I think this gives effect to what Robert Goff LJ said and gives some sensible added meaning to the clause. It gives substance to the fact that the reinsurer cannot require the insurer to prove that the assured's claim was in fact covered by the original policy, but requires him to show that the basis on which he settled it was one which fell within the terms of the reinsurance as a matter of law or arguably did so. This and the need for the insurer to have acted honestly and taken all reasonable and proper steps in settling the claim provide adequate protection for the reinsurer."

  124. However this may be, neither Mr. Kealey nor the Court of Appeal was considering the position where the insurance cover and the reinsurance cover are not back to back. Their judgments give no support for the proposition that a reinsurer agrees to be bound in a dispute about whether the reinsurance covers a claim by the views or assessment of the reassured when settling a claim when the question is relevant only to whether the loss falls within the reinsurance cover and not to whether it falls within the insurance cover. Indeed, at paragraph 44 of his judgment Mr. Kealey said,
  125. "... I should not ignore a further possible case where the insurers settle a claim on the basis of facts that demonstrate that the loss claimed by the assured is fair wear and tear in circumstances where the insurers overlook the legal implications of those facts and ignore the exclusion clause. In this last case, it seems to me (irrespective of the possible relevance of the second Scor proviso) that, according to the basis on which it was settled and in the absence of any recognition by admission or compromise that the exclusion in respect of fair wear and tear did not or arguably did not apply, the claim so recognised falls outside the contract of reinsurance as a matter of law."

  126. For my part, I cannot see any sensible basis for distinguishing a case where there was a clause in the insurance contract that was overlooked (and therefore the reassured did not consider whether it provided a defence under the insurance cover and so, in a case of fully back to back insurance and reinsurance covers, under the reinsurance cover), and a case where the reassured did not consider the question simply because there was no relevant clause in the insurance contract and the question comes into play only when a question of the applicability of the reinsurance cover arises.
  127. I cannot accept, therefore, that there is any support in the authorities for the Syndicated argument. Even if the factual premise for it were established and quite apart from my understanding of the effect of the phrase "as far as applicable", I would reject the Syndicate's submission based on the "follow the settlements" provision.
  128. Conclusion

  129. I therefore conclude that the Syndicate succeeds in respect of the D1AR claim and is to be dismissed in respect of the Visbreaker claim. I understand that the quantum of the D1AR claim is agreed to be US$826,875.


BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII
URL: http://www.bailii.org/ew/cases/EWHC/Comm/2007/1762.html