Rubicon Vantage International Pte Ltd v Krisenergy Ltd (Rev 1)

The Claimant (“Rubicon”) is a Singaporean Company and it owns a Floating Storage and Offloading Facility called the Rubicon Vantage (“the Vessel”). By a bareboat charter (“the Charter”) dated 13 October 2014, Rubicon chartered the FSO to Kris Energy (Gulf of Thailand) Limited (“Kegot”), a company incorporated in the Cayman Islands. Kegot is a wholly owned subsidiary of the Defendant (“Krisenergy”) which is also incorporated in the Cayman Islands.

By clause 22.2 of the Charter, Kegot was obliged to procure for the Claimant a “Charterer Guarantee”. The terms of the required “Charterer Guarantee” were set out in Exhibit E to the Charter. Krisenergy did indeed provide a guarantee to the Claimant, albeit not in fact quite in the terms of Exhibit E, and I shall refer to the document that was in fact provided, on or about 13 October 2014, as “the Guarantee”. The main issues in this case concern the proper interpretation of the terms of this Guarantee.

By the terms of the Charter Rubicon was to organise various works on the Vessel before the charter term commenced. Most of these works were directed to extending the working life of the Vessel, and in the Charter they are called the Life Extension Works. Rubicon was responsible for the cost of those works by clause 2.1, but there was a

On 21 October 2014 (shortly after the Charter was signed) the Vessel arrived at Unithai shipyard for the various works to be performed. The Life Extension Works and the skid integration works were all performed, and on 27 March 2015 a sail-away certificate was issued confirming the acceptance of the Vessel by Kegot.

The Vessel arrived at the oil field on 24 July 2015, whereupon charter hire began to fall due at the agreed rate of $30,650 per day. On the evidence before me, charter hire has been paid on time throughout the Charter.

On 3 June 2015 a series of invoices were sent by Rubicon to Kegot. There are four such invoices which give rise to this dispute. Each of them was said to relate to a Variation Order. The four disputed invoices, and the VOs to which they referred, are as follows:

Various discussions and negotiations ensued over the next three years. There was a meeting on 19 April 2018 at which Rubicon contends, but Krisenergy denies, that an admission was made by Kegot that some sums were due. In any event, nothing has been paid by Kegot in relation to any of the four disputed invoices.

The parties are agreed that the disputes engendered by these four invoices do not call for resolution in these proceedings.

On 3 September 2018 Rubicon made a demand on Krisenergy under the Guarantee for the total sum outstanding under the four invoices. By letter of 6 September 2018 Krisenergy declined to pay. These proceedings were commenced on 2 November 2018. A second demand was made on 29 January 2019, which added a claim for interest.

Rubicon contends that the Guarantee is, at least in part, an on-demand instrument, that it has made compliant demands, and that Krisenergy is therefore liable to pay notwithstanding that the underlying claims against Kegot are in dispute and have not been adjudicated upon. Krisenergy accepts that the Guarantee is, in part, an on-demand instrument, but says that it is only an on-demand instrument where liability has been admitted by Kegot (even if quantum remains in dispute), and Krisenergy says there has been no such admission of liability. Krisenergy further says that the demands do not comply with the terms of the Guarantee, so that no liability has arisen under it. Rubicon says in reply that there were admissions by Kegot, so that even if Krisenergy is right on the construction point, the Guarantee liabilities are triggered. Accordingly, the issues which arise are:

In the Guarantee, Kegot is the Company, Krisenergy is the Guarantor, and Rubicon is the Contractor. The main relevant provisions of the Guarantee are as follows:

The law of guarantees has in the past been bedevilled by difficulties in determining whether a particular instrument, on its proper construction, merely imposes a secondary obligation on the guarantor to “see-to-it” that the principal obligor’s obligations are met, or whether it imposes an autonomous obligation on the “guarantor” to pay, irrespective of the actual liability of the primary obligor. The first type of instrument is sometimes characterised as being a “true guarantee”, and such instruments are often issued by parties who are not banks or financiers but who have some other commercial relationship with the primary obligor. One particular type of such an instrument is a parent company guarantee, where a parent company guarantees the liabilities of its subsidiary. The second type of instrument often takes the form of an on-demand bond, or performance bond, and such instruments are typically issued by banks, who, though they are likely to be in a poor position independently to assess the true merits of an underlying claim, are prepared to take on an obligation to pay against documents. Typically in such instruments the bank must simply pay against a compliant demand, but if it turns out later that in fact the principal obligor is not liable, the bank is entitled to recover its money from the payee.

First instance judges and commercial parties who want to know into which category a particular instrument falls now have the benefit of two Court of Appeal cases which set out what are described as presumptions which can be applied if the appropriate preconditions for their application are made out.

In Marubeni Hong Kong v Mongolian Government [2005] EWCA Civ 395 the Court of Appeal held (per Carnwath LJ at #30) that, in a transaction outside the banking context, the absence of language appropriate to describe a demand bond or something of similar effect created a strong presumption against construing the document as an autonomous bond rather than merely as a see-to-it guarantee imposing only secondary liability.

In Wuhan Guoyu Logistics Group Co Lt v Emporiki Bank of Greece SA [2012] EWCA Civ 1629 the Court of Appeal held (per Longmore LJ, #26 to 29) that, where an instrument (i) related to an underlying transaction between the parties in different jurisdictions, (ii) is issued by a bank, (iii) contains an undertaking to pay “on demand”

Perhaps rather unusually, the Guarantee in this case falls into both categories. It is described as being a parent company guarantee, and it is as a matter of fact a guarantee provided by Krisenergy, which is the parent company of the principal obligor Kegot. The parties agree that its effect is to make Krisenergy liable to pay any amount which Kegot is liable to pay but has not paid, provided that a compliant demand is made. But the parties also agree that there are some circumstances in which Krisenergy can be liable to pay in response to a compliant demand even if there is an unresolved dispute between Rubicon and Kegot.

Where the parties differ is as to precisely what those circumstances are. The parties formulated the issue in these terms:

There is also a dispute as to how I should approach this question of construction. Mr Eaton QC, for Krisenergy, submits that I should apply the Marubeni presumption, and that that should lead me to construe restrictively the scope of Krisenergy’s autonomous on-demand liabilities. Although he accepts that Marubeni is directed to the question of whether an instrument is an autonomous on demand instrument or a see-to-it instrument, he submits that I should apply it by analogy. I am unable to accept this submission. The Marubeni presumption is directed to the question of whether a particular instrument should be construed as imposing autonomous obligations, or merely as a see-to-it guarantee. Once it is accepted, as it is in this case, that a particular instrument operates, to some extent, to impose autonomous liabilities, it seems to me that the role of the Marubeni presumption is spent. I cannot see any logical reason why the mere fact that the party which submits itself to an autonomous on-demand obligation is not a bank, is in general likely to assist in determining the extent of the obligation it has undertaken, and I accordingly reject the submission that the fact that Krisenergy is not a bank raises any form of presumption of construction that the ondemand obligation it undertook is to be construed narrowly rather than broadly. In my view the correct approach in determining the extent of the on-demand obligation in this Guarantee is to begin simply by considering the words the parties chose to use to record their agreement, free from any antecedent presumption as to what meaning they are likely to have, or as towards a wide or narrow construction.

Adopting that approach, the meaning of clauses 4 and 5 is, in my view, clear. Both, of course, are predicated on the assumption that there is a valid demand, in the sense of a demand complying with clause 3. But once that hurdle is overcome the position is as follows.

First, by clause 4, if the amounts demanded are not in dispute as between Rubicon and Kegot, then Krisenergy must pay them within 48 hours of receipt of the demand. The operative wording used is “... where the amount(s) demanded are not in dispute”. That must in my view mean that the amounts demanded are not in dispute either as to liability to pay, or as to the quantum of what must be paid. Krisenergy must pay only if and insofar as there is a dispute neither as to liability nor as to quantum. So, if a there is a demand for $3m, and Kegot does not dispute liability to pay $2m of that sum, it must, under clause 4, pay $2m, but it is not obliged, under clause 4, to pay the other $1m.

Clause 5, in my view, is directed to what is left over from clause 4. It is engaged if and insofar as there is a dispute between the parties as to Kegot’s liability to pay some part of the sum or sums demanded. Such a dispute might arise because of an in-principle or liability-based objection to the claim, or it might arise because Kegot contends that the claim, although good as to part, is overstated as to the balance. But there is an important limitation which applies to clause 5 liabilities, which is that Krisenergy is only obliged to pay, under clause 5, up to a maximum of $3m.

The opening words of clause 5 are “In the event of dispute(s) ... as to [Kegot’s] liability in respect of any amount(s) demanded under this Guarantee.” Mr Eaton submits that this limits clause 5 to disputes which are only about quantum, that is to say only about the quantum of the sum demanded. I do not consider that these words can bear that meaning. A “dispute ... as to liability ...” means what it says: a dispute as to liability. It is not limited to disputes as to quantum by the words that follow.

Mr Eaton further submits that the references to “disputed amounts” in clause 5 suggest that clause 5 is directed to liabilities which are disputed only because of a dispute as to quantum. In my view “disputed amounts” simply means a sum (or sums) as to which liability is disputed. It does not matter whether the dispute arises because the entirety of a particular claim is disputed on grounds of liability or whether the sum which is disputed is part of a larger sum in relation to which liability in principle is admitted, but subject to a dispute as to quantum which means that liability for this part of the larger sum is disputed.

Mr Eaton further submitted that there was a tension between clause 5 and clause 10. He submitted that the effect of clause 10 is that the Guarantor could not pay out more than had been admitted due, for otherwise it would, or at least might, find itself in a position in which it had a greater liability than the Contractor’s liability, and that is not allowed by clause 10. This argument fails because in my view clause 10 is directed to the Guarantor’s ultimate liability at the end of the day, after any adjustment has been made under clause 6 to achieve repayment of sums paid under clause 5 which turned out ultimately not to be due. But even if that were not the case, the clause 10 argument does not assist Mr Eaton. Suppose that clause 5 were, as he contends, limited to demands where liability has been admitted but quantum was in dispute. The Guarantor would still be placed in a position in which it might, pending a final clause 6 accounting, have paid out a greater sum than that for which Kegot was ultimately liable. In other

Finally Mr Eaton notes that under clause 13.2 of the Charter, where an item billed is disputed in good faith, it is not payable until any dispute has been resolved. He says it would make no commercial sense for the Guarantee to impose on Krisenergy an obligation to pay immediately on demand on account of items for which Kegot is expressly not, at that stage, liable to pay under the Charter. I disagree. On the contrary, it might well make good commercial sense that, although the subsidiary is not liable to pay on an interim basis, the parent is. The arrangement overall amounts to a pay-now argue-later regime in relation to the first $3m that is in dispute as between Kegot and Rubicon, except that the pay-now payment is to be made by Krisenergy, not Kegot. Imposing the pay-now obligation on the parent rather than the subsidiary has obvious attractions for the payee Rubicon, and may even be preferable from the point of view of Krisenergy and Kegot since it may well be easier for Krisenergy to weather the cash flow strain of a $3m payment.

I note for completeness that Appendix 1 to the Guarantee contains some worked examples to show how the $3m limit is intended to operate if there are a series of claims. It is entirely consistent with the Claimant’s construction, but it has not helped me decide as between the rival constructions.

Accordingly on this first issue I find that the on–demand liability arises in relation to the first $3m worth of claims in relation to which Krisenergy disputes that it is obliged to pay. It does not matter whether the dispute is as to liability or merely as to quantum.

The next issues are questions about the meaning of clause 3, the text of which I repeat for convenience.

The first point to note is that any demand made under clause 3 might lead to monies being paid under the see-to-it obligation of clause 4, or under the primary obligation under clause 5, or under a combination of the two. This particular provision, like the agreement as a whole, therefore covers both primary and secondary type obligations.

The parties agree that a demand is invalid unless it is accompanied by a sworn statement satisfying limb (a) and limb (c), and they agree that both the first and second demand

It is clear that something has gone wrong with the wording of clause 3 because it does not make good grammatical sense. It makes poor sense to require “a statement ... stating as follows: ... (b) the calculation of such sums together with any supporting documentation reasonably required”. Accordingly I cannot apply the literal meaning of the words used, and on the authority of Rainy Sky SA v Kookmin Bank 2011 UKSC 50, and given that this is a commercial contract, my task is to determine what the parties meant by the language used, which involves ascertaining what a reasonable person would have understood the parties to have meant. The relevant person for these purposes is one who has available all the background knowledge which would have been reasonably available to the parties.

I have no doubt that such a person would understand the parties to have intended a requirement that the demand actually be accompanied both by the calculation of the sums demanded and by any supporting documentation reasonably required to assess the demand. In other words the clause as whole should be construed as if it read like this: Any demand under this Guarantee shall be in writing and shall be accompanied by

Such a reading amounts to no more than a reordering of the actual words used, and it makes good commercial sense. When Krisenergy receives a demand it will need to decide within 48 hours whether to pay. That requires it to decide whether this is a demand for amounts that are not in dispute (triggering clause 4), or for amounts that are in dispute (triggering clause 5), or a mixture of the two. Krisenergy can ask Kegot whether and to what extent the claim is disputed, but it needs to know what the claim is for and how the sums demanded are calculated in order to ask that question. It may also need some supporting documentation in order to be clear what the claim is for. Krisenergy might also reasonably require some supporting documentation in relation to the underlying claim in order to satisfy itself that the claim, albeit it is disputed, is made bona fide. But whatever it is that is reasonably required in any particular case, Krisenergy needs the calculation and needs the documents: it is no use to Krisenergy to be told that the calculation or the documents are attached to the demand if in fact they are not, particularly given the tight timescale for payment.

For that reason Miss Selvaratnam’s alternative construction makes no commercial sense. It would also require a more radical rewriting of the words which the parties actually used.

Given that I have decided that clause 3 requires the actual provision of calculations and documents, the issues which were argued before me on the assumption that a mere certification was sufficient do not arise.

The issue which does, however, arise, is whether or not the first and/or second demands did in fact comply with clause 3 as I have construed it.

It is common ground that a calculation of the sums due did accompany the demands, but the parties disagree as to whether or not the demands were also accompanied by “any supporting documents reasonably required to assess such demands”.

Miss Selvaratnam submitted for Rubicon that supporting documents were limited, as a matter of construction of clause 3, to those going to issues of quantum. I do not accept that limitation. Although the requirement to provide documents is found immediately after the requirement to provide a calculation, the supporting documents are described as being those reasonably required to assess “such demand”, not “such calculation”.

Mr Eaton submitted that Rubicon was required to provide Krisenergy, on the facts of this case and in relation to these particular demands, with any documents reasonably required in order to ascertain (i) what work was done, so as to enable an assessment of whether that work was within the scope of works for which Rubicon was to be paid under the Charter and (ii) whether the costs of that work were reasonably incurred and/or reasonable in amount. I agree that such documents would be reasonably required for a full assessment of the merits of the underlying claims against Kegot, but I do not agree that such extensive documentation is reasonably required for the purpose of assessing the demand(s) made under the Guarantee against Krisenergy. As I have already indicated, in my view what is reasonably required to assess the demand(s) made under the Guarantee includes documents reasonably required for Krisenergy to be able quickly to find out from Kegot whether (and to what extent) the claim is admitted or disputed by Kegot, and I am prepared to assume without deciding that “documentation reasonably required” also extends to documents sufficient to allow Krisenergy to form a provisional view as to whether or not the claims which give rise to the demands are bona fide and not fraudulent claims. But the supporting documentation required does not in my view go beyond that.

This construction has the effect that if there is a compliant demand Krisenergy will be provided with all that it reasonably needs in order to decide what to do within the 48 hours allowed for its response.

So: did the first demand comply with those requirements? The first demand, dated 3 September 2018, was accompanied by the four Rubicon invoices, and each of the four Rubicon invoices was accompanied by a breakdown of the sums invoiced. Where the Rubicon invoices relied on sums invoiced to Rubicon by Unithai, the Unithai invoices were provided too. There were in all 270 odd pages of supporting documents, presented in a logical order. In my view those documents were amply sufficient to satisfy the requirements of clause 3, and they provided Krisenergy with all that it reasonably required to assess the first demand. Krisenergy can have been in no doubt about the basis on which the claim was advanced and how the sums were calculated, and had available to it (in the form of the Unithai and other third party invoices) all of the key supporting material.

Had Krisenergy prevailed on its construction of the extent of its on demand liabilities, it would have been necessary for me to decide whether, as contended by Rubicon, there had been an admission of liability sufficient to trigger a clause 5 liability. The point does not arise on the construction I have reached.

However, the evidence which was called on the first day of the two day hearing was directed entirely to this point. In the circumstances although it is not necessary to do so I shall make factual findings, but I shall express my reasons shortly.

The question is what precisely was said at a meeting on 19 April 2018. The meeting was attended by Mr Coombes, Mr Helyer and Ms Jirapojaporn of Krisenergy, and by Mr Sim, Mr MacLean, Mr Keesuwan and Mr Ali of Rubicon. It was agreed that Mr Ali would produce minutes after the meeting, and he did so.

The purpose of that meeting was to discuss various outstanding claims, and it began with Skid integration costs, which as I noted above had been advanced under VO-01, as though this were a claim under the variation order machinery of the Charter. It is common ground that Mr Coombes made the point that VO-01 had never been approved, and so could not be payable, and that Mr MacLean made the point that although a VO request form had been used, the Skid integration costs were in fact base scope work under the Charter, so that no variation was in fact required. It is also common ground that during this phase of the meeting Mr Helyer suggested that the cost of the work should have been in the region of $700-800,000. What is in dispute is what happened next and how matters were left.

The first draft of Mr Ali’s post meeting minutes (which was circulated internally at Rubicon) recorded this under a column entitled first “Action” and then (in the row beneath) “KE Position:”

A later internal version of the minutes showed these proposed amendments:

A version containing those amendments (but not showing the tracking) was sent to Mr Coombes on 14 May 2018. This was the first version to cross the line between Rubicon and Krisenergy. Mr Coombes responded later on 14 May 2018 saying that the minutes failed to cover many issues raised by Krisenergy, making a series of seven points by way of example of what had been omitted, and concluding that “the above are just a few of the points raised none of which appear in these minutes hence as a true recording there are huge gaps in what was discussed. They are therefore rejected in their entirety.” But Mr Coombes made no comment on the text I have set out above.

Undaunted by Mr Coombes’ wholesale rejection of the minutes, Mr Ali added Mr

I heard evidence from Mr MacLean (Rubicon’s Head of Engineering), Mr Helyer (Krisenergy’s Vice President of Operations) and Mr Coombes (Senior Production Manager for Kegot, but now retired). I am satisfied that each of them was doing their best to assist the court and to recollect what was said. None of them purported to be able to recall the precise words used at the meeting, which I take as an indication that they were all honest witnesses. The problem of course is that only the finest gradations of wording and meaning separate (a) a position in which Krisenergy accepts that it is likely that it will eventually have to pay a reasonable sum for such work as was done, recognises the difficulties with its defence that no VO was approved, and suggests that the parties discuss further with a view to agreeing a sum to be paid; and (b) a position in which Krisenergy accepts that it does have an obligation to pay a reasonable sum for such work as was done, gives up any defence based on the need for an approved VO, and agrees that the parties meet to try to agree what a reasonable sum is. Lawyers might find such distinctions easier to draw than would many commercial and technical people.

Doing the best I can, and giving significant but not determinative weight to the minutes, I find that the position that was reached in the meeting was that Krisenergy accepted that it was likely that it would eventually have to pay a reasonable sum for such work as was done, but did not formally give up its potential defence based on the absence of an approved VO, and that matters were left on the basis that there would be an attempt to agree a figure for the reasonable costs of the Skid integration works on the basis that, if a sum could be agreed, then Krisenergy or Kegot would pay it.

It is not necessary for me decide whether that is, or is sufficiently close to, an admission to trigger a clause 5 liability on some construction of clause 5 which is different to my construction, and I decline to do so, since the answer would depend on the precise construction of clause 5.

Finally and for completeness I should note that it was faintly suggested by the Claimant that various statements made in the Defendant’s witness statements themselves amounted to admissions of liability. In my view they did not go beyond the position I have described at paragraph 51 above.

I conclude that both of the demands were valid demands within the meaning of clause 3, and that Krisenergy was obliged to pay the sums demanded. I will hear Counsel as to the appropriate form of order since the argument before me did not deal with the question of whether judgment should be based on the first demand (which did not claim interest) or the second demand (which did).